ML14063A529

Attachment 1 to NL-14-030: Akrf Report - Update of Aquatic Impact Analyses Presented in Nrc'S FSEIS (December 2010) Regarding Potential Impacts of Operation of Indian Point Units 2 and 3

ML14063A529
Person / Time
Site:	Indian Point
Issue date:	02/19/2014
From:	AKRF
To:	Entergy Nuclear Operations, Office of Nuclear Reactor Regulation
Shared Package
ML14064A326	List: ML14063A529 NL-14-030, Final Supplemental Environmental Impact Statement NL-14-030, Indian Point Nuclear Generating Unit Nos. 2 & 3 - Final Supplemental Environmental Impact Statement
References
NL-14-030
Download: ML14063A529 (263)

Text

ATTACHMENT 1 TO NL-14-030 AKRF REPORT Update of Aquatic Impact Analyses Presented in NRC's FSEIS (December 2010) Regarding Potential Impacts of Operation of Indian Point Units 2 and 3 ENTERGY NUCLEAR OPERATIONS, INC.

INDIAN POINT NUCLEAR GENERATING UNIT NOS. 2 & 3 DOCKET NOS. 50-247 AND 50-286

Update of Aquatic Impact Analyses Presented in NRC's FSEIS (December 2010)

Regarding Potential Impacts of Operation of Indian Point Units 2 and 3 2/19/14 Prepared for:

INDIAN POINT ENERGY CENTER 450 Broadway, Suite 1 Buchanan, NY 10511 Prepared by AKRF, Inc.

7250 Parkway Drive, Suite 210 Hanover, MD 21076

Table of Contents-I. Introduction and O verview .................................................................................... 1 A. Background ................................................................................................... I B. Inter-annual Changes in LRS, FSS and BSS Sampling Designs .................. 2 C. Potential Confounding Effects of Sampling Design Changes ....................... 2 D. N ew ly A vailable D ata .................................................................................. 3 E. A nalysis Update ............................................................................................ 4 F. Conclusions ................................................................................................... 4 II. A nalysis Update M ethods .................................................................................. 4 A . Trends A nalysis M ethods ............................................................................. 4 B. SO C A nalysis M ethods ................................................................................ 5

1. A pparent Typographical Errors in FSEIS ................................................. 5 C . Independent Q uality Control Review .......................................................... 6 III. Results ............................................................................................................ 7 A . Updated Trends A nalyses ............................................................................. 7 B . Updated SO C A nalyses ................................................................................ 8 C . U pdated Im pact Conclusions ........................................................................ 8 IV . D iscussion ..................................................................................................... 9 A . N RC 's Precautionary M ethodology ............................................................. 9

1. SO C M ethods ............................................................................................ 9

2. Trends Analysis M ethods ........................................................................ 10

3. Conservative Results ................................................................................ 10 B . FSS G ear Change ......................................................................................... 10 C . Sum m ary of Changes in Im pact Conclusions .............................................. 12

1. H ogchoker, W eakfish and W hite Perch ................................................... 12

2. Striped Bass ............................................................................................. 12

3. B lueback H erring .................................................................................... 13

4. Rainbow Sm elt ......................................................................................... 14 V . Literature C ited ................................................................................................ 15 V I. Figures .......................................................................................................... 16 V II. Tables .......................................................................................................... 24 V III. A ppendix A ................................................................................................ 41 IX . A ppendix B .................................................................................................. 42

I. Introduction and Overview A. Background In December 2010, the Nuclear Regulatory Commission ("NRC") issued the "Generic Environmental Impact Statement for License Renewal of Nuclear Plants Supplement 38 Regarding Indian Point Nuclear Generating Unit Nos. 2 and 3 Final Report. NUREG-1437" ("FSEIS"). The FSEIS included a chapter entitled "Environmental Impacts of Operation" which described an assessment of potential impacts of entrainment and impingement at Indian Point Units 2 and 3 ("IP2 and IP3")

on fish populations in the Hudson River. That impact assessment was based on two sets of analyses: 1) an assessment of trends in young of year ("YOY") fish populations, and

2) and assessment of what NRC referred to as strength of connection ("SOC"). Both assessments were conducted using data provided to NRC by Entergy, which operates IP2 and IP3.

NRC's trends assessment had two components: 1) riverwide trends in fish abundance, and 2) trends in fish abundance in the sampling region adjacent to IP2 and IP3 (referred to as River Segment 4). For both components, NRC calculated indices of abundance using Entergy provided data from three Hudson River fish sampling programs: 1) Long River Survey ("LRL") which collected data on eggs, larvae and juvenile fish, 2) Fall Shoals Survey ("FSS") which collected data on juvenile and older fish, and 3) Beach Seine Survey ("BSS") which collected data on juvenile and older fish.

Separate indices of abundance were calculated for each species addressed by the assessment.

NRC's indices of abundance for the riverwide trends assessment were estimates of catch per unit effort ("CPUE"), i.e., the number of fish collected divided by the number of samples taken. In addition, NRC used a riverwide index of abundance from annual reports prepared by electric utility companies that operate power plants on the Hudson River and fund and manage the LRS, FSS and BSS sampling programs. For the River Segment 4 indices of abundance, NRC calculated estimates of CPUJE and estimates of density, i.e. the number of fish collected divided by the volume of water sampled.

NRC's SOC assessment used estimates of density from River Segment 4 from the BSS and FSS to characterize long-term linear trends in abundance and interannual variability in abundance. That information was coupled with NRC's estimates of entrainment and impingement mortality rates. NRC's estimates of entrainment and impingement mortality rates were based on annual estimates of total number of organisms entrained (1981-1987) and impinged (1984-1990) and estimates of the abundance of entrainable organisms within River Segment 4 from the LRS.

I

B. Inter-annual Changes in LRS, FSS and BSS Sampling Designs The data on fish population abundance in the Hudson River that Entergy provided to NRC in 2007 and that NRC used for the FSEIS were collected from the 27 year period 1979 through 2005. Over that period of years, the data were affected by inter-annual changes in sampling designs. Major changes in sampling designs included:

1) different sets of weeks of sampling in each year and sampling program, and 2) a change in the sampling gear used by the Fall Shoals Survey to sample the bottom stratum of the Hudson River. That gear change, from an epibenthic sled to a beam trawl, occurred in 1985.

The BSS sampling design saw a dramatic change in 1981 when the number of weeks of sampling was greatly curtailed (Figure 1). The number of weeks of sampling more than doubled by the late 1980's and then remained the same. Starting in 1998, sampling by the FSS beam trawl (which began in 1985) expanded to include the late fall (Figure 2). Like the BSS, weeks of sampling by the FSS epibenthic sled (which was terminated in 1984) were curtailed in 1981 (Figure 3). The weeks of sampling by the FSS tucker trawl have been fairly consistent although fewer weeks were sampled in the early 1980's (Figure 4). Starting in 1991, the LRS increased the weeks of sampling to include much of the fall (Figures 5 and 6).

C. Potential Confounding Effects of Sampling Design Changes Because the presence of early life stages of fish in the Hudson River is seasonal, changes in the weeks of sampling can introduce confounding effects to fish abundance data. For example, consider the hypothetical scenario of a species of fish whose larvae are only present in May of each year. If sampling only occurred during May, then an estimate of CPUE computed as the total number of those larvae collected divided by the total number of samples taken would be a valid index of abundance. Now consider the effect of doubling the sampling effort in the later years of the sampling program by extending the period of sampling to also include June (when the larvae are no longer present). Estimates of CPUE, computed as the total number of those larvae divided by the total number of samples taken, for the later years would not be comparable to the estimates of CPUE from the earlier years of the program. Even if the abundance of larvae did not change, it would appear as if the abundance had declined to half because the estimates of CPUE in the later years would be half the estimates from the earlier years.

Included in the data files provide to NRC by Entergy were data files that contained total counts of each species of fish (over all life stages) collected by each sampling program per year over all weeks of sampling. One data file of this type was provided for each of three sampling programs: LRS, FSS and BSS. An accompanying file for each sampling program was provided that listed the total number of samples collected by each program in each year. Those data files apparently were used by NRC to compute annual riverwide catch per unit of effort ("CPUE") indices of abundance.

2

For each sampling program, species and year, NRC apparently divided the total number of fish collected by the number of samples collected to compute an annual CPUE index of abundance. For the reasons discussed above, the historical changes in the weeks of sampling that occurred in the LRS, FSS and BSS appear to have introduced non-trivial confounding effects into NRC's CPUE indices of riverwide abundance.

For the River Segment 4 trends analyses, NRC apparently used a more detailed set of data files provided by Entergy that listed fish density by lifestage and week. For the River Segment 4 trends analyses, NRC subset the data to include a more consistent set of weeks in all years, and only included YOY fish (Table 1). Therefore, the River Segment 4 trends analyses likely were not as confounded by the changes in weeks sampled in each year or by changes in lifestage composition among years.

As previously provided in comments to NRC, the change in FSS sampling gear in 1985 appears to have also introduced non-trivial confounding effects to the FSEIS trends and SOC assessments. The gear change was substantial from the epibenthic sled with a I m2 mouth opening and 3 mm mesh collection net to the beam trawl with a 2.7 m2 mouth opening and 1.3 cm mesh collection net. The use of catch data from survey nets to address trends depends on the assumption of constant collection efficiency over all years of the survey. Collection efficiency can be thought of as the ratio of the average number of fish collected in a single sample to the underlying abundance of those fish in the portion of the river subject to sampling. For some species, like bay anchovy, the increase in mesh size of the beam trawl allowed YOY fish, which would have been retained by the smaller mesh of the epibenthic sled, to pass through the beam trawl net.

Accordingly, for bay anchovy the collection efficiency of the beam trawl was lower than the collection efficiency of the epibenthic sled. Similarly, for species like striped bass, the smaller epibenthic sled appeared to be more easily avoided than the larger beam trawl. Under those circumstances the beam trawl would have a higher collection efficiency than the epibenthic sled.

D. Newly Available Data As noted above, the data that Entergy provided and NRC used for the FSEIS were collected from the 27 year period of years 1979 through 2005. Since the time Entergy provided those data, the Hudson River Biological Monitoring Program has been continued, with fish data collected through the LRS, BSS and FSS. Data from those programs have been published in the annual series of reports titled, "Year Class Report for the Hudson River Estuary Monitoring Program" ("YCR"). Accordingly, data from the LRS, BSS and FSS now are available for the 27 year period 1985 through 2011.

During this 27 year period of years, there were no gear changes in any of the sampling programs, thus eliminating this confounding effect and bringing the data current. Other regulators also have performed analyses, which allow the dataset, if brought current, to be more readily compared to these regulatory findings.

3

E. Analysis Update This report describes an update to the trends and SOC analyses presented in the FSEIS. This update of the FSEIS analyses used the LRS, BSS and FSS data from 1985 through 2011. For this analysis update, the data were subset in every year to include only a consistent set of weeks for each sampling program (Table 2). Furthermore, the data used for the trends analyses were subset to include only YOY fish. These steps removed the confounding effects on riverwide CPUE indices of abundance due to changes in the weeks sampled in each year and due to the inclusion of all life stages collected. In addition, this analysis update avoids the confounding effects of the FSS gear change that occurred in 1985.

F. Conclusions In comparison to the conclusions reported in the FSEIS, results from the updated analyses changed the impact conclusions for seven (7) of the 18 aquatic species evaluated in the FSEIS:

- Alewife changed from Moderate to Small

- Blueback Herring changed from Large to Small

- Hogchoker changed from Large to Moderate

- Rainbow Smelt changed from Moderate-Large to Moderate

- Striped Bass changed from Small to Moderate

- Weakfish changed from Moderate to Small

- White Perch changed from Large to Small These changes in impact conclusions were due to a combination of changes in the results from the trends analyses and from the SOC analyses. The results from both sets of updated analyses were free from confounding effects due to inter-annual changes in the weeks of sampling by the LRS, BSS and FSS. The results from the updated analyses are also free from confounding effects of the FSS gear change that occurred in 1985. Also, results from the updated riverwide trends analyses were not affected by interannual changes in lifestage composition. The impact conclusion change for rainbow smelt is due to newly available information on the range contraction of rainbow smelt on the Atlantic coast.

II. Analysis Update Methods A. Trends Analysis Methods The updated trends analyses were conducted according to the methods described in section 1.2.1 of Appendix I of the FSEIS (pages 1-2 through 1-50). Because the update is based on data from the 27-year period 1985-2011, analysis steps that NRC used to 4

address the FSS gear change that occurred in 1985 did not have to be conducted. Steps described on the following pages of Appendix I were not performed:

1. pages 1-9 through 1-14: River Segment 4 trends in FSS density

2. pages 1-23 through 1-26: River Segment 4 trends in FSS CPUE

3. pages 1-34 through 1-37: Riverwide trends in FSS CPUE.

Not having to address the issue of the FSS gear change in 1985 greatly simplified the trends analyses and materially reduced the uncertainty in the results of the trends analyses.

B. SOC Analysis Methods The updated SOC analyses were conducted according to the methods described in section 1.2.2 of Appendix I of the FSEIS (pages 1-50 through 1-63). In the FSEIS, the coefficient of variation required for the SOC analyses was calculated from the first 12 years of data used in the FSEIS analyses, i.e., 1979-1990 (FSEIS Table 1-46). For the updated analyses, the coefficient of variation was calculated from the first 12 years of data used in the updated analyses, i.e., 1985-1996. The species-specific entrainment mortality rates ("EMR") and impingement mortality rates ("IMR") used in the FSEIS SOC analyses were also used for the updated SOC analyses. For spottail shiner, the EMR estimate used for the update was taken from NRC's June 2012 "NUREG-1437, Supplement 38, Volume 4, draft supplement to final - Draft Report for Comment".

As described below, some minor changes to the methods as documented were made for the analysis update to account for apparent typographical errors in the FSEIS.

1. Apparent TypographicalErrors in FSEIS Equation (2) on page 1-51 of the FSEIS indicates that the entrainment mortality rate (EMR) only affects the initial number of fish (No), and that the impingement mortality rate (IMR) only affects the slope parameter (r):

No = No( + EMR) and r* =rucL(I-IMR)/max(1,CV)) (2)

"where EMR and IMR are conditional mortality rates for entrainment and impingement; rucL is the upper 95 percent confidence limit of the linear slope; and CV is the coefficient of variation of the annual 75th percentiles from the weekly catch density."

Because the FSEIS SOC analysis was intended to address entrainment and impingement, it appears that the omission of IMR from the definition of No, and the omission of EMR from the definition of r* were typographical errors. Therefore, for the analysis update, 5

equation (2) was revised so that both the entrainment mortality rate and the impingement mortality rate affected the initial number of fish and the slope parameter:

No = No(I + EMR + IMR) and r* =rUCL(I-EVIR-IMiR)/max(1,CV)) (3)

Table 1-46 in the FSEIS lists values for the "Upper 95% Confidence Limit of the Slope" that were used with equation (2). As shown below, the values in the column labeled "Upper 95% Confidence Limit of the Slope" apparently were mislabeled.

Rather than the upper 95% confidence limits they are the slope estimates plus one standard error.

The "Linear Slope (r)" and "Upper 95% Confidence Limit of the Slope" entries in Table 1-46 were taken from Table 1-9 (for FSS) and Table 1-12 (for BSS) of the FSEIS. For each species, the entry in the column labeled "Upper 95% Confidence Limit of the Slope" in Table 1-46 is the corresponding linear regression slope estimate from Table 1-9 or 1-12 plus the undefined value to the right of the +/- symbol in the linear regression slope column of the table. It can be shown from the "p-value", also listed in Tables 1-9 and 1-12, that the undefined value to the right of the +/- symbol is the standard error of the linear slope estimate (Appendix A).

Therefore, the entries listed in Table 1-46 are, if fact, the slopes plus one standard error. If those entries had been upper 95% confidence limits, they would have been approximately equal to the slopes plus two standard errors.

Based on the values listed in Table 1-46, it appears that the SOC analyses presented in the FSEIS were conducted with rUcL in equation (2) set equal to the estimated slope plus the standard error of the slope. Accordingly, to be consistent with the SOC analyses presented in the FSEIS, the value of rucL in equation (2) was set to the estimated slope plus the standard error of the slope for this analysis update.

C. Independent Quality Control Review The updated analyses were conducted using data analysis programs written with SAS computer software, and all data inputs were in SAS format data files. The full set of computer programs and input data files used for the updated analyses were submitted to John Young, PhD of ASA Analysis & Communication, Inc. for a thorough quality control review. The purpose of the review was to determine whether the computer code was correctly written to accurately conduct the analyses documented in the FSEIS. Dr. Young has decades of experience working with data files from the Hudson River Biological Monitoring Program. Dr. Young's independent review of the computer programs and input data files used for the updated analyses confirmed the computer programs accurately reflected the analysis methods documented in the FSEIS and identified no computer programming errors (see Appendix B).

6

III. Results A. Updated Trends Analyses Following the trends analysis methods documented in the FSEIS, a total of nine sets of trends analyses were conducted: three for River Segment 4 density (FSS, BSS and LRS), two for River Segment 4 CPUE (FSS and LRS), three for riverwide CPUE (FSS, BSS and LRS), and one for the YCR abundance indices. Each set of trends analyses included analyses conducted using linear regression and analyses using segmented regression. For each species, one type of regression was selected using the decision rules documented in the FSEIS. Based on the results of the selected type of regression analysis, each species was assigned a trend score of either 1 (i.e., no decline detected) or 4 (i.e., decline detected).

For River Segment 4 trends, comparisons of results from the two types of regressions are summarized in the following tables:

- FSS Density (Table 3)

- BSS Density (Table 5)

- LRS Density (Table 7)

- FSS CPUE (Table 9)

- LRS CPUE (Table 11)

The corresponding trend conclusions (i.e., score of 1 or 4) for these five sets of analyses are summarized in the following tables:

- FSS Density (Table 4)

- BSS Density (Table 6)

- LRS Density (Table 8)

- FSS CPUE (Table 10)

- LRS CPUE (Table 12)

The River Segment 4 trends conclusions, based on the average score from the five sets of analyses are listed in Table 13.

For riverwide trends, comparisons of results from the two types of regressions are summarized in the following tables:

- FSS CPUE (Table 14)

- BSS CPUE (Table 16)

- LRS CPUE (Table 18)

- YCR Abundance Index (Table 20) 7

The corresponding trend conclusions (i.e., score of 1 or 4) for these four sets of analyses are summarized in the following tables:

- FSS CPUE (Table 15)

- BSS CPUE (Table 17)

- LRS CPUE (Table 19)

- YCR Abundance Index (Table 21)

The riverwide trends conclusions, based on the average score from the four sets of analyses are listed in Table 22.

The overall trends conclusions, which were based on weighted averages of the River Segment 4 scores and riverwide scores, are summarized in Table 23. In comparison to the conclusions reported in the FSEIS, results from the updated analyses changed the trends conclusions for 8 of the 13 species analyzed in the FSEIS:

- Alewife changed from Variable to Undetected Decline

- Bluefish changed from Detected Decline to Undetected Decline

- Hogchoker changed from Detected Decline to Variable

- Spottail Shiner changed from Detected Decline to Undetected Decline

- Striped Bass changed from Undetected Decline to Variable

- Weakfish changed from Variable to Undetected Decline

- White Catfish changed from Variable to Undetected Decline

- White Perch changed from Detected Decline to Undetected Decline B. Updated SOC Analyses Parameter values used in the updated SOC analyses are listed in Table 24. All parameter values except EMR and IMR (which remain the values that were used in the FSEIS) were computed using the same data files used for the updated trends analyses.

Results from the SOC Monte Carlo analyses, and corresponding SOC conclusions, are summarized in Table 25. In comparison to the conclusions reported in the FSEIS, results from the updated analyses changed the SOC conclusions for 3 of the 13 species analyzed in the FSEIS:

- Alewife changed from High to Low

- Blueback Herring changed from High to Low

- White Perch changed from High to Low C. Updated Impact Conclusions The overall impact conclusions based on the updated analyses (Table 26) were determined by combining the trends conclusions and SOC conclusions as described in Appendix H of the FSEIS. In comparison to the conclusions reported in the FSEIS, 8

results from the updated analyses changed the subsidiary impact conclusions for 7 of the 18 species analyzed in the FSEIS:

- Alewife changed from Moderate to Small

- Blueback Herring changed from Large to Small

- Hogchoker changed from Large to Moderate

- Rainbow Smelt changed from Moderate-Large to Moderate

- Striped Bass changed from Small to Moderate

- Weakfish changed from Moderate to Small

- White Perch changed from Large to Small The change for rainbow smelt was due to newly available information on the range contraction of Atlantic coast rainbow smelt (see Discussion section, below).

IV. Discussion A. NRC's Precautionary Methodology The methods applied by NRC to assess the magnitude of potential aquatic impacts due to the operation of IP2 and IP3 are highly conservative in that they include several components that tend to lead to conclusions of "Large" impacts. As discussed below, both the trends in YOY abundance analyses and the SOC analyses contain such components.

1. SOC Methods NRC's SOC analyses are based on the comparison of the magnitude of entrainment (and impingement) mortality rates to the magnitude of interannual variability in YOY abundance. For the purpose of the SOC analyses, NRC defined the magnitude of entrainment mortality as the difference between: 1) projected population abundance with entrainment and 2) projected population abundance without entrainment.

Key among the conservative components of the SOC analyses are:

I. Estimates of entrainment mortality rates were based on total annual entrainment in comparison to the number of entrainable organisms found in sampling River Segment 4 only, rather than to the entire Hudson River population. Because most entrainable organisms found in sampling River Segment 4 are transient, moving with tidal currents into and out of sampling River Segment 4, the number of fish in River Segment 4 severely underestimated the total number of fish from which those entrained were drawn. Therefore, entrainment mortality rates were overstated.

9

2. Projected abundance in the absence of entrainment was based on the upper confidence limit of the estimated historical trend in abundance; whereas, projected abundance with entrainment was based on the estimated trend itself. Therefore, even in the absence of entrainment, the method would show a purported reduction in abundance due to entrainment.

2. Trends Analysis Methods The methods applied by NRC to assess trends in abundance also contained conservative components. For each species, the trends assessment included a linear regression analysis and segmented regression analysis. If the residual error from the segmented regression was lower than the residual error from the linear regression analysis, NRC selected the segmented regression analysis for the trends assessment.

Because the segmented regression included four parameters (compared to two parameters for the linear regression) it was able to fit the data more closely and therefore was often selected over the linear regression.

The segmented regression analysis resulted in two connected line segments being fit to the data, each with an estimated duration, and each with an estimated slope. If either slope was negative (and statistically significant) NRC's method was to conclude a detected decline in abundance, regardless of the duration. For example, a short-term decline followed by a long term increase would be recorded as a detected decline.

3. Conservative Results For the reasons discussed above, the results from analyses conducted using NRC's methods can be viewed as being highly conservative. Therefore, even allowing for inherent uncertainties, the results from the updated analyses support the conclusion that the continued operation of IP2 and IP3 will not pose any meaningful risks of adverse impacts to fish populations in the Hudson River.

B. FSS Gear Change As noted above, in addition to using datasets with a consistent set of weeks of sampling in all years, this analysis update also used data from the 27-year period 1985-2011, whereas the FSEIS used data from the 27-year period 1979-2005. Data from the period 1985-2011 do not suffer from the potential confounding effects of the FSS gear change in 1985, which may be substantial. The FSS gear for sampling the bottom stratum changed from the epibenthic sled with a 1 m2 mouth opening and 3 mm mesh collection net to the beam trawl with a 2.7 m 2 mouth opening and 1.3 cm mesh collection net (i.e., 13 mm mesh). Those changes in gear specifications materially altered the collection efficiency of samples from the FSS, which necessarily affected estimates of CPUE and density.

10

In 1984, a gear comparison study was conducted that deployed over 250 paired epibenthic sled and beam trawl samples in the Tappan Zee, Croton-Haverstraw, and Indian Point regions of the Hudson River, during four alternate weeks of sampling in August and September (Normandeau Associates, Inc. 1986). Density estimates for striped bass young of the year ("YOY") were 4 times higher for the beam trawl than for the epibenthic sled, and the density estimates for YOY striped bass were higher for the beam trawl in all four weeks sampled. Density estimates for bay anchovy YOY were 46 times higher for the epibenthic sled, and the density estimates for YOY bay anchovy were higher for the epibenthic sled in all four weeks sampled. Comparisons for other species were not presented in the report.

The 1984 gear comparison study clearly demonstrated that the beam trawl and epibenthic sled had materially different collection efficiencies, and that the differences were species-specific. For that reason, data collected by the two gear types are not directly comparable and cannot be used together in a valid trends assessment without accounting for the species-specific differences in collection efficiencies.

NRC addressed the FSS gear change by conducting a series of statistical analysis that compared FSS densities to BSS densities before and after 1985, and FSS CPUE to BSS CPUE before and after 1985. Based on those analyses of densities in River Segment 4, NRC concluded that for the 12 species considered, the gear change only caused a biological difference to bay anchovy. For CPUE in River Segment 4, NRC concluded that for the 11 species considered, the gear change only caused a biological difference to bay anchovy and blueback herring. For riverwide CPUE, NRC concluded that for the 10 species considered, the gear change caused a biological difference to alewife, American shad, bay anchovy, blueback herring, and bluefish.

For these 8 out of 33 combinations of species and abundance indices, NRC conducted separate trends analyses for the period of year 1979-1984 and the period of years 1985-2005. For all other combinations of species (including striped bass) and abundance indices, NRC made no adjustments for the gear change. Furthermore, for the trends component of the SOC analyses (based on density estimates in River Segment 4),

no adjustments for the gear change were made for any species.

Although NRC made efforts to address the FSS gear change, the results presented in the FSEIS still contain uncertainties due to the FSS gear change that occurred in 1985. Because the updated analyses were based on a 27 years of data that were not affected by any gear changes, the results from the updated analyses do not contain that layer of uncertainty.

II

C. Summary of Changes in Impact Conclusions As noted above, impact conclusions from the updated analyses differed from the impact conclusions from the FSEIS for seven species: alewife, blueback herring, hogchoker, rainbow smelt, striped bass, weakfish and white perch. Each change is discussed below.

1. Hogchoker, Weakfish and White Perch For three of these seven species the change was to a lower potential impact level due to revised trends conclusions:

Species FSEIS Updated Analyses Trends Conclusion Trends Conclusion Hogchoker Detected Decline Variable Weakfish Variable Undetected Decline White Perch Detected Decline Undetected Decline These changes in the trends conclusions were largely due to changes in Riverwide Assessment Scores:

Species Riverwide Assessment River Segment 4 Assessment Score Score FSEIS Updated FSEIS Updated (Table H- 15) Analyses (Table H- 15) Analyses (Table 23) (Table 23)

Hogchoker 3.0 1.0 4.0 4.0 Weakfish 2.5 1.0 2.5 2.5 White Perch 4.0 1.0 3.0 2.0 This pattern of changes is consistent with the expected confounding effects of the inadvertent inclusion of all weeks of sampling in the FSEIS Riverwide Assessment.

2. Striped Bass For Striped Bass, the change in impact conclusion was to a higher level of potential impact (i.e., "Small" to "Variable"). This change was due to changes in both the Riverwide and River Segment 4 Assessments:

12

Species Riverwide Assessment River Segment 4 Assessment Score Score FSEIS Updated FSEIS Updated (Table H-15) Analyses (Table H-15) Analyses (Table 23) (Table 23)

Striped Bass 1.0 2.0 1.0 3.0 Because the updated analyses were based on a more recent period of years than the FSEIS analyses, these changes in trend scores reflect the recent decline in striped bass stock abundance that followed a period of abundance increases. Beginning in the mid-i 980's the Atlantic coast striped bass stock experienced a surge in abundance in response to reduced fishing pressure due to a coastwide fishing moratorium on striped bass. After the moratorium was lifted in 1990, the stock continued to increase in abundance through the late 1990's after which it began to decline (Atlantic States Marine Fisheries Commission, 2013).

3. Alewife and Blueback Herring For alewife and blueback herring, the change in impact conclusion from "Large" to "Small" is due to the change in the SOC conclusion from "High" (from the FSEIS) to "Low". The "Low" SOC conclusion from the updated analyses for alewife and blueback herring is consistent with the historical distribution patterns of entrainable life stages of river herring (i.e., collectively alewife and blueback herring). The vast majority of entrainable lifestages of river herring inhabit portions of the Hudson River that are far upstream of IP2 and IP3 (Figure 7). The documented distribution patterns of entrainable lifestages of river herring in the Hudson River, in comparison to the location of IP2 and IP3, are consistent with the "Low" SOC conclusion from the updated analyses.

The recent conclusion of the New York State Department of Environmental Conservation ("NYSDEC") that recruitment of river herring is variable but stable, despite the upsurge in the use of river herring as bait for striped bass (NYSDEC, 2011) is consistent with a finding of "Small" potential impacts as well. NYSDEC proposed to maintain the Hudson River and tributaries as a restricted river herring fishery because, under current conditions (including the operation of IP2 and IP3) the fishery was "sustainable" and would "not diminish potential future reproduction and recruitment of herring stocks." NYSDEC also noted that since the mid-1990's there has been an increasing trend in YOY alewife abundance.

In addition, in the National Marine Fishery Service ("NMFS") decision not to list blueback herring as threatened or endangered (Fed. Reg. Vol. 78, No. 155. August 12, 2013), NMFS concluded that water withdrawals and outfalls (including pumped storage, 13

irrigation, thermal discharges, industrial pollutants and atmospheric deposition) collectively posed only a "medium low" threat to blueback herring. The number one threat was listed as "dams and other barriers". Behind that, "climate change," "water quality (chemical)", "incidental catch", and "predation", ranked as medium threats. The NMFS's findings are consistent with the change in impact conclusion for blueback herring of "Large" to "Small" for IP2 and IP3.

4. Rainbow Smelt For Rainbow Smelt, NRC modified the conclusion of a "Moderate" impact, that was determined using the impact assessment methodology of the FSEIS, to a conclusion of "Moderate to Large":

"Although detectable population declines occurred in two of four river data sets, indicating population trend results were variable, the staff concluded that a MODERATE to LARGE, rather than just MODERATE, impact was present based on the dramatic population declines observed for this species over the past three decades." (FSEIS Section 4.1.3.3, page 4-24)

This position regarding rainbow smelt is not supported by recent evidence regarding large-scale changes in the distribution of rainbow smelt. The decline in abundance of rainbow smelt in the Hudson River has been due to a coastwide contraction of the range of rainbow smelt on the Atlantic coast. Several decades ago, rainbow smelt populations were found as far south as Chesapeake Bay. Now their range only includes waters north of Long Island Sound (Enterline and Chase, 2012; National Oceanic and Atmospheric Administration, 2010).

The decline in rainbow smelt abundance in the Hudson River occurred simultaneously with the decline in abundance in coastal streams in Connecticut, which supports the conclusion that the decline was not due to the operation of IP2 and IP3.

Because rainbow smelt is a cold water species, the cause of its range contraction may be related to global warming.

"The Hudson River population of rainbow smelt is at the southern extreme of the reproductive range (Lee et al. 1980), although historically it occurred farther south (Smith 1985). The abrupt decline in rainbow smelt early life stages in the ichthyoplankton may result from global warming. Ashizawa and Cole (1994) documented the trend of slowly increasing water temperature in the Hudson River. The rainbow smelt runs in the coastal streams of western Connecticut have drastically declined or disappeared simultaneously with the decline in the Hudson River population (S. Gephard, Connecticut Department of Environmental Protection, personal communication)." (Daniels, et al, 2005) 14

For these reasons, the impact conclusion for rainbow smelt from the updated analyses was kept at "Moderate", based on the results from applying the trends analysis and SOC methodologies of the FSEIS to the updated input data files.

V. Literature Cited Atlantic States Marine Fisheries Commission. 2013. Striped bass stock assessment for 2013; 57th SAW Assessment Report.

Daniels, R.A., K.E. Limburg, R.E. Schmidt, D.L. Strayer and R.C. Chambers. 2005.

Changes in fish assemblages in the tidal Hudson River, New York. American Fisheries Society Symposium 45:471-503.

Draper, N.R. and H. Smith. 1966. Applied Regression Analysis. John Wiley & Sons, Inc. New York. 407 pp.

Enterline C. and B. Chase. 2012. Recent range changes by rainbow smelt (Osmerus mordax) and current annual migrations. Maine Department of Marine Resources, Orno ME.

National Oceanic and Atmospheric Administration. 2010. Rainbow Smelt An Imperiled Fish in a Changing World. Support provided by the Maine Department of Marine Resources, Massachusetts Division of Marine Fisheries, New Hampshire Department of Fish and Game, and National Oceanic and Atmospheric Administration.

Normandeau Associates. Inc. 1986. Size selectivity and relative catch efficiency of a 3m beam trawl and a I m 2 epibenthic sled for sampling young of the year striped bass and other fishes in the Hudson River estuary.

NYSDEC. 2011. Sustainable Fishing Plan for New York River Herring Stocks.

15

VI. Figures 16

Figure 1. Number of Samples Collected per Week (proportionate to drde size) program = BSS samplinggear = Beach Seine location = Riverwide WEEK 49-45 000 O 0 0 0 0 0 0 000 00 0 0 0oo 0 0 0 0 0OO0000c 0 0 0 0 0 ,--, 00 0000000000000C 41- oo-"0 0 000ý-"ý -00 /-0000,-O000000000C 37- O00 0 0000 .00 0000000 000000 00OO000000C 33 O000000 O 000 0000 000. 0000 O:O 0000000000C 0000 000OO.OO 0.C 00000 00 " 000 000 0 O0OO00000.C 29- O0000 0 000 0 000 00OO

• 000O 00 "

25 00 0 000 0000 O0000000O0C 21-17 13 9-5-

1-1 111 111 1 11 11 11 1 11 111 1 11 1 11222222222222 99999999999999999999999999000000.000000 7777778888888888999999999900000000001 1 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 .5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 YEAR OF DATA COLLECTION 17

Figure 2. Number of Samples Collected per Week (proportionate to circle size) program=FSS sampling_gar= Beam Trawl location= Riverwide WEEK 49-1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 99999999999999999999999999000000000000 7777778888888888999999999900000000001 1 45678901 2345678901 2345678901 2345678901 YEAR OF DATA COLLECTION 18

Figure 3. Number of Samples Collected per Week (proportionate to dirde size) program= FSS samplinggear= Epibenthic Sled location =Riverwide WEEK 49 111111111111111111111111112222222222222 99999999999999999999999999000000000000 7777778888888888999999999900000000001 1 456789012345678901 23456789012345678901 YEAR OF DATA COLLECTION 19

Figure 4. Number of Samples Collected per Week (proportinae to circle size) program = FSS samplinggear = Tucker Trawl location = Riverwide WEEK 1 1 1 11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 222222222222 99999999999999999999999999000000000000 77777788888888 889999 9999 9900000 000001 1 45678901 2345678901 2345678901 2345678901 YEAR OF DATA COLLECTION 20

Figure 5. Number of Samples Collected per Week

(:oportiadate to drdve sze) program = LRS samplinggear= Epibenthic Sled locaton = Riverwide WEEK 49-45 41-37-33-29-25 21-17-13 9

5-1-

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 9 999 999 9999999999999 9999990 000000000 0 0 7 77 7778888888 ,88899999999 9 90000000 ,0001 1 4 5678901 2345678901 2345678901 23456 78901 YEAR OF DATA COLLECTION 21

Figure 6. Number of Samples Collected per Week (poportinae to drde size) program = LRS samplinggear = Tucker Trawl locatlon = Riverwide WEEK 49- 0 0 41 00 0 0 0 0 0 a0 000 0O00O00O0QO00O 37 00 0 0000000000000

. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 C 0 0 0 0 0 33 000000 oo° 00 0 00 0 000 0 0 0 0 0 0 0 0 0 C

- 0o o 000 00 000 00 25-21 w 111 13 0 0 0 13 o00 0060 0 00 00 00 0 0 0 0 00000 0000 9- Q0000 0 0 0 0 0 0 0 0 0 0 5

I l I II I I I I 'l ,

I-- 1 1 1 11 , 1 1I IT 1 1 1 1 1 1' 'I1 1 1 1 1 1 1 1 1 1 1 11111111 1 1 1 1 1 11111 1 2 2 2 2 2 2 222222 9 9 9 9 9 9 9 99.999999 9 9 9 9 9 99999 9 0 0 0 0 0 0 000000 7 7 7 7 7 7 8 88888888 8 9 9 9 9 99999 9 0 0 0 0 0 0 000011 4 5 6 7 8 9 0 12345678 9 0 1 2 3 45678 9 0 1 2 3 4 5 678901 YEAR OF DATA COLLECTION 22

Figure 7. Spatial distribution of early lifestages of river herring (Blueback herring and Alewife) in theHudson River based on LRS sampling (copy of Figure 4-46, 2011 Year Class Report for the Hudson River Estuary Monitoring Program).

Eggs 100%

80% -iL201i 60%

40%

20%

0% = A A = . -F] i'%'.1 YK TZ CH IP WP CW PK HP KG SG CS AL Yolk Sac Larvae 100%

T - r 2 01 1 80% ,.1974_2010 60%

40%

20%

0%

YK TZ CH tP WP CW PK HP KG SG CS AL Post Yolk Sac Larvae 100%

80% T-12011

--- 1974-2010 60%

40%

20%

0%

YK TZ CH IP WP CW PK HP KG SG CS AL Young-of-Year 100%

80% 2011 60% No young-of-year Alosa spp. were collected during the -,-172010 temporal limits (weeks 18 - 26) of this index in 2011.

40%

20%

0%

YK TZ CH 1P WP CW PK HP KG SG CS AL Region 23

VII. Tables 24

Table 1. Weeks, sampling gears and lifestages included in NRC FSEIS Trends Analyses (27 years: 1979-2005). Annual abundance indices confounded by inter-annual changes in sampling designs.

River Segment 4 Density and CPUE Riverwide CPUE BSS FSS LRS BSS FSS LRS Lifestage YOY YOY YOY All All All Weeks 22-43 27-43 20-40

• All All All Gears Beach Seine Tucker Trawl (1979-2005) Tucker Trawl Beach Seine Tucker Trawl (1979-2005) Tucker Trawl Epibenthic Sled (1979-1984) Epibenthic Sled Epibenthic Sled (1979-1984) Epibenthic Sled Beam Trawl (1985-2005) , Beam Trawl (1985-2005)

• Inferred from FSEIS Atlantic Tomcod indices of abundance.

Table 2. Weeks, sampling gears and lifestages included in Trends Analyses Update (27 years: 1985-2011). Consistent set of sampling conditions among years.

River Segment 4 Density and CPUE RiverwideCPUE BSS FSS LRS BSS FSS LRS Lifestage YOY YOY YOY YOY YOY YOY Weeks 28-42 29-42 17-27 28-42 29-42 17-27 Gears Beach Seine Tucker Trawl Tucker Trawl Beach Seine Tucker Trawl Tucker Trawl Beam Trawl Epibenthic Sled Beam Trawl Epibenthic Sled 25

Table 3. Competing Models Used To Characterize the Standardized River Segment 4 FSS Population Trends of YOY Fish Density Using a 3-Year Moving Average (updated FSEIS Table 1-9).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope 1 Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Alewife 0.877 -0.054 0.026 0.048 0.867 -0.827 1.912 1989 -0.135 -0.005 American Shad 0.232 -0.120 0.013 0.000 0.186 -2.198 0.339 1988 -0.139 -0.083 Atlantic Tomcod 0.678 -0.080 0.023 0.002 0.370 -1.075 -0.275 1991 -0.069 0.029 ay Anchovy 0.601 -0.088 0.022 0.000 0.527 -1.453 2.818 1989 -0.158 -0.063 Blueback Herring 0.878 -0.054 0.026 0.048 0.081 -5.254 -3.578 1988 -0.027 0.010 Bluefish 0.925 -0.046 0.027 0.100 0.591 0.079 1.508 1990 -0.160 -0.044 Hogchoker 0.434 -0.104 0.018 0.000 Failed to Converge Rainbow Smelt 0.623 -0.086 0.022 0.001 0.535 -0.193 0.534 1992 -0.188 -0.060 Striped Bass 0.776 -0.069 0.024 0.010 0.684 -0.720 4.145 1988 -0.144 -0.036 Weakfish 0.811 -0.064 0.025 0.017 0.459 -0.030 0.157 2001 -0.408 -0.139 White Catfish 0.945 -0.042 0.027 0.136 0.967 -0.257 0.374 1995 -0.181 0.022 White Perch 0.838 -0.060 0.025 0.026 0.656 -0.542 -0.023 1995 -0.062 0.105 Table 4. River Segment 4 Assessment of the Level of Potential Negative Impact Based on the Standardized FSS Density Using a 3-Year Moving Average (updated FSEIS Table 1-10).

Species Best Fit Slope Slope 1 Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Alewife SR S1=0 S2<0 4 American Shad SR S1=0 S2<0 4 Atlantic Tomcod SR S1<0 S2=0 4 Bay Anchovy SR S1=0 S2<0 4 Blueback Herring SR SI<0 S2=0 4 Bluefish SR S1>0 S2<0 4 Hogchoker LR S<0 4 Rainbow Smelt SR S1=0 S2<0 4 Striped Bass SR S1=0 S2<0 4 Weakfish SR S1=0 S2<0 4 White Catfish LR S=0 I White Perch SR S1<0 S2=0 4 26

Table 5. Competing Models Used To Characterize the Standardized River Segment 4 BSS Population Trends of YOY Fish Density Using a 3-Year Moving Average (updated FSEIS Table 1-12).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope 1 Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Alewife 0.725 0.075 0.024 0.004 0.698 -0.088 0.120 2002 -0.007 0.376 American Shad 0.235 -0.120 0.013 0.000 0.252 -0.149 -0.06 1 2005 -0.426 0.074 Bay Anchovy 0.844 0.059 0.025 0.029 Failed to Converge Blueback Herring 0.726 -0.075 0.024 0.004 0,665 -0.154 0.369 1994 -0.211 -0.043 Bluefish 1.034 0.013 0.028 0.646 0.915 -0.355 0.083 1997 -0.014 0.224 Hogchoker 0.776 0.069 0.024 0.010 0.331 -0.251 -0.023 1998 0.152 0.310 Spottail Shiner 0.989 -0.031 0.028 0.271 0.932 -0.556 2.283 1989 -0.123 0.012 Striped Bass 1.016 0.022 0.028 0.436 0.454 0.107 0.342 1999 -0.284 -0.076 White Perch 1.035 -0.012 0.028 0.663 0.873 -1.114 0.115 1991 -0.042 0.109 Table 6. River Segment 4 Assessment of the Level of Potential Negative Impact Based on the Standardized BSS Density Using a 3-Year Moving Average (updated FSEIS Table 1-13).

Species Best Fit Slope Slope 1 Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Alewife SR SI=0 S2=0 1 American Shad LR S<0 4 Bay Anchovy LR S>0 I Blueback Herring SR S1=0 S2<0 4 Bluefish SR S1=0 S2=0 1 Hogchoker SR S1<0 S2>0 4 Spottail Shiner SR SI=0 S2=0 1 Striped Bass SR SI>0 S2<0 4 White Perch SR S1 =0 S2=0 1 27

Table 7. Competing Models Used To Characterize the Standardized River Segment 4 LRS Population Trends of YOY Atlantic Tomcod Density Using a 3-Year Moving Average (updated FSEIS Table I-15).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope I Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Atlantic Tomcod 1.030 -0.015 0.028 0.590 0.471 -2.721 -0.702 1989 -0.007 0.089 Table 8. River Segment 4 Assessment of the Level of Potential Negative Impact Based on the Standardized LRS Atlantic Tomcod YOY Density Using a 3-Year Moving Average (updated FSEIS Table 1-16).

Species Best Fit Slope Slope I Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Atlantic Tomcod SR S1<0 S2=0 4 28

Table 9. Competing Models Used To Characterize the Standardized River Segment 4, FSS Population Trends of YOY Fish CPUE (updated FSEIS Table 1-19).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope 1 Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Alewife 0.955 -0.036 0.024 0.149 -0.097 0.009 2010 American Shad 0.753 -0.066 Q.021 0.005 0.677 -0.106 0.179 1996 -0.234 -0.030 Atlantic Tomcod 0.791 -0.062 0.022 0.010 Failed to Converge Bay Anchovy 1.039 0.003 0.025 0.905 0.880 -0.030 0.227 1999 -0.265 0.023 Blueback Herring 0.936 -0.040 0.024 0.108 0.596 -2.448 -0.190 1987 -0.045 0.050 Bluefish 0.861 -0.052 0.023 0.031 0.848 -0.099 0.090 2002 -0.371 0.049 Hogchoker 0.832 -0.056 0.023 0.019 0.805 -1.988 3.263 1987 -0.125 -0.022 Rainbow Smelt 0.839 -0.055 0.023 0.022 0.837 -0.265 0.451 1992 -0.163 -0.016 Striped Bass 0.952 -0.037 0.024 0.141 0.895 -0.560 2.207 1987 -0.115 0.001 Weakfish 1.014 -0.020 0.025 0.432 0.968 -0.091 0.130 2000 -0.296 0.092 White Perch 0.948 -0.038 0.024 0.132 0.943 -0.318 0.065 1996 -0.091 0.127 Table 10. River Segment 4 Assessment of the Level of Potential Negative Impact Based on the Standardized FSS CPUE (updated FSEIS Table 1-20).

Species Best Fit Slope Slope 1 Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Alewife LR S=0 I American Shad SR S1=0 S2<0 4 Atlantic Tomcod LR S<0 4 Bay Anchovy SR S1=0 S2=0 1 Blueback Herring SR S1<0 S2=0 4 Bluefish SR SI=0 S2=0 1 Hogchoker SR S1=0 S2<0 4 Rainbow Smelt SR S1=0 S2<0 4 Striped Bass SR S1--0 S2=0 I Weakfish SR S1=0 S2=0 1 White Perch SR SI=0 S2=0 I 29

Table 11. Competing Models Used To Characterize the Standardized River Segment 4 LRS Population Trends of YOY Atlantic Tomcod CPUE Using a 3-Year Moving Average (updated FSEIS Table I-22).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope 1 Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL tlantic Tomcod 1.012 -0.021 0.025 0.410 0.842 -1.609 0.089 1988 -0.044 0.076 Table 12. River Segment 4 Assessment of the Level of Potential Negative Impact Based 7 on the Standardized LRS Atlantic Tomcod YOY CPUE Using a 3-Year Moving Average (updated FSEIS Table 1-23).

Species Best Fit Slope Slope I Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Atlantic Tomcod SR I S1=0 S2=0 1 30

Table 13. Assessment of Population Impacts for IP2 and IP3 River Segment 4 (updated FSEIS Table 1-24).

Species Density CPUE River Segment FSS BSS LRS FSS LRS Assessment Alewife 4 1 N/A 1 N/A 2.0 American Shad 4 4 N/A 4 N/A 4.0 Atlantic Menhaden N/A N/A N/A N/A N/A Unknown Atlantic Sturgeon N/A N/A N/A N/A N/A Unknown Atlantic Tomcod 4 N/A 4 4 1 3.3 Bay Anchovy 4 1 N/A 1 N/A 2.0 Blueback Herring 4 4 N/A 4 N/A 4.0 Bluefish 4 1 N/A 1 N/A 2.0 Gizzard Shad N/A N/A N/A N/A N/A Unknown Hogchoker 4 4 N/A 4 N/A 4.0 Rainbow Smelt 4 N/A N/A 4 N/A 4.0 Shortnose Sturgeon N/A N/A N/A N/A N/A Unknown Spottail Shiner N/A 1 N/A N/A N/A 1.0 Striped Bass 4 4 N/A 1 N/A 3.0 Weakfish 4 N/A N/A 1 N/A 2.5 White Catfish 1 N/A N/A N/A N/A 1.0 White Perch 4 1 N/A 1 N/A 2.0 Blue Crab N/A N/A N/A N/A N/A Unknown 31

Table 14. Competing Models Used To Characterize the Standardized Riverwide FSS Population Trends of YOY Fish CPUE (updated FSEIS Table 1-27).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope 1 Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Alewife 1.006 0.023 0.025 0.370 1.011 -0.038 0.188 2000 -0.266 0.130 American Shad 0.553 -0.086 0.018 0.000 0.556 -0.380 0.596 1989 -0.151 -0.048 Atlantic Tomcod 0.725 -0.069 0.021 0.003 0.768 -0.414 0.146 1993 -0.125 0.026 Bay Anchovy 1.036 0.008 0.025 0.763 0.902 -0.027 0.348 1995 -0.175 0.038 Blueback Herring 0.701 -0.072 0.021 0.002 0.746 -0.394 0.460 1991 -0.154 -0.025 Bluefish 0.822 -0.058 0.022 0.016 0.828 -0.121 0.104 1999 -0.281 0.034 Hogchoker 0.921 -0.043 0.024 0.084 0.912 -0.222 0.014 1999 -0.126 0.204 Spottail Shiner 0.827 -0.057 0.022 0.018 0.880 -0.174 0.019 2002 -0.218 0.209 Striped Bass 0.833 -0.056 0.023 0.020 0.614 0.088 2.381 1987 -0.135 -0.039 White Perch 1.004 -0.023 0.025 0.352 0.988 -0.479 0.155 1993 -0.066 0.106 Table 15. Riverwide Assessment of the Level of Potential Negative Impact Based on the Standardized FSS CPUE (updated FSEIS Table 1-28).

Species Best Fit Slope Slope 1 Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Alewife LR S=0 I American Shad LR S<0 4 Atlantic Tomcod LR S<0 4 Bay Anchovy SR S1=O S2=0 1 Blueback Herring LR S<0 4 Bluefish LR S<0 4 Hogchoker SR S1=0 S2=0 I Spottail Shiner LR S<0 4 Striped Bass SR SI>0 S2<0 4 White Perch SR S 1=0 S2=0 I 32

Table 16. Competing Models Used To Characterize the Standardized Riverwide BSS Population Trends of YOY Fish CPUE (updated FSEIS Table 1-30).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope I Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Alewife 0.744 0.067 0.021 0.004 0.726 -0.054 0.107 2002 -0.053 0.402 American Shad 0.551 -0.086 0.018 0.000 0.554 -0.285 0.451 1990 -0.162 -0.051 Atlantic Tomcod 0.543 -0.087 0.018 0.000 0.341 -1.704 0.004 1988 -0.087 -0.016 Bay Anchovy 0.8 13 0.059 0.022 0.014 0.607 -0.046 0.062 2006 -0.026 0.994 Blueback Herring 1.036 -0.008 0.025 0.760 1.072 -0.515 0.839 1990 -0.101 0.043 luefish 1.040 0.003 0.025 0.919 1.073 -0.076 0.180 1999 -0.240 0.119 ogchoker 1.034 0.010 0.025 0.695 1.068 -0.204 0.113 1998 -0.076 0.208 ainbow Smelt 0.972 -0.032 0.024 0.199 1.002 -0.269 0.370 1993 -0.139 0.034 Spottail Shiner 0.743 0.067 0.021 0.004 0.805 -0.124 0.230 1996 -0.026 0.176 Striped Bass 1.020 0.018 0.025 0.488 0.932 -0.017 0.127 2005 -0.704 0.251 Weakfish 0.918 -0.043 0.024 0.081 1986 -0.055 0.024 White Catfish 1.034 -0.010 0.025 0.699 1.010 -1.315 0.545 1989 -0.047 0.083 White Perch 1.033 -0.010 0.025 0.691 1.015 -0.367 0.092 1994 -0.063 0.144 Table 17. Riverwide Assessment of the Level of Potential Negative Impact Based on the BSS CPUE (updated FSEIS Table 1-31).

Species Best Fit Slope Slope I Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Alewife SR S1=0 S2=0 1 American Shad LR S<0 4 Atlantic Tomcod SR S1=0 S2<0 4 Bay Anchovy SR S1=0 S2=0 I Blueback Herring LR S=0 I Bluefish LR S=0 1 Hogchoker LR S=0 1 Rainbow Smelt LR S=0 I Spottail Shiner LR S>0 1 Striped Bass SR SI=0 S2=0 I Weakfish LR S=0 I White Catfish SR S1=0 S2=0 I White Perch SR S1=0 S2=0 I 33

Table 18. Competing Models Used To Characterize the Standardized Riverwide LRS Population Trend of YOY Atlantic Tomcod CPUE (updated FSEIS Table 1-33).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope 1 Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Atlantic Tomcod 0.938 -0.039 0.024 0.112 -0.089 0.010 2016 Table 19. Riverwide Assessment of the Level of Potential Negative Impact Based on the Standardized LRS CPUE of Atlantic Tomcod (updated FSEIS Table 1-34).

Species Best Fit Slope Slope I Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Atlantic Tomcod LR S=0 1 34

Table 20. Competing Models Used To Characterize the Standardized Riverwide YOY Abundance Index Trends (updated FSEIS Table 1-36).

Species Linear Regression Segmented Regression MSE Slope Std Err p-value MSE Slope 1 Join Slope 2 of Slope Point Estimate Lower Upper Lower Upper 95% CL 95% CL 95% CL 95% CL Alewife 1.017 0.019 0.025 0.458 1 American Shad 0.596 -0.082 0.019 0.000 0.594 -0.588 0.838 1989 -0.150 -0.050 Atlantic Tomcod 0.576 -0.084 0.019 0.000 0.547 -1.637 2.690 1987 -0.141 -0.056 Bay Anchovy 0.744 -0.067 0.021 0.004 0.786 -0.194 0.005 2000 -0.198 0.152 Blueback Herring 0.792 -0.062 0.022 0.010 0.794 -0.154 -0.021 2006 -0.300 0.581 Bluefish 1.035 0.009 0.025 0.731 0.967 -0.038 0.205 1999 -0.259 0.081 Hogchoker 0.902 -0.046 0.023 0.062 0.942 -0.165 0.017 2003 -0.219 0.299 Rainbow Smelt 0.971 -0.033 0.024 0.193 0.960 -0.216 0.409 1992 -0.148 0.022 Spottail Shiner 0.844 0.055 0.023 0.024 0.879 -0.008 0.168 2003 -0.273 0.227 Striped Bass 1.039 0.005 0.025 0.855 0.925 -0.024 0.131 2005 -0.657 0.095 Weakfish 0.647 -0.077 0.020 0.001 0.576 -0.561 0.032 1992 -0.095 0.027 White Catfish 0.833 -0.056 0.023 0.020 0.863 -0.198 0.011 2001 -0.173 0.193 White Perch 1.039 -0.003 0.025 0.906 1.093 -0.079 0.103 2003 -0.424 0.243 Table 21. Riverwide Assessment of the Level of Potential Negative Impact Based in the Abundance Index (updated FSEIS Table 1-37).

Species Best Fit Slope Slope 1 Slope 2 Final from from from Decision Linear Segmented Segmented Regression Regression Regression Alewife LR S=0 I American Shad SR S 1O S2<0 4 Atlantic Tomcod SR S1=0 S2<0 4 Bay Anchovy LR S<0 4 Blueback Herring LR S<0 4 Bluefish SR SI=0 S2=0 I Hogchoker LR S=0 I Rainbow Smelt SR S1=0 S2=0 I Spottail Shiner LR S>0 1 Striped Bass SR S1=0 S2=0 I Weakfish SR S1=0 S2=0 I White Catfish LR S<0 4 White Perch LR S=0 I 35

Table 22. Assessment of Riverwide Population Impacts (updated FSEIS Table 1-38).

Species CPUE Abundance Riverwide Index Assessment FSS BSS LRS Alewife 1 1 N/A 1 1.0 American Shad 4 4 N/A 4 4.0 Atlantic Menhaden N/A N/A N/A N/A Unknown Atlantic Sturgeon N/A N/A N/A N/A Unknown Atlantic Tomcod 4 4 1 4 3.3 Bay Anchovy 1 1 N/A 4 2.0 Blueback Herring 4 1 N/A 4 3.0 Bluefish 4 1 N/A 1 2.0 Gizzard Shad N/A N/A N/A N/A Unknown Hogchoker I I N/A 1 1.0 Rainbow Smelt N/A 1 N/A 1 1.0 Shortnose Sturgeon N/A N/A N/A N/A Unknown Spottail Shiner 4 1 N/A 1 2.0 Striped Bass 4 1 N/A 1 2.0 Weakfish N/A I N/A 1 1.0 White Catfish N/A I N/A 4 2.5 White Perch I I N/A 1 1.0 Blue Crab N/A N/A N/A N/A Unknown 36

Table 23. Weight of Evidence Results for the Population Trend Line of Evidence (updated FSEIS Table H-15).

Species River Riverwide WOE Impact Segment Assessment Score Conclusion Assessment Score Score Alewife 2.0 1.0 1.6 Undetected Decline American Shad 4.0 4.0 4.0 Detected Decline Atlantic Menhaden Unknown Unknown Unknown Unresolved Atlantic Sturgeon Unknown Unknown Unknown Unresolved Atlantic Tomcod 3.3 3.3 3.3 Detected Decline Bay Anchovy 2.0 2.0 2.0 Undetected Decline Blueback Herring 4.0 3.0 3.6 Detected Decline Bluefish 2.0 2.0 2.0 Undetected Decline Gizzard Shad Unknown Unknown Unknown Unresolved Hogchoker 4.0 1.0 2.8 Variable Rainbow Smelt 4.0 1.0 2.8 Variable Shortnose Sturgeon Unknown Unknown Unknown Unresolved Spottail Shiner 1.0 2.0 1.4 Undetected Decline Striped Bass 3.0 2.0 2.6 Variable Weakfish 2.5 1.0 1.9 Undetected Decline White Catfish 1.0 2.5 1.6 Undetected Decline White Perch 2.0 1.0 1.6 Undetected Decline Blue Crab Unknown Unknown Unknown Unresolved 37

Table 24. Parameter Values Used in the Monte Carlo Simulation (updated FSEIS Table 1-46).

RIS Survey Linear Slope Error CV of EMR IMR Used Slope plus Mean Density (r) Standard Square Data Error of from (1985-the Slope Regression 1996)

Estimate Alewife BSS 0.075 0.099 0.725 1.294 0.095 0.0020 American Shad BSS -0.120 -0.106 0.235 0.510 0.042 0.0005 Atlantic Tomcod FSS -0.080 -0.058 0.678 0.794 0.036 0.0300 Bay Anchovy FSS -0.088 -0.067 0.601 0.511 0.213 0.0040 Blueback Herring BSS -0.075 -0.051 0.726 1.034 0.095 0.0040 Bluefish BSS 0.013 0.041 1.034 0.754 0.003 0.0005 Hogchoker FSS -0.104 -0.086 0.434 1.225 0.386 0.0005 Rainbow Smelt FSS -0.086 -0.064 0.623 1.211 0.258 0.0005 Spottail Shiner BSS -0.031 -0.004 0.989 1.182 0.031 0.0070 Striped Bass BSS 0.022 0.050 1.016 0.523 0.106 0.0080 Weakfish FSS -0.064 -0.039 0.811 0.698 0.544 0.0005 White Catfish FSS -0.042 -0.015 0.945 2.566 0.114 0.0005 White Perch BSS -0.012 0.016 1.035 1.005 0.076 0.0320 38

Table 25. Quartiles of the Relative Difference in Cumulative Abundance and Conclusions for the Strength-of-Connection From the Monte Carlo Simulation (updated FSEIS Table 1-47).

Taxa Number No= 1000 No= 1 x 108 Strength of of Median Q1 Q3 Median Q1 Q3 Connection Years Conclusion Alewife 20 -0.07 -1.19 1.03 -0.07 -1.17 1.01 Low 27 -0.32 -1.63 1.02 -0.32 -1.69 1.04 American Shad 20 0.07 -0.01 0.14 0.07 -0.01 0.14 Low 27 0.06 0.00 0.11 0.06 0.00 0.11 Atlantic Tomcod 20 0.15 -0.03 0.34 0.16 -0.03 0.35 Low 27 0.16 0.01 0.30 0.15 0.01 0.30 Bay Anchovy 20 0.29 0.13 0.44 0.29 0.13 0.44 High 27 0.27 0.15 0.39 0.27 0.15 0.39 Blueback Herring 20 0.21 -0.03 0.46 0.22 -0.02 0.46 Low 27 0.22 0.04 0.41 0.23 0.04 0.42 Bluefish 20 0.45 -0.09 0.99 0.45 -0.09 0.98 Low 27 0.67 0.11 1.21 0.69 0.15 1.23 Hogchoker 20 0.58 0.31 0.85 0.57 0.30 0.86 High 27 0.56 0.35 0.78 0.56 0.35 0.78 Rainbow Smelt 20 0.45 0.16 0.74 0.46 0.16 0.76 High 27 0.45 0.23 0.68 0.45 0.23 0.68 Spottail Shiner 20 0.27 -0.14 0.68 0.27 -0.13 0.69 Low 27 0.34 -0.00 0.69 0.34 0.00 0.68 Striped Bass 20 0.84 0.25 1.43 0.83 0.24 1.42 High 27 1.27 0.64 1.93 1.28 0.64 1.92 Weakfish 20 0.74 0.42 1.07 0.75 0.42 1.07 High 27 0.76 0.49 1.02 0.76 0.50 1.02 White Catfish 20 0.42 -0.26 1.10 0.44 -0.27 1.13 Low 27 0.49 -0.07 1.06 0.47 -0.10 1.06 White Perch 20 0.40 -0.08 0.90 0.40 -0.07 0.88 Low 1 27 0.51 0.10 0.93 0.51 0.08 0.93 39

Table 26. Impingement and Entrainment Impact Summary for Hudson River YOY RIS (updated FSEIS Table H- 17).

Species Population Trend Strength of Impacts of IP2 and Line of Evidence Connection IP3 Cooling Systems Line of Evidence on YOY RIS Alewife Undetected Decline Low Small American Shad Detected Decline Low Small Atlantic Menhaden Unresolved Low(b) Small Atlantic Sturgeon Unresolved Low(bJ Small Atlantic Tomcod Detected Decline Low Small Bay Anchovy Undetected Decline High Small Blueback Herring Detected Decline Low Small Bluefish Undetected Decline Low Small Gizzard Shad Unresolved Low(b) Small Hogchoker Variable High Moderate Rainbow Smelt Variable High Moderate Shortnose Sturgeon Unresolved Low(b) Small Spottail Shiner Undetected Decline Low Small Striped Bass Variable High Moderate Weakfish Undetected Decline High Small White Catfish Undetected Decline Low Small White Perch Undetected Decline Low Small Blue Crab Unresolved Low(b) Small (b Strength of connection could not be established using Monte Carlo Simulation; therefore, strength of connection was based on the rate of entrainment and impingement.

40

VIII. Appendix A The "p-value" is the probability level for the significance test of the estimated slope (i) from the linear regression. It is the probability that the absolute value of a random variable from a t-distribution is greater than the ratio:

P se(i) where se(i) is the standard error of the estimated slope (Draper and Smith, 1966). For Tables 1-9 and 1-12 it is a t-distribution with 23 degrees of freedom because the time series of 3-year averages of River Segment 4 density estimates contained 25 index values and the linear regression model had 2 parameters.

For each linear slope estimate listed in Table 1-46, the corresponding "p-value" listed in Table 1-9 or Table 1-12 was equal (allowing for round-off errors with 3 significant digits listed in Tables 1-9 and 1-12) to the probability that the absolute value of a random variable with a t-distribution with 23 degrees of freedom was greater than the ratio:

estimated slope value to the right of the +/- symbol This demonstrates that the undefined value to the right of the +/- symbol in the slope column of Tables 1-9 and 1-12 was, in fact, the standard error of the estimated slope.

41

IX. Appendix B Report on QC Review of Analysis Update Prepared by John Young, PhD ASA Analysis & Communication, Inc.

921 Pike Street, PO Box 303 Lemont, PA 16851-0303 October 18, 2013 ASA reviewed the SAS programs used to analyze clean data files (1985-20 t 1) from the Hudson River Biological Monitoring Program with NRC's trend assessment methods. The first step in the review process was to create SAS datasets from 1974-2011 level files for BSS, FSS, and LRS programs. These datasets contained one observation for each of the target species for each sample from each program. A separate dataset was constructed for each program each year. Once the datasets had been created, the next step was to run the following series of SAS programs used for the analyses:

1. NRC Region 4 Indices Corrected vl0

2. NRC Riverwide Indices Corrected v l0

3. NLIN and REG NRC trends Corrected v13

4. NRC Trends summary Corrected vi 6

5. SOC input updated trends results v 1

6. SOC update vl0 Each line of the resulting SAS log files was evaluated for error, warning and other unexpected messages. The presence of such messages indicates an error is present in the program that may lead to inaccurate results. All six programs ran successfully and were found to be free of errors. The draft results presented in table H-15, H-17, 1-24, and 1-38 were reproduced.

42

After confirming that the programs were running successfully, the methodology applied within the programs was compared to that used by NRC to ensure that the analysis was accurately reproduced. The following steps were taken during the methodology review:

1. Reviewed all input and output datasets of the program

2. Evaluated sort order of all datasets

3. Evaluated all macros

4. Evaluated code logic These steps ultimately tested the accuracy and integrity of the program logic and output which it produces.

The methodology review did not identify deviances from the NRC data analysis methodology. All input and output datasets were accurate, sorting was not found to be an issue, macros ran without error, and code logic mirrored that used by NRC.

In summary, the programs used to apply NRC trend assessment methodology to the clean data files from the Hudson River Biological Monitoring Program were found to be free of errors. The results produced by the programs are considered to be accurate.

43

ATTACHMENT 2 TO NL-14-030 78 Fed. Reg. 48944 (August 12, 2013);

Sustainable Fishing Plan for New York River Herring Stocks (2011);

Rainbow Smelt: An Imperiled Fish in a Changing World (2010);

A Regional Conservation Plan for Anadromous Smelt (2012); and, Correspondence from Mark D. Sanza, Assistant Counsel for NYSDEC to ALJs Villa and O'Connell, Administrative Law Judges for NYSDEC, re: Entergy Nuclear Indian Point Units 2 and 3, CWA Section 401 WQC Application Proceeding.

ENTERGY NUCLEAR OPERATIONS, INC.

INDIAN POINT NUCLEAR GENERATING UNIT NOS. 2 & 3 DOCKET NOS. 50-247 AND 50-286

FEDERAL REGISTER Vol. 78 Monday, No. 155 August 12, 2013 Part II Department of Commerce National Oceanic and Atmospheric Administration Endangered and Threatened Wildlife and Plants; Endangered Species Act Listing Determination for Alewife and Blueback Herring; Notice

48944 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices DEPARTMENT OF COMMERCE Administrator, Protected Resources Northeast Fisheries Science Center staff)

Division, Northeast Region, NMFS, 55 to compile the best commercial and National Oceanic and Atmospheric Great Republic Drive, Gloucester, MA scientific data available for alewife and Administration 01930. blueback herring throughout their

[Docket No. 111024651-3630-02] FOR FURTHER INFORMATION CONTACT: Kim ranges.

Damon-Randall, NMFS Northeast In May 2012, the ASMFC completed RIN 0648-XA739 a river herring stock assessment, which Regional Office, (978) 282-8485; or Marta Nammack, NMFS, Office of covers over 50 river-specific stocks Endangered and Threatened Wildlife throughout the ranges of the species in and Plants; Endangered Species Act Protected Resources (301) 427-8469.

the United States (ASMFC, 2012; Listing Determination for Alewife and SUPPLEMENTARY INFORMATION: hereafter referred to in this Blueback Herring Background determination as "the stock AGENCY: National Marine Fisheries On August 5, 2011, we, the National assessment"). In order to avoid Service (NMFS), National Oceanic and Marine Fisheries Service (NMFS), duplicating this extensive effort, we Atmospheric Administration (NOAA), worked cooperatively with ASMFC to received a petition from the Natural Commerce. Resources Defense Council (NRDC), use this information in the review of the status of these two species and identify ACTION: Notice of a listing requesting that we list alewife (Alosa information not in the stock assessment determination. pseudoharengus)and blueback herring (Alosa aestivalis)under the ESA as that was needed for our listing

SUMMARY

We, NMFS, have completed a determination. We identified the threatened throughout all or a comprehensive review of the status of missing required elements and held significant portion of their ranges. In the river herring (alewife and blueback workshops/working group meetings alternative, they requested that we herring) in response to a petition focused on addressing information on designate DPSs of alewife and blueback submitted by the Natural Resources stock structure, extinction risk analysis, herring as specified in the petition and climate change.

Defense Council (NRDC) requesting that (Central New England, Long Island we list alewife (Alosa pseudoharengus) Reports from each workshop/working Sound, Chesapeake Bay, and Carolina group meeting were compiled and and blueback herring (Alosa oestivalis) for alewives, and Central New England, as threatened under the Endangered independently peer reviewed (the stock Long Island Sound, and Chesapeake Bay structure and extinction risk reports Species Act (ESA) throughout all or a for blueback herring). The petition significant portion of their range or as were peer reviewed by reviewers contained information on the two selected by the Center for Independent specific distinct population segments species, including the taxonomy, (DPS) identified in the petition. The Experts, and the climate change report historical and current distribution, was peer reviewed by 4 experts Atlantic States Marine Fisheries physical and biological characteristics Commission (ASMFC) completed a identified during the workshops). These of their habitat and ecosystem reports did not contain any listing comprehensive stock assessment for relationships, population status and river herring in May 2012 which covers advice or reach any ESA listing trends, and factors contributing to the conclusions-such synthesis and over 50 river specific stocks throughout species' decline. The petition also the range of the species in the United analysis for river herring is solely included information regarding within the agency's purview. We used States. The ASMFC stock assessment potential DPSs of alewife and blueback contained much of the information this information to determine which herring as described above. The extinction risk method and stock necessary to make an ESA listing following five factors identified in determination for both species; structure analysis would best inform the section 4(a)(1) of the ESA were listing determination, as well as however, any deficiencies were addressed in the petition: (1) Present or addressed through focused workshops understand how climate change may threatened destruction, modification, or impact river herring, and ultimately, we and working group meetings and review curtailment of habitat or range; (2) over-of additional sources of information. are using these reports along with the utilization for commercial, recreational, stock assessment and all other best Based on the best scientific and scientific, or educational purposes; (3) commercial information available, we available information in this listing disease or predation; (4) inadequacy of determination.

have determined that listing alewife as existing regulatory mechanisms; and (5) Alewife and blueback herring are threatened or endangered under the other natural or man-made factors ESA is not warranted at this time. collectively referred to as "river affecting the species' continued herring." Due to difficulties in Additionally, based on the best existence. distinguishing between the species, they scientific and commercial information We reviewed the petition and are often harvested together in available, we have determined that determined that, based on the listing blueback herring as threatened or commercial and recreational fisheries, information in the petition and in our and managed together by the ASMFC.

endangered under the ESA is not files at the time we received the warranted at this time. Throughout this finding, where there petition, the petitioned action may be are similarities, they will be collectively DATES: This finding is effective on warranted. Therefore, we published a referred to as river herring, and where August 12, 2013. positive 90-day finding on November 2, there are distinctions, they will be ADDRESSES: The listing determination, 2011, and as a result, we were required identified by species.

list of references used in the listing to review the status of the species (e.g.,

determination, and other related anadromous alewife and blueback Range materials regarding this determination herring) to determine if listing under the River herring can be found along the can be obtained via the Internet at: ESA is warranted. We formed an Atlantic coast of North America, from http://www.nero.nooo.gov/prot res/ internal status review team (SRT) the Southern Gulf of St. Lawrence, CandidateSpeciesProgram/River comprised of nine NMFS staff members Canada to the southeastern United HerringSOC.htmor by submitting a (Northeast Regional Office (NERO) States (Mullen et al., 1986; Schultz et request to the Assistant Regional Protected Resources Division and al., 2009). The coastal ranges of the two

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013 / Notices 48945 species overlap. Blueback herring range including the ocean, estuaries, rivers, matures earlier, has a smaller adult body from Nova Scotia south to the St. John's and freshwater lakes and ponds. The size, and reduced fecundity (Palkovacs River, Florida; and alewife range from substrate preferred for spawning varies et al., 2007). At this time, there is no Labrador and Newfoundland south to greatly and can include gravel, detritus, substantive information that would South Carolina, though their occurrence and submerged aquatic vegetation. suggest that landlocked populations can in the extreme southern range is less Blueback herring prefer swifter moving or would revert back to an anadromous common (Collette and Klein-MacPhee, waters than alewives do (ASMFC, life history if they had the opportunity 2002; ASMFC, 2009a; Kocik et al., 2009a). Nursery areas include to do so (Gephard, CT DEEP, Pers.

2009). freshwater and semi-brackish waters. comm. 2012; Jordaan, UMASS Amherst, In Canada, river herring (i.e., Little is known about their habitat Pers. comm. 2012).

gaspereaul are most abundant in the preference in the marine environment The discrete life history and Miramichi, Margaree, LaHave, Tusket, (Meadows, 2008; ASMFC, 2009a). morphological differences between the Shubenacadie and Saint John Rivers two life history variants (anadromous (Gaspereau Management Plan, 2001). Landlocked Populations and landlocked) provide substantial They are proportionally less abundant Landlocked populations of alewives evidence that upon becoming in smaller coastal rivers and streams and blueback herring also exist. landlocked, landlocked populations (Gaspereau Management Plan, 2001). Landlocked alewife populations occur become largely independent and Generally, blueback herring in Canada in many freshwater lakes and ponds separate from anadromous populations occur in fewer rivers than alewives and from Canada to North Carolina as well and occupy largely separate ecological are less abundant in rivers where both as the Great Lakes (Rothschild, 1966; niches (Palkovacs and Post, 2008).

species coexist (DFO 2001). Boaze & Lackey, 1974). Many There is the possibility that landlocked landlocked populations occur as a result alewife and blueback herring may have Habitat and Migration of stocking to provide a forage base for the opportunity to mix with River herring are anadromous, game fish species (Palkovacs et al., anadromous river herring during high meaning that they mature in the marine 2007). discharge years and through dam environment and then migrate up Landlocked blueback herring occur removals which could provide passage coastal rivers to estuarine and mostly in the southeastern United States over dams and access to historic freshwater rivers, ponds, and lake and the Hudson River drainage. The spawning habitats restored for habitats to spawn (Collette and Klein- occurrence of landlocked blueback anadromous populations, where it did MacPhee, 2002; ASMFC, 2009a; Kocik herring is primarily believed to be the not previously exist. The implications of et al., 2009). In general, adult river result of accidental stockings in this are not known at this time.

herring are most often found at depths reservoirs (Prince and Barwick, 1981), In summary, genetics indicate that less than 328 feet (ft) (100 meters (m)) unsanctioned stocking by recreational anadromous alewife populations are in waters along the continental shelf anglers to provide forage for game fish, discrete from landlocked populations, (Neves, 1981; ASMFC, 2009a; Schultz et and also through the construction of and that this divergence can be al., 2009). They are highly migratory, locks, dams and canal systems that have estimated to have taken place from 300 pelagic, schooling species, with subsequently allowed for blueback to 5,000 years ago. Some landlocked seasonal spawning migrations that are herring occupation of several lakes and populations of blueback herring do cued by water temperature (Collette and ponds along the Hudson River drainage occur in the Mid-Atlantic and Klein-MacPhee, 2002; Schultz et al., up to, and including Lake Ontario southeastern United States. Given the 2009). Depending upon temperature, (Limburg et al., 2001). similarity in life histories between blueback herring typically spawn from Recent efforts to assess the anadromous alewife and blueback late March through mid-May. However, evolutionary origins of landlocked herring, we assume that landlocked they spawn in the southern parts of alewives indicate that they rapidly populations of blueback herring would their range as early as December or diverged from their anadromous cousins exhibit a similar divergence from January, and as late as August in the between 300 and 5,000 years ago, and anadromous blueback herring, as has northern portion of their range (ASMFC, now represent a discrete life history been documented with alewives.

2009a). Alewives have been variant of the species, Alosa A Memorandum of Understanding documented spawning as early as pseudoharengus(Palkovacs et al., 2007). (MOU) between the U.S. Fish and February in the southern portion of their Though given their relatively recent Wildlife Service (USFWS) and NMFS range, and as late as August in the divergence from anadromous (collectively, the Services) regarding northern portion of the range (ASMFC, populations, one plausible explanation jurisdictional responsibilities and listing 2009a). The river herring migration in for the existence of landlocked procedures under the ESA was signed Canada extends from late April through populations may be the construction of August 28, 1974. This MOU states that early July, with the peak occurring in dams by either native Americans or NMFS shall have jurisdiction over late May and early June. Blueback early colonial settlers that precluded the species "which either (1) reside the herring generally make their spawning downstream migration of juvenile major portion of their lifetimes in runs about 2 weeks later than alewives herring (Palkovacs et al., 2007). Since marine waters; or (2) are species which do (DFO, 2001). River herring conform their divergence, landlocked alewives spend part of their lifetimes in estuarine to a metapopulation paradigm (e.g., a have evolved to a point they now waters, if the major portion of the group of spatially separated populations possess significantly different remaining time (the time which is not of the same species which interact at mouthparts than their anadromous spent in estuarine waters) is spent in some level) with adults frequently cousins, including narrower gapes and marine waters."

returning to their natal rivers for smaller gill raker spacings to take Given that landlocked populations of spawning but with some limited advantage of year round availability of river herring remain in freshwater straying occurring between rivers (Jones, smaller prey in freshwater lakes and throughout their life history and are 2006; ASMFC, 2009a). ponds (Palkovacs et al., 2007). genetically divergent from the Throughout their life cycle, river Furthermore, the landlocked alewife, anadromous species, pursuant to the herring use many different habitats, compared to its anadromous cousin, aforementioned MOU, we did not

48946 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices include the landlocked populations of segment to the remainder of the species The subcommittee also noted that alewife and blueback herring in our (or subspecies) to which it belongs. As most state landings records listed review of the status of the species and further stated in the joint policy, if a alewife and blueback herring together as do not consider landlocked populations population segment is discrete and 'river herring' rather than identifying by in this listing determination in response significant (i.e., it meets the DPS policy species. These landings averaged 30.5 to the petition to list these anadromous criteria), its evaluation for endangered million pounds (lbs) (13,847 metric tons species. or threatened status will be based on the (mt)) per year from 1889 to 1938, and Listing Species Under the Endangered ESA's definitions of those terms and a severe declines were noted coast-wide review of the five factors enumerated in starting in the 1970s. Beginning in 2005, Species Act section 4(a)(1) of the ESA. states began enacting moratoria on river We are responsible for determining As provided in section 4(a) of the herring fisheries, and as of January whether alewife and blueback herring ESA, the statute requires us to 2012, all directed harvest of river are threatened or endangered under the determine whether any species is herring in state waters is prohibited ESA (16 U.S.C. 1531 et seq.). endangered or threatened because of unless states have submitted and Accordingly, based on the statutory, any of the following five factors: (1) The obtained approved sustainable fisheries regulatory, and policy provisions present or threatened destruction, management plans (FMP) under described below, the steps we followed modification, or curtailment of its ASMFC's Amendment 2 to the Shad and in making our listing determination for habitat or range; (2) overutilization for River Herring FMP.

alewife and blueback herring were to: commercial, recreational, scientific, or The subcommittee summarized its (1) Determine how alewife and blueback educational purposes; (3) disease or findings for trends in commercial catch-herring meet the definition of "species"; predation; (4) the inadequacy of existing per-unit-effort (CPUE); run counts; (2) determine the status of the species regulatory mechanisms; or (5) other young-of-the-year (YOY) seine surveys; and the factors affecting them; and (3) natural or manmade factors affecting its juvenile-adult fisheries independent identify and assess efforts being made to continued existence (section seine, gillnet and electrofishing surveys; protect the species and determine if 4(a)(1)(A)(E)). Section 4(b)(1)(A) of the juvenile-adult trawl surveys; mean these efforts are adequate to mitigate ESA further requires that listing length; maximum age; mean length-at-existing threats. determinations be based solely on the age; repeat spawner frequency; total To be considered for listing under the best scientific and commercial data mortality (Z) estimates; and exploitation ESA, a group of organisms must available after taking into account rates. Because the stock assessment constitute a "species." Section 3 of the efforts being made to protect the contains the most recent and ESA defines a "species" as "any comprehensive description of this species.

subspecies of fish or wildlife or plants, information and the subcommittee's and any distinct population segment of Distribution and Abundance conclusions, the following sections were any species of vertebrate fish or wildlife taken from the stock assessment which interbreeds when mature." United States (ASMFC, 2012).

Section 3 of the ESA further defines an The stock assessment (described endangered species as "any species above) was prepared and compiled by Commercial CPUE which is in danger of extinction the River Herring Stock Assessment Since the mid-1990s, CPUE indices throughout all or a significant portion of Subcommittee, hereafter referred to as for alewives showed declining trends in its range" and a threatefied species as the 'subcommittee,' of the ASMFC Shad the Potomac River and James River one "which is likely to become an and River Herring Technical Committee. (VA), no trend in the Rappahannock endangered species within the Data and reports used for this River (VA), and increasing trends in the foreseeable future throughout all or a assessment were obtained from Federal York River (VA) and Chowan River significant portion of its range." Thus, and state resource agencies, power (NC). CPUE indices available for we interpret an "endangered species" to generating companies, and universities. blueback herring showed a declining be one that is presently in danger of The subcommittee conducted its trend in the Chowan River and no trend extinction. A "threatened species," on assessment on the coastal stocks of in the Santee River (SC). Combined the other hand, is not presently in alewife and blueback herring by species CPUE indices showed declining danger of extinction, but is likely to individual rivers as well as coast-wide trends in Delaware Bay and the become so in the foreseeable future (that depending on available data. The Nanticoke River, but CPUE has recently is, at a later time). In other words, the subcommittee concluded that river increased in the Hudson River (ASMFC, primary statutory difference between a herring should ideally be assessed and 2012).

threatened and endangered species is managed by individual river system, but the timing of when a species may be in that the marine portion of their life Run Counts danger of extinction, either presently history likely influences survival Major declines in run sizes occurred (endangered) or in the foreseeable future through mixing in the marine portion of in many rivers from 2001 to 2005. These (threatened). their range. However, coast-wide declines were followed by increasing On February 7, 1996, the Services assessments are complicated by the trends (2006 to 2010) in the adopted a policy to clarify our complex life history of these species as Androscoggin River (ME), Damaraiscotta interpretation of the phrase "distinct well, given that factors influencing River (ME), Nemasket River (MA),

population segment of any species of population dynamics for the freshwater Gilbert-Stuart River (RI), and Nonquit vertebrate fish or wildlife" (61 FR 4722). portion of their life history can not River (RI) for alewife and in the The joint DPS policy describes two readily be separated from marine Sebasticook River (ME), Cocheco River criteria that must be considered when factors. In addition, it was noted that (NH), Lamprey River (NH), and identifying DPSs: (1) The discreteness of data quality and availability varies by Winnicut River (NH) for both species the population segment in relation to river and is mostly dependent upon the combined. No trends in run sizes were the remainder of the species (or monitoring efforts that each state evident following the recent major subspecies) to which it belongs; and (2) dedicates to these species, which further declines in the Union River (ME),

the significance of the population complicated the assessment. Mattapoisett River (MA), and

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48947 Monument River (MA) for alewife and The electrofishing indices showed Mean Length-at-Age in the Exeter River (NH) for both species opposing trends and then declining combined. Run sizes have declined or trends in the Rappahannock River Declines in mean length of at least are still declining following recent and (alewife and blueback herring) with one age were observed in most rivers historical major declines in the Oyster catch rates of blueback herring peaking examined. The lack of significance in River (NH) and Taylor River (NH) for during 2001-2003, and catch rates of some systems is likely due to the both species, in the Parker River (MA) alewives lowest during the same time absence of data prior to 1990 when the for alewife, and in the Monument River period (ASMFC, 2012). decline in sizes began, similar to the (MA) and Connecticut River for pattern observed for mean length.

blueback herring (ASMFC, 2012). Juvenile and Adult Trawl Surveys Declines in mean lengths-at-age for most Trends in trawl survey indices varied ages were observed in the north (NH)

Young-of-the-Year Seine Surveys and the south (NC). There is little greatly with some surveys showing an The young-of-the-year (YOY) seine increase in recent years, some showing indication of a general pattern of size surveys were quite variable and showed a decrease, and some remaining stable. changes along the Atlantic coast differing patterns of trends among Trawl survey data were available from (ASMFC, 2012).

rivers. Maine rivers showed similar 1966-2010 (for a complete description trends in alewife and blueback herring Repeat Spawner Frequency of data see ASMFC (2012)). Trawl YOY indices after 1991, with peaks surveys in northern areas tended to Examination of percentage of repeat occurring in 1995 and 2004. YOY spawners in available data revealed show either an increasing or stable trend indices from North Carolina and in alewife indices, whereas trawl significant, declining trends in the Connecticut showed declines from the Gilbert-Stuart River (RI-combined surveys in southern areas tended to 1980s to the present. New York's species), Nonquit River (RI-combined Hudson River showed peaks in YOY show stable or decreasing trends.

Patterns in trends across surveys were species), and the Nanticoke River indices in 1999, 2001, 2005, and 2007. (blueback herring). There were no New Jersey and Maryland YOY indices less evident for blueback herring. The NMFS surveys showed a consistent trends in the remaining rivers for which showed peaks in 1994, 1996, and 2001. data are available, although scant data increasing trend coast-wide and in the Virginia YOY surveys showed peaks in suggest that current percentages of 1993, 1996, 2001, and 2003 (ASMFC, northern regions for alewife and the combined river herring species group repeat spawners are lower than 2012). historical percentages in the Monument (ASMFC, 2012).

Juvenile-Adult Fisheries-Independent River (MA) and the Hudson River (NY)

Seine, Gillnet and Electrofishing Mean Length (ASMFC, 2012).

Surveys Mean sizes for male and female Total Mortality (Z) Estimates The juvenile-adult indices from alewife declined in 4 of 10 rivers, and fisheries-independent seine, gillnet and mean sizes for female and male With the exception of male blueback electrofishing surveys showed a variety blueback herring declined in 5 of 8 herring from the Nanticoke River, which of trends in the available datasets for the rivers. Data were available from 1960- showed a slight increase over time, Rappahanock River (1991-2010), James 2010 (for a complete description of data there were no trends in the Z estimates River (2000-2010), St. John's River, FL see ASMFC (2012)). The common trait produced using age data (ASMFC, (2001-2010), and Narragansett Bay among most rivers in which significant 2012).

(1988-2010). The gillnet indices from declines in mean sizes were detected is Exploitation Rates the Rappahannock River (alewife and that historical length data were available blueback herring) showed a low and for years prior to 1990. Mean lengths Exploitation of river herring appears stable or decreasing trend after a major started to decline in the mid to late to be declining or remaining stable. In-decline after 1995 and has remained low 1980s; therefore, it is likely that declines river exploitation estimates have since 2000 (except for a rise in alewife in other rivers were not detected fluctuated, but are lower in recent years.

CPUE during 2008). The gillnet and because of the shortness of their time A coast-wide index of relative electrofishing indices in the James River series. Mean lengths for combined sexes exploitation showed a decline following (alewife and blueback herring) showed in trawl surveys were quite variable a peak in the 1980s, and the index a stable or increasing trend. Blueback through time for both alewives and indicates that exploitation has remained herring peak catch rates occurred in blueback herring. Despite this fairly stable over the past decade. The 2004, and alewife peak catch rates variability, alewife mean length tended majority of depletion-based stock occurred in 2005. The blueback herring to be lowest in more recent surveys. reduction analysis (DB-SRA) model index from electrofishing in the St. This pattern was less apparent for runs showed declining exploitation John's River, FL, showed no trend after blueback herring. Trend analysis of rates coast-wide. Exploitation rates a major decline from 2001-2002. The mean lengths indicated significant estimated from the statistical catch-at-seine indices in Narragansett Bay, RI declines in mean lengths over time for age model for blueback herring in the (combined species) and coastal ponds alewives coast-wide and in the northern Chowan River also showed a slight (combined species) showed no trends region in both seasons, and for blueback declining trend from 1999 to 2007, at over the time series. The CPUE for coast-wide and in the northern region in which time a moratorium was Narragansett Bay fluctuated without fall (ASMFC, 2012). instituted. There appears to be a trend from 1988-1997, increased consensus among various assessment through 2000, declined and then Maximum Age methodologies that exploitation has remained stable from 2001-2004. The Except for Maine and New decreased in recent times. The decline pond survey CPUE increased during Hampshire, maximum age of male and in exploitation over the past decade is 1993-1996, declined through 1998, female alewife and blueback herring not surprising because river herring increased in 1999, declined through during 2005-2007 was 1 or 2 years populations are at low levels and more 2002, peaked in 2003 and then declined lower than historical observations restrictive regulations or moratoria have and fluctuated without trend thereafter. (ASMFC, 2012). been enacted by states (ASMFC, 2012).

48948 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Summary of Stock Assessment less intensively, though harvest rates are belongs include, but are not limited to, Conclusions monitored throughout Atlantic Canada the following: (1) Persistence of the Of the in-river stocks of alewife and through license sales, reporting discrete population segment in an blueback herring for which data were requirements, and a logbook system that ecological setting unusual or unique for available and were considered in the was enacted in 1992 (DFO, 2001). the taxon; (2) evidence that loss of the stock assessment, 22 were depleted, 1 At the time DFO conducted their last discrete population segment would was increasing, and the status of 28 stock assessment in 2001, they result in a significant gap in the range stocks could not be determined because identified river herring harvest levels as of the taxon; (3) evidence that the the time-series of available data was too being low (relative to historical levels) discrete population segment represents short. In most recent years, 2 in-river and stable, to low and decreasing across the only surviving natural occurrence of stocks were increasing, 4 were most rivers where data were available a taxon that may be more abundant decreasing, and 9 were stable, with 38 (DFO, 2001). With respect to the elsewhere as an introduced population rivers not having enough data to assess commercial harvest of river herring, outside its historic range; or (4) recent trends. The coast-wide meta- reported landings of river herring evidence that the discrete population complex of river herring stocks in the peaked in 1980 at slightly less than 25.5 segment differs markedly from other United States is depleted to near million lbs (11,600 mt) and declined to populations of the species in its genetic less than 11 million lbs (5,000 mt) in characteristics.

historical lows. A depleted status If a population segment is deemed indicates that there was evidence for 1996. Landings data reported through DFO indicate that river herring harvests discrete and significant, then it qualifies declines in abundance due to a number have continued to decline through 2010. as a DPS.

of factors, but the relative importance of these factors in reducing river herring Consideration as a Species Under the Information Related to Discreteness stocks could not be determined. ESA To obtain expert opinion about Commercial landings of river herring anadromous alewife and blueback peaked in the late 1960s, declined DistinctPopulationSegment Background herring stock structure, we convened a rapidly through the 1970s and 1980s working group in Gloucester, MA, on and have remained at levels less than 3 According to Section 3 of the ESA, the June 20-21, 2012. This working group percent of the peak over the past term "species" includes "any meeting brought together river herring decade. Estimates of run sizes varied subspecies of fish or wildlife or plants, experts from state and Federal fisheries among rivers, but in general, declining and any distinct population segment of management agencies and academic trends in run size were evident in many any species of vertebrate fish or wildlife institutions. Participants presented rivers over the last decade. Fisheries- that interbreeds when mature." information to inform the presence or independent surveys did not show Congress included the term "distinct absence of stock structure such as consistent trends and were quite population segment" in the 1978 genetics, life history, and variable both within and among amendments to the ESA. On February 7, morphometrics. A public workshop was surveys. Those surveys that showed 1996, the Services adopted a policy to held to present the expert working declines tended to be from areas south clarify their interpretation of the phrase group's findings on June 22, 2012, and of Long Island. A problem with the "distinct population segment" for the during this workshop, additional majority of fisheries-independent purpose of listing, delisting, and information on stock structure was surveys was that the length of their time reclassifying species (61 FR 4721). The sought from the public. Subsequently, a series did not overlap the period of peak policy described two criteria a summary report was developed (NMFS, commercial landings that occurred prior population segment must meet in order 2012a), and a peer review of the to 1970. There appears to be a to be considered a DPS (61 FR 4721): (1) document was completed by three consensus among various assessment It must be discrete in relation to the independent reviewers. The summary methodologies that exploitation has remainder of the species to which it report and peer review reports are decreased in recent times. The decline belongs; and (2) it must be significant to available on the NMFS Web site (see the in exploitation over the past decade is the species to which it belongs. ADDRESSES section above).

not surprising because river herring Determining if a population is Steve Gephard of the Connecticut populations are at low levels and more discrete requires either one of the Department of Energy and restrictive regulations or moratoria have following conditions: (1) It is markedly Environmental Protection (CT DEP) been enacted by states (ASMFC, 2012). separated from other populations of the presented a preliminary U.S. coast-wide same taxon as a consequence of genetic analysis of alewife and blueback Canada physical, physiological, ecological, or herring data (Palkovacs et al., 2012, The Department of Fisheries and behavioral factors. Quantitative unpublished report). Palkovacs et al.,

Oceans (DFO) monitors and manages measures of genetic or morphological (2012, unpublished report) used 15 river herring runs in Canada. River discontinuity may provide evidence of novel microsatellite markers on samples herring runs in the Miramichi River in this separation; or (2) it is delimited by collected from Maine to Florida. For New Brunswick and the Maragree River international governmental boundaries alewife, 778 samples were collected in Cape Breton, Nova Scotia were within which differences in control of from spawning runs in 15 different monitored intensively from 1983 to exploitation, management of habitat, rivers, and 1,201 blueback herring 2000 (DFO, 2001). More recently (1997 conservation status, or regulatory samples were collected from 20 rivers.

to 2006) the Gaspereau River alewife mechanisms exist that are significant in Bayesian analyses identified five run and harvest has been intensively light of section 4(a)(1)(D) of the ESA. genetically distinguishable stocks for monitored and managed partially in If a population is deemed discrete, alewife with similar results using both response to a 2002 fisheries then the population segment is STRUCTURE and Bayesian Analysis of management plan that had a goal of evaluated in terms of significance. Population Structure (BAPS) software increasing spawning escapement to Factors to consider in determining models. The alewife stock complexes 400,000 adults (DFO, 2007). Elsewhere, whether a discrete population segment identified were: (1) Northern New river herring runs have been monitored is significant to the species to which it England; (2) Southern New England; (3)

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48949 Connecticut River; (4) Mid-Atlantic; and blueback herring dataset; and (5) of basin stocking for the purpose of (5) North Carolina. For blueback hybridization may be occurring between enhancement, recolonization of herring, no optimum solution was alewife and blueback herring and may extirpated populations, and stock reached using STRUCTURE, while influence the results of the species- introduction. Alewife stocking in Maine BAPS suggested four genetically specific analyses. dates back at least to 1803 when identifiable stock complexes. The stock Following the Stock Structure alewives were reportedly moved from complexes identified for blueback Workshop, additional analyses were run the Pemaquid and St. George Rivers to herring were: (1) Northern New on the alewife dataset to examine the create a run of alewives in the England; (2) Southern New England; (3) uniqueness of the (tentatively) Damariscotta River (Atkins and Goode, Mid Atlantic; (4) and Southern. designated Connecticut River alewife 1887). These efforts were largely However, it should be noted that these stock complex. Hybrids and responsive to considerable declines in Bayesian inferences of population misidentified samples were found and alewife populations following the structure provide a minimum number of subsequently removed for this analysis, construction of dams, over exploitation genetically distinguishable groups. In and the results were refined. By and pollution. Although there has been the future, in order to better define removing these samples from the considerable alewife stocking and potential stock complexes, further tests Connecticut River alewife dataset, relocation throughout Maine, there are examining structure within designated Palkovacs et aJ. (2012, unpublished very few records documenting these stocks should be conducted using report) found that, for alewife, the efforts. In contrast, considerably less hierarchical clustering analysis and Connecticut and Hudson Rivers belong stocking of alewives has occurred in genetic tests. to the Southern New England stock. The Maritime Canada. These genetic The study also examined the effects of analyses were further refined and analyses suggest that river herring from geography and found a strong effect of Palkovacs et al. (2012, unpublished Canadian waters are genetically distinct latitude on genetic divergence, report) provided an updated map of the from Maine river herring.

suggesting a stepping stone model of alewife genetic stock complexes, All of the expert opinions we received population structure, and a strong combining the tentative North Carolina during the Stock Structure Workshop pattern of isolation by distance, where stock with the Mid-Atlantic stock. This suggested evidence of regional stock gene flow is most likely among information and analysis is complete structure exists for both alewife and neighboring spawning populations. The and is currently being prepared for blueback herring as shown by the recent preliminary results from the study publication. Thus, the refined genetic genetics data (Palkovacs et al., 2012, found significant differentiation among stock complexes for alewife in the unpublished report; Bentzen et al.,

spawning rivers for both alewife and coastal United States include Northern unpublished data). However, the blueback herring. Based on the results of New England, Southern New England, suggested boundaries of the regional their study, the authors' preliminary and the Mid-Atlantic. For blueback stock complexes differed from expert to management recommendations suggest herring, the identified genetic stocks expert. Migration and mixing patterns of that river drainage is the appropriate include Northern New England, alewives and blueback herring in the level of management for both of the Southern New England, Mid-Atlantic ocean have not been determined, though species. This inference was also and Southern (Palcovacs et al., 2012, regional stock mixing is suspected.

supported by genetic tests which were unpublished report). Therefore, the experts suggested that the conducted later. These tests suggest that Bentzen et al. (2012) implemented a ocean phase of alewives and blueback there is substantial population structure two-part genetic analysis of river herring herring should be considered a mixed at the drainage scale. to evaluate the genetic diversity of stock until further tagging and genetic The authors noted a number of alewives in Maine and Maritime data become available. There is caveats for their study including: (1) Canada, and to assess the regional evidence to support regional differences Collection of specimens on their effects of stocking on alewives and in migration patterns, but not at a level upstream spawning run may pool blueback herring in Maine. The genetic of river-specific stocks.

samples from what are truly distinct analysis of alewives and blueback In the mid-1980s, Rulifson et al.

spawning populations within the major herring along mid-coast Maine revealed (1987) tagged and released river drainages sampled, thereby, significant genetic differentiation among approximately 19,000 river herring in underestimating genetic structure populations. Despite significant the upper Bay of Fundy, Nova Scotia within rivers (Hasselman, 2010); (2) a differentiation, the patterns of with an overall recapture rate of 0.39 more detailed analysis of population correlation did not closely correspond percent. Alewife tag returns were from structure within the major stocks with geography or drainage affiliation. freshwater locations in Nova Scotia, and identified (i.e., using hierarchical The genetic analysis of alewives from marine locations in Nova Scotia and Bayesian clustering methods and genic rivers in Maine and Atlantic Canada Massachusetts. Blueback herring tag test) would be useful for identifying any detected isolation by distance, returns were from freshwater locations substructure within these major stocks; suggesting that homing behavior in Maryland and North Carolina and (3) neutral genetic markers used in this indicative of alewives' metapopulation marine locations in Nova Scotia.

study represent the effects of gene flow conformance does produce genetically Rulifson et al. (1987) suspected from and historical population isolation, but distinguishable populations. Further recapture data that alewives and not the effects of adaptive processes, testing also suggested that there may be blueback herring tagged in the Bay of which are important to consider in the interbreeding between alewives and Fundy were of different origins, context of stock identification; (4) the blueback herring (e.g., hybrids), hypothesizing that alewives were likely analysis is preliminary, and there are a especially at sample sites with regional fish from as far away as New number of issues that need to be further impassible dams. England, while the blueback herring investigated, including the effect of The unusual genetic groupings of recaptures were likely not regional fish, deviations in the Hardy-Weinberg river herring in Maine are likely a result but those of U.S. origin from the mid-Equilibrium model encountered in four of Maine's complex stocking history, as Atlantic region. However, the low tag alewife loci and the failure of alewife populations in Maine have been return numbers (n = 2) made it difficult STRUCTURE to perform well on the subject to considerable within and out to generalize about the natal rivers of

48950 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices blueback herring caught in the Bay of alewife and blueback herring stocks species that interact, recolonize vacant Fundy. The results of this tagging study within U.S. rivers using 15 neutral loci habitats, and occupy new habitats show that river herring present in and documented that there are at least through dispersal mechanisms (Hanski Canadian waters may originate from three stock complexes of alewife in the and Gilpin, 1991)).

U.S. waters and vice versa. United States and four stock complexes Metapopulations of river herring are DPS Determination of blueback herring in the United States.

believed to exist, with adults frequently Palkovac et al. (2012, unpublished Evidence for genetic differentiation returning to their natal rivers for report) showed a strong effect of latitude exists for both alewife and blueback spawning and some straying occurring on genetic divergence, suggesting that herring, allowing for preliminary between rivers-straying rates have although most populations are identification of stock complexes; been estimated up to 20 percent (Jones, genetically differentiated, gene flow is however, available data are lacking on 2006; ASMFC, 2009a; Gahagan et al., greater among neighboring runs than the significance of each of these 2012). Given the available information among distant runs. The genetic data are individual stock complexes. Therefore, on genetic differentiation coast-wide for consistent with the recent results of the we have determined that there is not alewife and blueback herring, it appears ASMFC stock assessment (2012), which enough evidence to suggest that the that stock complexes exist for both noted that even among rivers within the stock complexes identified through species. same state, there are differences in genetics should be treated under the River herring originating from trends in abundance indices, size-at-age, DPS policy as separate DPSs. The stock Canadian rivers are delimited by age structure and other metrics, complexes may be discrete, but under international governmental boundaries. indicating there are localized factors the DPS policy, they are not significant Differences in control of exploitation, affecting the population dynamics of to the species as a whole. Furthermore, management of habitat, conservation both species. given the unknown level of intermixing status, or regulatory mechanisms exist Neutral genetic markers such as between Canadian and U.S. river and, therefore, meet the discreteness microsatellites have a longstanding herring in coastal waters, the Canadian criterion under the DPS policy; history of utilization in stock stock complex should also not be however, intermixing between both designation for many anadromous fish considered separately under the DPS alewife and blueback herring from U.S. species (Waples, 1998). However, these policy.

and Canadian coastal waters occurs, and markers represent the effects of gene Throughout the rest of this the extent of this mixing is unknown. flow and historical population isolation determination, the species will be Given the best available information, and not the effects of adaptive referred to by species (alewife or it is possible to determine that the processes. The effects of adaptive blueback herring), as river herring various stocks of both alewife and genetic and phenotypic diversity are where information overlaps, and by the blueback herring are discrete. The best also extremely important to consider in identified stock complexes (Palkovacs et available information suggests that the the context of stock designation, but are al., 2012, unpublished report) for each delineation of the stock complexes is as not captured by the use of neutral species as necessary. While the described above; however, future work genetic markers. Therefore, the available individual stock complexes do not will likely further refine these genetic data are most appropriately used constitute separate DPSs, they are preliminary boundaries. Additionally, in support of the discreteness criterion, important components of the overall further information is needed on the rather than to determine significance. species and relevant to the evaluation of oceanic migratory patterns of both Determining whether a gap in the whether either species may be species. range of the taxon would be significant threatened or endangered in a if a stock were extirpated is difficult to significant portion of their overall range.

Information Related to Significance determine with anadromous fish such as Therefore, we have evaluated the threats If a population is deemed discrete, the river herring. River herring are to, and extinction risk of the overall population is evaluated in terms of suspected to migrate great distances species and each of the individual stock significance. Significance can be between their natal rivers and complexes as presented below. For this determined using the four criteria noted overwintering areas, and therefore, analysis, the identified stock complexes above. Since the best available estuarine and marine populations are for alewife (Figure 1) in the coastal information indicates that the stock comprised of mixed stocks. United States for the purposes of this complexes identified for alewives and Consequently, the loss of a stock finding will include Northern New blueback herring are most likely complex would mean the loss of England, Southern New England, the discrete, the SRT reviewed the available riverine spawning subpopulations, Mid-Atlantic, and Canada; and stock information to determine if they are while the marine and estuarine habitat complexes for blueback herring (Figure significant. would most likely still be occupied by 2) will include Northern New England, In evaluating the significance migratory river herring from other stock Southern New England, Mid-Atlantic, criterion, the SRT considered all of the complexes. As it has been shown that Southern Atlantic, and Canada. While above criteria. As indicated earlier, both gene flow is greater among neighboring the SRT concluded that there was not alewives and blueback herring occupy a runs than among distant runs, we might sufficient information at this time to large range spanning almost the entire expect that river herring would re- determine with any certainty whether East Coast of the United States and into colonize neighboring systems over a alewife or blueback herring stock Canada. They appear to migrate freely relatively short time frame. Thus, the complexes constitute separate DPSs, throughout their oceanic range and loss of one stock complex in itself may they recognized that future information return to freshwater habitats to spawn in not be significant; the loss of contiguous on behavior, ecology and genetic streams, lakes and rivers. Therefore, stock complexes may be. The goal then population structure may reveal they occupy many different ecological for river herring stock complexes is to significant differences, showing fish to settings throughout their range. maintain connectivity between genetic be uniquely adapted to each stock As described earlier, the Palkovacs et groups to support proper complex. We agree with this conclusion.

al. (2012, unpublished report) study metapopulation function (spatially Thus, we are not identifying DPSs for assessed the genetic composition of separated populations of the same either species.

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48951 Figure 1. Alewife stock structure identified in Palkovacs et al., 2012, unpublished report.

48952 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Stock Complex..,

• Northern New England A Southern New England Mid-Altantic Southern 0": 90 *180 540 A~ 60 ~Kilom~eters Figure 2. Blueback herring stock structure identified in Palkovacs et al., 2012, unpublished report.

Foreseeable Future and Significant endangered species within the December 9, 2011). The draft policy Portion of Its Range foreseeable future throughout all or a provides that: (1) If a species is found significant portion of its range." NMFS to be endangered or threatened in only The ESA defines an "endangered and the U.S. Fish and Wildlife Servce a significant portion of its range, the species" as "any species which is in (USFWS) recently published a draft entire species is listed as endangered or danger of extinction throughout all or a policy to clarify the interpretation of the threatened, respectively, and the ESA's significant portion of its range," while a phrase "significant portion of the range" protections apply across the species' "threatened species" is defined as "any in the ESA definitions of "threatened" entire range; (2) a portion of the range species which is likely to become an and "endangered" (76 FR 76987; of a species is "significant" if its

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48953 contribution to the viability of the also evaluated the threat from climate educational purposes; (C) disease or species is so important that, without change from 2060 to 2100 and climate predation; (D) inadequacy of existing that portion, the species would be in variability in the near term (as described regulatory mechanisms; and (E) other danger of extinction; (3) the range of a in detail below). natural or man-made factors affecting species is considered to be the general Highly productive species with short the species' continued existence. This geographical area within which that generation times are more resilient than section briefly summarizes the findings species can be found at the time USFWS less productive, long lived species, as regarding these factors.

or NMFS makes any particular status they are quickly able to take advantage determination; and (4) if the species is of available habitats for reproduction A. The Present or Threatened not endangered or threatened (Mace et al., 2002). Species with shorter Destruction, Modification, or throughout all of its range, but it is generation times, such as river herring Curtailmentof Its Habitator Range endangered or threatened within a (4 to 6 years), experience greater Past, present, and reasonably significant portion of its range, and the population variability than species with foreseeable future factors that have the population in that significant portion is long generation times, because they potential to affect river herring habitat a valid DPS, we will list the DPS rather maintain the capacity to replenish include, but are not limited to, dams than the entire taxonomic species or themselves more quickly following a and hydropower facilities, dredging, subspecies. period of low survival (Mace et 0l., water quality (including land use The Services are currently reviewing 2002). Given the high population change, water withdrawals, discharge public comment received on the draft variability among clupeids, projecting and contaminants), climate change and policy. While the Services' intent is to out further than three generations could climate variability. As noted above, establish a legally binding interpretation lead to considerable uncertainty in the river herring occupy a variety of of the term "significant portion of the probability that the model will provide different habitats including freshwater, range," the draft policy does not have an accurate representation of the estuarine and marine environments legal effect until such time as it may be population trajectory for each species. throughout their lives, and thus, they adopted as final policy. Here, we apply Thus, a 12 to 18 year timeframe (e.g., are subjected to habitat impacts the principles of this draft policy as 2024-2030), or a three-generation time occurring in all of these different non-binding guidance in evaluating period, for each species was determined habitats.

whether to list alewife or blueback by the Team to be appropriate for use Dams and Other Barriers herring under the ESA. If the policy as the foreseeable future for both alewife changes in a material way, we will and blueback herring. We agree with the Dams and other barriers to upstream revisit the determination and assess Team that a three-generation time and downstream passage (e.g., culverts) whether the final policy would result in period (12-18 years) is a reasonable can block or impede access to habitats a different outcome. foreseeable future for both alewife and necessary for spawning and rearing; can While we have determined that DPSs blueback herring. cause direct and indirect mortality from cannot be defined for either of these Connectivity, population resilience injuries incurred while passing over species based on the available and diversity are important when dams, through downstream passage information, the stock complexes do determining what constitutes a facilities, or through hydropower represent important groupings within significant portion of the species' range turbines; and can degrade habitat the range of both species. Thus, in our (Waples et al., 2007). Maintaining features necessary to support essential analysis of extinction risk and threats connectivity between genetic groups river herring life history functions. Man-assessment below, we have evaluated supports proper metapopulation made barriers that block or impede whether either species is at risk function, in this case, anadromy. access to rivers throughout the entire rangewide and within any of the Ensuring that river herring populations historical range of river herring have individual stock complexes so that we are well represented across diverse resulted in significant losses of can evaluate whether either species is habitats helps to maintain and enhance historical spawning habitat for river threatened or endangered in a genetic variability and population herring. Dams and other man-made significant ortion of its range. resilience (McElhany et al., 2000). barriers have contributed to the We estabflished that the appropriate Additionally, ensuring wide geographic historical and current declines in period of time corresponding to the distribution across diverse climate and abundance of both blueback and alewife foreseeable future is a function of the geographic regions helps to minimize populations. While estimates of habitat particular type of threats, the life-history risk from catastrophes (e.g., droughts, loss over the entire range of river characteristics, and the specific habitat floods, hurricanes, etc.; McElhany et al., herring are not available, estimates from requirements for river herring. The 2000). Furthermore, preventing isolation studies in Maine show that less than 5 timeframe established for the of genetic groups protects against percent of lake spawning habitat and 20 foreseeable future takes into account the population divergence (Allendorf and percent of river habitat remains time necessary to provide for the Luikart, 2007). accessible for river herring (Hall et al.,

conservation and recovery of each 2010). As described in more detail species and the ecosystems upon which Threats Evaluation below, dams are also known to impact they depend, but is also a function of As described above, Section 4(a)(1) of river herring through various the reliability of available data regarding the ESA and NMFS implementing mechanisms, such as habitat alteration, the identified threats and extends only regulations (50 CFR 424) states that we fish passage delays, and entrainment as far as the data allow for making must determine whether a species is and impingement (Ruggles 1980; NRC reasonable predictions about the endangered or threatened because of 2004). River herring can undergo species' response to those threats. As any one or a combination of the indirect mortality from injuries such as described below, the SRT determined following factors: (A) Current or scale loss, lacerations, bruising, eye or that dams and other impediments to threatened habitat destruction or fin damage, or internal hemorrhaging migration have already created a clear modification or curtailment of habitat or when passing through turbines, over and present threat to river herring that range; (B) overutilization for spillways, and through bypasses will continue into the future. The SRT commercial, recreational, scientific, or (Amaral et al., 2012).

48954 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices The following summary of the effects where higher revolution, Francis-type Such conditions have occurred along of dams and other barriers on river runners were used (RMC, 1992; ASMFC, the Susquehanna River at the herring is taken from Amendment 2 to 2009). Conowingo Dam, Maryland, from late the Interstate Fishery Management Plan Additional studies reported that spring through early fall, and have for Shad and River Herring (hereafter, changes in pressure had a more historically caused large fish kills below referred to as "Amendment 2" and cited pronounced effect on juveniles with the dam (Krauthamer and Richkus, as "ASMFC, 2009"). Because it includes thinner and weaker tissues as they 1987; ASMFC, 2009).

a detailed description of barriers to moved through turbines (Taylor and Disruption of seasonal flow rates in upstream and downstream passage, it is Kynard, 1984). Furthermore, some fish rivers can impact upstream and the best source of comprehensive may die later from stress, or become downstream migration patterns for adult information on this topic. Please refer to weakened and more susceptible to and juvenile alosines (ASMFC, 1985; Amendment 2 for more information. predation, and as such, losses may not Limburg, 1996; ASMFC, 1999; USFWS Dams and spillways impeding rivers be immediately apparent to researchers et al., 2001). Changes to natural flows along the East Coast of the United States (Gloss, 1982) (ASMFC, 2009). can also disrupt natural productivity have resulted in a considerable loss of Changes to the river system, resulting and availability of zooplankton that historical spawning habitat for shad and in delayed migration among other larval and early juvenile alosines feed river herring. Permanent man-made things, were also identified in on (Crecco and Savoy, 1987; Limburg, structures pose an ongoing barrier to Amendment 2 as impacting river 1996; ASMFC, 2009).

fish passage unless fishways are herring. Amendment 2 notes that when Although most dams that impact installed or structures are removed. juvenile alosines delay out-migration, diadromous fish are located along the Low-head dams can also pose a they may concentrate behind dams and lengths of rivers, fish can also be problem, as fish are unable to pass over become more susceptible to actively affected by hydroelectric projects at the them except when tides or river feeding predators. They may also be mouths of rivers, such as the large tidal discharges are exceptionally high more vulnerable to anglers that target hydroelectric project at the Annapolis (Loesch and Atran, 1994). Historically, alosines as a source of bait. Delayed out- River in the Bay of Fundy, Canada. This major dams were often constructed at migration can also make juvenile particular basin and other surrounding the site of natural formations conducive alosines more susceptible to marine waters are used as foraging areas during to waterpower, such as natural falls. predators that they may have avoided if summer months by American shad from Diversion of water away from rapids at they had followed their natural all runs along the East Coast of the the base of falls can reduce fish habitat, migration patterns (McCord, 2005a). In United States (Dadswell et al., 1983).

and in some cases cause rivers to run open rivers, juvenile alosines gradually Because the facilities are tidal dry at the base for much of the summer move seaward in groups that are likely hydroelectric projects, fish may move in (MEOEA, 2005; ASMFC, 2009). spaced according to the spatial and out of the impacted areas with each Prior to the early 1990s, it was separation of spawning and nursery tidal cycle. While turbine mortality is thought that migrating shad and river grounds (Limburg, 1996; J. McCord, relatively low with each passage, the herring suffered significant mortality South Carolina Department of Natural repeated passage in and out of these going through turbines during Resources, personal observation). facilities may cumulatively result in downstream passage (Mathur and Releasing water from dams and substantial overall mortalities (Scarratt Heisey, 1992). Juvenile shad emigrating impoundments (or reservoirs) may lead and Dadswell, 1983; ASMFC, 2009).

from rivers have been found to to flow alterations, altered sediment Additional man-made structures that accumulate in larger numbers near the transport, disruption of nutrient may obstruct upstream passage include:

forebay of hydroelectric facilities, where availability, changes in downstream tidal and amenity barrages (barriers they become entrained in intake flow water quality (including both reduced constructed to alter tidal flow for areas (Martin et a0., 1994). Relatively and increased temperatures), aesthetic purposes or to harness energy);

high mortality rates were reported (62 streambank erosion, concentration of tidal flaps (used to control tidal flow);

percent to 82 percent) at a hydroelectric sediment and pollutants, changes in mill, gauging, amenity, navigation, dam for juvenile American shad and species composition, solubilization of diversion, and water intake weirs; fish blueback herring, depending on the iron and manganese and their absorbed counting structures; and earthen berms power generation levels tested (Taylor or chelated ions, and hydrogen sulfide (Durkas, 1992; Solomon and Beach, and Kynard, 1984). In contrast, Mathur in hypolimnetic (water at low level 2004). The impact of these structures is and Heisey (1992) reported a mortality outlets) releases (Yeager, 1995; Erkan, site-specific and will vary with a rate of 0 percent to 3 percent for 2002; ASMFC, 2009). number of conditions including head juvenile American shad (2 to 6 in fork Many dams spill water over the top of drop, form of the structure, length (55 to 140 mm)), and 4 percent the structure where water temperatures hydrodynamic conditions upstream and for juvenile blueback herring (3 to 4 in are the warmest, essentially creating a downstream, condition of the structure, fork length (77 to 105 mm)) through series of warm water ponds in place of and presence of edge effects (Solomon Kaplan turbines. Mortality rate the natural stream channel (Erkan, and Beach, 2004). Road culverts are also increased to 11 percent in passage 2002). Conversely, water released from a significant source of blockage.

through a low-head Francis turbine deep reservoirs may be poorly Culverts are popular, low-cost (Mathur and Heisey, 1992). Other oxygenated, at below-normal seasonal alternatives to bridges when roads must studies reported less than 5 percent water temperature, or both, thereby cross small streams and creeks.

mortality when large Kaplan and fixed- causing loss of suitable spawning or Although the amount of habitat affected blade, mixed-flow turbines were used at nursery habitat in otherwise habitable by an individual culvert may be small, a facility along the Susquehanna River areas (ASMFC, 2009). the cumulative impact of multiple (RMC, 1990; RMC, 1994). At the same Reducing minimum flows can reduce culverts within a watershed can be site, using small Kaplan and Francis the amount of water available and cause substantial (Collier and Odom, 1989; runners, the mortality rate was as high increased water temperature or reduced ASMFC, 2009).

as 22 percent (NA, 2001). At another dissolved oxygen levels (ASMFC, 1985; Roads and culverts can also impose site, mortality rate was about 15 percent ASMFC, 1999; USFWS et al., 2001). significant changes in water quality.

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48955 Winter runoff in some states may to river herring habitat. The following channelization of Grindle Creek, North include high concentrations of road salt, section, taken from Amendment 2, Carolina removed in-creek vegetation while stormwater flows in the summer describes these threats. and woody debris, which had served as may cause thermal stress and bring high Channelization can cause significant substrate for fertilized eggs (ASMFC, concentrations of other pollutants environmental impacts (Simpson et al., 2009).

(MEOEA, 2005; ASMFC, 2009). 1982; Brookes, 1988), including bank Channelization can also reduce the Sampled sites in North Carolina erosion, elevated water velocity, amount of pool and riffle habitat revealed river herring upstream and reduced habitat diversity, increased (Hubbard, 1993), which is an important downstream of bridge crossings, but no drainage, and poor water quality food-producing area for larvae (Keller, herring were found in upstream sections (Hubbard, 1993). Dredging and disposal 1978; Wesche, 1985; ASMFC, 2009).

of streams with culverts. Additional of spoils along the shoreline can also Dredging can negatively affect alosine study is underway to determine if river create spoil banks, which block access populations by producing suspended herring are absent from these areas to sloughs, pools, adjacent vegetated sediments (Reine et al., 1998), and because of the culverts (NCDENR, 2000). areas, and backwater swamps migrating alosines are known to avoid Even structures only 8 to 12 in (20 to 30 (Frankensteen, 1976). Dredging may also waters of high sediment load (ASMFC, cm) above the water can block shad and release contaminants, resulting in 1985; Reine et al., 1998). Fish may also river herring migration (ASMFC, 1999; bioaccumulation, direct toxicity to avoid areas that are being dredged ASMFC, 2009). aquatic organisms, or reduced dissolved because of suspended sediment in the Rivers can also be blocked by non- oxygen levels (Morton, 1977). water column. Filter-feeding fishes, anthropogenic barriers, such as beaver Furthermore, careless land use practices such as alosines, can be negatively dams, waterfalls, log piles, and may lead to erosion, which can lead to impacted by suspended sediments on vegetative debris. These blockages may high concentrations of suspended solids gill tissues (Cronin et al., 1970).

hinder migration, but they can also (turbidity) and substrate (siltation) in Suspended sediments can clog gills that benefit by providing adhesion sites for the water following normal and intense provide oxygen, resulting in lethal and eggs, protective cover, and feeding sites rainfall events. This can displace larvae sub-lethal effects to fish (Sherk et al.,

(Klauda et al., 1991b). Successful and juveniles to less desirable areas 1974 and 1975; ASMFC, 2009).

passage at these natural barriers often downstream and cause osmotic stress Nursery areas along the shorelines of depends on individual stream flow (Klauda et al., 1991b; ASMFC, 2009). the rivers in North Carolina have been characteristics during the fish migration Spoil banks are often unsuitable affected by dredging and filling, as well season (ASMFC, 2009). habitat for fishes. Suitable habitat is as by erection of bulkheads; however, often lost when dredge disposal material the degree of impact has not been Dredging is placed on natural sand bars and/or measured. In some areas, juvenile Wetlands provide migratory corridors point bars. The spoil is too unstable to alosines were unable to enter and spawning habitat for river herring. provide good habitat for the food chain. channelized sections of a stream due to The combination of incremental losses Draining and filling, or both, of high water velocities caused by of wetland habitat, changes in wetlands adjacent to rivers and creeks dredging (ASMFC, 2000 and 2009).

hydrology, and nutrient and chemical in which alosines spawn has eliminated inputs over time, can be extremely Water Quality spawning areas in North Carolina harmful, resulting in diseases and (NCDENR, 2000; ASMFC, 2009). Nutrient enrichment has become a declines in the abundance and quality. Secondary impacts from channel major cumulative problem for many Wetland loss is a cumulative impact formation include loss of vegetation and coastal waters. Nutrient loading results that results from activities related to debris, which can reduce habitat for from the individual activities of coastal dredging/dredge spoil placement, port invertebrates and result in reduced development, marinas and recreational development, marinas, solid waste quantity and diversity of prey for boating, sewage treatment and disposal, disposal, ocean disposal, and marine juveniles (Frankensteen, 1976). industrial wastewater and solid waste mining. In the late 1970s and early Additionally, stream channelization disposal, ocean disposal, agriculture, 1980s, the United States was losing often leads to altered substrate in the and aquaculture. Excess nutrients from wetlands at an estimated rate of 300,000 riverbed and increased sedimentation land based activities accumulate in the acres (1,214 sq km) per year. The Clean (Hubbard, 1993), which in turn can soil, pollute the atmosphere, pollute Water Act and state wetland protection reduce the diversity, density, and ground water, or move into streams and programs helped decrease wetland species richness of aquatic insects coastal waters. Nutrient inputs are losses to 117,000 acres (473 sq km) per (Chutter, 1969; Gammon, 1970; Taylor, known to have a direct effect on water year, between 1985 and 1995. Estimates 1977). Suspended sediments can reduce quality. For example, nutrient of wetlands loss vary according to the feeding success in larval or juvenile enrichment can stimulate growth of different agencies. The U.S. Department fishes that rely on visual cues for phytoplankton that consumes oxygen of Agriculture (USDA) attributes 57 plankton feeding (Kortschal et al., when they decay, which can lead to low percent of wetland loss to development, 1991). Sediment re-suspension from dissolved oxygen that may result in fish 20 percent to agriculture, 13 percent to dredging can also deplete dissolved kills (Correll, 1987; Tuttle et al., 1987; the creation of deepwater habitat, and oxygen, and increase bioavailability of Klauda et al., 1991b); this condition is 10 percent to forest land, rangeland, and any contaminants that may be bound to known as eutrophication.

other uses. Of the wetlands lost between the sediments (Clark and Wilber, 2000; In addition to the direct cumulative 1985 and 1995, the USFWS estimates ASMFC, 2009). effects incurred by development that 79 percent of wetlands were lost to Migrating adult river herring avoid activities, inshore and coastal habitats upland agriculture. Urban development channelized areas with increased water are also threatened by persistent and other types of land use activities velocities. Several channelized creeks in increases in certain chemical were responsible for 6 percent and 15 the Neuse River basin in North Carolina discharges. The combination of percent of wetland loss, respectively. have reduced river herring distribution incremental losses of wetland habitat, Amendment 2 identifies and spawning areas (Hawkins, 1979). changes in hydrology, and nutrient and channelization and dredging as a threat Frankensteen (1976) found that the chemical inputs produced over time can

48956 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices be extremely harmful to marine and adhesive eggs to adhere to substrates While juvenile mortality rates are estuarine biota, including river herring, (Mansueti, 1962; ASMFC, 2009). generally low at well-screened facilities, resulting in diseases and declines in the From the 1950s to the present, large numbers of juveniles can be abundance and quality of the affected increased nutrient loading has made entrained (Hauck and Edson, 1976; resources. hypoxic conditions more prevalent Robbins and Mathur, 1976; ASMFC, Amendment 2 identified land use (Officer et al., 1984; Mackiernan, 1987; 2009).

changes including agriculture, logging/ Jordan et al., 1992; Kemp et al., 1992; Fish impinged against water filtration forestry, urbanization and non-point Cooper and Brush, 1993; Secor and screens can die from asphyxiation, source pollution as threats to river Gunderson, 1998). Hypoxia is most exhaustion, removal from the water for herring habitat. The following section, likely caused by eutrophication, due prolonged periods of time, removal of taken from Amendment 2, describes mostly to non-point source pollution protective mucous, and descaling (DBC, these threats. (e.g., industrial fertilizers used in 1980). Studies conducted along the The effects of land use and land cover agriculture) and point source pollution Connecticut River found that larvae and on water quality, stream morphology, (e.g., urban sewage). early juveniles of alewife, blueback and flow regimes are numerous, and Logging activities can modify herring, and American shad suffered may be the most important factors hydrologic balances and in-stream flow 100-percent mortality when determining quantity and quality of patterns, create obstructions, modify temperatures in the cooling system of a aquatic habitats (Boger, 2002). Studies temperature regimes, and add nutrients, power plant were elevated above 82 'F have shown that land use influences sediments, and toxic substances into (280 C); 80 percent of the total mortality dissolved oxygen (Limburg and river systems. Loss of riparian was caused by mechanical damage, 20 vegetation can result in fewer refuge percent by heat shock (Marcy, 1976).

Schmidt, 1990), sediments and turbidity areas for fish from fallen trees, fewer Ninety-five percent of the fish near the (Comeleo et al., 1996; Basnyat et al.,

1999), water temperature (Hartman et insects for fish to feed on, and reduced intake were not captured by the screen, aL., 1996; Mitchell, 1999), pH (Osborne shade along the river, which can lead to and Marcy (1976) concluded that it did increased water temperatures and not seem possible to screen fish larvae and Wiley, 1988; Schofield, 1992), reduced dissolved oxygen (EDF, 2003).

nutrients (Peterjohn and Correll, 1984; effectively (ASMFC, 2009).

Threats from deforestation of swamp The physical characteristics of Osborne and Wiley, 1988; Basnyat et aL., forests include: siltation from increased streams (e.g., stream width, depth, and 1999), and flow regime (Johnston et al.,

erosion and runoff; decreased dissolved current velocity; substrate; and 1990; Webster et al., 1992; ASMFC, oxygen (Lockaby et al., 1997); and temperature) can be altered by water 2009). disturbance of food-web relationships in withdrawals (Zale et al., 1993). River Siltation, caused by erosion due to adjacent and downstream waterways herring can experience thermal stress, land use practices, can kill submerged (Batzer et al., 2005; ASMFC, 2009). direct mortality, or indirect mortality aquatic vegetation (SAV). SAV can be Urbanization can cause elevated when water is not released during times adversely affected by suspended concentrations of nutrients, organics, or of low river flows and water sediment concentrations of less than 15 sediment metals in streams (Wilber and temperatures are higher than normal.

ppm (15 mg/L) (Funderburk et al., 1991) Hunter, 1977; Kelly and Hite, 1984; Water flow disruption can also result in and by deposition of excessive Lenat and Crawford, 1994). More less freshwater input to estuaries sediments (Valdes-Murtha and Price, research is needed on how urbanization (Rulifson, 1994), which are important 1998). SAV is important because it affects diadromous fish populations; nursery areas for river herring and other improves water quality (Carter et al., however, Limburg and Schmidt (1990) anadromous species (ASMFC, 2009).

1991). SAV consumes nutrients in the found that when the percent of Industrial discharges may contain water and as the plants die and decay, urbanized land increased to about 10 toxic chemicals, such as heavy metals they slowly release the nutrients back percent of the watershed, the number of and various organic chemicals (e.g.,

into the water column. Additionally, alewife eggs and larvae decreased insecticides, solvents, herbicides) that through primary production and significantly in tributaries of the are harmful to aquatic life (ASMFC, respiration, SAV affects the dissolved Hudson River, New York (ASMFC, 1999). Many contaminants can have oxygen and carbon dioxide 2009). harmful effects on fish, including concentrations, alkalinity, and pH of the reproductive impairment (Safe, 1990; waterbody. SAV beds also bind Water Withdrawal/Outfall Mac and Edsall, 1991; Longwell et al.,

sediments to the bottom resulting in Water withdrawal facilities and toxic 1992). Chemicals and heavy metals can increased water clarity, and they and thermal discharges have also been move through the food chain, producing provide refuge habitat for migratory. fish identified as impacting river herring, sub-lethal effects such as behavioral and and planktonic prey items (Maldeis, and the following section is summarized reproductive abnormalities (Matthews et 1978; Monk, 1988; Killgore et al., 1989; from Amendment 2. al., 1980). In fish, exposure to ASMFC, 2009). Large volume water withdrawals (e.g., polychlorinated biphenyls (PCBs) can Decreased water quality from drinking water, pumped-storage cause fin erosion, epidermal lesions, sedimentation became a problem with hydroelectric projects, irrigation, and blood anemia, altered immune response, the advent of land-clearing agriculture snow-making) can alter local current and egg mortality (Post, 1987; Kennish in the late 18th century (McBride, 2006). characteristics (e.g., reverse river flow), et al., 1992). Steam power plants that Agricultural practices can lead to which can result in delayed movement use chlorine to prevent bacterial, fungal, sedimentation in streams, riparian past a facility or entrainment in water and algal growth present a hazard to all vegetation loss, influx of nutrients (e.g., intakes (Layzer and O'Leary, 1978). aquatic life in the receiving stream, even inorganic fertilizers and animal wastes), Planktonic eggs and larvae entrained at at low concentrations (Miller et al.,

and flow modification (Fajen and water withdrawal projects experience 1982; ASMFC, 2009).

Layzer, 1993). Agriculture, silviculture, high mortality rates due to pressure Pulp mill effluent and other oxygen-and other land use practices can lead to changes, shear and mechanical stresses, consuming wastes discharged into rivers sedimentation, which reduces the and heat shock (Carlson and McCann, and streams can reduce dissolved ability of semi-buoyant eggs and 1969; Marcy, 1973; Morgan et al., 1976). oxygen concentrations below what is

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48957 required for river herring survival. Low decrease the quantity of both spawning North Atlantic Oscillation, and the El dissolved oxygen resulting from and nursery habitat for anadromous Nifio Southern Oscillation. During the industrial pollution and sewage fish. Reduced streamflow can reduce workshop, it was noted that impacts discharge can also delay or prevent water quality by concentrating from global climate change induced by upstream and downstream migrations. pollutants and/or increasing water human activities are likely to become Everett (1983) found that during times temperature (ASMFC, 1985). O'Connell more apparent in future years of low water flow when pulp mill and Angermeier (1999) found that in (Intergovernmental Panel on Climate effluent comprised a large percentage of some Virginia streams, there was an Change (IPCC), 2007). Results presented the flow, river herring avoided the inverse relationship between the from the North American Regional effluent. Pollution may be diluted in the proportion of a stream's watershed that Climate Change Assessment Program fall when water flows increase, but fish was agriculturally developed and the (NARCCAP-a group that uses fields that reach the polluted waters overall tendency of the stream to from the global climate models to downriver before the water has flushed support river herring runs. In North provide boundary conditions for the area will typically succumb to Carolina, cropland alteration along regional atmospheric models covering suffocation (Miller et aI., 1982; ASMFC, several creeks and rivers significantly most of North America and extending 2009). reduced river herring distribution and over the adjacent oceans) suggest that Effluent may also pose a greater threat spawning areas in the Neuse River basin temperature will warm throughout the during times of drought. Such (Hawkins, 1979; ASMFC, 2009). years over the northeast, mid-Atlantic conditions were suspected of interfering Atmospheric deposition occurs when and Southeast United States (comparing with the herring migration along the pollutants (e.g. nitrates, sulfates, 1968-1999 to 2038-2069; NMFS, Chowan River, North Carolina, in 1981. ammonium, and mercury) are 2012b). Additionally, it was noted that In the years before 1981, the effluent transferred from the air to the earth's there is an expected but less certain from the pulp mill had passed prior to surface. Pollutants can get from the air increase in precipitation over the the river herring run, but drought into the water through rain and snow, northeast United States during fall and conditions caused the effluent to remain falling particles, and absorption of the winter during the same years (NMFS, in the system longer that year. Toxic gas form of the pollutants into the water. 2012b). In conjunction with increased effects were indicated, and researchers Atmospheric pollutants can result in evaporation from warmer temperatures, suggested that growth and reproduction increased eutrophication (Paerl et al., the Northeast and mid-Atlantic may might have been disrupted as a result of 1999) and acidification of surface waters experience decrease in runoff and eutrophication and other factors (Haines, 1981). Atmospheric nitrogen decreased stream flow in late winter and (Winslow et al., 1983; ASMFC, 2009). deposition in coastal estuaries can lead early spring (NMFS, 2012b).

Klauda et al. (1991a) provides an to accelerated algal production (or Additionally, enhanced ocean extensive review of temperature eutrophication) and water quality thresholds for alewife and bluback stratification could be caused by greater declines (e.g., hypoxia, toxicity, and fish warming at the ocean surface than at herring. In summary, the spawning kills) (Paerl et al., 1999). Nitrate and migration for alewives most often occurs depth (NMFS, 2012b).

sulfate deposition is acidic and can when water temperatures range from reduce stream pH (measure of the Many observed changes in river 50-64 0F (10-18 °C), and for bluebacks hydronium ion concentration) and herring biology related to environmental when temperatures range from 57-77 0F elevate toxic forms of aluminum conditions were noted at the workshop, (14-25 'CQ. Alewife egg deposition most (Haines, 1981). When pH declines, the but few detailed analyses were available often occurs when temperatures range normal ionic salt balance of the fish is to distinguish climate change from between 50-72 OF (10 and 22 0C), and compromised and fish lose body salts to climate variability. One analysis by for bluebacks when temperatures range the surrounding water (Southerland et Massachusetts Division of Marine between 70-77 OF (21 and 25 0C). al., 1997). Sensitive fish species can Fisheries showed precipitation effects Alewife egg and larval development is experience acute mortality, reduced on spawning run recruitment at optimal when temperatures range from growth, skeletal deformities, and Monument River, MA (1980-2012; 63-70 OF (17-21 °C), and for bluebacks reproductive failure (Haines, 1981). NMFS, 2012b). Jordaan and Kritzer when temperatures range from 68-75 OF (unpublished data) showed normalized (20-24 'C) (temperature ranges were Climate Change and Climate Variability run counts of alewife and blueback also presented and discussed at the Possible climate change impacts to herring have a stronger correlation with Climate Workshop (NMFS, 2012b)). river herring were noted in the stock fisheries and predators than various Thermal effluent from power plants assessment (ASMFC, 2012) based on climate variables at broad scales (NMFS, outside these temperature ranges when regional patterns in trends (e.g., trawl 2012b). Once fine-scale (flow related to river herring are present can disrupt surveys in southern regions showed fishways and dams) data were used, schooling behavior, cause declining trends more frequently results indicate that summer and fall disorientation, and may result in death. compared to those in northern regions). conditions were more important. Nye et Sewage can directly and indirectly However, additional information was al. (2012) investigated climate-related affect anadromous fish. Major needed on this topic to inform our mechanisms in the marine habitat of the phytoplankton and algal blooms that listing decision, and as noted above, we United States that may impact river reduced light penetration (Dixon, 1996) held a workshop to obtain expert herring. Their preliminary results and ultimately reduced SAV abundance opinion on the potential impacts of indicate the following: (1) A shift in (Orth et al., 1991) in tidal freshwater climate change on river herring (NMFS, northern ocean distribution for both areas of the Chesapeake Bay in the 2012b). blueback herring and alewife depending 1960s and early 1970s may have been As discussed at the workshop, both on the season; (2) decrease in ocean caused by ineffective sewage treatment natural climate variability and habitat within the preferred temperature (ASMFC, 2009). anthropogenic-forced climate change for alewife and blueback herring in the Water withdrawal for irrigation can will affect river herring (NMFS, 2012b). spring; and (3) effects of climate change cause dewatering or reduced streamflow Natural climate variability includes the on river herring populations may of freshwater streams, which can Atlantic Multidecadal Oscillation, the depend on the current condition (e.g.,

48958 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices abundance and health) of the these species are maintained, declines have shown that the climate change population, assumptions, and due to the effects of climate change will signal is readily apparent by the end of temperature tolerances (e.g., blueback be reduced. Their specific results the 21st century (Hare et al., 2010; Hare herring have a higher temperature include the following: et al., 2012). At intermediate time tolerance than alewife). e Alewife: At low population size, periods (e.g., 2024-2030), the signal of Although preliminary, Nye et a0. coast-wide abundance is projected to natural climate variability is likely (2012) indicate that climate change will decrease with less suitable habitat and similar to the signal of climate change.

impact river herring. The results (also patchy areas of high density in the Gulf Thus, a large component of the climate supported by Nye et al., 2009) indicate of Maine and Georges Bank in 2060- effect on river herring in 2024-2030 will that both blueback herring and alewife 2100. At high population size, be composed of natural climate have and will continue to shift their abundance is projected to increase variability, which could be either distribution to more northerly waters in slightly from 2020-2060 (+4.64 percent) warming or cooling.

the spring, and blueback herring has but is projected to decrease (- 39.14 percent) and become more patchy in Summary and Evaluation of Factor A also shifted its distribution to more northerly waters in the fall (1975-2010) 2060-2100. Dams and hydropower facilities, (Nye et al., 2012). Additionally, Nye et e Blueback herring: Abundance is water quality and water withdrawals projected to increase at both high and from urbanization and agricultural al. (2012) found a decrease in habitat low population size throughout the runoff, dredging and other wetland (bottom waters) within the preferred Northeast United States, especially in alterations are likely the causes of temperature for alewife and blueback the mid-Atlantic and Georges Bank. historical and recent declines in herring in the spring under future climate predictions (2020-2060 and However, at low abundance the increase abundance of alewife and blueback is minimal and remains at a level below herring populations. Climate variability 2060-2100). They concluded that an the 40-year mean. The percentage rather than climate change is expected expected decrease in optimal marine change due to climate change (factoring habitat and natal spawning habitat will to have more of an impact on river only temperature) is +29.93 percent for herring from 2024-2030 (NMFS' negatively affect river herring the time period 2020-2060 and +55.81 foreseeable future for river herring). Nye populations at the southern extent of percent from 2060-2100.

their range. Additionally, Nye et al. et al., (2012) conducted a preliminary We hoped to obtain information analysis investigating climate-related (2012) infer that this will have negative during the workshop on potential population level effects and cause mechanisms in the marine habitat of the impacts of climate change by region, United States that may impact river population declines in southern rivers, including information on species, life resulting in an observed shift in herring, and found that changes in the stage, indicators, potential impacts, and distribution which has already been amount of preferred habitat and a available data/relevant references observed. Nye et al. (2012) also found potential northward shift in distribution (NMFS, 2012b). Although we did obtain that the effects of climate change on as a result of climate change may affect information on each of these categories, river herring populations may depend river herring in the future (e.g., 2020-substantial data gaps in the species 2100). Thus, the level of threat posed by on the current condition (e.g., information were apparent (NMFS, abundance and health) of the these potential stressors is evaluated 2012b). For example, although no population, assumptions, and further in the qualitative threats specific information on impacts of assessment as described below.

temperature tolerances. Using the ocean acidification on river herring was model, projections of alewife presented, possible effects on larval B. Overutilization for Commercial, distribution and abundance can be development, chemical signaling Recreational,Scientific, or Educational predicted for each year, but for ease of (olfaction), and de-calcification of prey Purposes interpretation, 2 years of low and high were noted (NMFS, 2012b). Additional relative abundance were chosen to research is needed to identify the Directed Commercial Harvest illustrate the effects of population limiting factor(s) for river herring This following section on river abundance and temperature on alewife populations. As Nye et al. (2012) noted, herring fisheries in the United States is distribution. The low and high the links between climate and river from the stock assessment (ASMFC, abundance years were objectively herring biology during freshwater stages 2012).

chosen as the years closest to - 1 and are unclear and will require additional Fisheries for anadromous species

+1 standard deviation from overall time to research and thoroughly have existed in the United States for a mean abundance. Two years closest to analyze. This conclusion is supported very long time. They not only provided the - 1 and +1 standard deviation from by the results of the workshop, which sustenance for early settlers but a source mean population abundance were noted numerous potential climate of income as the fisheries were selected to reflect the combined effect of effects on the freshwater stages, but commercialized. It is difficult to fully warming with low and high abundance little synthesis has been accomplished describe the characteristics of these of blueback herring. The difference in to date. The preliminary analysis of Nye early fisheries because of the lack of species response (as noted below) may et al. (2012) indicates that water quantifiable data.

reflect the different temperature temperatures in the rivers will be The earliest commercial river herring tolerances (9-11 °C for blueback herring warmer, and there will be a decrease in data were generally reported in state and 4-11 °C for alewife) as indicated by the river flow in the northeast and Mid- and town reports or local newspapers.

the southern limit of their ranges. Atlantic in late winter/early spring. In 1871, the U.S. Fish Commission was Blueback herring may be able to tolerate Although current information founded (later became known as the higher temperature as their range indicates climate change is and will U.S. Fish and Fisheries Commission in extends as far south as Florida, but the continue to impact river herring (e.g., 1881). This organization collected southern extent of the alewife's range is Nye et al., 2012), climate variability fisheries statistics to characterize the limited to North Carolina. For both rather than climate change is expected biological and economic aspects of species, the Nye et al. (2012) analysis to have more of an impact on river commercial fisheries. Data describing indicates that, if robust populations of herring from 2024-2030. Several studies historical river herring fisheries were

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48959 available from two of this organization's approximately 2.87 million dollars or jurisdictions without an approved publications-the Bulletin of the U.S. (2010 USD) (ASMFC, 2012). sustainable fisheries management plan, Fish Commission (renamed Fishery Domestic commercial landings of as required under ASMFC Amendment Bulletin in 1971; Collins and Smith, river herring were presented in the stock 2 to the Shad and River Herring FMP, 1890; Smith, 1891) and the U.S. Fish assessment by state and by gear from were closed. As a result, prohibitions on Commission Annual Report (USFC, 1887 to 2010 where available. Landings harvest (commercial or recreational) 1888-1940). In the stock assessment, the of alewife and blueback herring were were extended to the following states:

river herring data were transcribed and collectively classified as "river herring" New Jersey, Delaware, Pennsylvania, when available, dollar values were by most states. Only a few states had Maryland, DC, Virginia (for all waters),

converted to 2010 dollar values using species-specific information recorded Georgia and Florida (ASMFC, 2012).

conversion factors based on the annual for a limited range of years. Commercial Pound nets were identified as the average consumer price index (CPI) landings records were available for each dominant gear type used to harvest river values, which were obtained from the state since 1887 except for Florida and herring from 1887 through 2010. Seines U.S. Bureau of Labor Statistics. Note the Potomac River Fisheries were more prevalent prior to the 1960s, that CPI values are not available for Commission (PRFC), which began but by the 1980s, they were rarely used.

years prior to 1913 so conversion factors recording landings in 1929 and 1960, Purse seines were used only for herring could not be calculated for years earlier respectively. It is important to note that landed in Massachusetts, but made up than 1913 (ASMFC, 2012). historical landings presented in the a large proportion of the landings in the There are several caveats to using the stock assessment do not include all 1950s and 1960s. Historically, gill nets historical fisheries data. There is an landings for all states over the entire made up a small percentage of the apparent bias in the area sampled. In time period and are likely overall harvest. However, even though most cases, there was no systematic underestimated, particularly for the first the actual pounds landed continued to sampling of all fisheries; instead, third of the time series, since not all decline, the proportion of gill nets that sampling appeared to be opportunistic, river landings were reported (ASMFC, contributed to the overall harvest has concentrating on the mid-Atlantic 2012). increased in recent years (ASMFC, States. It is also difficult to assess the Total domestic coast-wide landings 2012).

accuracy and precision of these data. In averaged 18.5 million lb (8,399 mt) from Foreign fleet landings of river herring some instances, the pounds were 1887 to 1928 (See table 2.2 in ASMFC (reported as alewife and blueback shad) reported at a fine level of detail (e.g., at (2012)). During this early time period, are available through the Northwest the state/county/gear level), but details landings were predominately from Atlantic Fisheries Organization (NAFO).

regarding the specific source of the data Maryland, North Carolina, Virginia, and Offshore exploitation of river herring were often not described. The level of Massachusetts (overall harvest is likely and shad (generally <7.5 in (190 mm) in detail provided in the reports varied underestimated because landings were length) by foreign fleets began in the late among states and years. Additionally, not recorded consistently during this 1960s and landings peaked at about 80 not all states and fisheries were time). Virginia made up approximately million lbs (36,320 mt) in 1969 canvassed in all years, so absence of half of the commercial landings from (ASMFC, 2012).

landings data does not necessarily 1929 until the 1970s, and the majority Total U.S. and foreign fleet harvest of indicate the fishery was not active as it of Virginia's landings came from the river herring from the waters off the is possible that the data just were not Chesapeake Bay, Potomac River, York coast of the United States (NAFO areas collected. For these reasons, these River, and offshore harvest. Coast-wide 5 and 6) peaked at about 140 million lb historical river herring landings should landings started increasing sharply in (63,560 mt) in 1969, after which not be considered even minimum values the early 1940s and peaked at over 68.7 landings declined dramatically. After because of the variation in detail and million lb (31,160 mt) in 1958 (See 1977 and the formation of the Fishery coverage over the time series. No Table 2.2, ASMFC, 2012). In the 1950s Conservation Zone, foreign allocation of attempt was made to estimate missing and 1960s, a large proportion of the river herring (to both foreign vessels and river herring data since no benchmark harvest came from Massachusetts purse joint venture vessels) between 1977 and or data characteristics could be found, seine fisheries that operated offshore on 1980 was 1.1 million lb (499 mt). The and the stock assessment subcommittee Georges Bank targeting Atlantic herring foreign allocation was reduced to also did not attempt to estimate missing (G. Nelson, Massachusetts Division of 220,000 lb (100 mt) in 1981 because of data in a time series at a particular Marine Fisheries, Pers. comm., 2012). the condition of the river herring location because of the bias associated Landings from North Carolina were also resource. In 1985, a bycatch cap of no with these estimates (ASMFC, 2012). at their highest during this time and more than 0.25 percent of total catch During 1880 to 1938, reported originated primarily from the Chowan was enacted for the foreign fishery. The commercial landings of river herring River pound net fishery. Severe declines cap was exceeded once in 1987, and this along the Atlantic Coast averaged in landings began coast-wide in the shut down the foreign mackerel fishery.

approximately 30.5 million lbs (13,835 early 1970s and domestic landings are In 1991, area restrictions were passed to mt) per year. The majority of river now a fraction of what they were at their exclude foreign vessels from within 20 herring landed by commercial fisheries peak, having remained at persistently miles (32.2 kin) of shore for two reasons:

in these early years are attributed to the low levels since the mid-1990s. 1) In response to the increased mid-Atlantic region (NY-VA). The Moratoria were enacted in occurrence of river herring bycatch dominance of the mid-Atlantic region is, Massachusetts (commercial and closer to shore and 2) to promote in part, due to the apparent bias in the recreational in 2005), Rhode Island increased fishing opportunities for the spatial coverage of the canvass (see (commercial and recreational in 2006), domestic mackerel fleet (ASMFC, 2012).

above). From 1920 to 1938, the average Connecticut (commercial and annual weight of reported commercial recreational in 2002), Virginia (for In-river Exploitation river herring landings was about 22.8 waters flowing into North Carolina in The stock assessment subcommittee million lbs (10,351 mt). The value of the 2007), and North Carolina (commercial calculated in-river exploitation rates of commercial river herring landings and recreational in 2007). As of January the spawning runs for five rivers during this same time period was 1, 2012, river herring fisheries in states (Damariscotta River (ME-alewife),

48960 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Union River (ME-alewife), Monument abundance. Fishing effort has been catch of approximately 71,290 lb (32.4 River (MA-both species combined), shown to increase variation in fish mt) in the Atlantic herring fishery for Mattapoisett River (MA-alewife), and abundance through truncation of the age 2005-2007, and the corresponding Nemasket River (MA-alewife)) by structure, and recruitment becomes coefficient of variation (CV) was 0.56.

dividing in-river harvest by total run primarily governed by environmental Cournane et al. (2010) extended this size (escapement plus harvest) for a variation (Hsieh et al., 2006; Anderson analysis with additional years of data.

given year. Exploitation rates were et al., 2008). When fish species are at Further work is needed to elucidate how highest (range: 0.53 to 0.98) in the very low abundances, as is believed for the landed catch of river herring in the Damariscotta River and Union River river herring, it is possible that the only directed Atlantic herring fishery prior to 1985, while exploitation was population regulatory processes compares to total incidental catch across lowest (range: 0.26 to 0.68) in the operating are stochastic fluctuations in all fisheries. Since this analysis only Monument River. Exploitation declined the environment (Shepherd and quantified kept river herring in the in all rivers through 1991 to 1992. Cushing, 1990) (ASMFC, 2012). Atlantic herring fishery, it Exploitation rates of both species in the underestimates the total catch (kept plus Monument River and of alewives in the Canadian Harvest discarded) of river herring across all Mattapoisett River and Nemasket River Fisheries in Canada for river herring fishing fleets. Wigley et al. (2009) were variable (average = 0.16) and, are regulated through limited seasons, quantified river herring discards across except for the Nemasket River, declined gears, and licenses. Licenses may cover fishing fleets that had sufficient generally through 2005 until the different gear types; however, few new observer coverage from July 2007-Massachusetts moratorium was licenses have been issued since 1993 August 2008. Wigley et al. (2009) imposed. Exploitation rates of alewives (DFO, 2001). River-specific management estimated that approximately 105,820 lb in the Damariscotta River were low plans include closures and restrictions. (48 mt) were discarded during the 12

(<0.05) during 1993 to 2000, but they River herring used locally for bait in months (July 2007 to August 2008), and increased steadily through 2004 and other fisheries are not accounted for in the estimated precision was low (149 remained greater than 0.34 through river-specific management plans (DFO, percent CV). This analysis estimated 2008. Exploitation in the Damariscotta 2001). DFO estimated river herring only river herring discards (in contrast dropped to 0.15 in 2009 to 2010. landings at just under 25.5 million lb to total incidental catch), and noted that Exploitation rates of alewives in the (11,577 mt) in 1980, 23.1 million lb midwater trawl fleets generally retained Union River declined through 2005 but (10,487 mt) in 1988, and 11 million lb river herring while otter trawls typically have remained above 0.50 since 2007 (4,994 mt) in 1996 (DFO, 2001). The discarded river herring.

(ASMFC, 2012). largest river herring fisheries in According to the stock assessment, Canadian waters occur in the Bay of Lessard and Bryan (2011) estimated exploitation of river herring appears to Fundy, southern Gulf of Maine, New an average incidental catch of river be declining or remaining stable. In- Brunswick, and in the Saint John and herring and American shad of 3.3 river exploitation was highest in Maine Miramichi Rivers where annual harvest million lb (1,498 mt)/yr from 2000-rivers (Damariscotta and Union) and has estimates often exceed 2.2 million lb 2008. The methodology used in this fluctuated, but it is currently lower than (1,000 mt) (DFO, 2001). Recreational study differed from the Standardized levels seen in the 1980s. Also, in-river fisheries in Canada for river herring are Bycatch Reporting Methodology (SBRM) exploitation in Massachusetts rivers limited by regulations including area, (the method used by NOAA's Northeast (Monument and Mattapoisett) was gear and season closures with limits on Fisheries Science Center (NEFSC) to declining at the time a moratorium was the number of fish that can be harvested quantify bycatch in stock assessments) imposed in 2005. The coast-wide index per day; however, information on (Wigley et al., 2007; Wigley et al., 2012).

of relative exploitation also declined recreational catch is limited. Licenses Data from NEFOP were analyzed at the following a peak in the late 1980s and and reporting are not required by haul level; however, the sampling unit has remained fairly stable over the past Canadian regulations for recreational for the NEFOP database is at the trip decade. Exploitation rates declined in fisheries, and harvest is not well level. Within each gear and region, all the DB-SRA model runs except when documented. data, including those from high volume the input biomass-to-K ratio in 2010 was fisheries, appeared to be aggregated 0.01. Exploitation rates estimated from Incidental Catch across years from 2000 through 2008.

the statistical catch-at-age model for The following section on river herring However, substantial changes in NEFOP blueback herring in the Chowan River incidental catch in the United States is sampling methodology for high volume (see the NC state report in the stock from the stock assessment (ASMFC, fisheries were implemented in 2005, assessment) also showed a slight 2012). limiting the interpretability of estimates declining trend from 1999 to 2007, at Three recent studies estimated river from these fleets in prior years. Total which time a moratorium was herring discards and incidental catch number of tows from the fishing vessel instituted. There appears to be a (Cieri et al., 2008; Wigley et al., 2009; trip report (VTR) database was used as consensus among various assessment Lessard and Bryan, 2011). The discard the raising factor to estimate total methodologies that exploitation has and incidental catch estimates from incidental catch. The use of effort decreased in recent times. The stock these studies cannot be directly without standardization makes the assessment indicates that the decline in compared as they used different ratio implicit assumption that effort is exploitation over the past decade is not estimators based on data from the constant across all tows within a gear surprising because river herring Northeast Fishery Observer Program type, potentially resulting in a biased populations are at low levels and more (NEFOP), as well as different raising effort metric. In contrast, the total kept restrictive regulations or moratoria have factors to obtain total estimates. Cieri et weight of all species is used as the been enacted by states (ASMFC, 2012). al. (2008) estimated the kept (i.e., raising factor in SBRM. When Past high exploitation may also be a landed) portion of river herring quantifying incidental catch across reason for the high amount of variation incidental catch in the Atlantic herring multiple fleets, total kept weight of all and inconsistent patterns observed in fishery. Cieri et al. (2008) estimated an species is an appropriate surrogate for fisheries-independent indices of average annual landed river herring effective fishing power because it is

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48961 likely that all trips will not exhibit the percent during each of the other three believed to have minimal impacts on same attributes. Lessard and Bryan quarters. river herring populations.

(2011) also did not provide precision Recreational Harvest Summary and Evaluation of Factor B estimates, which are imperative for estimation of incidental catch. The Marine Recreational Fishery Historical commercial and Statistics Survey (MRFSS) provided recreational fisheries for river herring The total incidental catch of river estimates of numbers of fish harvested likely contributed to the decline in herring was estimated as part of the and released by recreational fisheries abundance of both alewife and blueback work for Amendment 14 to the Atlantic along the Atlantic coast. The stock herring populations. Current directed Mackerel, Squid and Butterfish (MSB) assessment subcommittee extracted commercial and recreational alewife Fishery Management Plan, that includes state harvest and release estimates for and blueback herring fisheries, as well measures to address incidental catch of alewives and blueback herring from the as commercial fishery incidental catch river herring and shads, From 2005- MRFSS catch and effort estimates files may continue to pose a threat to these 2010, the total annual incidental catch available on the web (http:// species. Since the 1970s, regulations of alewife ranged from 41,887 lb (19.0 www.sefsc.noaa.gov/about/mrfss.htm). have been enacted in the United States mt) to 1.04 million lb (472 mt) in New Historically, there were few reports of on the directed harvest of river herring England and 19,620 lb (8.9 mt) to river herring taken by recreational in an attempt to halt or reverse their 564,818 lb (256.4 mt) in the Mid- anglers for food. Most often, river decline with the most recent regulations Atlantic. The dominant gear varied herring were taken for bait. MRFSS being imposed in January 2012.

across years between paired midwater estimates of the numbers of river herring Additionally, there are regulations in trawls and bottom trawls. harvested and released by anglers are Canada on river herring harvest.

Corresponding estimates of precision very imprecise and show little trend. Historical landings data and current (COV) exhibited substantial interannual Thus, the stock assessment concluded fishery effort is the best available variation and ranged from 0.28 to 3.12 that these data are not useful for information to describe the impact that across gears and regions. Total annual management purposes. MRFSS the commercial fishery may be having blueback herring incidental catch from concentrates their sampling strata in on river herring.

2005 to 2010 ranged from 30,643 lb coastal water areas and does not capture Moratoria are in place on directed (13.9 mt) to 389,111 lb (176.6 mt) in any data on recreational fisheries that catch of these species throughout most New England and 2,645 lb (1.2 mt) to occur in inland waters. Few states of the United States; however, they are 843,479 lb (382.9 mt} in the Mid- conduct creel surveys or other taken as incidental catch in several Atlantic. Across years, paired and single consistent survey instruments (diary or fisheries. The extent to which incidental midwater trawls exhibited the greatest log books) in their inland waters to catch is affecting river herring has not blueback herring catches, with the collect data on recreational catch of been quantified and is not fully exception of 2010 in the mid-Atlantic river herring. Some data are reported in understood. Thus, the level of threat where bottom trawl was the most the state chapters in the stock posed by directed and indirect catch is dominant gear. Corresponding estimates assessment; but the stock assessment evaluated further in the qualitative of precision ranged from 0.27 to 3.65. committee concluded that data are too threats assessment as described below.

The temporal distribution of incidental sparse to conduct any systematic Scientific collections or collections for catches was summarized by quarter and comparison of trends (ASMFC, 2012). educational purposes do not appear to fishing region for the most recent 6-year Scientific Monitoring and Educational be significantly affecting the status of period (2005 to 2010). River herring Harvest river herring, as they result in low catches occurred primarily in midwater Maine, New Hampshire, mortality.

trawls (76 percent, of which 56 percent Massachusetts and Rhode Island C. Disease and Predation were from paired midwater trawls and estimate run sizes using electronic the rest from single midwater trawls), counters or visual methods. Various Disease followed by small mesh bottom trawls counting methods are used at the Little information exists on diseases (24 percent). Catches of river herring in Holyoke Dam fish lift and fishways on that may affect river herring; however, gillnets were negligible. Across gear the Connecticut River. Young of year there are reports of a variety of parasites types, catches of river herring were (YOY) surveys are conducted through that have been found in both alewife greater in New England (56 percent) fixed seine surveys capturing YOY and blueback herring. The most than in the Mid-Atlantic (44 percent). alewife and blueback herring generally comprehensive report is that of Landry The percentages of midwater trawl during the summer and fall in Maine, et al. (1992) in which 13 species of catches of river herring were similar Rhode Island, Connecticut, New York, parasites were identified in blueback between New England (37 percent) and New Jersey, Maryland, District of herring and 12 species in alewives from the Mid-Atlantic (38 percent). However, Columbia, Virginia and North Carolina. the Miramichi River, New Brunswick, catches in New England small mesh Rhode Island conducts surveys for Canada. The parasites found included bottom trawls were three times higher juvenile and adult river herring at large one monogenetic trematode, four (18 percent) than those from the Mid- fixed seine stations. Virginia samples digenetic trematodes, one cestode, three Atlantic (6 percent). Overall, the highest river herring using a multi-panel gill net nematodes, one acanthocephalan, one quarterly catches of river herring survey and electroshocking surveys. annelid, one copepod and one mollusk.

occurred in midwater trawls during Florida conducts electroshocking The same species were found in both Quarter I in the Mid-Atlantic (35 surveys to sample river herring. Maine, alewife and blueback herring with the percent), followed by catches in New New Hampshire, Massachusetts, Rhode exception of the acanthocephalan, England during Quarter 4 (16 percent) Island, Maryland, and North Carolina which was absent from alewives.

and Quarter 3 (11 percent). Quarterly collect age data from commercial and In other studies, Sherburne (1977) catches in small mesh bottom trawls fisheries independent sampling reported piscine erythrocytic necrosis were highest in New England during programs, and length-at-age data. All of (PEN) in the blood of 56 percent of Quarter 1 (7 percent) and totaled 3 to 4 these scientific monitoring efforts are prespawning and 10 percent of

48962 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices postspawning alewives in Maine coastal speculation that increased predation species are a particular concern because streams. PEN was not found in juvenile may be contributing to the decline of of the lack of native predators, parasites, alewives from the same locations. river herring and American shad and competitors to keep their Coccidian parasites were found in the (Hartman, 2003; Crecco et al., 2007; populations in check.

livers of alewives and other finfish off Heimbuch, 2008). Quantifying the Predation and multispecies models, the coast of Nova Scotia (Morrison and impacts of predation on alewife and such as the MS-VPA (NEFSC, 2006),

Marryatt, 1990). Marcogliese and blueback herring is difficult. The diet of have tremendous data needs, and more Compagna (1999) reported that most striped bass has been studied research needs to be conducted before fish species, including alewife, in the St. extensively, and the prevalence of they can be applied to river herring.

Lawrence River become infected with alosines varies greatly depending on However, given the potential magnitude trematode metacercariae during the first location, season, and predator size of predatory interactions, it is an area of years of life. Examination of Great Lakes (Walter et al., 2003). Studies from the research worth pursuing (ASMFC, fishes in Canadian waters showed larval northeast U.S. continental shelf show 2012).

Diplostomum (trematode) commonly in low rates of consumption by striped Two papers have become available the eyes of alewife in Lake Superior bass (alewife and blueback herring each since the ASMFC (2012) stock (Dechtiar and Lawrie, 1988) and Lake make up less than 5 percent of striped assessment that discuss striped bass Ontario (Dechtiar and Christie, 1988), bass diet by weight) (Smith and Link, predation on river herring in though intensity of infections was low 2010), while studies that sampled Massachusetts and Connecticut

(<9/host). Heavy infections of striped bass in rivers and estuaries estuaries and rivers, showing temporal Saprolegnia,a fresh and brackish water during the spring spawning runs found and spatial patterns in predation (Davis fungus, were found in 25 percent of much higher rates of consumption et al., 2012; Ferry and Mather, 2012).

Lake Superior alewife examined, and (greater than 60 percent of striped bass Davis et al. (2012) estimated that light infections were found in 33 diet by weight in some months and size approximately 400,000 blueback herring percent of Lake Ontario alewife classes) (Walter and Austin, 2003; are consumed annually by striped bass (Dechtiar and Lawrie, 1988). Larval Rudershausen et al., 2005). Translating in the Connecticut River spring acanthocephala were also found in the these snapshots of diet composition into migration. In this study, striped bass guts of alewife from both lakes. estimates of total removals requires were found in the rivers during the Saprolegniatypically is a secondary additional data on both annual per spring spawning migrations of blueback infection, invading open sores and capita consumption rates and estimates herring and had generally left the wounds, and eggs in poor of annual abundance for predator system by mid-June (Davis et al., 2012).

environmental conditions, but under the species. Many blueback herring in the right conditions it can become a primary The diets of other predators, Connecticut River are thought to be pathogen. Saprolegniainfections including other fish (e.g., bluefish, spiny consumed prior to ascending the river usually are lethal to the host. dogfish), along with marine mammals on their spawning migration, and are, More recently, alewives were found (e.g., seals) and birds (e.g., double- therefore, being removed from the positive for Cryptosporidium for the crested cormorant), have not been system before spawning. Alternatively, first time on record by Ziegler et al. quantified nearly as extensively, making Ferry and Mather (2012) discuss the (2007). Mycobacteria, which can result it more difficult to assess the results of a similar study conducted in in ulcers, emaciation, and sometimes importance of river herring in the Massachusetts watersheds with death, have been found in many freshwater and marine food webs. As a drastically different findings for striped Chesapeake Bay fish, including result, some models predict a significant bass predation. Striped bass were blueback herring (Stine et al., 2010). negative effect from predation (Hartman, collected and stomach contents 2003; Heimbuch, 2008), while other analyzed during three seasons from May Predation studies did not find an effect through October (Ferry and Mather, Information on predation of river (Tuomikoski et al., 2008; Dalton et al., 2012). The stomach contents of striped herring was compiled and published in 2009). bass from the survey were examined Volume I of the River Herring In addition to predators native to the and less than 5 percent of the clupeid Benchmark Assessment (2012) by Atlantic coast, river herring are category (from 12 categories identified ASMFC. The following section on vulnerable to invasive species such as to summarize prey) consisted of predation was compiled by Dr. Katie the blue catfish (Ictalurusfurcatus)and anadromous alosines (Ferry and Mather, Drew from this assessment. the flathead catfish (Pylodictis olivaris). 2012). Overall, the Ferry and Mather Alewife and blueback herring are an These catfish are large, opportunistic (2012) study observed few anadromous important forage fish for marine and predators native to the Mississippi River alosines in the striped bass stomach anadromous predators, such as striped drainage that were introduced into contents during the study period. These bass, spiny dogfish, bluefish, Atlantic rivers on the Atlantic coast. They have two recent studies echo similar cod, and pollock (Bowman et al., 2000; been observed to consume a wide range contradictory findings from previous Smith and Link, 2010). Historically, of species, including alosines, and studies showing a wide variation in river herring and striped bass landings ecological modeling on flathead catfish predation by striped bass with spatial have tracked each other quite well, with suggests they may have a large impact and temporal effects; however, they highs in the 1960s, followed by declines on their prey species (Pine, 2003; exhibit no consistent trends along the through the 1970s and 1980s. Although Schloesser et al., 2011). In August 2011, coast.

populations of Atlantic cod and pollock ASMFC approved a resolution calling are currently low, the populations of for efforts to reduce the population size Summary and Evaluation for Factor C striped bass and spiny dogfish have and ecological impacts of invasive While data are limited, the best increased in recent years (since the early species and named blue and flathead available information indicates that 1980s for striped bass and since 2005 for catfish specifically, as species of river herring are not likely affected to a spiny dogfish), while the landings and concern, due to their increasing large degree by diseases caused by run counts of river herring remain at abundance and potential impacts on viruses, bacteria, protozoans, historical lows. This has led to native anadromous species. Non-native metazoans, or microalgae. Much of the

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48963 information on diseases in alewife or harvested by various gear types (e.g., Federal blueback herring comes from studies on gillnet, dip nets, trap) and the ASMFC and Enabling Legislation landlocked species; therefore, even if regulations depend upon the river and studies indicated that landlocked associated location (DFO, 2001). The Authorized under the terms of the alewife and blueback herring were primary management measures are Atlantic States Marine Fisheries highly susceptible to diseases and weekly closed periods and limiting the Compact, as amended (Pub. L.81-721),

suffered high mortality rates, it is not number of licenses to existing levels in the purpose of the ASMFC is to promote known whether anadromous river all areas (DFO, 2001). Logbooks are the better utilization of the fisheries herring would be affected in the same issued to commercial fishermen in some (marine, shell, and anadromous) of the way. While it may be possible that areas as a condition of the license, and Atlantic seaboard "by the development disease threats to river herring could pilot programs are being considered in of a joint program for the promotion and increase in prevalence or magnitude other areas (DFO, 2001). The protection of such fisheries, and by the under various climate change scenarios, prevention of the physical waste of the management objective is to maintain there are currently no data available to harvest near long-term mean levels fisheries from any cause."

support this supposition. We have Given management authority in 1993 when no specific biological and under the Atlantic Coastal Fisheries included disease as a threat in the fisheries information is available (DFO, qualitative threats assessment described Cooperative Management Act (16 U.S.C.

2001). 5101-5108), the ASMFC may issue in detail below.

Alewife and blueback herring are DFO (2001) stated that additional interstate FMPs that must be considered to be an important forage management measures may be required administered by state agencies. If the fish for many marine and anadromous if increased effort occurs in response to ASMFC believes that a state is not in predators, and therefore, may be stock conditions or favorable markets. compliance with a coastal FMP, it must affected by predation, especially if some There has been concern as fishery notify the Secretaries of Commerce and populations of predators (e.g., striped exploitation rates have been above Interior. If the Secretaries find the state bass, spiny dogfish) continue to reference levels and fewer licenses are not in compliance with the management increase. There may also be effects from fished than have been issued (DFO, plan, the Secretaries must declare a predation by invasive species such as 2001). In 2001, DFO reported that in moratorium on the fishery in question.

the blue and flathead catfish. Some some rivers river herring were being Atlantic Coastal Fisheries Cooperative predation and multispecies models have harvested at or above reference levels Management Act estimated an effect of predation on river (e.g., Miramichi), while in other rivers herring, while others have not. In river herring were harvested at or below We manage river herring stocks under general, the effect of predation on the the reference point (e.g., St. John River the authority of section 803(b) of the persistence of river herring is not fully at Mactaquac Dam). DFO (2001) believes Atlantic Coastal Fisheries Cooperative understood; however, predation may be precautionary management involving no Management Act (Atlantic Coastal Act) affecting river herring populations and increase or decrease in exploitation is 16 U.S.C. 5101 et seq., which states, in consequently, it is included as a threat important for Maritime river herring the absence of an approved and in the qualitative threats assessment fisheries, given that biological and implemented FMP under the Magnuson-described below. harvest data are not widely available. Stevens Act (MSA, 16 U.S.C. 1801 et Additionally, DFO (2001) added that seq.) and, after consultation with the D. Inadequacy of Existing Regulatory appropriate Fishery Management Mechanisms river-specific management plans based on stock assessments should be Council(s), the Secretary of Commerce As wide-ranging anadromous species, prioritized over general management may implement regulations to govern alewife and blueback herring are subject initiatives. fishing in the Exclusive Economic Zone to numerous Federal (U.S. and (EEZ), i.e., from 3 to 200 nautical mi Canadian), state and provincial, Tribal, Eastern New Brunswick is currently (nm) offshore. The regulations must be:

and inter-jurisdictional laws, the only area in the Canadian Maritimes (1) Compatible with the effective regulations, and agency activities. These with a river herring integrated fishery implementation of an Interstate Fishery regulatory mechanisms are described in management plan (DFO, 2006). The Management Plan for American Shad detail in the following section. DFO uses Integrated Fisheries and River Herring (ISFMP) developed Management Plans (IFMPs) to guide the by the ASMFC; and (2) consistent with International conservation and sustainable use of the national standards set forth in The Canadian DFO manages alewife marine resources (DFO, 2010). An IFMP section 301 of the MSA.

and blueback herring fisheries that manages a fishery in a given region by The ASMFC adopted Amendment 2 to occur in the rivers of the Canadian combining the best available science on the ISFMP in 2009. Amendment 2 Maritimes under the Fisheries Act the species with industry data on establishes the foundation for river (R.S.C., 1985, c. F-14). The Maritime capacity and methods for harvesting herring management. It was developed Provinces Fishery Regulations includes (DFO, 2010). The 6-year management to address concerns that many Atlantic requirements when fishing for or plan (2007-2012) for river herring for coast populations of river herring were catching and retaining river herring in Eastern New Brunswick is implemented in decline or are at depressed but stable recreational and commercial fisheries in conjunction with annual updates to levels, and that the ability to accurately (DFO, 2006; http://laws- specific fishery management measures assess the status of river herring stocks lois.justice.gc.ca). (e.g., seasons). For example, it notes a is complicated by a lack of fishery Commercial and recreational river management problem of gear congestion independent data.

herring fisheries in the Canadian in some rivers and an approach to Amendment 2 requires states to close Maritimes are regulated by license, establish a carrying capacity of the river their waters to recreational and fishing gear, season and/or other and find a solution to the gear limit by commercial river herring harvest, unless measures (DFO, 2001). Since 1993, DFO working with fishermen (DFO, 2006). At they have an approved sustainable has issued few new licenses for river this time, an updated Eastern New management plan in place. To be herring (DFO, 2001). River herring are Brunswick IFMP is not available. approved, a state's plan must clearly

48964 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices meet the Amendment's standard of a amendment mandated the use of benefit to river herring over any other sustainable fishery defined as "a Annual Catch Limits (ACL) and EFH designation. Habitat features used commercial and/or recreational fishery Accountability Measures (AM) to end for spawning, breeding, feeding, growth that will not diminish the potential overfishing, provided for widespread and maturity by these two species future stock reproduction and market-based fishery management encompasses many of the habitat recruitment." The plans must meet the through limited access privilege features selected by river herring to definition of sustainability by programs, and called for increased carry out their life history. The developing and maintaining international cooperation. geographic range in which river herring sustainability targets. States without an The MSA requires that Federal FMPs may benefit from the designation of approved plan were required to close contain conservation and management Atlantic salmon EFH extends from their respective river herring fisheries as measures that are consistent with the Connecticut to the Maine/Canada of January 1, 2012, until such a plan is ten National Standards. National border. The geographic range in which submitted and approved by the Standard #9 states that conservation and river herring may benefit from the ASMFC's Shad and River Herring management measures shall, to the designation of Atlantic herring EFH Management Board. Proposals to re- extent practicable, (A) minimize bycatch designation extends from the Maine/

open closed fisheries may be submitted and (B) to the extent bycatch cannot be Canada border to Cape Hatteras.

annually as part of a state's annual avoided, minimize the mortality of such The Atlantic salmon EFH includes compliance report. Currently, the states bycatch. The MSA defines bycatch as most freshwater, estuary and bay of ME, NH, RI, NY, NC, and SC have fish that are harvested in a fishery, but habitats historically accessible to approved river herring management which are not sold or kept for personal Atlantic salmon from Connecticut to the plans (see "State section of Factor D" for use. This includes economic discards Maine/Canada border (NEFMC, 2006).

more information). and regulatory discards. River herring is Many of the estuary, bay and freshwater In addition to the state sustainability encountered both as bycatch and habitats within the current and plan mandate, Amendment 2 makes incidental catch in Federal fisheries. historical range of Atlantic salmon recommendations to states for the While there is no directed fishery for incorporate habitats used by river conservation, restoration, and protection river herring in Federal waters, river herring for spawning, migration and of critical river herring habitat. The herring co-occur with other species that juvenile rearing. Among Atlantic Amendment also requires states to have directed fisheries (Atlantic herring EFHs are the pelagic waters in implement fisheries-dependent and mackerel, Atlantic herring, whiting, the Gulf of Maine, Georges Bank, independent monitoring programs, to squid and butterfish) and are either Southern New England, and middle provide critical data for use in future discarded or retained in those fisheries. Atlantic south to Cape Hatteras out to river herring stock assessments. the offshore U.S. boundary of the EEZ While these measures address Essential Fish Habitat Under the MSA (see NEFMC 1998). These areas problems to the river herring Under the MSA, there is a incorporate nearly all of the U.S. marine populations in coastal areas, incidental requirement to describe and identify areas most frequently used by river catch in small mesh fisheries, such as EFH in each Federal FMP. EFH is herring for growth and maturity.

those for sea herring, occurs outside defined as ". . . those waters and Subsequently, in areas where EFH state jurisdiction and remains a substrate necessary to fish for spawning, designations for Atlantic salmon and substantial source of fishing mortality breeding, feeding, or growth to Atlantic herring overlap with freshwater according to the ASMFC. Consequently, maturity." The rules promulgated by the and marine habitats used by river the ASMFC has requested that the New NMFS in 1997 and 2002 further clarify herring, conservation benefits afforded England and Mid-Atlantic Fishery EFH with the following definitions: (1) through the designation of EFH for these Management Councils (NEFMC and Waters-aquatic areas and their species may provide similar benefits to MAFMC) increase efforts to monitor associated physical, chemical, and river herring.

river herring incidental catch in small- biological properties that are used by mesh fisheries (See section on "NEFMC fish and may include aquatic areas Federal Power Act (FPA) (16 U.S.C.

and MAFMC recommendations for historically used by fish where 791-828) and Amendments future river herring bycatch reduction appropriate; (2) substrate-sediment, The FPA, as amended, provides for efforts"). hard bottom, structures underlying the protecting, mitigating damages to, and waters, and associated biological enhancing fish and wildlife resources Magnuson-Stevens Fishery communities; (3) necessary-the habitat (including anadromous fish) impacted Conservation and Management Act required to support a sustainable fishery by hydroelectric facilities regulated by (MSA) and the managed species' contribution the Federal Energy and Regulatory The Magnuson-Stevens Fishery to a healthy ecosystem; and (4) Commission (FERC). Applicants must Conservation and Management Act spawning, breeding, feeding, or growth consult with state and Federal resource (MSA) is the primary law governing to maturity-stages representing a agencies who review proposed marine fisheries management in Federal species' full life cycle. hydroelectric projects and make waters. The MSA was first enacted in EFH has not been designated for recommendations to FERC concerning 1976 and amended in 1996 and 2006. alewife or blueback herring, though EFH fish and wildlife and their habitat, e.g.,

Most notably, the MSA aided in the has been designated for numerous other including spawning habitat, wetlands, development of the domestic fishing species in the Northwest Atlantic. instream flows (timing, quality, industry by phasing out foreign fishing. Measures to improve habitats and quantity), reservoir establishment and To manage the fisheries and promote reduce impacts resulting from those regulation, project construction and conservation, the MSA created eight EFH designations may directly or operation, fish entrainment and regional fishery management councils. indirectly benefit river herring. mortality, and recreational access.

A 1996 amendment focused on Conservation measures implemented in Section 10(j) of the FPA provides that rebuilding overfished fisheries, response to the designation of Atlantic licenses issued by FERC contain protecting Essential Fish Habitat (EFH), salmon EFH and Atlantic herring EFH conditions to protect, mitigate damages and reducing bycatch. A 2006 likely provide the most conservation to, and enhance fish and wildlife based

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48965 on recommendations received from state agency is not bound by the primarily by the Department of and Federal agencies during the recommendations. Agriculture.

licensing process. With regard to fish The FWCA applies to water-related Like the Fish and Wildlife passage, Section 18 requires a FERC activities proposed by non-Federal Coordination and River and Harbors licensee to construct, maintain, and entities for which a Federal permit or Acts, Sections 401 and 404 of the operate fishways prescribed by the license is required. The most significant FWPCA have played a role in reducing Secretary of the Interior or the Secretary permits or licenses required are Section discharges of pollutants, restricting the of Commerce. Under the FPA, others 404 and discharge permits under the timing and location of dredge and fill may review proposed projects and make Clean Water Act and Section 10 permits operations, and affecting other changes timely recommendations to FERC to under the Rivers and Harbors Act. The that have improved river herring habitat represent additional interests. Interested USFWS and NMFS may review the in many rivers and estuaries over the parties may intervene in the FERC proposed permit action and make last several decades. Examples include proceeding for any project to receive recommendations to the permitting reductions in sewage discharges into the pertinent documentation and to appeal agencies to avoid or mitigate any Hudson River (A. Kahnle, New York an adverse decision by FERC. potential adverse effects on fish and State DEC, Pers. comm. 1998) and While the construction of wildlife habitat. These nutrient reduction strategies hydroelectric dams contributed to some recommendations must be given full implemented in the Chesapeake Bay (R.

historical losses of river herring consideration by the permitting agency, St. Pierre, USFWS, Pers. comm. 1998).

spawning habitat, only a few new dams but are not binding. Rivers and Harbors Act of 1899 have been constructed in the range of Federal Water Pollution Control Act, these species in the last 50 years. In Section 10 of the Rivers and Harbors and amendments (FWPCA) (33 U.S.C. Act requires a permit from the ACOE to some areas, successful fish passage has 1251-1376) place structures in navigable waters of been created; thus, restoring access to many habitats once blocked. Thus, river Also called the "Clean Water Act," the United States or modify a navigable herring may often benefit from FPA the FWPCA mandates Federal stream by excavation or filling activities.

fishway requirements when protection of water quality. The law also National Environmental Policy Act of prescriptions are made to address provides for assessment of injury, 1969 (NEPA) (42 U.S.C. 4321-4347) anadromous fish passage and during the destruction, or loss of natural resources re-licensing of existing hydroelectric caused by discharge of pollutants. The NEPA requires an environmental review process of all Federal actions.

dams when anadromous species are Of major significance is Section 404 of This includes preparation of an considered. the FWPCA, which prohibits the discharge of dredged or fill material into environmental impact statement for Anadromous Fish Conservation Act (16 major Federal actions that may affect the U.S.C. 757a-757f) as Amended navigable waters without a permit.

quality of the human environment. Less Navigable waters are defined under the rigorous environmental assessments are This law authorizes the Secretaries of FWPCA to include all waters of the reviewed for most other actions, while Interior and Commerce to enter into cost United States, including the territorial some actions are categorically excluded sharing with states and other non- seas and wetlands adjacent to such from formal review. These reviews Federal interests for the conservation, waters. The permit program is provide an opportunity for the agency development, and enhancement of the administered by the Army Corps of and the public to comment on projects nation's anadromous fish. Engineers (ACOE). The Environmental that may impact fish and wildlife Investigations, engineering, biological Protection Agency (EPA) may approve habitat.

surveys, and research, as well as the delegation of Section 404 permit construction, maintenance, and authority for certain waters (not Coastal Zone Management Act (16 operations of hatcheries, are authorized. including traditional navigable waters) U.S.C. 1451-1464) and Estuarine Areas This Act was last authorized in 2002, to a state agency; however, the EPA Act which provided 5 million dollars for the retains the authority to prohibit or deny Congress passed policy on values of fiscal years 2005 and 2006 (Pub. L. 107- a proposed discharge under Section 404 estuaries and coastal areas through these 372). There was an attempt to of the FWPCA. Acts. Comprehensive planning reauthorize the Act in 2012; however, The FWPCA (Section 401) also programs, to be carried out at the state this action has not yet been authorized. authorizes programs to remove or limit level, were established to enhance, Fish and Wildlife Coordination Act the entry of various types of pollutants protect, and utilize coastal resources.

(FWCA) (16 U.S.C. 661-666) into the nation's waters. A point source Federal activities must comply with the permit system was established by the individual state programs. Habitat may The FWCA is the primary law EPA and is now being administered at be protected by planning and regulating providing for consideration of fish and the state level in most states. This development that could cause damage wildlife habitat values in conjunction system, referred to as the National to sensitive coastal habitats.

with Federal water development Pollutant Discharge Elimination System activities. Under this law, the (NPDES), sets specific limits on Federal Land Management and Other Secretaries of Interior and Commerce discharge of various types of pollutants Protective Designations may investigate and advise on the from point source outfalls. A non-point Protection and good stewardship of effects of Federal water development source control program focuses lands and waters managed by Federal projects on fish and wildlife habitat. primarily on the reduction of agencies, such as the Departments of Such reports and recommendations, agricultural siltation and chemical Defense, Energy and Interior (National which require concurrence of the state pollution resulting from rain runoff into Parks and National Wildlife Refuges, as fish and wildlife agency(ies) involved, the nation's streams. This effort well as state-protected park, wildlife must accompany the construction currently relies on the use of land and other natural areas), contributes to agency's. request for congressional management practices to reduce surface the health of nearby aquatic systems authorization, although the construction runoff through programs administered that support important river herring

48966 Federal Register / Vol. 78, No. 155 / Monday, August 12, 2013 / Notices spawning and nursery habitats. Relevant identified in the critical habitat harvest river herring resources, examples include the Great Bay, Rachel designation for Atlantic salmon was cooperatively manage municipal Carson's and ACE Basin National freshwater and estuary migration sites fisheries. Each town must submit an Estuarine Research Reserves, with abundant, diverse native fish annual harvesting plan to DMR for Department of Defense properties in the communities to serve as a protective approval that includes a 3-day per week Chesapeake Bay, and many National buffer against predation. Co-evolved escapement period or biological Wildlife Refuges. diadromous fish species such as equivalent to ensure conservation of the Marine Protection, Research and alewives and blueback herring are resource. In some instances, an included in this native fish community. escapement number is calculated and Sanctuaries Act of 1972 (MPRSA), Titles I and III and the Shore Protection Act of Because the ESA also requires that any the harvester passes a specific number Federal agency that funds, authorizes, or upstream to meet escapement goals.

1988 (SPA) carries out an action ensure that the River herring runs not controlled by a The MPRSA protects fish habitat action does not adversely modify or municipality and not approved as through establishment and maintenance destroy designated critical habitat, the sustainable by the ASMFC River Herring of marine sanctuaries. The MPRSA and impacts to alewife and blueback herring and American Shad Management Board, the SPA regulate ocean transportation populations must be considered during as required under Amendment 2, are and dumping of dredge materials, consultation with NMFS to ensure that closed. Each run and harvest location is sewage sludge, and other materials. Atlantic salmon critical habitat is not unique, either in seasonality, fish Criteria that the ACOE uses for issuing adversely affected by a Federal action. composition, or harvesting limitations.

permits include considering the effects Some runs have specific management dumping has on the marine Atlantic Sturgeon ESA Listing plans that require continuous environment, ecological systems and In 2012, five distinct population escapement and are more restrictive fisheries resources. segments of Atlantic sturgeon were than the 3-day closed period. Others Atlantic Salmon ESA Listing and listed under the ESA (77 FR 5914; 77 FR have closed periods shorter than the 3-Critical Habitat Designation 5880). The Chesapeake Bay, New York day requirement, but require an Bight, Carolina, and South Atlantic escapement number, irrespective of the In 2009, the Gulf of Maine (GOM) DPS DPSs of Atlantic sturgeon are listed as of Atlantic salmon was listed as number harvested during the season.

endangered, while the Gulf of Maine Maine increased the weekly fishing endangered under the Endangered DPS is listed as threatened.

Species Act (74 FR 29344). The GOM closure from a 24-hour closure in the Measures to improve habitats and 1960s to a 48-hour closure beginning in DPS includes all anadromous Atlantic reduce impacts to Atlantic sturgeon may salmon whose freshwater range occurs 1988. The closed period increased to 72 directly or indirectly benefit river hours beginning in 1995 to protect in the watersheds from the herring. Atlantic sturgeon are Androscoggin River northward along spawning fish. Most towns operate a anadromous; adults spawn in freshwater weir at one location on each stream and the Maine coast to the Dennys River. in the spring and early summer and Concurrently in 2009, critical habitat prohibit fishing at any other location on migrate into estuarine and marine the stream. The state landings program was designated for the Atlantic salmon waters where they spend most of their GOM DPS pursuant to section 4(b)(2) of compiles in-river landings of river lives. As with Atlantic salmon, many of herring from mandatory reports the ESA (74 FR 29300; August 10, 2009). the habitats that Atlantic sturgeon The critical habitat designation includes occupy are also habitats that river provided by the municipality under 45 specific areas occupied by Atlantic each municipal harvest plan or they lose herring use for spawning, migration and exclusive fishing rights. The state salmon at the time of listing, and juvenile rearing. The geographic range includes approximately 12,160 miles permitted 22 municipalities to fish for in which river herring may benefit from river herring in 2011. The river specific (19,600 kin) of perennial river, stream, Atlantic sturgeon ESA protections and estuary habitat and 308 square management plans require the extends from the Maine/Canada border remaining municipalities to close their miles (495 sq kin) of lake habitat within to Florida. Therefore, any protection the range of the GOM DPS in the State runs for conservation and not harvest.

measures within this range such as There are several reasons for these state/

of Maine. improved fish passage or a reduction of Measures to improve habitats and municipal imposed restrictions on the water withdrawals may also provide a reduce impacts to Atlantic salmon as a fishery. Many municipalities voluntarily benefit to river herring. restrict harvest to increase the numbers result of the ESA listing may directly or indirectly benefit river herring. Atlantic State Regulations of fish that return in subsequent years.

salmon are anadromous and spend a A historical review of state Some of these runs are large but have portion of their life in freshwater and regulations was compiled and published the potential to become even larger. The the remaining portion in the marine in Volume I of the stock assessment. commercial fishery does not exploit the environment. River herring occupy a lot The following section on state estimated 1.5 to 2.0 million river herring of the same habitats as listed Atlantic regulations includes current that return to the East Machias River salmon for spawning, breeding, feeding, requirements only and is cited from annually. These regulations have been growth and maturity. Therefore, Volume I of the assessment as compiled approved through a sustainable fisheries protection measures such as improved by Dr. Gary Nelson and Kate Taylor management plan, as required under fish passage or reduced discharge (ASMFC, 2012). Otherwise, updates are ASMFC Amendment 2 to the Shad and permits may benefit river herring. provided by Kate Taylor, supplemental River Herring FMP (Taylor, Pers.

The critical habitat designation information from state river herring Comm., 2013).

provides additional protections beyond plans or state regulations. Recreational fishermen are allowed to classifying a species as endangered by fish for river herring year-round. The preserving the physical and biological Maine limit is 25 fish per day and gear is features essential for the conservation of In Maine, the Department of Marine restricted to dip net and hook-and-line.

the species in designated waters in Resources (DMR), along with Recreational fishermen may not fish in Maine. One of the biological features municipalities granted the rights to waters, or in waters upstream, of a

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48967 municipality that owns fishing rights. fishery in Federal waters. RIDFW will comm). This will allow bait shops to Recreational fishermen are not required also implement a mandatory permitting sell, and fishermen to possess, river to report their catch. The MRFSS and process that will require vessels wanting herring for bait that was harvested from MRIP programs do sample some of these to fish in the Rhode Island waters a state whose fishery remains open, as fishermen based on results queried from Atlantic herring fishery to, amongst an ASMFC approved sustainable fishery the database. Recreational fishing for other requirements, integrate in to the (Taylor, Pers. Comm).

river herring in Maine is limited and University of Massachusetts Dartmouth, landings are low. These regulations Potomac River Fisheries Commission School for Marine Science and (PRFC)/District of Columbia have been approved through a Technology, river herring bycatch sustainable fisheries management plan, monitoring program to ensure The PRFC regulates only the as required under ASMFC Amendment monitoring of the fishery and minimize mainstem of the river, while the 2 to the Shad and River Herring FMP bycatch. As of Jan 1, 2013, there is a tributaries on either side are under (Taylor, Pers. Comm., 2013). prohibition to land, catch, take, or Maryland and Virginia jurisdiction. The attempt to catch or take river herring District of Columbia's Department of the New Hampshire Environment (DDOE) has authority for which is a continuation of measures that The current general regulations are: RIDFW has had in place since 2006 the Potomac River to the Virginia shore (1) No person shall take river herring, when a moratorium was originally and other waters within District of alewives and blueback herring, from the established (Taylor, Pers. comm., 2013). Columbia. Today, the river herring waters of the state, by any method, harvest in the Potomac is almost between sunrise Wednesday and sunrise Connecticut exclusively taken by pound nets. In Thursday of any week; (2) any trap or Since April 2002, there has been a 1964, licenses were required to weir used during a specified time prohibition on the commercial or commercially harvest fish in the period, shall be constructed so as to recreational taking of migratory Potomac River. After Maryland and allow total escapement of all river alewives and blueback herring from all Virginia established limited entry herring; and (3) any river herring taken marine waters and most inland waters. fisheries in the 1990s, the PRFC by any method during the specified time As of January 1, 2012, commercial and responded to industry's request and, in period shall be immediately released recreational harvest of river herring was 1995, capped the Potomac River pound back into the waters from which it was prohibited in Connecticut, as required net fishery at 100 licenses. As of January taken. Specific river regulations are: by ASMFC Amendment 2 to the Shad 1, 2010, harvest of river herring was Taylor River-from the railroad bridge and River Herring FMP (Taylor, Pers. prohibited in the Potomac River, with a to the head of tide dam in Hampton Comm., 2013). minimal bycatch provision of 50 lb (22 shall be closed to the taking of river kg) per licensee per day for pound nets.

herring by netting of any method; and New York These regulations have been approved Squamscott River-during April, May Current regulations allow for a through a sustainable fisheries and June, the taking of river herring in restricted river herring commercial and management plan, as required under the Squamscott River and its tributaries recreational fishery in the Hudson River ASMFC Amendment 2 to the Shad and from the Rt. 108 Bridge to the Great Dam and tributaries, while all other state River Herring FMP.

in Exeter is open to the taking of river waters prohibit river herring fisheries.

herring by netting of any method only These regulations have been approved Virginia on Saturdays and Mondays, the daily through a sustainable fisheries Virginia's Department of Game and limit shall be one tote per person ("tote" management plan, as required under Inland Fisheries (VDGIF) is responsible means a fish box or container measuring ASMFC Amendment 2 to the Shad and for the management of fishery resources 31.5 in (80.01cm) x 18 in (45.72 cm) x River Herring FMP. in the state's inland waters. As of 11.5 in (29.21cm)) and the tote shall January 1, 2008, possession of alewives New Jersey/Delaware and blueback herring was prohibited on have the harvester's coastal harvest permit number plainly visible on the As of January 1, 2012, commercial rivers draining into North Carolina (4 outside of the tote. These regulations harvest of river herring was prohibited VAC 15-320-25). The Virginia Marine have been approved through a in New Jersey and Delaware, as required Resources Commission (VMRC) is sustainable fisheries management plan, by ASMFC Amendment 2 to the Shad responsible for management of fishery as required under ASMFC Amendment and River Herring FMP. Additionally, resources within the state's marine 2 to the Shad and River Herring FMP. only commercial vessels fishing waters. As of January 1, 2012, exclusively in Federal waters while commercial and recreational harvest of Massachusetts operating with a valid Federal permit river herring was prohibited in all As of January 1, 2012, commercial for Atlantic mackerel and/or Atlantic waters of Virginia, as required by and recreational harvest of river herring herring may possess river herring up to ASMFC Amendment 2 to the Shad and was prohibited in Massachusetts, as a maximum of five percent by weight of River Herring FMP. Additionally, it is required by ASMFC Amendment 2 to all species possessed (Taylor, Pers. unlawful for any person to possess river the Shad and River Herring FMP Comm.). herring aboard a vessel on Virginia tidal (Taylor, Pers. Comm., 2013). The waters, or to land any river herring in Maryland exception is for federally permitted Virginia (4 VAC 20-1260-30).

vessels which are allowed to land up to As of January 1, 2012, commercial 5 percent of total bait fish per trip harvest of river herring was prohibited North Carolina (Taylor, Pers. Comm., 2013). in Maryland, as required by ASMFC A no harvest provision for river Amendment 2 to the Shad and River herring, commercial and recreational, Rhode Island Herring FMP. However, an exception is within North Carolina was approved in The Rhode Island Division of Fish provided for anyone in possession of 2007. A limited research set aside of and Wildlife (RIDFW) will implement a river herring as bait, as long as a receipt 7,500 lb (3.4 mt) was established, and to 5 percent bycatch allowance for Federal indicating where the herring was implement this harvest, a Discretionary vessels fishing in the Atlantic herring purchased is in hand (Taylor, Pers. Herring Fishing Permit (DHFP) was

48968 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013 / Notices created. Individuals interested in in Florida. As of January 1, 2012, states with regulations that have been participating had to meet the following harvest of river herring was prohibited, approved through a sustainable fisheries requirements: (1) Obtain a DHFP, (2) as required by ASMFC Amendment 2 to management plan, as required under harvest only from the Joint Fishing the Shad and River Herring FMP. ASMFC Amendment 2 to the Shad and Waters of Chowan River during the River Herring FMP. All other states had Tribal and First Nation Fisheries harvest period, (3) must hold a valid previously established moratoria or, as North Carolina Standard Commercial We have identified thirteen federally of January 1, 2012, harvest of river Fishing License (SCFL) or a Retired recognized East Coast tribes from Maine herring was prohibited, as required by SCFL, and (4) participate in statistical to South Carolina that have tribal rights ASMFC Amendment 2 to the Shad and information and data collection to sustenance and ceremonial fishing, River Herring FMP. However, river programs. Sale of harvested river and which may harvest river herring for herring are incidentally caught in herring had to be to a licensed and sustenance and ceremonial purposes several commercial fisheries, but the permitted River Herring Dealer. Each and/or engage in other river herring extent to which this is occurring has not permit holder was allocated 125-250 lb conservation and management been fully quantified. The New England (56-113 kg) for the 4-day season during activities. The Mashpee Wampanoag and Mid-Atlantic Fishery Management Easter weekend. These regulations were tribe is the only East Coast tribe that Councils have adopted measures for the approved through a sustainable fisheries voluntarily reported harvest numbers to Atlantic herring and mackerel fisheries management plan, as required under the State of MA that were incorporated intended to decrease incidental catch ASMFC Amendment 2 to the Shad and into the ASMFC Management Plan as and bycatch of alewife and blueback River Herring FMP. The North Carolina subsistence harvest. The reported herring. In the United States, thirteen Wildlife Resources Commission harvest for 2006 and 2008 ranged federally recognized East Coast tribes (NCWRC) has authority over the Inland between 1,200 and 3,500 fish per year, from Maine to South Carolina have Waters of the state. Since July 1, 2006, with removals coming from several tribal rights to sustenance and harvest of river herring, greater than 6 rivers. Aside from the harvest reported ceremonial fishing, and may harvest inches (15.24 cm) has been prohibited by ASMFC for the Mashpee Wampanoag river herring for sustenance and in the inland waters of North Carolina's tribe, information as to what tribes may ceremonial purposes and/or engage in coastal systems. harvest river herring for sustenance and/ other river herring conservation and or ceremonial purposes is not available. management activities. We have further South Carolina Letters have been sent to all 13 evaluated the existing international, In South Carolina, the South Carolina potentially affected tribes to solicit any Federal, and state management Division of Natural Resources (SCDNR) input they may have on the measures in the qualitative threats manages commercial herring fisheries conservation status of the species and/ assessment section below.

using a combination of seasons, gear or health of particular riverine restrictions, and catch limits. Today, the populations, tribal conservation and E. Other Naturalor Manmade Factors commercial fishery for blueback herring management activities for river herring, Affecting the Continued Existence of the has a 10-bushel daily limit (500 lb (226 biological data for either species, and Species kg)) per boat in the Cooper and Santee comments and/or concerns regarding Competition Rivers and the Santee-Cooper the status review process and potential Rediversion Canal and a 250-lb-per-boat implications for tribal trust resources Intra- and inter-specific competition (113 kg) limit in the Santee-Cooper and activities. To date, we have not were considered as potential natural lakes. Seasons generally span the received any information from any threats to alewife and blueback herring.

spawning season. All licensed tribes. The earlier spawning time of alewife fishermen have been required to report may lead to differences in prey selection Summary and Evaluation for Factor D from blueback herring, given that they their daily catch and effort to the SCDNR since 1998. As described in Factor A, there are become more omnivorous with The recreational fishery has a 1- multiple threats to habitat that have increasing size (Klauda et al., 1991a).

bushel (49 lb (22.7 kg)) fish aggregate affected and may continue to affect river This could lead to differences in prey daily creel for blueback herring in all herring including dams/culverts, selection given that juvenile alewife rivers; however, very few recreational dredging, water quality, water would achieve a greater age and size anglers target blueback herring. These withdrawals and discharge. However, earlier than blueback herring. Juvenile regulations have been approved through many of these threats are being American shad are reported to focus on a sustainable fisheries management addressed to some degree through different prey than blueback herring plan, as required under ASMFC existing Federal legislation such as the (Klauda et al., 1991b). However, Smith Amendment 2 to the Shad and River Federal Water Pollution Control Act, and Link (2010) found few differences Herring FMP. also known as the Clean Water Act, the between American shad and blueback Coastal Zone Management Act, the herring diets across geographic areas Georgia Rivers and Harbors Act, the FPA, and size categories; therefore, The take of blueback herring is illegal Marine Protection, Research and competition between these two species in freshwater in Georgia. As of January Sanctuaries Act of 1972, the Shore may be occurring. Cannibalism has been 1, 2012, harvest of river herring was Protection Act of 1988, EFH observed (rarely) in landlocked systems prohibited in Georgia, as required by designations for other species and ESA with alewife. Additionally, evidence of ASMFC Amendment 2 to the Shad and listings for Atlantic salmon and Atlantic hybridization exists between alewife River Herring FMP. sturgeon. and blueback herring, but the Commercial harvest of alewife and implications of this are unknown.

Florida blueback herring is occurring in Canada Competition for habitat or resources has The St. Johns River, Florida, harbors with regulations, closures, and quotas in not been documented with alewife/

the southernmost spawning run of effect. In the United States, commercial blueback herring hybrids, as there is blueback herring. There is currently no harvest of alewife and blueback herring little documentation of hybridization in active management of blueback herring is also currently occurring in a few published literature, but given the

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48969 unknowns about their life history, it is been deposited into five to ten systems. morphological differences between the possible that competition between non- Many of the recent efforts have been two life history variants provide hybrids and hybrids could be occurring. within system, moving fish upstream substantial evidence that upon past multiple obstructions to the becoming landlocked, landlocked Artificial Propagation and Stocking headwater spawning habitat. Rhode herring populations become largely Genetics data have shown that Island's Department of Environmental independent and separate from stocking alewife and blueback herring Management (DEM) has been stocking anadromous populations. Landlocked within and out of basin in Maine has the Blackstone River with adult populations and anadromous had an impact on the genetic groupings broodstock which was acquired from populations occupy largely separate within Maine (Bentzen, 2012, existing Rhode Island river herring runs ecological niches, especially in respect unpublished data); however, the extent and other sources out of state. In April to their contribution to freshwater, to which this poses a threat to river 2012, over 2,000 river herring pre- estuary and marine food-webs herring locally or coast-wide is spawned adults were stocked into the (Palkovacs and Post, 2008). Thus, the unknown. Stocking river herring Blackstone River. A small number of existence of landlocked life forms does directly impacts a specific river/ alewife (200-400 fish) were stocked in not appear to pose a significant threat to watershed system for river herring in the Bronx River, NY, in 2006 and 2007 the anadromous forms.

that it can result in passing fish above from Brides Brook in East Lyme, CT.

barriers into suitable spawning and Furthermore, an experimental stocking Interbreeding Among Alewife and rearing habitat, expanding populations program exists in Virginia where Blueback Herring (Hybridization) into other watersheds, and introducing hatchery broodstock are marked and Recent genetic studies indicate that fish to newly accessible spawning stocked into the Kimages Creek, a hybridization may be occurring in some habitat. tributary to the James River. A total of instances among alewife and blueback The alewife restoration program in 319,856 marked river herring fry were herring where populations overlap Merrymeeting Bay, Maine, focuses on stocked in this creek in 2011. (discussed in the River Herring Stock stocking lakes and ponds in the The Edenton National Fish Hatchery Structure Working Group Report, Sebasticook River watershed and Seven (NFH) in North Carolina and the NMFS, 2012a). Though interbreeding Mile Stream drainage. The highest Harrison Lake NFH in Virginia have among closely related species is number of stocked fish was 2,211,658 in propagated blueback herring for 2009 in the Sebasticook River and uncommon, it does occasionally occur restoration purposes. Edenton NFH is (Levin, 2002). Most often, different 93,775 in 2008 in the Kennebec River. currently rearing blueback herring for reproductive strategies, home ranges, The annual stocking goal of the stocking in Indian Creek and Bennett's restoration projects range from 120,000 and habitat differences of closely related Creek in the Chowan River watershed in species either prevent interbreeding, or to 500,000 fish, with most fish stocked Virginia. This is a pilot project to see if keep interbreeding at very low levels. In in the Androscoggin and Sebasticook hatchery contribution makes a watersheds. The Union River fishery in circumstances where interbreeding does significant improvement in runs of occur, natural selection often keeps Ellsworth, Maine, is sustained through returning adults (S. Jackson, USFWS, the stocking of adult alewives above the hybrids in check because hybrids are Pers. comm., 2012). Artificial less fit in terms of survival or their hydropower dam at the head-of-tide. propagation through the Edenton NFH Fish passage is not currently required at ability to breed successfully (Levin, for the pilot program in the Chowan 2002). Other times, intermediate this dam, but fish are transported River watershed is intended for around the dam to spawning habitat in environmental conditions can provide restoration purposes, and it is not two lakes. The annual adult stocking an environment where hybrids can thought that negative impacts to rate (from 2011 forward) is 150,000 fish. thrive, and when hybrids breed with the anadromous blueback herring Adult river herring are trapped at a member of the parent species, this can populations will be associated with lead to "mongrelization" of one or both commercial harvest sites below the dam these efforts.

and trucked to waters upstream of the parent species; a process referred to as dam. The highest number of stocked Landlocked Alewife and Blueback introgressive hybridization (Arnold, fish in the Union River was 1,238,790 in Herring 1997). Introgressive hybridization can 1986. In the Penobscot River watershed, As noted above, alewives and also occur as a result of introductions of over 48,000 adult fish were stocked into blueback herring maintain two life closely related species, or man-made or lakes in 2012, using fish collected from history variants; anadromous and natural disturbances that create the Kennebec (39,650) and Union Rivers landlocked. It is believed that they environments more suitable for the (8,998). The New Hampshire Fish and diverged relatively recently (300 to hybrid offspring than for the parents Game stocks river herring into the 5,000 years ago) and are now discrete (e.g., the introduction of mallards has Nashua River, the Pine Island Pond, and from each other. Landlocked alewife led to the decline of the American black the Winnisquam Lake using fish from populations occur in many freshwater duck through hybridization and various rivers which have included the lakes and ponds from Canada to North introgression) (Anderson, 1949; Rhymer, Connecticut, Cocheco, Lamprey, Carolina as well as the Great Lakes 2008).

Kennebec, and Androscoggin Rivers. (Rothschild, 1966; Boaze & Lackey, Though evidence has come forward MA Division of Marine Fisheries (DMF) 1974). Landlocked blueback herring that indicates that some hybridization conducts a trap and transport stocking occur mostly in the southeastern United may be occurring between alewife and program for alewife and blueback States and the Hudson River drainage. blueback herring, there is not enough herring. Prior to the moratorium in the At this time, there is no substantive evidence to conclude whether or not state, the program transported between information that would suggest that hybridization poses a threat to one or 30,000 and 50,000 fish per year into 10- landlocked populations can or would both species of river herring. Most 15 different systems. Since the revert back to an anadromous life importantly, there is not enough moratorium, effort has been reduced to history if they had the opportunity to do evidence to show whether hybrids protect donor populations and so (Gephard and Jordaan, Pers. comm., survive to maturity and, if so, whether approximately 20,000 fish per year have 2012). The discrete life history and they are capable of breeding with each

48970 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices other or breeding with either of the rockfish, and eulachon) and East Coast its range or for individual stock parent species. (e.g., Atlantic sturgeon, cusk, Atlantic complexes. When a Team member chose wolffish), and the River Herring SRT either N/A (Not Applicable) or 0 Summary and Evaluation of Factor E developed a qualitative ranking system (Unknown) for a threat, all 5 likelihood The potential for inter- and intra- that was adapted from these types of points had to be assigned to that rank specific competition has been qualitative analyses. The results from only. Qualitative descriptions of ranks investigated with respect to alewife and the threats assessment have been for the threats listed for alewife and blueback herring. Differences have been organized and described according to blueback herring (Table 1, 2) are:

observed in the diet activity patterns the above mentioned section 4(a)(1)

• N/A-Not Applicable.

and in spawning times of anadromous factors. They were used in combination

• 0-Unknown.

alosids, and this may reduce inter- and with the results of the extinction risk

• 1 Low-It is likely that this threat intra- specific competition. However, it modeling to make a determination as to is not significantly affecting the species is possible that competition is whether listing is warranted. now and into the foreseeable future, and occurring, as similarities in prey choice When ranking each threat, Team that this threat is limited in geographic have been identified. Stocking is a tool members considered how various scope or is localized within the species/

that managers have used for hundreds of demographic variables (e.g., abundance, stock complex' range.

years with many different species of population size, productivity, spatial

• 2 Moderately Low-Threat falls fish. This tool has been used as a means structure and genetic diversity) may be between rankings 1 and 3.

of supporting restoration (e.g., passing affected by a particular threat. While

• 3 Moderate-It is likely that this fish above barriers into suitable Factor D, "inadequacy of existing threat has some effect on the species spawning and rearing habitat, regulatory mechanisms," is a different now and into the foreseeable future, and expanding populations into other type of factor, the impacts on the it is widespread throughout the species/

watersheds, and introducing fish to species resulting from unregulated or stock complex' range.

newly accessible spawning habitat). In inadequately regulated threats should be

• 4 Moderately High-Threat falls addition, stocking has been used to evaluated in the same way as the other between rankings 3 and 5.

introduce species to a watershed for four factors. e 5 High-It is likely that this threat recreational purposes. Stocking of river is significantly affecting the species now herring has occurred for many years in QTA Methods and into the foreseeable future, and it is Maine watersheds, but is less common All nine SRT members conducted an widespread in geographic scope and throughout the rest of the range of both independent, qualitative ranking of the pervasive throughout the species/stock species. Stocking in the United States severity of each of the 22 identified complex' range.

has consisted primarily of trap and threats to alewives and blueback The SRT identified and ranked 22 truck operations that move fish from herring. NERO staff developed fact threats to both species both rangewide one river system to another or over an sheets for the SRT that contained and for the individual stock complexes.

impassible dam. Artificial propagation essential information about the Threats included dams and barriers, of river herring is not occurring to a particular threats under each of the five dredging, water quality and water significant extent, though blueback ESA section 4(a)(1) factors, attempts to withdrawals, climate change/variability, herring are being reared on a small scale ameliorate these threats, and how the harvest (both directed and incidental),

for experimental stocking in North threats are or may be affecting both disease, predation, management Carolina. species. These fact sheets were reviewed internationally, federally, and at the We have considered natural or by various experts within NMFS to state level, competition, artificial manmade factors that may affect river ensure that they contained all of the best propagation and stocking, hybrids, and herring, including competition, artificial available information for each of the from landlocked populations.

propagation and stocking, landlocked factors.

river herring, and hybrids. Several Team members ranked the threats QTA Results potential natural or manmade threats to separately for both species at a The SRT unequivocally identified river herring were identified, and we rangewide scale and at the individual dams and barriers as the most important have considered the effects of these stock complex level. Each Team threat to alewife and blueback herring potential threats further in the member was allotted five likelihood populations both rangewide and across qualitative threats assessment described points to rank each threat. Team all stock complexes (the qualitative below. members ranked the severity of each ranking for dams and barriers was threat through the allocation of these between moderately high and high).

Threats Evaluation for Alewife and five likelihood points across five ranks Incidental catch, climate change, Blueback Herring ranging from "low" to "high." Each dredging, water quality, water During the course of the Status Team member could allocate all five withdrawal/outfall, predation, and Review for river herring, 22 potential likelihood points to one rank or existing regulation were among the threats to alewife and blueback herring distribute the likelihood points across more important threats after dams for were identified that relate to one or several ranks to account for any both species, and for all stock more of the five ESA section 4(a)(1) uncertainty. Each individual Team complexes (qualitative rankings for factors identified above. The SRT member distributed the likelihood these threats ranged between conducted a qualitative threats points as he/she deemed appropriate moderately low and moderate). Water assessment (QTA) to help evaluate the with the condition that all five quality, water withdrawal/outfall, significance of the threats to both likelihood points had to be used for predation, climate change and climate species of river herring now and into the each threat. Team members also had the variability were generally seen as greater foreseeable future. NMFS has used option of ranking the threat as "0" to threats to both species in the southern qualitative analyses to estimate indicate that in their opinion there were portion of their ranges than in the extinction risk in previous status insufficient data to assign a rank, or "N/ northern portion of their ranges. In reviews on the West Coast (e.g., Pacific A" if in their opinion the threat was not addition, the Team identified salmon, Pacific herring, Pacific hake, relevant to the species either throughout commercial harvest as being notably

Federal Register / Vol. 78, No. 155 / Monday, August 12, 2013 / Notices 48971 more important in Canada than in the and South Atlantic diverged the most model to complete a biomass estimate.

United States. The results of the threats from the other stock complexes with Therefore, it is difficult to accurately analysis for alewives are presented in respect to certainty of threats. In Canada quantify the declines from historical Tables 1-5 and Figure 3. The results of there was more certainty surrounding biomass to present-day biomass, though the threats analysis for blueback herring the threats of climate change and significant declines have been noted.

are presented in Tables 6-10 and Figure climate variability for both species, and Studies from Maine show that dams

4. less certainty surrounding the threat of have reduced accessible habitat to a QTA Conclusion directed commercial harvest and fraction of historical levels, 5 percent for incidental catch for alewives compared alewives and 20 percent for blueback The distribution of rankings across to the certainty surrounding these herring (Hall et al., 2011).

threat levels provides a way to evaluate threats for the other stock complexes. In Rangewide, for alewife and blueback certainty in the threat level for each of the mid-Atlantic for alewives and herring, no other threats rose to the level the threats identified. The amount of South-Atlantic for bluebacks, there was of dams, but several other stressors certainty for a threat is a reflection of more uncertainty surrounding climate ranked near the moderate threat level.

the amount of evidence that links a variability and climate change The Team ranked incidental catch, particular threat to the continued compared to the certainty surrounding water quality, and predation as threats survival of each species. For threats these threats for the other stock likely to have some effect on the species with more data, there tended to be more complexes. now and into the foreseeable future that certainty surrounding the threat level, Based on the Team member rankings, are widespread throughout the species' whereas threats with fewer data tended dams and other barriers present the range. Incidental catch is primarily from to have more uncertainty. The same greatest and most persistent threat fisheries that use small-mesh mobile holds true for datasets that were limited rangewide to both blueback herring and gear, such as bottom and midwater over space and/or time. alewife (Tables 12-13). Dams and trawls. Sources of water quality The results of the threats assessment culverts block access to historical problems vary from river to river and rangewide and for all stock complexes migratory corridors and spawning are therefore unique to each of the stock reveal strong agreement and low locations, in some instances, even when complexes. And finally, predation by uncertainty among the reviewers that fish passage facilities are present. striped bass, seals, double-crested dams and barriers are the greatest threat Centuries of blocked and reduced access cormorants (and other fish-eating avian to both alewives and blueback herring. to spawning and rearing habitat have species, e.g., northern gannets) and There was also strong agreement that resulted in decreased overall production other predators is known to exist, but tribal fisheries, scientific monitoring, potential of watersheds along the data are lacking on the overall and educational harvest currently pose Atlantic coast for alewives and blueback magnitude. Overall, the degree of little threat to the species. For the herring (Hall et al., 2012). This reduced certainty associated with these mid-threats of state, Federal and production potential has likely been one level threats is much lower, primarily international management, dredging, of the main drivers in the decreased due to lack of information on how these climate change, climate variability, abundance of both species. The recent stressors are affecting both species.

predation, and incidental catch, there ASMFC Stock Assessment (2012) The SRT's qualitative rankings and was more uncertainty. attempted to quantify biomass estimates analysis of threats for alewife rangewide Among alewife and blueback stock for both alewife and blueback herring and for each stock complex:

complexes, Canada, the Mid-Atlantic, but was unable to develop an acceptable BILLING CODE 3510-22-P

48972 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices High 5 4

Co 3

Low 2 1

0 0 Canada

• 1Northern New England

• Southern New England

• Mid-Atlantic

• Overall Range Figure 3. Median qualitative ranking of threats to alewives range-wide and for each stock complex.

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48973 Table 1. Qualitative ranking of threats for the alewife rangewide. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat.

N=number of Team members who ranked the threat between 1 and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 4.3 0.7 5 3-5 9 Water Quality (chemical) 2.8 1.0 3 1-5 9 Incidental Catch 2.7 0.9 3 1-5 9 Predation 2.6 1.1 3 1-5 9 Water Withdrawal/Outfall (physical and temp.) 2.4 0.8 2 1-5 9 Dredging 2.4 1.0 2 1-4 9 Climate change 2.4 0.9 3 1-4 8 Climate variability 2.3 1.1 2 1-5 9 Federal Management 2.3 1.1 2 1-5 9 International Management 2.3 1.1 2 1-5 9 State Management 2.2 1.2 1 1-5 9 Directed Commercial Harvest 1.8 0.8 2 1-3 9 Competition 1.6 0.7 1 1-4 9 Artificial Propagation and Stocking 1.5 0.7 1 1-3 9 Hybrids 1.5 0.7 1 1-3 2 Recreational Harvest 1.4 0.6 1 1-3 9 Tribal/First Nation Fisheries Management 1.3 0.5 1 1-3 7 Disease 1.3 0.4 1 1-2 8 Landlocked Populations 1.2 0.5 1 1-3 8 Tribal/First Nation Fisheries Utilization 1.2 0.4 1 1-2 8 Scientific Monitoring 1.0 0.2 1 1-2 9 Educational Harvest 1.0 0.1 1 1-2 9

48974 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Table 2. Qualitative ranking of threats for the Canadian stock complex of alewife. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks:

1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between 1 and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 4.0 0.9 5 2-5 8 State Management 2.4 0.9 2 1-4 6 Incidental Catch 2.4 1.2 1 1-5 6 Federal Management 2.4 0.9 2 1-4 6 Water Withdrawal/Outfall (physical and temp.) 2.3 0.7 2 1-3 6 Directed Commercial Harvest 2.2 0.9 2 1-4 8 International Management 2.2 0.9 2 1-4 8 Water Quality (chemical) 2.1 0.7 2 1-3 7 Predation 2.1 1.0 2 1-5 8 Dredging 2.0 0.7 2 1-4 6 Climate variability 1.9 0.9 2 1-5 8 Climate change 1.6 0.7 1 1-4 8 Hybrids 1.5 0.7 1 1-3 2 Competition 1.4 0.5 1 1-3 9 Disease 1.3 0.5 1 1-2 7 Artificial Propagation and Stocking 1.3 0.5 1 1-3 7 Tribal/First Nation Fisheries Management 1.2 0.4 1 1-2 5 Recreational Harvest 1.2 0.4 1 1-2 6 Tribal/First Nation Fisheries Utilization 1.2 0.4 1 1-2 6 Landlocked Populations 1.1 0.3 1 1-2 7 Scientific Monitoring 1.0 0.2 1 1-2 6 Educational Harvest 1.0 0.0 1 1 6

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48975 Table 3. Qualitative ranking of threats for the Northern New England stock complex of alewife. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between I and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 4.3 0.7 5 3-5 9 Incidental Catch 2.9 0.8 3 1-5 7 Water Withdrawal/Outfall (physical and temp.) 2.5 0.9 3 1-5 8 Dredging 2.4 0.9 2,3 1-4 8 State Management 2.4 1.1 2 1-5 9 Predation 2.4 1.2 2 1-5 9 Federal Management 2.4 1.1 2 1-5 9 International Management 2.2 0.9 2 1-4 9 Water Quality (chemical) 2.1 1.0 1 1-5 9 Climate variability 2.0 1.0 2 1-5 9 Directed Commercial Harvest 1.9 0.9 1 1-4 9 Climate change 1.8 0.8 1 1-4 8 Artificial Propagation and Stocking 1.6 0.7 1 1-3 9 Hybrids 1.5 0.7 1 1-3 2 Competition 1.5 0.6 1 1-3 9 Recreational Harvest 1.3 0.5 1 1-3 9 Disease 1.3 0.5 1 1-2 8 Landlocked Populations 1.2 0.4 1 1-2 8 Tribal/First Nation Fisheries Management 1.2 0.4 1 1-2 7 Tribal/First Nation Fisheries Utilization 1.1 0.3 1 1-2 8 Scientific Monitoring 1.0 0.1 1 1-2 9 Educational Harvest 1.0 0.0 1 1 9

48976 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Table 4. Qualitative ranking of threats for the Southern New England stock complex of alewife. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between I and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 4.2 0,7 4 3-5 9 Incidental Catch 2.9 0.8 3 1-5 7 Water Withdrawal/Outfall (physical and temp.) 2.7 0.8 3 1-5 8 Water Quality (chemical) 2.5 0.9 3 1-5 9 Predation 2.5 1.1 2 1-5 9 Dredging 2.5 0.9 3 1-4 8 Federal Management 2.2 1.1 2 1-5 9 Climate variability 2.2 1.0 2 1-5 9 State Management 2.2 1.1 2 1-5 9 Climate change 2.2 1.0 1,3 1-4 8 International Management 2.0 0.8 2 1-4 9 Directed Commercial Harvest 1.7 0.8 1 1-3 9 Hybrids 1.5 0.7 1 1-3 2 Artificial Propagation and Stocking 1.5 0.6 1 1-3 9 Competition 1.4 0.6 1 1-3 9 Disease 1.3 0.5 1 1-2 8 Recreational Harvest 1.3 0.5 1 1-3 9 Landlocked Populations 1.2 0.4 1 1-2 8 Tribal/First Nation Fisheries Management 1.2 0.4 1 1-2 7 Tribal/First Nation Fisheries Utilization 1.2 0.4 1 1-2 8 Scientific Monitoring 1.0 0.1 1 1-2 9 Educational Harvest 1.0 0.0 1 1 9

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48977 Table 5. Qualitative ranking of threats for the Mid-Atlantic stock complex of alewife.

Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between I and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 3.8 1.0 4 3-5 9 Incidental Catch 2.9 0.8 3 1-5 7 Water Quality (chemical) 2.9 0.9 3 1-5 9 Water Withdrawal/Outfall (physical and temp.) 2.8 0.8 3 1-5 8 Climate change 2.7 1.2 3 1-5 8 Climate variability 2.6 1.2 2 1-5 9 Predation 2.5 1.1 2 1-5 9 Dredging 2.5 0.9 3 1-4 8 Federal Management 2.3 1.1 2 1-5 9 State Management 2.2 1.1 2 1-5 9 International Management 1.8 0.8 1 1-4 9 Directed Commercial Harvest 1.7 0.8 1 1-3 9 Hybrids 1.5 0.7 1 1-3 2 Artificial Propagation and Stocking 1.5 0.7 1 1-3 9 Competition 1.4 0.6 1 1-3 9 Disease 1.4 0.5 1 1-3 8 Recreational Harvest 1.3 0.5 1 1-3 9 Landlocked Populations 1.2 0.4 1 1-2 8 Tribal/First Nation Fisheries Management 1.2 0.4 1 1-3 7 Tribal/First Nation Fisheries Utilization 1.1 0.3 1 1-2 7 Scientific Monitoring 1.0 0.1 1 1-2 9 Educational Harvest 1.0 0.0 1 1 9

sqalttverninso Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48978 The SRT's qualitative rankings of threats for blueback herring rangewide and for each stock complex:

Figure 4. Median qualitative ranking of threats to blueback herring rangewide and for each stock complex.

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48979 Table 6. Qualitative ranking of threats for blueback herring rangewide. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: I- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between 1 and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 4.2 0.8 4,5 3-5 9 Water Quality (chemical) 2.8 1.0 3 1-5 9 Incidental Catch 2.7 0.9 3 1-5 9 Climate change 2.7 1.2 3,4 1-5 8 Predation 2.6 1.1 3 1-5 9 Climate variability 2.4 1.2 1,2,3 1-5 9 Water Withdrawal/Outfall (physical and temp.) 2.4 0.8 2 1-5 9 Dredging 2.4 1.0 2 1-4 9 Hybrids 2.4 1.0 3 1-4 2 Federal Management 2.3 1.1 2 1-5 9 International Management 2.3 1.1 2 1-5 8 State Management 2.2 1.1 1 1-5 9 Directed Commercial Harvest 1.8 0.8 1 1-3 9 Competition 1.5 0.6 1 1-3 9 Artificial Propagation and Stocking 1.5 0.7 1 1-3 9 Tribal/First Nation Fisheries Management 1.3 0.5 1 1-3 7 Recreational Harvest 1.3 0.5 1 1-3 9 Disease 1.3 0.5 1 1-2 8 Landlocked Populations 1.2 0.5 .1 1-3 7 Tribal/First Nation Fisheries Utilization 1.2 0.4 1 1-2 8 Scientific Monitoring 1.0 0.2 1 1-2 9 Educational Harvest 1.0 0.1 1 1-2 9

48980 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Table 7. Qualitative rankings of threats for the Canadian stock complex of blueback herring. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between I and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 3.9 0.9 4 2-5 8 Incidental Catch 2.4 1.2 1,3 1-5 6 State Management 2.4 0.9 2 1-4 6 Hybrids 2.4 1.0 3 1-4 2 Water Withdrawal/Outfall (physical and temp.) 2.4 0.6 2 1-3 6 Federal Management 2.4 0.9 2 1-4 6 Directed Commercial Harvest 2.4 0.8 3 1-4 8 Water Quality (chemical) 2.2 0.7 2 1-3 7 Climate variability 2.2 1.2 1 1-5 8 Predation 2.1 1.0 2 1-4 8 International Management 2.1 0.9 2 1-4 8 Dredging 2.0 0.7 2 1-3 6 Climate change 2.0 1.0 1 1-4 8 Competition 1.5 0.6 1 1-4 9 Recreational Harvest 1.3 0.5 1 1-2 6 Disease 1.3 0.5 1 1-2 7 Artificial Propagation and Stocking 1.3 0.5 1 1-3 7 Tribal/First Nation Fisheries Utilization 1.2 0.4 1 1-2 6 Tribal/First Nation Fisheries Regulation 1.2 0.4 1 1-3 5 Landlocked Populations 1.A 0.3 1 1-2 6 Scientific Monitoring 1.0 0.2 1 1-2 6 Educational Harvest 1.0 0.0 1 1 6

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48981 Table 8. Qualitative ranking of threats for the Northern New England stock complex of blueback herring. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between 1 and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 4.3 0.7 5 3-5 9 Incidental Catch 2.8 0.9 3 1-5 7 Dredging 2.6 1.0 3 1-4 8 Water Withdrawal/Outfall (physical and temp.) 2.5 0.9 3 1-5 8 State Management 2.4 1.1 2 1-5 9 Hybrids 2.4 1.0 3 1-4 2 Water Quality (chemical) 2.4 1.1 3 1-5 9 Predation 2.4 1.2 2 1-5 9 Federal Management 2.4 1.1 2 1-5 9 Climate variability 2.2 1.2 2 1-4 9 Climate change 2.1 1.0 2 1-4 8 International Management 2.0 0.9 2 1-4 9 Directed Commercial Harvest 1.9 0.9 1 1-3 9 Artificial Propagation and Stocking 1.6 0.7 1 1-3 9 Competition 1.5 0.7 1 1-3 9 Recreational Harvest 1.3 0.5 1 1-3 9 Disease 1.3 0.5 1 1-2 8 Landlocked Populations 1.2 0.4 1 1-2 7 Tribal/First Nation Fisheries Management 1.1 0.4 1 1-2 7 Tribal/First Nation Fisheries Utilization 1.1 0.3 1 1-2 8 Scientific Monitoring 1.0 0.1 1 1-2 9 Educational Harvest 1.0 0.0 1 1 9

48982 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Table 9. Qualitative ranking of threats for the Southern New England stock complex of blueback herring. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between I and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 4.3 0.7 4,5 3-5 9 Incidental Catch 2.8 0.9 3 1-5 7 Water Withdrawal/Outfall (physical and temp.) 2.6 0.8 2,3 1-5 8 Dredging 2.6 1.0 3 1-4 8 Water Quality (chemical) 2.6 1.0 3 1-5 9 Predation 2.4 1.1 2 1-5 9 Hybrids 2.4 1.0 3 1-4 2 Climate change 2.3 1.0 2 1-4 8 Climate variability 2.3 1.1 2 1-5 9 Federal Management 2.2 1.1 2 1-5 9 State Management 1.9 1.1 1 1-5 9 International Management 1.9 0.8 2 1-4 9 Directed Commercial Harvest 1.6 0.8 1 1-3 9 Artificial Propagation and Stocking 1.6 0.7 1 1-3 9 Competition 1.5 0.7 1 1-4 9 Disease 1.3 0.5 1 1-2 8 Recreational Harvest 1.2 0.4 1 1-2 9 Landlocked Populations 1.2 0.4 1 1-2 7 Tribal/First Nation Fisheries Management 1.1 0.4 1 1-2 7 Tribal/First Nation Fisheries Utilization 1.1 0.3 1 1-2 8 Scientific Monitoring 1.0 0.1 1 1-2 9 Educational Harvest 1.0 0.0 1 1 9

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48983 Table 10. Qualitative ranking of threats for the Mid-Atlantic stock complex of blueback herring. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between 1 and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 3.9 1.0 4 3-5 9 Water Quality (chemical) 3.0 0.9 3 1-5 9 Incidental Catch 2.8 0.9 3 1-5 7 Water Withdrawal/Outfall (physical and temp.) 2.7 0.8 3 1-5 8 Climate change 2.7 1.2 3 1-5 8 Dredging 2.6 1.0 3 1-4 8 Climate variability 2.6 1.2 2,3 1-5 9 Predation 2.4 1.1 2 1-5 9 Hybrids 2.4 1.0 3 1-4 2 Federal Management 2.3 1.1 2 1-5 9 State Management 2.2 1.1 2 1-5 9 Directed Commercial Harvest 1.9 0.9 1 1-4 9 International Management 1.7 0.8 1 1-4 9 Competition 1.5 0.7 1 1-3 9 Artificial Propagation and Stocking 1.5 0.7 1 1-3 9 Disease 1.4 0.5 1 1-3 8 Recreational Harvest 1.3 0.5 1 1-3 9 Landlocked Populations 1.2 0.4 1 1-2 7 Tribal/First Nation Fisheries Management 1.1 0.4 1 1-2 7 Tribal/First Nation Fisheries Utilization 1.1 0.3 1 1-2 7 Scientific Monitoring 1.0 0.1 1 1-2 9 Educational Harvest 1.0 0.0 1 1 9

48984 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Table 11. Qualitative ranking of threats for the Southern Atlantic stock complex of blueback herring. Status Review Team members ranked threats by distributing 5 likelihood points among 5 ranks: 1- low, 2-medium/low, 3- medium, 4-medium/high, 5-high. The mean represents the overall Team average rank, mode represents the rank which received the most likelihood points, and range represents the range of ranks that were assigned likelihood points for each threat. N=number of Team members who ranked the threat between 1 and 5; likelihood points for threats that Team members ranked as either unknown or not applicable are not included.

Threats Mean SD Mode Range N Dams and Other Barriers 3.8 1.1 4 3-5 8 Water Quality (chemical) 3.0 0.9 3 1-5 9 Climate change 3.0 1.3 4 1-5 8 Climate variability 2.8 1.4 2,4 1-5 9 Water Withdrawal/Outfall (physical and 8 2.8 0.8 3 1-5 temp.)

Dredging 2.7 1.0 3 1-4 Incidental Catch 2.6 1.0 3 1-5 Predation 2.6 1.2 3 1-5 Federal Management 2.3 1.1 2 1-5 State Management 2.2 1.1 2 1-5 Hybrids 1.9 0.7 2 1-3 Directed Commercial Harvest 1.8 0.8 1 1-3 International Management 1.7 0.8 1 1-4 Competition 1.6 0.7 1 1-3 Disease 1.5 0.6 1 1-3 Artificial Propagation and Stocking 1.5 0.7 1 1-3 Recreational Harvest 1.3 0.5 1 1-3 Landlocked Populations 1.2 0.4 1 1-2 1.1 0.3 1 1-2 Tribal/First Nation Fisheries Management Tribal/First Nation Fisheries Utilization 1.1 0.3 1 1-2 Scientific Monitoring 1.0 0.1 1 1-2 Educational Harvest 1.0 0.0 1 1

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48985 Table 12. Summary table of threat ranking for alewife rangewide.

Threat Threat Level Section 4 Factor Dams and Other Barriers Medium High A Water Quality (chemical) Medium A Incidental Catch Medium B Predation Medium C Dredging Medium Low A Water Withdrawal/Outfall (physical and temp.) Medium Low A Climate change Medium Low A Climate variability Medium Low A Directed Commercial Harvest Medium Low B International Management Medium Low D Federal Management Medium Low D State Management Medium Low D Competition Medium Low E Artificial Propagation and Stocking Medium Low E Recreational Harvest Low B Tribal/First Nation Fisheries Management Low B Scientific Monitoring Low B Educational Harvest Low B Disease Low C Tribal/First Nation Fisheries Utilization Low D Hybrids Low E Landlocked Populations Low E

48986 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Table 13. Summary table of threat ranking for blueback herrin rangewide.

Threat Threat Level Section 4 Factor Dams and Other Barriers Medium High A Climate change Medium A Water Quality (chemical) Medium A Incidental Catch Medium B Predation Medium C Water Withdrawal/Outfall (physical and temp.) Medium Low A Dredging Medium Low A Climate variability Medium Low A Directed Commercial Harvest Medium Low B International Management Medium Low D Federal Management Medium Low D State Management Medium Low D Competition Medium Low E Hybrids Medium Low E Recreational Harvest Low B Tribal/First Nation Fisheries Management Low B Scientific Monitoring Low B Educational Harvest Low B Disease Low C Tribal/First Nation Fisheries Utilization Low D Artificial Propagation and Stocking Low E Landlocked Populations Low E BILLING CODE 3510-22-C abundance by estimating the population with the available data and a significant Extinction Risk Analysis growth rate for both species both number of data deficiencies for both rangewide and for each individual stock species, it was not necessary to have In order to assess the risk of complex. The SRT established two tiers information under both tiers in order to extinction for alewife and blueback that could be used separately or in make a risk determination, and we herring, trends in the relative combination to interpret the results of concur with this decision.

abundance of alewife and blueback the modeling in order to assess risk to The goal of Tier A was to maintain herring were assessed for each species alewife and blueback herring rangewide three contiguous stock complexes that rangewide, as well as for each species- and for the individual stock complexes. are stable or increasing as this: (1) specific stock complex. As noted We concur that these tiers are Satisfies the need to maintain both previously, for alewife, the stock appropriate. Tier A relates to what is geographic closeness and geographic complexes include Canada, Northern known about the geographic distance for a properly functioning New England, Southern New England distribution, habitat connectivity and metapopulation (see McElhany et al.,

and the mid-Atlantic. For blueback genetic diversity of each species, and 2000); (2) ensures that the recovered herring, the stock complexes are Tier B relates to the risk thresholds population does not include isolated Canada, Northern New England, established for the trend analysis that genetic groups that could lead to genetic Southern New England, mid-Atlantic was conducted by the NEFSC. These divergence (McDowall, 2003, Quinn, and Southern. tiers are subject to change in the future 1984); (3) provides some assurance that Criteria Established by SRT for as more information becomes available. the species persists across a relatively Evaluating Risk For example, Tier A is based on wide geographic area supporting diverse preliminary genetic data addressing environmental conditions and diverse Prior to conducting the trend analysis possible stock complexes, which could habitat types; and (4) ensures that the modeling, the SRT established criteria change in the future. Data related to entire population does not share the that would be used to evaluate the risk both tiers were assessed to determine if same risk from localized environmental to both species as well as to the sufficient information was available to catastrophe (McElhany et al., 2000).

individual stock complexes. At the make a conclusion under one or both of Tier B information was used to SRT's request, the NEFSC conducted the tiers. The SRT decided that, because directly interpret the results of the modeling to develop trends in relative of significant uncertainties associated trends in relative abundance modeling

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48987 conducted by the NEFSC. As described 2012). Following completion of the the risk falls between moderate and below, relative abundance of both model results, we determined that the high; High-it is likely that the threats alewife and blueback herring was used plus/minus 10-percent change in are significantly affecting the species' to estimate growth rate (along with the population growth rate would not continued existence now and/or into the 95 percent confidence intervals for the provide additional information that foreseeable future). If the coast wide growth rates) for each species rangewide would change the conclusions as to population growth rate is stable or and for each stock complex. Tier B whether the populations are significantly increasing and one stock established risk criteria depending on significantly increasing, stable or complex is significantly decreasing but the outcomes of the population growth decreasing. Without the projections of all others are stable or significantly rate modeling. As indicated in the the population growth rate into the increasing, the species is at a moderate-foreseeable future section above, a 12- to foreseeable future, the plus/minus 10- low risk. A significantly decreasing 18-year timeframe (e.g., 2024-2030) for percent would merely provide an population growth rate for several stock each species was determined to be additional set of bounds around the complexes would be an indicator that appropriate. After subsequent population growth rate estimate, and, the current abundance may not be discussions, the SRT decided that the therefore, we determined that running sustainable relative to current projections into the foreseeable future the model with the plus/minus 10- management measures and, therefore, would not provide meaningful percent was not necessary. may warrant further protections. Thus, information for the extinction risk The population growth rates derived if the population growth rates for two of analysis. As noted previously, the trend from the analysis help identify whether the stock complexes are significantly analysis provides a steady population stability exists within the population. decreasing but the coast-wide index is growth rate. If the population growth Mace et al. (2002) and Demaster et al. significantly increasing, the species is at rate is positive and everything else (2004) recognized that highly fecund, moderate-high risk. If the growth rates remains the same into the foreseeable short generation time species like river for three or more of the stock complexes future (e.g., natural and anthropogenic herring may be able to withstand a 95 are significantly decreasing and/or the mortality rates do not change), the to 99 percent decline in biomass. Both coast-wide index is significantly abundance into the foreseeable future alewives and blueback herring may decreasing, the species is at high risk.

will continue to increase. If the already be at or less than two percent of population growth rate is negative, then the historical baseline (e.g., Limburg Risk Scenarios the abundance into the foreseeable and Waldman, 2009), though these estimates are based on commercial

• Low risk future will continue to decline.

landings data, which are dependent o Coast wide trajectory-Stable to Currently, there is insufficient significantly increasing information available to modify any of upon management and are not a reliable the factors that may change the growth estimate of biomass. However, o Stock complex trajectories-All rates into the foreseeable future, and recognizing historical declines for both stable to significantly increasing thus, performing these projections will species, the modeled population growth " Moderate-Low risk not provide meaningful information for rates were used to gauge whether these o Coast wide trajectory-Stable to the extinction risk of either of these stock complexes are stable, significantly increasing or decreasing. Relative significantly increasing species. o Stock complex trajectories-One abundance of a stock is considered to be The baseline for the overall risk significantly increasing or decreasing if significantly decreasing,all others assessment assumes that there has the 95-percent confidence intervals of stable to significantly increasing already been a significant decline in the population growth rate do not abundance in both species due to a

• Moderate-High risk include zero. In contrast, if the 95-reduction in carrying capacity and o Coast wide trajectory-Stable to percent confidence intervals do contain overfishing as indicated in various zero, then the population is considered significantly increasing publications (Limburg and Waldman, to be stable, as the increasing or o Stock complex trajectories-Two or 2009; Hall et al., 2012), as well as other decreasing trend in abundance is not more significantly decreasing threats. The estimated population statistically significant.

• High risk growth rates reflect the impacts from the The SRT determined and we agree o Coast wide trajectory-Significantly various threats to which the species are that a stable or significantly increasing currently exposed. The SRT decreasing trajectory suggests that these species recommended that NEFSC use data from may be within the margins of being self- o Stock complex trajectories-Three 1976 through the present to minimize sustainable and thus, if all of the growth or more significantly decreasing the overfishing influence from distant rates for the coast-wide distribution and Trend Analysis Modeling water fleets that occurred in earlier the stock complexes are stable or years but has since been curtailed by significantly increasing, the species is at The sections below include fisheries management measures. The low risk of extinction (the risk summaries/excerpts from the NEFSC SRT recommended that the NEFSC also categories were defined by adapting the Report to the SRT, "Analysis of Trends run a trajectory using a plus/minus 10- categories described above for the in Alewife and Blueback Herring percent growth rate to test model QTA-Low risk-it is likely that the Relative Abundance," June 17, 2013, 42 sensitivity with respect to changes in threats to the species' continued pp. (NEFSC, 2013). For detailed the model variables. This approach has existence are not significant now and/or information on the modeling conducted, been used in analyses for other species into the foreseeable future; Moderately please see the complete report which (e.g., Atlantic croaker, Atlantic cod) and Low-risk falls between low and can be found at http://

can serve as a means of showing moderate rankings; Moderate-it is www.nero.noaa.gov/prot resi sensitivities in the model to potential likely that the threats are having some CandidateSpeciesProgram/

variables (e.g., population growth rate effect on the species continued RiverHerringSOC.htmor see FOR changes, climate change) (Hare and existence now and/or into the FURTHER INFORMATION CONTACT section Able, 2007; Hare, NMFS Pers. comm., foreseeable future; Moderately High- above for contacts.

48988 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Data Used in the Trend Analysis these trawl surveys using both inshore In 2009, the survey changed primary Modeling and offshore strata for 1976-2012 in the research vessels from the RV Albatross spring and 1975-2011 in the fall. IV to the RV Henry B. Bigelow. Due to Rangewide Data Additional relative abundance indices the deeper draft of the RV Henry B.

Relative abundance indices from were derived using only offshore strata Bigelow, the two shallowest series of multiple fishery-independent survey for 1968-2012 in the spring and 1967- inshore strata (8-18 m depth) are no time series were considered as possible 2011 in the fall (from 1963-1967 the fall longer sampled. Concurrent with the data inputs for the rangewide analysis. survey did not extend south of Hudson change in fishing vessel, substantial These time series included the NEFSC Canyon). These time series were changes to the characteristics of the spring, fall, and winter bottom trawl developed following the same sampling protocol and trawl gear were surveys as well as the NEFSC shrimp methodology used in the ASMFC river survey. For alewife, two additional time made, including tow speed, net type herring stock assessment (ASMFC, and tow duration (NEFSC, 2007).

series were available: Canada's DFO 2012).

summer research vessel (RV) survey of Through 2008, standard bottom trawl Calibration experiments, comprising the Scotian Shelf and Bay of Fundy tows were conducted for 30 minutes at paired standardized tows of the two (1970-present), and DFO's Georges 6.5 km/hour with the RV Albatross IV fishing vessels, were conducted to Bank RV survey (1987-present, as the primary survey research vessel measure the relative catchability conducted during February and March). (Despres-Patanjo et al., 1988). However, between the two vessel-gear For the NEFSC spring and fall bottom vessel, door and net changes did occur combinations and develop calibration trawl surveys, inshore strata from 8 to during this time, resulting in the need factors to convert Bigelow survey 27 m depth and offshore strata from 27 for conversion factors to adjust survey catches to RV Albatross equivalents to 366 m depth have been most catches for some species. Conversion (Miller et al., 2010). In the modeling, the consistently sampled by the RV factors were not available for net and NEFSC developed species-specific Albatross IV and RV Delaware II since door changes, but a vessel conversion calibration coefficients which were the fall of 1975 and spring of 1976. Prior factor for alewife was available to estimated for both catch numbers and to these time periods, either only a account for years where the RV weights using the method of Miller et al.

portion of the survey area was sampled Delaware II was used. A vessel (2010) (Table 14). The calibration factors or a different vessel and gear were used conversion factor of 0.58 was applied to were combined across seasons due to to sample the inshore strata (Azarovitz, alewife weight-per-tow indices from the low within-season sample sizes from the 1981). Accordingly, seasonal alewife RV Delaware IL.Alewife number-per- 2008 calibration studies (fewer than 30 and blueback herring relative tow indices did not require a conversion tows with positive catches by one or abundance indices were derived from factor (Byrne and Forrester, 1991). both vessels).

Table 14. Coefficients and associated standard errors used to convert RV Bigelow catches of alewife and blueback herring to RV Albatross IV equivalents for the 2009-2011 NEFSC bottom trawl surveys.

Number Biomass Species Coefficient SE Coefficient SE Alewife 1.05 0.16 0.72 0.11 Blueback herring 0.87 0.17 1.59 0.45 Bottom trawl catches of river herring locations may have substantially survey was merged with the NEFSC tend to be higher during the daytime different irradiance levels that could spring survey and discontinued.

due to diel migration patterns (Loesch et influence survey catchability (NEFSC, Alewife and blueback herring indices of al., 1982; Stone and Jessop, 1992). 2011). Preliminary analyses (Lisa relative abundance were developed for Accordingly, only daytime tows were Hendrickson, NMFS, 2012- the winter survey from 1992-2007 using used to compute relative abundance and unpublished data) confirmed that river daytime tows from all sampled inshore biomass indices. In addition, the herring catches were generally greater and offshore strata. The shrimp survey calibration factors used to convert RV during daylight hours compared to is conducted during the summer (July/

Bigelow catches to RV Albatross nighttime hours. August) in the western Gulf of Maine equivalents were estimated using only In addition to the NEFSC spring and during daylight hours. Relative catches from daytime tows. Daytime fall trawl surveys, the NEFSC winter abundance indices were derived for tows, defined as those tows between and shrimp surveys were considered for alewife and blueback herring from sunrise and sunset, were identified for inclusion in the analysis. For the winter 1983-2011 using all strata that were each survey station based on sampling survey (February), the sampling area consistently sampled across the survey date, location, and solar zenith angle extended from Cape Hatteras, NC, time series in the NEFSC winter and using the method of Jacobson et al. through the southern flank of Georges shrimp surveys.

(2011). Although there is a clear general Bank, but did not include the remaining Stratified mean indices of relative relationship between solar zenith and portion of Georges Bank or the Gulf of abundance of alewife from Canada's time of day, tows carried out at the same Maine. With the arrival of the RV summer RV survey and Georges Bank time but at different geographic Bigelow in late 2007, the NEFSC winter RV survey were provided by Heath

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48989 Stone of Canada's DFO. In these exception of run counts from the St. after 1976. In these cases, the first surveys, alewife is the predominant Croix and Union Rivers. These datasets modeled year coincided with the first species captured; however, some were excluded due to the artificial running sum of the earliest survey.

blueback herring are likely included in impacts of management activities on run the alewife indices because catches are sizes. The closure of the Woodland Dam MARRS Model Description not always separated by river herring and Great Falls fishways in the St. Croix Multivariate Autoregressive State-species (Heath Stone, DFO Pers. comm., River prevented the upstream passage of Space models (MARSS) were developed 2012). Furthermore, some Georges Bank alewives to spawning habitat. In using the MARSS package in R (Holmes strata were not sampled in all years of contrast, fluctuations in Union River et al., 2012a). This package fits linear the survey due to inclement weather run counts were likely impacted by MARSS models to time series data using and vessel mechanical problems (Stone lifting and stocking activities used to a maximum likelihood framework based and Gross, 2012). maintain a fishery above the Ellsworth on the Kalman smoother and an Due to the restricted spatial coverage Dam. In the southern Gulf of St. Expectation Maximization algorithm of the winter, shrimp and Canadian Lawrence trawl survey, all river herring (Holmes et al., 2012b).

Georges Bank surveys, these surveys were considered to be alewife because Each MARSS model is comprised of were not used in the final rangewide survey catches were not separated by a process model and an observation analyses. Accordingly, relative river herring species (Luc Savoie DFO, model (Holmes and Ward, 2010; Holmes abundance (number-per-tow) from the Pers. comm., 2012). No blueback herring et al., 2012b). The model is described in NEFSC spring and fall surveys was used abundance indices were available for detail in the NEFSC (2013) final report in the rangewide models for blueback the Canadian stock. Select strata were to the SRT (posted on the Northeast herring, and number-per-tow from the not used to estimate stock-specific Regional Office's Web site-http://

NEFSC spring survey, NEFSC fall indices from the NEFSC trawl surveys www.nero.noao.gov/prot res!

survey, and the Canadian summer because mixing occurs on the CandidateSpeciesProgram!

survey were used in the rangewide continental shelf. Accordingly, any RiverHerringSOC.htm).Population models for alewife. NEFSC trawl survey indices, even projectionsand model analysis.

Data from 1976 through the present estimated using only particular strata, For each stock complex, the estimated were incorporated into the trend would likely include individuals from population growth rate and associated analysis. This time series permitted the more than one stock. 95 percent confidence intervals were inclusion of the spring and fall surveys' Each available dataset in the stock-used to classify whether the stock's inshore strata. In addition, with this specific analyses represented a relative abundance was stable, time series, the required assumption particular age or stage (spawners, significantly increasing or decreasing.

that the population growth rate will young-of-year, etc.) of fish. As noted previously, relative abundance remain the same was reasonable. Prior Consequently, each time series was of a stock was considered to be to 1976, fishing intensity was much transformed using a running sum over 4 significantly increasing or decreasing if greater due to the presence of distant years. The selection of 4 years for the the 95 percent confidence intervals of water fleets on the East Coast of the running sum was based on the the population growth rate did not United States. generation time of river herring. For age- include zero. In contrast, if the 95 Years with zero catches were treated and stage-specific data, a running sum percent confidence intervals included as missing data. For alewife, there were transformation is recommended to zero, the population was considered to no years with zero catches in the spring, obtain a time series that more closely be stable because the increasing or fall and Scotian shelf surveys. Zero approximates the total population decreasing trend in abundance was not catches of blueback herring occurred in (Holmes, 2001). In order to compute the running sums for each dataset, missing significant.

the fall survey in 1988, 1990, 1992 and 1998. data were imputed by computing the Model Results means of immediately adjacent years.

Stock-Specific Data For both species 4 years were imputed Rangewide Analyses Stock-specific time series of alewife for the Monument River, and 1 year was For the rangewide analysis, as shown and blueback herring relative imputed for the DC seine survey. For in Table 15 below, the preferred model abundance were obtained from the alewife, 1 year was also imputed for the run for alewife indicates that the 95-ASMFC and Canada's DFO. Available Mattapoisett River, Nemasket River, and percent confidence intervals spanning time series varied among stocks and the southern Gulf of St. Lawrence trawl the estimated population growth rate do included run counts, as well as young- survey. For blueback herring, 1 year was not include 0 and are statistically of-year (YOY), juvenile and adult also imputed for the Long Island Sound significantly increasing. For blueback surveys that occurred solely within the (LIS) trawl survey and Santee-Cooper herring rangewide, however, the 95-bays or sounds of the stock of interest catch-per-unit-effort (CPUE). percent confidence intervals do include (for alewife see Table 15 in the NEFSC's If possible data from 1976 through the 0, and thus, it is not possible to state "Analysis of Trends in Alewife and present were incorporated into each that the trend rangewide for this species Blueback Herring Relative Abundance," stock-specific model, with the first is increasing. We, therefore, conclude and for blueback herring, see Table 16). running sum incorporating data from based on our criteria described above All available datasets were included in 1976 through 1979. However, for some that blueback herring rangewide are the stock-specific analyses, with the stocks, observation time series began stable.

48990 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices Table 15. Population growth rate maximum likelihood estimates (ML.Est), associated standard errors (Std.Err) and lower and upper 95 percent confidence intervals (low.CI, up.CI) for each rangewide model run. The preferred model run (lowest AIC) for each species is highlighted in grey.

Species Run ML.Est Std.Frr Iow.Cl u.ClI Independent with equal variances 0.034 0.006 0.022 0.046 Independent with unequal variances ,; 0.0321Ofý *, 0 , 2, '. P0..4Y)

Alewife Unconstrained 0.030 0.005 0.020 0.041 Unequal variances with one covariance term 0.035 0.013 0.009 0.062 Equal variance and covariance 0.034 0.005 0.023 0.045 Independent with equal variances `0. . 0X t104*09 0.,1i Blueback herring Independent with unequal variances 0.022 0.036 -0.047 0.093 Unconstrained 0.026 0.045 -0.063 0.112 Equal variance and covariance 0.040 0.052 -0.064 0.144 Stock-Specific Analyses Southern New England and mid- estimated population growth rate do Atlantic alewife stock complexes are include 0 and thus, the trend for these As shown in Table 16 below, the 95- stable. stock complexes is stable. For the mid-percent confidence intervals spanning As Canada does not separate alewife Atlantic stock complex, the population the estimated population growth rate for and blueback herring in their surveys growth rate and both 95-percent the Canadian stock complex do not (e.g., they indicate that all fish are confidence intervals are all statistically include 0 and are statistically alewife), we were unable to obtain data significantly decreasing. Thus, we significantly increasing. For the other from Canada specifically for blueback conclude that this stock complex is three stock complexes, however, the herring. For three of the remaining four significantly decreasing.

confidence intervals do include 0, and stock complexes, the 95-percent BILLING CODE 3510-22-P thus, the Northern New England, confidence intervals spanning the

Table 16. Population growth rate maximum likelihood estimates (ML.Est), associated standard errors (Std.Err) and lower and upper 95-percent confidence intervals (low.CI, up.CI) for each stock-specific model run. The preferred model run (lowest AIC) for each stock is highlighted in grey.

Species Stock Run MLEst StdErr low.Cl up.CI Alewife Mid Atlantic Independent with equal variances 0.004 0.034 -0.061 0.073 Independent with unequal variances W

p-O021:~*.3 P 9~ 0 Unconstrained -0.013 0.029 -0.071 0.044 Unequal variances with one covariance term -0.021 0.035 -0.088 0.054 Cn Equal variance and covariance -0.004 0.046 -0.092 0.088 00 Southern New England Independent with equal variances 0.008 0.032 -0.052 0.072 Independent with unequal variances Equal variance and covariance 0.005 0.032 -0.057 0.069 Northern New England Independent with equal variances CD Unconstrained 0.038 0.036 -0.034 0.108 0

Equal variance and covariance 0.036 0.041 -0.048 0.114 Canada Independent with equal variances z 0D Blueback herring Southern Independent with equal variances -0.004 0.047 -0.091 0.091 CJI Independent with unequal variances O0: ',02 04L Unconstrained 0.024 0.042 -0.058 0.103 Equal variance and covariance -0.001 0.046 -0.091 0.092 Mid Atlantic Independent with equal variances -0.070 0.008 -0.085 -0.055 Independent with unequal variances 05 Equal variance and covariance -0.072 0.013 -0.097 -0.046 Southem New England Independent with equal variances . ,,

Northern New England Independent with equal variances -4

48992 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices BILLING CODE 3510-22-C wide and for the individual stock Fishery Management Councils have Model Assumptions and Limitations complexes, and thus, for providing recommended management measures information to be used in assessing the under the MSA that are expected to The available data for each analysis varied considerably among species and risk to these species and stock decrease the risk from this particular complexes. threat. Under both the Atlantic Herring stocks. Some stocks such as Southern New England blueback herring had only Extinction Risk Conclusion Fishery Management Plan and the Mackerel/Squid/Butterfish Fishery one available data set; however, other In performing our analysis of the risk stocks such as Southern New England Management Plan, the Councils have of extinction to the species, we recommended a suite of reporting, alewife and mid-Atlantic blueback considered the current status and trends herring had eight or more available time vessel operation, river herring catch cap and the threats as they are impacting the provisions, and observer provisions that series. Within each analysis, all input species at this time. Currently, neither time series must be weighted equally, would improve information on the species is experiencing high rates of amount and extent of river herring catch regardless of the variability in the decline coast-wide as evidenced by the in the Atlantic herring and mackerel dataset. Furthermore, only the annual rangewide trends (significantly point estimates of relative abundance fisheries. NMFS has partially approved increasing for alewife and stable for the measures as recommended by the are inputs to the model; associated blueback herring). Thus, using the standard errors for the time series are New England Council and will be extinction risk tiers identified by the implementing the measures in not inputted. SRT, we have concluded the following:

However, some observation time September or October 2013. Another Alewife- threat that has been identified for both series may be more representative of the

• Tier A: There is sufficient stock of interest than other time series. species is loss of habitat or loss of access information available to conclude that to spawning habitats. We have been For example, for Northern New England there are at least three contiguous alewife, available datasets included run working to restore access to spawning populations that are stable to habitats for river herring and other counts from five rivers and Maine's significantly increasing.

juvenile alosine seine survey. Each time diadromous fish species through habitat

• Tier B: The species is at "Low risk" restoration projects. While several series of run counts represents the as the coast-wide trajectory is spawning population in one particular threats may lessen in the future, given significantly increasing and all of the the extensive decline from historical river, whereas the juvenile seine survey stock complexes are stable or samples six Maine rivers including levels, neither species is thought to be significantly increasing. capable of withstanding continued high Merrymeeting Bay (ASMFC, 2012). Blueback herring-Accordingly, it is possible that the rates of decline.

s Tier A: There is insufficient juvenile seine survey provides a better information available to make a Research Needs representation of Northern New England conclusion under Tier A as we were As noted above, there is insufficient alewife than the run counts from any unable to obtain data from Canada to information available on river herring in particular river because the seine survey determine the population growth rate many areas. Research needs were samples multiple populations. Likewise, for rivers in Canada. Thus, we were only recently identified in the ASMFC River for Southern New England alewife, able to obtain information for four of the Herring Stock Assessment Report available datasets included the Long five stock complexes identified for the (ASMFC, 2012); NMFS Stock Structure, Island Sound (LIS) trawl survey, New species. Climate Change and Extinction Risk York juvenile seine survey, and run

• Tier B: The species is at "Moderate- Workshop/Working Group Reports counts from six rivers. The LIS trawl low risk "as the coast-wide trajectory is (NMFSa, 2012; NMFSb, 2012; NMFSc, survey samples Long Island Sound from stable and three of the four stock 2012) and associated peer reviews; and New London to Greenwich Connecticut complexes are stable. The estimated New England and Mid-Atlantic Fishery with stations in both Connecticut and population growth rate of the mid- Management Council documents New York state waters, including the Atlantic stock complex is significantly (NEFMC, 2012; MAFMC, 2012). We mouths of several rivers including the decreasing based on the available have identified below some of the most Thames, Connecticut, Housatonic, East information. However, the relative critical and immediate research needs to and Quinnipiac (CTDEP, 2011; ASMFC, abundance of the species throughout its conserve river herring taking the 2012). The NY juvenile seine survey range (as demonstrated through the recently identified needs into samples the Hudson River estuary coast-wide population growth rate) is consideration, as well as information (ASMFC, 2012), and run counts are stable, and thus, the SRT concluded that from this determination. However, these specific to particular rivers. As a the mid-Atlantic stock complex does not are subject to refinement as a consequence, the LIS trawl survey may constitute a significant portion of the coordinated and prioritized coast-wide be more representative of the Southern species range. We concur with this approach to continue to fill in data gaps New England alewife stock because it conclusion. In other words, the data and conserve river herring and their samples not only a greater proportion of indicate that the mid-Atlantic stock habitat is developed (see "Listing the stock, but also samples LIS where complex does not contribute so much to Determination" below).

mixing of river-specific populations the species that, without it, the entire 9 Gather additional information on likely occurs. species would be in danger of life history for all stages and habitat Several sources of uncertainty are extinction. areas using consistent and described in detail in the modeling Many conservation efforts are comprehensive coast-wide protocols report. It is important to understand and underway that may lessen the impact of (i.e., within and between the United document these sources of uncertainty. some of these threats into the States and Canada). This includes However, even with several foreseeable future. One of the significant information on movements such as assumptions and these sources of threats identified for both species is straying rates and migrations at sea.

uncertainty, we are confident that the bycatch in Federal fisheries, such as the Improve methods to develop biological model results are useful in determining Atlantic herring and mackerel fisheries. benchmarks used in assessment the population growth rates both coast- The New England and Mid Atlantic modeling.

Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices 48993 9 Continue genetic analyses to assess climate change workshops/working of the potential differences in the genetic diversity, determine population groups, the population growth rates magnitude of the threats to specific stock structure along the coast (U.S. and from the trends in relative abundance areas or populations, we next evaluated Canada) and determination of river estimates and qualitative threats whether alewife or blueback herring origin of incidental catch in non- assessment, the Center for Independent might be threatened or endangered in targeted ocean fisheries. Also, obtain Experts peer reviewers' comments, other any significant portion of its range. In information on hybridization and qualified peer reviewer submissions, accordance with our draft policy on understand the effects of stocking on and consulted with scientists, "significant portion of its range," our genetic diversity. fishermen, fishery resource managers, first step in this evaluation was to e Further assess human impacts on and Native American Tribes familiar review the entire supporting record for river herring (e.g., quantifying bycatch with river herring and related research this listing determination to "identify through expanded observer and port areas, and all other information any portions of the range[s] of the sampling coverage to quantify fishing encompassing the best available species that warrant further impact in the ocean environment and information on river herring. Based on consideration" (76 FR 77002; December improve reporting of commercial and the best available information, the SRT 9, 2011). Therefore, we evaluated recreational harvest by waterbody and concluded that alewife are at a low risk whether there is substantial information gear, ocean acidification) of extinction from the threats identified suggesting that the hypothetical loss of a Continue developing models to in the QTA (e.g., dams and other any of the individual stock complexes predict the potential impacts of climate barriers to migration, incidental catch, for either species (e.g., portions of the change on river herring. This includes, climate change, dredging, water quality, species' ranges) would reasonably be as needed to support these efforts, water withdrawal/outfall, predation, expected to increase the demographic environmental tolerances and and existing regulation), and blueback risks to the point that the species would thresholds (e.g., temperature) for all life herring are at a moderate-low risk of then be in danger of extinction, (i.e.,

stages in various habitats. extinction from similar threats whether any of the stock complexes

• Develop and implement monitoring identified and discussed in the QTA within either species' range should be protocols and analyses to determine discussion above. We concur with this considered "significant"). As noted in river herring population responses and conclusion, and we have determined the extinction risk analysis section, all targets for rivers undergoing restoration that as a result of the extinction risk of the alewife stock complexes as well (e.g., dam removals, fishways, analysis for both species, these two as the coastwide trend are either stable supplemental stocking). Also, estimate species are not in danger of extinction or increasing. For blueback herring, 3 of spawning habitat by watershed (with or likely to become so in the foreseeable the stock complexes and the coastwide and without dams). future. Therefore, listing alewife and trend are all stable, but the mid-Atlantic

• Assess the frequency and blueback herring as either endangered stock complex is decreasing. The SRT occurrence of hybridization between or threatened throughout all of their determined that the mid-Atlantic stock alewife and blueback herring and ranges is not warranted at this time. complex is not significant to the species, possible conditions that contribute to its given that even though it is decreasing, occurrence (e.g., occurs naturally or in Significant Portionof the Range Evaluation the overall coastwide trend is stable.

response to climate change, dams, or Thus, the loss of this stock complex other anthropogenic factors). Under the ESA and our implementing a Continue investigating predator regulations, a species warrants listing if would not place the entire species at prey relationships. it is threatened or endangered risk of extinction. We concur with this throughout all or a significant portion of conclusion. Because the portion of the Listing Determination its range. In our analysis for this listing blueback herring stock complex residing The ESA defines an endangered determination, we initially evaluated in the mid-Atlantic is not so significant species as any species in danger of the status of and threats to the alewife that its hypothetical loss would render extinction throughout all or a significant and blueback herring throughout the the species endangered, we conclude portion of its range, and a threatened entire range of both species. As stated that the mid-Atlantic stock complex species as any species likely to become previously, we have concluded that does not constitute a significant portion an endangered species within the there was not sufficient evidence to of the blueback herring's range.

foreseeable future throughout all or a suggest that the genetically distinct Consequently, we need not address the significant portion of its range. Section stock complexes of alewife or blueback question of whether the portion of the 4(b)(1) of the ESA requires that the constitute DPSs. We also then assessed species occupying this portion of the listing determination be based solely on the status of each of the individual stock range of blueback herring is threatened the best scientific and commercial data complexes in order to determine or endangered.

available, after conducting a review of whether either species is threatened or Conclusion the status of the species and after taking endangered in a significant portion of its into account those efforts, if any, that range. Our review of the information are being made to protect such species. As noted above in the QTA section, pertaining to the five ESA section 4(a)(1)

We have considered the available the SRT determined that the threats to factors does not support the assertion information on the abundance of alewife both species are similar and the threats that there are threats acting on either and blueback herring, and whether any to each of the individual stock alewife or blueback herring or their one or a combination of the five ESA complexes are similar with some slight habitat that have rendered either species factors significantly affect the long-term variation based on geography. Water to be in danger of extinction or likely to persistence of these species now or into quality, water withdrawal/outfall, become so in the foreseeable future, the foreseeable future. We have predation, climate change and climate throughout all or a significant portion of reviewed the information received variability were generally seen as greater its range. Therefore, listing alewife or following the positive 90-day finding on threats to both species in the southern blueback herring as threatened or the petition, the reports from the stock portion of their ranges than in the endangered under the ESA is not structure, extinction risk analysis, and northern portion of their ranges. In light warranted at this time.

48994 Federal Register/Vol. 78, No. 155/Monday, August 12, 2013/Notices While neither species is currently unlikely that a detrimental impact to should be sent to us (see ADDRESSES endangered or threatened, both species either species could occur within this section above).

are at low abundance compared to period. Additionally, it allows for time historical levels, and monitoring both References Cited to complete ongoing scientific studies species is warranted. We agree with the (e.g., genetic analyses, ocean migration A complete list of all references cited SRT that there are significant data patterns, climate change impacts) and in this rulemaking can be found on our deficiencies for both species, and there for the results to be fully considered. Web site at http://www.nero.noaa.govl is uncertainty associated with available Also, it allows for the assessment of data prot res/CandidateSpeciesProgram/

data. There are many ongoing to determine whether the preliminary RiverHerringSOC.htmand is available restoration and conservation efforts and upon request from the NMFS office in reports of increased river counts in new management measures that are Gloucester, MA (see ADDRESSES).

many areas along the coast in the last 2 being initiated/considered that are expected to benefit the species; years represent sustained trends. During Authority: The authority for this action is however, it is not possible at this time this 3- to 5-year period, we intend to the Endangered Species Act of 1973, as to quantify the positive benefit from coordinate with ASMFC on a strategy to amended (16 U.S.C. 1531 et seq.).

these efforts. Given the uncertainties develop a long-term and dynamic Dated: August 6, 2013.

and data deficiencies for both species, conservation plan (e.g., priority Alan D. Risenhoover, we commit to revisiting both species in activities and areas) for river herring Director,Office of Sustainable Fisheries, 3 to 5 years. We have determined that considering the full range of both performingthe functions and duties of the this is an appropriate timeframe for species and with the goal of addressing Deputy Assistant Administratorfor considering this information in the many of the high priority data gaps for Regulatory ProgramsNationalMarine future as a 3- to 5-year timeframe river herring. We welcome input and FisheriesService.

equates to approximately one generation involvement from the public. Any [FR Doc. 2013-19380 Filed 8-9-13; 8:45 am]

time for each species, and it is therefore information that could help this effort BILLING CODE 3510-22-P

NOTE: This is a REVISED version of the plan, originallyposted to the DEC website in A ugust 2011. Changes were made as a result of public comment received by Sept 22, 2011.

SNew York State aDepartment of Environmental Conservation Sustainable Fishing Plan for New York River Herring Stocks Kathryn A. Hattala, Andrew W. Kahnle Bureau of Marine Resources, Hudson River Fisheries Unit and Robert D. Adams Hudson River Estuary Program September 2011 Submitted for review to the Atlantic State Marine Fisheries Commission

REVISED VERSION: September 2011, based on public comment received.

EXECUTIVE

SUMMARY

Amendment 2 to the Atlantic States Marine Fisheries Commission Shad and river Herring Interstate Fishery Management Plan requires member states to demonstrate that fisheries for river herring (alewife and blueback herring) within their state waters are sustainable. A sustainable fishery is defined as one that will not diminish potential future reproduction and recruitment of herring stocks. If states cannot demonstrate sustainability to the Atlantic States Marine Fisheries Commission (ASMFC), they must close their herring fisheries.

New York State proposes to maintain a restricted river herring (alewife and blueback herring) fishery in the Hudson River and tributaries and to close river herring fisheries elsewhere in the State. This proposal conforms to Goal 1 of the New York State Hudson River Estuary Action Agenda.

Stock Status Blueback herring and alewife are known to occur and spawn in New York State in the Hudson River and tributaries, the Bronx River, and several streams on Long Island. The Hudson River is tidal to the first dam at Troy, NY (rkm 245). Data on stock status are available for the Hudson River and tributaries. Few data are available on river herring in streams in Bronx County, southern Westchester County, or on Long Island. River herring are absent in the New York portion of the Delaware River.

Hudson River: Commercial and recreational fisheries exploit the spawning populations of river herring in the Hudson River and tributaries. Fixed and drifted gill, cast and scap/lift nets are used in the main stem Hudson, while scap/lift and cast nets are used in the tributaries. Recreational fishers often use commercial net gears because permit fees remain at 1911 levels. Anglers also are allowed take of river herring with variety of small nets and hook and line. In the last ten years, about 250 fishers annually purchased commercial gill net permits and approximately 240 purchased commercial scap net permits. However only 84 gill net and 93 scap/lift fishers reported using the gear licensed. Fishers using commercial gears are required to report landings annually. Most river herring taken in the Hudson and tributaries are used as bait in the recreational striped bass fishery. Anglers and subsistence fishers take a few river herring from Long Island streams.

Data on commercial harvest of river herring are available since the early 1900s. Landings peaked in the early 1900s and in the 1930s and then declined through the 1980s. Landings increased again through 2003, but have since declined. Reported commercial harvest has remained below 50,000 river herring per year since the early 1990s. A series of creel surveys and estimates since 2001 indicated substantial and increasing harvest of river herring by recreational anglers from the Hudson River and tributaries. We estimated that approximately 240,000 river herring were harvested by recreational anglers in 2007. The extent of the loss of river herring through bycatch in ocean commercial fisheries remains largely unknown but is expected to be significant.

2

REVISED VERSION: September 2011, basedon public comment received.

Fishery dependent data on river herring status since 2000 are available from commercial reports and from on-board monitoring. Catch per unit effort (CPUE) in fixed (anchored) gill nets fished in the main stem river has increased. Conversely, CPUE in scap nets fished in tributaries initially declined, but then varied without trend. Mean length of river herring observed in the commercial harvest has declined slightly since 2000. We feel that the CPUE in fixed gear below the Bear Mountain Bridge provides the best annual measure of abundance because it intercepts river herring migrating past the gear to upriver spawning locations..

Fishery independent data on size and age composition of river herring spawning in the Hudson River Estuary are available from 1936 and intermittently since the late 1970s. Sample size has been small in most years. The largest fish were collected in the 1930s. Size of both blueback herring and alewife has declined over the last 30 years. Age data were obtained from scales in 1936 and the late 1980s. Since then, ages were estimated from age length keys developed by Maine, Massachusetts, and Maryland. Observed and estimated age at length of Hudson River fish varied substantially among methods and thus age can only be used for trends within method.

Annual mean age since the late 1980s has remained stable in blueback herring and female alewife, but declined in male alewife. Because of the uncertainty with estimated ages, we estimated annual mortality with length-based methods. Estimates varied substantially depending on assumed model inputs and therefore actual total mortality on the stocks remains unknown.

However, we should emphasize that mortality on stocks must have been high in the last 30 years to have so consistently reduced mean size and presumably mean age. Within method, estimates of total mortality generally increased for both species since 1980. This increase was most pronounced in alewife.

Young of year production has been measured annually by beach seine since 1980. CPUE of alewife remained low through the late 1990s and has since increased erratically. CPUE of young of year blueback herring has varied with a very slight downward trend since 1980.

Streams on Long Island,Bronx and south shore of Westchester County: Limited data have been collected for some of the river herring populations in these areas. The data are not adequate to characterize stock condition.

Delaware River in New York: No records exist to document the presence of river herring in this portion of the river.

Proposed Fishery for the Hudson River Given the inconsistent measures of stock status described above, we do not feel that the data warrant a complete closure of the Hudson River fishery at this time. New York State proposes a five year restricted fishery in the main-stem Hudson River, a partial closure of the fishery in tributaries, and annual stock monitoring. We set a sustainability target for juvenile indices. We will monitor, but not set targets for mean length from fishery independent spawning stock sampling and CPUE in the commercial fixed gill net fishery in the lower river below the Bear Mountain Bridge. We will also monitor age structure, frequency of repeat spawning, and total mortality from fishery independent sampling if we can resolve problems with age determination 3

REVISED VERSION: September 2011, based on public comment received.

and mortality estimation.

A summary of existing and proposed restrictions is provided. Proposed restrictions to the recreational fishery include: a ten fish per day creel limit for individual anglers with a boat limit of 50, and a 10 fish creel limit per day for paying customers with a boat limit of 50 for charter vessels, no fishing within 825 ft (250m) of any man made or natural barrier in the main river and tributaries, no use of nets in tributaries, and the continuation of various small nets in the main river. Proposed restrictions to the commercial fishery and use of commercial gears include: a commercial verification requirement; a net ban in the upper 28 km of the main-stem estuary, shad spawning flats, or tributaries; gill net mesh and size restrictions; a ban on fixed gears or night fishing above the Bear Mountain Bridge; seine and scap/lift net size restrictions; extension of existing 36 hour4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> lift period to all commercial net gears; increased net fees to account for inflation since 1911 when fees were set or the preferred option of creation of a new Hudson River Commercial Fish Permit; extension of the current Marine and Coastal District Charter

/Party boat license to the tidal Hudson and tributaries at a cost of $250.00 annually; and monthly mandatory reporting of catch and harvest.

We should note that Draft Addendum 3 to Amendment 6 of the ASMFC Interstate Management Plan for striped bass stipulates that states should reduce fishing mortality on spawning stocks by 50%. If this draft is approved by the ASMFC Striped Bass Management Board, we may have to restrict effort in the recreational striped bass fishery. Restrictions may include a reduction in use of bait such as river herring. Any reduction in effort will likely reduce demand for river herring and thus reduce losses in the Hudson stocks.

Proposed Moratorium for streams on Long Island, Bronx County, the southern shore of Westchester County, and the Delaware River and its tributaries north of Port Jervis NY. Due to the inability to determine stock condition for these areas, the ASMFC Amendment 2 requires that a moratorium on river herring fishing be implemented.

This SFP does not directly address ocean bycatch but focuses on fisheries in New York State waters. New York is working with the National Marine Fisheries Service, the New England Fishery Management Council and the Mid-Atlantic Fishery Management Council to deal with this issue. Both councils are in the process of amending the Atlantic Herring and the Atlantic Mackerel, Squid and Butterfish Plans to reduce bycatch of river herring.

4

REVISED VERSION: September 2011, based on public comment received.

1 CONTENTS 2 INTRODUCTION ........................................................................................................... 6 3 MANAGEM ENT UNITS ................................................................................................... 6 3.1 Description of the M anagement Unit Habitat ............................................................. 7 3.1.1 Hudson River and tributaries ................................................................................ 7 3.1.2 Long Island and Westchester County ................................................................. 8 3.1.3 Delaware River ..................................................................................................... 9 3.2 Habitat Loss and Alteration .......................................................................................... 9 3.3 Habitat W ater Quality ................................................................................................ 10 4 STOCK STATUS ................................................................................................................. 10 4.1 Fisheries Dependent Data .......................................................................................... 11 4.1.1 Commercial Fishery ............................................................................................ 11 4.1.2 Recreational Fishery ........................................................................................... 15 4.2 Fishery Independent Surveys .................................................................................... 16 4.2.1 Spawning Stock Surveys - Hudson River .......................................................... 16 4.2.2 Hudson River Spawning Stock - Characteristics ................................................. 17 4.2.3 Spawning Stock Surveys - Long Island ............................................................ 19 4.2.4 Volunteer and Other river herring monitoring ................................................... 19 4.2.5 Young-of-the-Year Abundance .......................................................................... 20 4.2.6 Conclusion ......................................................................................................... 21 5 PROPOSED FISHERY CLOSURES ................................................................................ 21 5.1 Long Island, Bronx County and W estchester County .............................................. 21 5.2 Delaware River .......................................................................................................... 21 6 PROPOSED SUSTAINABLE FISHERY ........................................................................ 22 6.1 Hudson River and Tributaries .................................................................................... 22 6.1.1 Proposed Restrictions - Recreational Fishery ................................................... 23 6.1.2 Proposed Restrictions - Commercial Fishery ................................................... 25 7 PROPOSED M EASURES OF SUSTAINABILITY ........................................................ 27 7.1 T argets ............................................................................................................................ 27 7.2 Sustainability M easures .............................................................................................. 28 8 REFERENCES ..................................................................................................................... 29 5

REVISED VERSION: September 2011, based on public comment received.

2 INTRODUCTION Amendment 2 to the Atlantic States Marine Fisheries Commission Shad and River Herring Interstate Fishery Management Plan was adopted in 2009. It requires member states to demonstrate that fisheries for river herring (alewife and blueback herring) within state waters are sustainable. A sustainable fishery is defined as one that will not diminish potential future reproduction and recruitment of herring stocks. If states cannot demonstrate sustainability to ASMFC, they must close their herring fisheries.

The following proposes a plan for a sustainable fishery for river herring in waters of New York State. The goal of this plan is to ensure that river herring resources in New York provide a source of forage for New York's fish and wildlife and provide opportunities for recreational and commercial fishing now and in the future.

The fisheries that existed back in colonial days in the Hudson Valley of New York undoubtedly included river herring among the many species harvested. River herring, comprised of both alewife (Alosa pseudoharengus),and blueback herring (Alosa aestivalis) were among the fish mentioned by early explorers and colonists - the French Jesuits, Dutch and English.

Archaeological digs along the Hudson in Native American middens indicates that the fishery resources in the river provided an important food source to Native Americans.

Written records for river herring harvest in New York begin in the early 1900. Landings peaked in the early 1900s and in the 1930s and then declined through the 1980s. Landings increased again through 2003, but have since declined. Factors in addition to fishing have affected the stocks: habitat destruction (filling of shallow water spawning habitat) and water quality problems associated with pollution that caused oxygen blocks in major portions of the river (Albany and New York City). Water quality has improved over the last 30 years.

New York State does not augment wild river herring stocks with hatchery progeny. The New York City Parks Department initiated an experimental restoration program in which alewife were captured in a Long Island Sound tributary in Connecticut and released in the Bronx River above the first barrier. Limited returns to the river suggest that some reproduction has occurred from these stockings. A variety of non-governmental organizations along with state and federal agencies are working on development of fish passage for alewife in Long Island streams 3 MANAGEMENT UNITS The management unit for river herring stocks in New York State comprises three sub-units. All units extend throughout the stock's range on the Atlantic coast.

• The largest consists of the Hudson River Estuary from the Verrazano Narrows at New York City to the Federal Dam at Troy including numerous tributary streams (Figure 1).

" The second is made up of all Long Island streams that flow into waters surrounding Long Island and streams on the New York mainland (Bronx and Westchester Counties) that 6

REVISED VERSION: September 2011, based on public comment received.

flow into the East River and/or Long Island Sound (Figure 2).

• The third subunit consists of the non-tidal Delaware River and tributaries upriver of Port Jervis, NY.

Range of the New York river herring along the Atlantic coast is from the Bay of Fundy, Canada and Gulf of Maine south to waters off Virginia (NAI 2008).

A listing of most Hudson River tributaries, and streams on Long Island, and the Bronx and southern Westchester Counties are in Appendix Table A.

3.1 Description of the Management Unit Habitat 3.1.1 Hudson River and tributaries HabitatDescription The Hudson River Estuary is tidal its entire length of 246 km from the Battery (tip of Manhattan Island) in New York City to the Federal Dam at Troy (Figure 1). The estuary is fresh water above Newburgh (km 90).

The estuarine portion of the Hudson River is considered a "drowned" river valley in that the valley slopes steeply into the river. Many of the tributaries below the Troy Dam are tidal for a short distance (usually about a kilometer) ending at a natural or man-made barrier, often built on a natural barrier. There are approximately 67 primary and secondary, both named and unnamed, tributaries to the tidal portion of the Hudson River Estuary (Figure 1). Schmidt and Cooper (1996) catalogued 62 of these tributaries for the presence or absence of barriers to migratory fish.

They found that only one had no barrier for migratory fish, 31 were blocked (either partially or completely) by natural barriers, and the remaining 30 had artificial barriers, dams or culverts, that reduced or eliminated access for fish. We estimated stream length of all these tributaries to be about 97 km that is accessible to river herring below the first impassable man-made or natural barrier.

The Mohawk River is the largest tributary to the Hudson River. It enters the Hudson 2 km north of the Troy Dam. Cohoes Falls, a large scenic waterfall of 20 m is the first natural barrier on the Mohawk just upriver of the confluence with the Hudson. Access into the Mohawk system was created through the Waterford Flight - a series of five locks and dams, built as part of the Erie Canal to circumvent the falls. The canal lock and dam system was built in 1825, to connect the Hudson to central New York and Lakes Ontario and Erie. The Canal parallels and/or is part of the Mohawk River for the river's entire length to Rome, a distance of 183 km. A series of permanent and seasonal pools make up the canal where it intertwines with the Mohawk River.

Permanent pools created from hydro-power dams are found in the Waterford section. Temporary pools are created each year in early spring by removable dams (series of gates) that increase water levels to 14 feet (4.3 m) while the canal is in operation (May through November). During the winter months, the river is returned to its natural state of riffles and pools.

7

REVISED VERSION: September 2011, based on public comment received.

Habitat Use Hudson River alewife, blueback herring and American shad are spring spawners. Alewives are the first of the herring to enter the estuary, arriving as early as mid-March with continued spawning through early May. Blueback herring prefer slightly warmer temperatures and arrive later, usually in April.

Adults of both species spawn in Hudson River tributaries and in the shallow waters of the main stem Hudson. Alewife prefer to spawn over gravel, sand and stone in back water and eddies whereas bluebacks tend to spawn in fast moving water over a hard bottom. Herring spawn in the tidal freshwater Hudson from Kingston (km 144) to Troy (km 256) (Figure 1) and its tributaries for approximately six to ten weeks, dependent on water temperature (Smith 1985, Hattala et al.

2011). Once spawning ends, most mature fish quickly return to ocean waters. The nursery area includes the spawning reach and extends south to Newburgh Bay (km 90), encompassing the freshwater portion of the Estuary.

Some blueback herring of the Hudson River migrate above the Federal Dam at Troy. A few continue upriver in the non-tidal Hudson as far as Lock 4 on the Champlain Canal (NAI 2007).

However, most fish turn west into the Mohawk River. This larger portion migrates as far inland as Rome (439 km inland), via the Erie Canal and the Mohawk River. The canal system opens in New York on or about May st. Since most alewives are already spawning by then, they do not move into the system (J. Hasse, NYSDEC retired, personal communication).

Blueback herring began colonizing the Mohawk River in the 1970s. By 1982, they had migrated into Oneida Lake in the Great Lakes drainage. The number of herring using the Mohawk increased through the 1990s, but since 2000 herring have rarely occurred in the upper end of the River. Blueback herring were historically unable to access the Mohawk River until the locks of the Erie Canal provided upstream passage into the system. Now that they are established, however, they have become important forage for local sport fish populations.

3.1.2 Long Island and Westchester County The herring runs in streams on Long Island are comprised almost exclusively of alewife (B.

Young, NYSDEC retired, personal communication). Most streams are relatively short runs to saltwater from either head ponds (created by dammed streams) or deeper kettle-hole lakes. Either can be fed by a combination of groundwater, run-off or area springs. Spawning occurs in April through May in the tidal freshwater below most of the barriers. Natural passage for spawning adults into the head ponds or kettle lakes is present in very few streams.

There have been limited efforts to understand river herring runs on Long Island since 1995.

Several known runs of alewives on Long Island occur in East Hampton, Southampton, Riverhead and Brookhaven. With the advent of a more aggressive restoration effort in Riverhead on the Peconic River other runs have come to light. Since 2006, an annual volunteer alewife spawning 8

REVISED VERSION: September 2011, based on public comment received.

run survey has been conducted. This volunteer effort basically documents the presence or absence of alewives in Long Island Coastal Streams. In 2010 a volunteer investigation was initiated to quantify the Peconic River alewife run. Size and sex data have been collected for 2010 and 2011. A crude estimate of the runs size was also made in 2010, this effort was improved during 2011 with the placement of a video camera for recording alewife passage through the fish passage. These efforts have been undertaken to understand the Long Island Coastal streams and to improve the runs that exist there.

We have no record of river herring in any of the streams in southern Westchester County. In the Bronx River (Bronx County) alewives were introduced to this river in 2006 and 2008 and some adult fish returned in 2010. Monitoring of this run is in its early stages.

3.1.3 Delaware River No records exist to document the presence of river herring in the New York portion of the Delaware River.

3.2 Habitat Loss and Alteration Hudson River: Much spawning and nursery habitat in the upper half of the tidal Hudson was lost due to dredge and fill operations to maintain the river's shipping channel to Albany. Most of this loss occurred between the end of the 19th century (NYS Department of State 1990) and the first half of the 20th century. Preliminary estimates are that approximately 57% of the shallow water habitat (1,821 hectares or 4,500 acres) north of Hudson (km 190) was lost to filling (Miller and Ladd 2004). Work is in progress to map the entire bottom of the Hudson River. Data from this project will be used to characterize and quantify existing spawning and nursery habitat. While most of the dredge and fill loss affected American shad, it is suspected that herring were also affected as they spawn along the shallow water beaches in the river.

Very little, or no, habitat has been lost due to dam construction. The first major dam was constructed in 1826 at Rkm 256 at Troy. Prior to the dam, the first natural barrier occurred at Glens Falls, 32 km above the Troy Dam. The construction of the dam is not known to have reduced spawning or nursery habitat.

The introduction of zebra mussels in the Hudson in 1991, and their subsequent explosive growth in the river, quickly caused pervasive changes in the phytoplankton (80% drop) and micro- and macro- zooplankton (76% and 50% drop respectively) communities (Caraco et al. 1997). Water clarity improved dramatically (up by 45%) and shallow water zoobenthos increased by 10%.

Given these massive changes, (Strayer et al. 2004) explored potential effects of zebra mussel impact on young-of-year (YOY) fish species. Most telling was a decrease in observed growth rate and abundance of YOY fishes, including both alewife and blueback herring. It is not yet clear how this constraint affects annual survival and subsequent recruitment.

Long Island: Most all streams on Long Island have been impacted by human use as the 9

REVISED VERSION: September 2011, based on public comment received.

population expanded. Many streams were blocked off with dams to create head ponds, initially used to contain water for power or irrigation purposes. The dams remain; only a few with passage facilities. Many streams were also impacted by the construction of highways, with installations of culverts or other water diversions which impact immigrating fish.

Recent efforts at restoration look to provide fish passage over or around these barriers, or even removal of small obstructions. Permanent fish passage was recently installed on the Carmans River in the South Shore Estuary near Shirley, NY. This project was the result of advocacy and cooperation by environmental groups and local, state and federal agencies. Additional protections for the River are assured due to legislation enacted in 2011, and community awareness is building. An earlier cooperative effort resulted in the installation of a rock ramp passage in the Peconic River within the Peconic Bays Estuary. Local citizens monitor the spring alewife run in this river. As awareness of these successful efforts spreads, interest in replicating that success on other systems grows.

3.3 Habitat Water Quality The Hudson has a very long history of abuse by pollution. New York City Department of Environmental Protection recognized pollution, primarily sewage, as a growing problem as early as 1909. By the 1930s over a billion gallons a day of untreated sewage were dumped into New York Harbor. (NYCDEP http:/!/home2.nvc.gov/htmni/dep/htm l/news/hwqs.shtml )

New York City was not the only source of sewage. Most major towns and cities along the Hudson added their share. It was so prevalent that the Hudson was often referred to as an open sewer. Biological demand created by the sewage created oxygen blocks that occurred seasonally (generally mid to late summer) in some sections of the river. One of the best known blocks occurred near Albany in the northern section of the tidal estuary in the 1960s through the 1970s.

This block often developed in late spring and remained through the summer months. It essentially cut off the upper 40 km of the Hudson for use as spawning and nursery habitat. A second oxygen block occurred in the lower river in the vicinity of New York City in late summer. This block could potentially have affected emigrating age zero river herring. This summer oxygen-restricted area occurred for decades until 1989 when a major improvement in a sewage treatment plant came on line in upper Manhattan. It took decades, but water quality in general has greatly improved in both areas since the implementation of the Clean Water Act in the 1970s and subsequent reduced sewage loading to the river.

4 STOCK STATUS Following is a description of all available data for the Hudson's river herring stocks, plus a brief discussion of their usefulness as stock indicators. Sampling data are summarized in Tables 1 and

2. Sampling was in support of Goal 1 of the Hudson River Estuary Action Agenda and has been partially funded by the Hudson River Estuary Program.

10

REVISED VERSION: September 2011, based on public comment received.

4.1 Fisheries Dependent Data 4.1.1 Commercial Fishery Commercial fisheries for river herring in New York State waters occur in the Hudson River Estuary and in marine waters around Long Island. Current commercial fishing restrictions for New York waters are listed in Appendix Table B.

The present commercial fishery in the Hudson River and tributaries exploits the spawning migration of both alewife and blueback herring. The primary use of commercially caught herring is for bait in the recreational striped bass fishery. The herring fishery occurs from March into early June annually, although some fishers report catching herring as late as July.

Ocean bycatch River herring occur as bycatch in many commercial fisheries which are in the known migratory range of the Hudson stock from North Carolina up to the Gulf of Maine. Fishery bycatch is mostly un-documented but has the potential to harvest Hudson stock and many other stocks along the coast. In some years, estimated bycatch of river herring in the Atlantic herring fishery equaled or exceed the total of all coastal in-river landings (Cieri et al. 2008). More recent analyses by the National Marine Fisheries Service's Northeast Fisheries Science Center (2011) indicated that total annual incidental catch of river herring in all fishing fleets sampled by the Northeast Fisheries Observer Program during 1989-20 10 ranged from 108 to 1867 mt. It is not known how much of current ocean river herring bycatch consists of Hudson River fish.

This SFP does not directly address ocean bycatch but focuses on fisheries in New York State waters. New York is working with the National Marine Fisheries Service, the New England Fishery Management Council (www.nefinc.org) and the Mid-Atlantic Fishery Management Council (w-,w.mafinc.org) to deal with this issue. Both councils are in the process of amending the Atlantic Herring (Amendment 5) and the Atlantic Mackerel, Squid and Butterfish (Amendment 14) Plans to reduce bycatch of river herring.

Gear Use in the Hudson River and Tributaries The fixed gill net fishery occurs in the mainstem river from km 40 to km 75 (Piermont to Bear Mountain Bridge, Figure 1). In this stretch, the river is fairly wide (up to 5.5 km) with wide, deepwater (- six to eight m) shoals bordering the channel. Fishers use particular locations within this section away from the main shipping channel. Over the past ten years, an average of 22 active fishers participated in this lower river fixed gill net fishery annually. Nets are 3.7 to 183 m (12 to 600 ft) long. Above the Bear Mountain Bridge gill net fishers use both drift (-58%) and fixed gill nets (-42%). These gears are used up to km 225 (Castleton) where the river is much narrower (1.6 to 2 km wide). Approximately 60 fishers participate in this mid river gill net fishery. Nets range in size from 7.6 to 183 m (25 to 600 ft).

11

REVISED VERSION: September 2011, based on public comment received.

The other major gear used in the river herring fishery is scap nets (also known as lift and/or dip nets). The scap/lift net fishery occurs from km 70 to km 130 (Peekskill to New Baltimore),

primarily in the major river herring spawning tributaries. Scap/lift nets range in size from 0.2 to 121.9 m 2 (0.5 to 400ft2). On average, about 96 fishers participate annually.

Marine permits are required of fishers to use seines or scap nets greater than 36 ft2, dip or scoop nets exceeding 14 in. in diameter, and all gill nets. Marine permit holders are required to report effort and harvest annually to the Department. Many marine permit holders are recreational anglers taking river herring for personal use as bait or food. It should be noted that over the last ten years, an average of over 260 gill net and 260 scap nets permits were sold annually.

According to the required annual reports, however, only 36% of the permitees actively catch fish.

In addition to Marine permits, New York has a bait license that allows the take and sale of bait fish (river herring included) using seines and cast nets. As no reporting is required for this license, harvest of river herring using this license is unknown.

Commercial Landings andLicense Reporting Recorded landings of river herring in New York State began in the early 1900s. Anecdotal reports indicate that herring only played a small part in the historic commercial fishing industry in the Hudson River. Total New York commercial landings for river herring include all herring caught in all gears and for both marine and inland waters. Several different time series of data are reported including several state sources, National Marine Fisheries Service (NMFS), and more currently Atlantic Coastal Cooperative Statistics Program (ACCSP). NMFS data do not specify river or ocean source(s) and landings are often reported as either alewife or blueback herring, but not both in a given year. It is unlikely that only one species was caught. From 1995 to the present, the Department has summarized landings and fishing effort information from mandatory state catch reports required for Hudson River marine permits. Full compliance for this reporting started in 2000. All Hudson River data are sent to NMFS and ACCSP for incorporation into the national databases.

Because of the discrepancies among the data series and the lack of information to assign the landings to a specific water body source, only the highest value from all sources is used to avoid double counting. Several peaks occur in the river herring landings for New York (Figure 3). The first peak occurred in the early 1900s followed by a lull (with some gaps) until the period prior to, during, and after World War II when landing peaked a second time. By the 1950s landings were in a serious decline. A few unusual peaks occurred in the NMFS data series. In 1966, 1.9 million kg were landed (omitted on Figure 3), followed by a series of years of low landings with another peak in 1982. Landings were low, with some data gaps during the rest of the 1980s through 1994.

Hudson River landings 12

REVISED VERSION: September 2011, based on public comment received.

Since 1995, landings have been separated between the Hudson and other water (marine). Harvest in the river was relatively low in 1995, but grew in response to the need for bait for the expanding striped bass recreational fishery. In-river landings peaked in 2003 and have slowly declined since then (Figure 4). The reason for the decline is unknown. The striped bass fishery and the need for bait have not diminished. It is possible that recreational fishers have shifted harvest to non commercial gears which do not have a mandatory reporting requirement. The landings from these "personal use" gears are unknown. Reporting rate from fishers using commercial gear is unknown.

The primary outlet for harvest taken by Hudson River marine permits is for the in-river bait industry. Since 2000, most commercially caught river herring have been taken by scap/lift nets (10 year mean of 48% of the catch) (Figure 5). The remaining 52% was split between drift and fixed gill nets.

Commercial Discards From 1996 to 2010, river herring were not reported as discards on any mandatory reports targeting herring in the Hudson River or tributaries. Our commercial fisheries monitoring data, however, (See program description below) suggests otherwise. Since 1995, we have observed a 0.12% rate of discard in the anchored gill net fishery. Reasons for discards are unspecified.

Discard rates are unknown for ocean fisheries.

Hudson River Commercial Catch Rates - MandatoryReports Relative abundance of river herring is tracked through catch per unit effort (CPUE) statistics of fish taken from the targeted river herring commercial fishery in the Estuary. All commercial fishers annually fill out mandatory reports. Data reported include catch, discards, gear, effort, and fishing location for each trip. Data within week is summarized as total catch divided by total effort (square yards of net x hours fished), separately by gear type (fixed gill nets, drift gill nets, and scap nets). Annual means are summarized in two ways. Above the Bear Mountain Bridge and within the spawning reach, annual CPUE is calculated as total catch/total effort. Below the Bear Mountain Bridge (km 75) and thus below the spawning reach, annual CPUE is calculated as an annual sum of weekly CPUE. Here, nets capture fish moving through to reach upriver spawning locations and run size is determined by number (density) of spawners each week as well as duration (number of weeks) of the run. The sum of weekly CPUE mimics area under the curve calculations where sampling occurs in succeeding time periods. The downside of using reported CPUE to monitor relative abundance is that results can be influenced by inter-annual, location, and inter-gear differences in reporting rate.

We use the CPUE of the fixed gear fishery below the Bear Mountain Bridge for estimating relative abundance because effort expended by the fishery below this bridge is much greater

(-70% of fixed gill net effort) than in the river above this point (remaining 30%). Moreover, fixed gear below the bridge (rkm 40 to 75) is always fished in relatively the same location each year, is passive in nature, and intercepts fish that pass by. Annual CPUE for the lower river fixed gill net remained relatively flat until 2006 and has since increased (Figure 6).

13

REVISED VERSION: September 2011, based on public comment received.

We do not consider the CPUE of gears fished above the Bear Mountain Bridge and within the spawning reach as reliable an annual abundance indicator as that from fixed gill nets below the bridge. Upriver gears catch fish that are either staging (getting ready to spawn) or moving into areas to spawn and gears are generally not employed until fish are present. The gears include drift gill nets, scap nets and some fixed gill nets (Figure 5). Drift gill net CPUE is also more variable as it can be actively fished - set directly into a school of fish. Drifted gill net CPUE varied widely without trend through the time period. Scap net CPUE declined slightly from 2000 through 2003, and has since remained relatively stable (Figure 6). Fixed gill nets fished within the spawning reach show the same recent increasing trend as lower in the river, but effort expended is much less than below Bear Mountain Bridge.

Hudson River Commercial Catch Rates - Monitoring Program Up until the mid-1990s, the Department's commercial fishery monitoring program was directed at the American shad gill net fishery, a culturally historic and economically important fishery.

We expanded monitoring to the river herring fishery in 1996, but were limited by available manpower and the ability to connect with the fishers. Monitoring focused on the lower river fixed gill net fishery since we considered it to be a better measure of annual abundance trends (see section above).

Data were obtained by observers onboard fishing vessels. Technicians recorded data on numbers of fish caught, gear type and size, fishing time and location. Scale samples, lengths and weights are taken from a subsample of the fisher's catch. CPUE was calculated by the method used for summarizing mandatory report data (above).

Since 1996, 66 trips targeting river herring (lower river: 53; mid and upper river: 13) have been monitored. These trips were sporadic and sample size is low, from one to 11 trips per year.

Because of these few samples, the resulting CPUE is considered unreliable for tracking relative abundance. However, active monitoring provided the only data on catch composition of the commercial harvest and we consider these data to be useful.

Commercial Catch Monitoring-Size andAge Structure Commercial fixed gill net fishers use I 3/4 to 2 3/4 inch stretch mesh sizes to target herring. Catch composition include fish caught in all meshes. For trend analysis of size change, we subset the data to include only fish caught in similar size mesh each year; these include gill nets of 2 1/22 and 2 3/4 inch mesh.

Catch composition varied annually most likely due to the low number of monitored trips each year, and the timing of when the trips occurred. Annual sample size was relatively low, ranging from 40 to 185 fish from 2001 to 2007 (Table 3). Alewives were observed more often than blueback herring. The species difference may be the result of when the samples occurred (early or late in the run). The sex ratio of alewife in the observed catch was nearly equal (- 50:50) in all years; more blueback herring females were caught than males (60:30 ratio). From 2001 to 2010, 14

REVISED VERSION: September 2011, based on public comment received.

a slight decline was observed in mean total length (mm) for both alewife and blueback herring (Figure 7).

Age data for samples collected during the commercial monitoring program are yet to be analyzed (see discussion in Age section under FI programs below).

4.1.2 Recreational Fishery Hudson River and tributaries:The recreational river herring fishery exists throughout the main-stem Hudson River, and its tributaries including those in the tidal section and above the Troy Dam (Mohawk River). Herring are sought from shore and boat by angling (jigging) and multiple net gears (see Appendix B). Boat fishers utilize all allowable gears while shore fishers predominantly use scap/lift nets, or angling (jigging). Some recreational herring fishers use their catch as food (smoking/pickling). However, the recreational herring fishery is driven primarily by the need for bait in the striped bass fishery.

The magnitude of the recreational fishery for river herring is unknown for most years. NYSDEC contracted with Normandeau Associates, Inc. to conduct creel surveys on the Hudson River in 2001 and 2005 (NAI 2003 and 2007). Estimated catch of river herring in 2001 was 34,777 fish with a 35.2% retention rate. When the 2001 data were analyzed, NAI found that the total catch and harvest of herring was underestimated due to the angler interview methods. In the 2001 survey, herring caught by fishers targeting striped bass were only considered incidental catch, and not always included in herring total catch and harvest data. Fishers were actually targeting herring and striped bass simultaneously. Corrections were made to the interview process for the 2005 survey and estimated catch increased substantially to 152,117 herring with an increased retention rate of 75.1% (Table 4). Although some fish were reported as released, we consider these mortalities due to the herring's fragile nature. We also adjusted the 2001 catch using the 2005 survey data. The adjusted catch rose to 93,157 fish.

We also evaluated river herring use by striped bass anglers using data obtained from our Cooperative Angler Program (CAP). The CAP was designed to gather data from recreational striped bass anglers through voluntary trip reports. Volunteer anglers log information for each striped bass fishing trip including fishing time, location, bait use, and fish caught, including length, and weight, and bycatch. In 2006 through 2010, volunteer anglers were asked to provide specific information about herring bait use. The annual proportion of angler days where herring was used for bait ranged from 71% to 93 % with a mean of 77%. The proportion of herring used by anglers that were caught rather than purchased increased through the time period (Table 4).

Herring caught per trip varied from 1.6 to 4.8 and with the highest values in the last two years.

Herring purchased per trip ranged from 0.63 to 1.5 with the lowest value in 2009. We calculated the total number of herring caught or purchased by striped bass anglers in 2007 as the estimated number of striped bass trips from a statewide creel survey (90,742)

• average proportion of angler days using herring in the CAP in 2007 (0.77)

• number of herring caught or purchased per trip in the CAP (1.8 and 1.7). The result was 125,502 caught and 115,816 bought for a total of 241,318 herring used.

15

REVISED VERSION: September 2011, based on public comment received.

The number of river herring taken from the Hudson River and tributaries for personal use as food by anglers is unknown.

Long Island. Alewives can be caught in many of the small streams on Long Island, though only the Peconic River sees more than occasional effort. No creel data are available but anecdotal information (B. Young, NYSDEC retired, personal communication) suggests that harvest is rising in the more easily accessible streams. Herring taken are used for personal consumption as well as for bait.

The town of Southampton, on Long Island's East End, has local ordinances in place to prevent fishing (dipping) during the alewife spawning runs.

Bronx and Westchester Counties: We do not know if any fishery occurs in the streams in Bronx and Westchester Counties that empty into the East River and Long Island Sound.

4.2 Fishery Independent Surveys 4.2.1 Spawning Stock Surveys - Hudson River Several surveys have sampled the alewife and blueback herring spawning stocks of the Hudson River and tributaries. The spawning stocks are made up of the fish which have escaped from coastal and in-river commercial and recreational fisheries.

The earliest data is from a biological survey of the Hudson in 1936 by the then New York State Conservation Department (Greeley 1937). The sample size was small (25 fish) but indicates the fish were relatively large compared to recent data. More recent data on river herring come from several Department surveys. The longest dataset (1975-2000) is from an annual survey of chemical contaminants in fish that targeted multiple species within the Hudson River estuary.

Fish were collected by electro-fishing and river herring sample size varied among years. In most years, length data were recorded for a sub sample of herring. The Department also conducted a two-year electro-fishing survey in 1989 and 1990, to examine the population characteristics of blueback herring in the Hudson and the Mohawk River, the Hudson's largest tributary. Data were obtained on length, age, and sex.

Limited data on river herring stock characteristics have also been collected during annual monitoring of American shad and striped bass spawning stocks. Sampling occurs in the main-stem Hudson River between km 145 and 232 from late April through early June. Fish are collected by haul seines and electro-fishing. The 10.2 cm stretch mesh in the haul seines was specifically designed to catch shad and striped bass and avoid river herring, but some large (>

280mm) herring were occasionally retained in these gears. Herring were an incidental catch of the electro-fishing. Data were collected on length, age, and sex of river herring caught in both gears.

16

REVISED VERSION: September 2011, based on public comment received.

In 1987, the Department began to target adult river herring during the spring spawning stock survey. From 1987 to 1990, two small mesh (9.5 mm) beach seines (30.5 and 61m) were occasionally used with some success. In 1998, we specifically designed a small haul seine (91 m) with an appropriate mesh size (5.1 cm) to target herring. It was designed to capture all sizes of herring present with the least amount of size, and age, bias. We have used this gear since 1999.

Sampling occurs during the shad and bass survey within the area described above, using the same field crew.

We only use data from the least size-biased gears to describe characteristics of the herring spawning stock: electro-fishing, the beach seine (61m) and the herring haul seine (91 m). As sample size varied among years, all data were combined to characterize size and weight composition of the spawning population. Mean total length and weight data are summarized for adults only (>=170mm TL).

4.2.2 Hudson River Spawning Stock - Characteristics Mean Size and Growth Mean size of fish has been calculated for all years that samples were obtained (Figure 8). Sample size is relatively small, however, in most years presented (n<34 fish). Adequate samples (n>34),

following the method described by Lynch and Kim (2010) to characterize length (depicted with an X over the graph's data point) were collected in the late 1980s, early 1990s, then occasionally since 2001 for both species. Lengths have declined since the early 1980s. Since 2000, mean size of female alewife has been stable, but declined slightly in males (Figure 8). Mean size of blueback herring has declined for both sexes from 1989 to the present.

Age The Department samples from the 1989-1990 were primarily blueback herring. The aging method used was that of Cating (1954), developed for American shad. More recent scale samples from Department surveys remain un-aged and therefore we have limited age or repeat spawn data directly from scales of Hudson River fish. In attempting to age Hudson River herring scales, we relied on techniques used by other state agencies. As an alternative, and for a very general picture of potential age structure, we estimated annual age structure using length at age keys from datasets provided by Maine, Massachusetts, and Maryland for alewife and Massachusetts and Maryland for blueback herring. We found that three state agencies differ enough in their technique to produce variation in the results.

Blueback herring: Age estimates using length-age keys differed from ages assigned by the Department for the 1989- 1990 samples and from each other for most years (Figure 9). In general, keys from MD and MA were mostly in agreement for male blueback herring in most years, but MA aged females slightly older (Figure 10). Ages from two through eight were present in the spawning stock. Most fish were ages three, four, and five. Mean age remained 17

REVISED VERSION: September 2011, based on public comment received.

relatively stable among years within method (Figure 11).

Alewife: Age estimates using length-age keys from the three states differed from each other for alewife (Figure 12). In general, the ME key resulted in the youngest ages, followed by older ages from MA, then MD. Ages from two through eight or nine were present in the spawning stock.

Peak age varied with key used and by sex; most fish were ages three or four for males and four or five for females. Mean age was youngest for the ME key, older for MA, and oldest for MD age key (Figure 13). Mean age for males was greater in 2001 and 2003, then dropped and remained relatively stable for 2005 through 2010. Mean age for females was slightly lower in 2008 and 2009 but by 2010 returned to the same level as estimated for 2001 and 2003.

Maximum age that the Hudson River herring stock can attain is unknown. Jessop (B. Jessop DFO retired, personal communication) reported a maximum age of 12 for both alewife and blueback herring for the St. John's River in New Brunswick.

Given current uncertainty about aging methods and age of Hudson River river herring, we suggest that available estimates should only be used for a general discussion of age structure and for trends within estimate method. We do not feel that age estimates should be used to monitor changes in stock status or to set sustainable fishing targets until aging methods can be verified.

This issue is currently being discussed in the ongoing ASMFC River Herring stock assessment where resolution to the differences in ageing methods is being sought..

Mortality Estimates The variation in annual age structure translated into comparable variation in estimates of total mortality when various age-based estimation methods were used. This difficulty in estimating ages precluded the use of age-based mortality estimators. As an alternative, we explored use of the Beverton-Holt length-based method (Gedamke and Hoenig 2006) using growth parameters for length calculated from the 1936 length at age data (see section above). Since the definition of length at full recruitment (Lc) given by Nelson et al. (2010) seemed arbitrary, we estimated total mortality using the Nelson et al. (2010) and two additional Lc values. Results from the length based method were also influenced by Loo. The Beverton-Holt method also relies on several population assumptions including continuous recruitment to the stock that the population is in equilibrium. Neither of these assumptions are true for Hudson herring stocks.

Total mortality estimates for alewife of both sexes varied tremendously within and among years depending on assumed model inputs (Figure 14). Estimates increased until 2006, after which a decline occurred to 2010. An even greater variation occurred for blueback herring (Figure 15) with a series of very high peaks followed by low values. Given this demonstrated sensitivity to model inputs, we suggest that total mortality of Hudson River river herring stocks remains unknown. However, we should emphasize that mortality on stocks must have been high in the last 30 years to have so consistently reduced mean size and presumably mean age. We do not feel that estimates of total mortality should be used to monitor stock change during the proposed experimental fishery unless uncertainty in estimation methodology can be resolved. Current uncertainty precludes use of total mortality to set sustainability targets.

18

REVISED VERSION: September 2011, based on public comment received.

4.2.3 Spawning Stock Surveys - Long Island Young (2011) sampled alewife in the Peconic River 32 times throughout the spawning season in 2010. Sampling occurred by dip net just below the second barrier to migration at the lower end of a tributary stream. A rock ramp fish passage facility was completed at the first barrier near the end of February 2010. The author collected data on total length and sex and estimated the number of fish present based on fish that could be seen below the barrier. Peak spawning occurred during the last three weeks of April. The minimum estimate of run size was 25,000 fish and was the total of the minimal visual estimates made during each sample event. Males ranged from 243- 300 mm with a mean length of 263 mm. Females ranged from 243-313 mm with a mean of 273 mm.

4.2.4 Volunteer and Other river herring monitoring The Department's Hudson River Fisheries Unit (HRFU), Hudson River Estuary Program and the Environmental Defense's South Shore Estuary Reserve Diadromous Fish Workgroup (SSER) have begun to incorporate citizen volunteers into the collection of data on temporal variation of and physical characteristics associated with spawning of river herring in tributaries. These data were not provided by the fishery dependent and independent sample programs discussed above.

The volunteer programs also bring public awareness to environmentally important issues.

Long Island Streams The SSER began a volunteer survey of alewife spawning runs on the south shore of Long Island in 2006. The survey is designed to identify alewife spawning in support of diadromous fish restoration projects. The survey also evaluates current fish passage projects (i.e. Carmans River fish ladder), and sets a baseline of known spawning runs. Data were available for surveys in 2006 - 2008. Monitoring occurred on six to nine targeted streams annually, with volunteer participation ranging from 24 to 68 individuals. Monitoring takes place from March through May. Alewife were seen as early as March 5 (2006) and as late as May 31 (2008). Data indicated that alewife use multiple streams in low numbers. It is not clear whether each stream supports a spawning population since total sightings were very low. The Carmans and Swan Rivers showed the most alewife activity and likely support yearly spawning migrations. The first permanent fish ladder on Long Island was installed in 2008 on the Carmans River. Information gathered during this study will aid in future construction of additional fish passage (Kritzer et al. 2007a, 2007b and Hughes and O'Reilly 2008).

In addition to the SSER, other interested individuals have also monitored Long Island runs (see Appendix Table A). Anecdotal data provides valuable information on tracking existing in-stream conditions, whether streams hold active or suspected runs, interaction with human land uses and suggestions for improvement (L. Penney, Town of East Hampton, personal communication). A rock ramp was constructed around the first barrier to migration on the Peconic River in early 19

REVISED VERSION: September 2011, based on public comment received.

2010 (B. Young, retired, NYS Dept of Environmental Conservation, personal communication).

The Peconic River Fish Restoration Commission set up an automated video counting apparatus at the upriver end of this ramp. Data are still being analyzed.

The Department has conducted a similar river herring volunteer monitoring program annually since 2008 for tributaries of the Hudson River Estuary (Dufour et al. 2009, NYSDEC 2010, Hattala et al. 2011). We designed this project to gather presence-absence and temporal information about river herring spawning runs from the lower, middle and upper tributaries of the Estuary. Between nine and 11 tributaries were monitored annually by 70 to 213 volunteers in 2008, 2009, and 2010. Herring were seen as early as 31 March and as late as 1 June. River herring were observed in all but one of the tributaries. However, several tributaries with known strong historical runs had very few sightings. Water temperature seemed to be the most important factor determining when herring began to run up a given tributary. Sightings of herring were most common at water temperature above 50 F. Tributaries in the middle part of the estuary warmed the fastest each spring and generally had the earliest runs.

4.2.5 Young-of-the-Year Abundance Since 1980, the Department has obtained an annual measure of relative abundance of young-of-the-year (YOY) alewife and blueback herring in the Hudson River Estuary. Although the program was designed to sample YOY American shad, it also provides data on the two river herring species. Blueback herring appear more commonly than alewife. In the first four years of the program, sampling occurred river-wide (rkm 0-252), bi-weekly from August through October, beginning after the peak in YOY abundance occurred. The sampling program was altered in 1984 to concentrate in the freshwater middle and upper portions of the Estuary (km 88-225), the major nursery area for young herring. Timing of samples was changed to begin in late June or early July and continue biweekly through late October each year. Gear is a 30.5 m by 3.1 m beach seine of 6.4 mm stretch mesh. Collections are made during the day at approximately 28 standard sites in preferred YOY herring habitat. Catch per unit effort is expressed as annual geometric and arithmetic means of number of fish per seine haul for annual weeks 26 through 42 (July through October). This period encompasses the major peak of use in the middle and upper estuary.

From 1980 to 1998, the Department's geometric mean YOY annual index for alewife was low, with only one year (1991) over one fish per haul. Since 1998, the index has increased erratically (Figure 16).

From 1980 through 1994, the Department's geometric mean YOY annual index for blueback herring averaged about 24 fish per haul, with only one year (1981) dropping below 10 fish per haul (Figure 16). After 1994, the mean dropped to around 17 fish per haul, and then began the same high-low pattern observed for alewife.

The underlying reason for the wide inter-annual variation in YOY river herring indices is not clear. The same erratic trend that occurred since 1998 has also occurred in American shad 20

REVISED VERSION: September 2011, based on public comment received.

(Hattala and Kahnle 2007). The increased inter-annual variation in relative abundance indices of all three Alosines may indicate a change in overall stability in the system.

4.2.6 Conclusion Over the last 30 years, the Hudson River stocks of alewife and blueback herring have shown inconsistent signs in stock status trends. Calculated CPUE for commercial gill net gears has increased in recent years, while CPUE in scap nets fished in tributaries initially declined, but has remained relatively stable since 2003. Apparent mortality increased on mature fish and as mortality rose, mean total length and weight declined. Similar trends occur in the both the fishery dependent and independent data. Recruitment has become extremely variable since the mid-1990s for both species. Some decline is occurring for YOY blueback herring while, counter-intuitively, there has been an increasing trend for YOY alewife. Anecdotal evidence from anglers and commercial fishermen suggest a decline in abundance in tributaries yet a dramatic increase of herring in the main-stem river in the last few years.

The upsurge in river herring used as bait for striped bass has placed herring in a tenuous position.

With this continuing demand, declining size, and increasing mortality, careful management is needed despite variable but stable recruitment.

5 PROPOSED FISHERY CLOSURES 5.1 Long Island, Bronx County and Westchester County Limited data that have been collected for Long Island river herring populations are not adequate to characterize stock condition or to choose a measure of sustainability. Moreover, there are no long-term monitoring programs in place that could be used to monitor future changes in stock condition. In 2010, the Peconic River Fish Restoration Commission installed a rock ramp to provide fish passage at the first dam on the Peconic River system. In the spring of 2011, a fish counting apparatus was installed upriver of this ramp. In addition, the Commission initiated biological fish sampling of species, sex, length and scales. If these operations continue in the future and if these provide information that could be used to set and monitor a sustainability target, we will consider a fishery for this river. Little data have been collected for river herring populations in the Bronx and Westchester Counties.

For the above reasons, New York State will close all fisheries for river herring in Long Island streams and in the Bronx and Westchester County streams that empty into the East River and Long Island Sound.

5.2 Delaware River 21

REVISED VERSION: September 2011, based on public comment received.

We have no data that suggest river herring occur in New York waters of the Delaware River.

New York State proposes to close fishing for river herring in New York waters of the Delaware River to prevent future harvest should the Delaware stock rebound and expand upriver. This closure conforms to similar closures planned for the Delaware River and Bay by Pennsylvania, New Jersey, and Delaware.

6 PROPOSED SUSTAINABLE FISHERY 6.1 Hudson River and Tributaries Given the mixed picture of stock status provided by available data on Hudson River herring, New York State proposes a restricted fishery in the main-stem Hudson River coupled with a partial closure of the fishery in all tributaries. We do not feel that the data warrant a complete closure of all fisheries. We propose that the restricted fishery would continue for five years concurrent with annual stock monitoring. We propose a five-year period because the full effect of our proposed restrictions will not become apparent until all age classes in the population have been exposed to the change. Most of the fish in the Hudson River herring spawning stocks are estimated to be three through seven years old and these ages predominate in the fishery.

Sustainability targets would be set juvenile indices. We would monitor, but not yet set targets for mean length from fishery independent spawning stock sampling and CPUE in the commercial fixed gill net fisheries in the lower river below Bear Mountain Bridge. We will also monitor age structure, frequency of repeat spawning, and total mortality (Z) if we can resolve uncertainties about aging methods and mortality estimate methodology. Stock status would be evaluated during and after the five year period and a determination made whether to continue or change restrictions. Moreover, we do not know how much of the apparent high mortality is caused by bycatch in ocean fisheries and thus outside current scope of restrictions proposed in this plan.

Recreational harvest of river herring is much greater than reported harvest from commercial gears. Data from a creel survey in 2005 estimated approximately 152,000 herring were taken in the recreational fishery (NAI 2007) while some 31,000 herring were reported from commercial gears (Table 2). For this reason, we feel that restrictions to the recreational fishery will likely have a greater impact on take of herring than commercial restrictions.

We should note that Draft Addendum 3 to Amendment 6 of the ASMFC Interstate Management Plan for striped bass stipulates that states should reduce fishing mortality on spawning stocks by 50%. If this draft is approved by the ASMFC Striped Bass Management Board, we may have to restrict effort in the recreational striped bass fishery. Restrictions may include a reduction in use of bait such as river herring. Any reduction in effort will likely reduce demand for river herring and thus reduce losses in the Hudson stocks.

A summary of the following fishery restrictions are contained in Tables 5 and 6. These restrictions were based on public comments received from public information meetings held in 22

REVISED VERSION: September 2011, based on public comment received.

the Hudson valley in 2010 in addition to the need to reduce harvest. Public suggestions for restrictions are listed in Appendix C.

6.1.1 Proposed Restrictions - Recreational Fishery Recreationalfishing season Currently none; proposed season is March 15 to June 15.

RecreationalCreelLimit Currently there are no restrictions on daily take of river herring in the Hudson and its tributaries.

To reduce harvest and waste, we propose to implement a restrictive recreational creel limit often river herring per day, or a total maximum boat limit of 50 per day for a group of boat anglers, whichever is less. A Charter boat captain (see Commercial Fishery Restrictions) will be responsible for a possession limit of 10 river herring per paying customer or a total maximum boat limit of 50 herring per day, whichever is less. Charter boat captains are required, at minimum, to hold a US Coast Guard "six pack" license, i,e. a maximum number of six passengers can be on board. However, most vessels fishing the Hudson relatively small (20 to 30 ft) with an average of four fares maximum.

Most of the river herring harvest is driven by striped bass fishermen catching herring for bait.

Anecdotal reports and comments at public meetings suggest that many anglers take many more herring than they need for a day's fishing. The proposed creel limit will prevent such overharvest and avoid waste. We obtained an idea of potential harvest reduction from the proposed creel survey from data in the Cooperative Angler Program described in Section 2.1.3. Data were available on herring harvest during 502 trips. Since trip level reports often included more than one angler, we divided the reported herring catch by the number of anglers for an estimate of catch per angler trip. These data indicated that 56 percent of the catch per angler trips caught six or more herring suggesting that a five fish limit could reduce harvest by 56 percent.

To track harvest, New York will implement the on line creel survey/ diary program coordinated by ACCSP. It is scheduled to go live by Jan. 1, 2012. New York will increase public outreach to encourage angler use of this program. We will also continue the Cooperative Angler Program for comparison and for individuals not savvy with on-line tools.

ProhibitHarvest by Nets in Tributaries Recreational anglers generally use hook and line jigging) in the main-stem river and are allowed to use personal use gears (without a license) of scap/lift nets (36 sq ft or less), small dip nets, and cast nets. They are not required to report this catch and the number of herring taken by these gears is unknown. Anecdotal reports and observations suggest tributaries are popular locations for recreational harvest by these net gears, especially in the middle section of the estuary (Figure 1).

23

REVISED VERSION: September 2011, based on public comment received.

Information from the volunteer angler program along with anecdotal data on recreational harvest suggests that abundance of river herring, mostly alewife, has declined in some spawning tributaries. This may be due to the increased vulnerability to harvest as herring often concentrate in these tributaries in large schools to spawn. Tributaries with an impassable barrier close to the mouth confine fish to even smaller areas. For these reasons, we feel it prudent to close recreational harvest by nets from tributaries until measures of stock condition improve. We did not feel that it was feasible or desirable to enforce a closure on angling for river herring in tributaries.

In the main-stem Hudson, personal use nets will be allowed to continue but with a reduced size for scap/ lift nets (16 sq ft instead of 36 sq ft); seine, cast, and dip nets sizes will remain the same (Table 5).

Closed areas Although personal-use net fishing by recreational anglers will not be allowed in tributaries, angling will continue. However, to further relieve fishing pressure in areas of fish concentration, in addition to the net ban, no fishing will be allowed within the River Herring Conservation Area (RHCA) defined as stream length within 250 m (825 feet) of any type of barrier, natural or man-made. This is similar to a fishing ban within 50 rods of fishways instituted in New York in 1895.

Many of the Hudson's tributaries have natural (rapids) or man-made barriers a short distance in from the main river. River herring concentrate in great numbers below these barriers making them very vulnerable to any fishing. This closed area will allow them to spawn in this undisturbed stretch. The RHCA closure will effectively end all fishing in the eight smallest tributaries, or 14% of the tributaries in the estuary.

Above the Troy Dam, an area closure is already in effect for the "Waterford Flight", Lock 2 to Guard Gate 2, a series of dams and locks at the entrance to the Mohawk River. Within the Mohawk, a RHCA will be in effect below any of the remaining locks and dams up to Lock 21 in Rome.

Escapementperiod None are proposed.

Licensing and reporting In 2011, New York State implemented a recreational marine fishing registration. All anglers fishing for anadromous fish must register prior to fishing for migratory fish of the sea. For the Hudson this includes river herring and striped bass. The recreational and commercial fisheries for American shad were closed in the Hudson River in 2010.

24

REVISED VERSION: September 2011, based on public comment received.

By Jan 1 2012 New York, in cooperation with ACCSP, will start up an online angler survey. The Department will increase public outreach to strongly encourage fishers to use this new tool to aid in understanding recreational catch and harvest.

6.1.2 Proposed Restrictions - Commercial Fishery License Required:

Currently, fishers using commercial, non-personal use size gears to take and /or sell fish must be in possession of a Marine Permit for that gear. Marine permits have an annual reporting requirement, but no requirements for proof that harvest was for commercial purposes.

Recreational fishermen commonly purchase marine permits and use commercial gears because of the low cost. We propose to strengthen the commercial aspects of these gears by requiring proof that harvest was sold as a requirement for license renewal.

The overlap with gears licensed under the NY bait license will be minimized by requiring a Marine Permit to take river herring. Cast nets will be included under the Marine Permit licensing system.

Closed area We propose to continue the current closures as listed in Table 6 and implement a new closure:

ProhibitHarvest by Nets in Tributaries:Closing the tributaries to harvest by nets will likely reduce overall harvest, but the actual size of this reduction is not known. We do not know the size of recreational net harvest from tributaries. We can infer current commercial harvest from tributaries by the number of fish taken in scap nets since most river herring taken in tributaries are taken by this gear and most scap nets are fished in tributaries. Mean annual reported harvest by commercial scap nets in the last five years was about 15,000 river herring or 48% of the total reported commercial harvest. The mean number of commercial fishing trips using scap nets during this time period was 611 trips which were about 59% of all reported trips in the estuary and tributaries. Elimination of commercial net harvest from these waters will eliminate commercial fishing in 175 miles, or approximately 65% of linear spawning streams in the Estuary and above the Troy Dam.

Gear Restrictions All current gear restrictions will remain in place (Table 6). Other changes include:

Gill nets: Currently both anchor and drift gill nets are used in the mid and upper estuary above the Bear Mountain Bridge (> rkm75). Both gears catch herring, but losses can be higher in anchored nets because they are often not tended as frequently as drifted nets. This is especially the case with recreational fishermen who are often not experienced in use of gill nets. We propose to ban use of fixed gill nets in the Hudson River above Bear Mountain Bridge; drift gill 25

REVISED VERSION: September 2011, based on public comment received.

nets are required to be tended by owners as they are fished. We don't know what reduction in harvest would result, but some will occur and the change will certainly reduce waste of fish.

Scap /Lift nets: Currently there are no limits on size of scap nets to be used. Mandatory reports indicate that the largest nets in use are 400 sq ft (20 by 20 ft). The proposed maximum net size is 10 ft by 10 ft.

Fyke and Trap nets: Although currently legal for the take of river herring, no commercial harvest is reported from these gears. We propose that their use not be allowed for harvest of river herring.

CommercialNet Permit and Fees Commercial gears in the main-stem Hudson and tributaries are licensed under a NYSDEC Bureau of Marine Resources Marine Permit. Access to obtain a Marine Permit remains open, with no prior requirements. These commercial gears are often used by recreational fishermen because current permit fees are very low. Most fees were set in 1911 by the then New York Forest, Fish and Game Commission and no fee increases have occurred through the present time.

Commercial gears such as gill nets can take high numbers of herring and are not considered to be recreational gear in New York. For the purposes of harvest in ocean waters (Marine and Coastal District), gill nets are considered commercial gear and their use for recreational purposes is not permitted.

We propose regulations to increase fees to account for inflation, to emphasize that nets are commercial gears, and to discourage casual use by recreational anglers. Current fee structure can be found in New York Code of Rules and Regulation- Part 35 (see hittp:iwxN.vwv.dec.ny.gov/regs/4019.html ). We considered two alternatives.

1. Increased gear and fishing vessel fees.

a. In 1911, fees were $5.00 per each trap, seine or gill net, and $1.00 per scap net.

These fees would translate to $115.00 per gill net or seine and $25.00 per scap net in today's (2011) dollars.

b. Gill nets and seines can also be licensed by the linear foot of net rather than as a type of net. We propose that the current $ 0.05 per foot be increased to $1.00 per foot. Data from the mandatory reports indicates that the most recent (2010) licensed gill net lengths ranged from 10 ft ($10 fee) to 600 ft ($600 fee). Seines have no maximum length restriction in place; current use is 50 ft ($50 fee) to 100 ft ($100 fee).

c. Another way to differentiate between recreational and commercial fishermen is to reinstitute the 1911 fishing vessel registration for the Hudson River, which is still active for other waters of NY. The 1911 fee of $15.00 for the smallest motorized vessel translates to $350.00 per vessel in today's dollars.

2. A single commercial gear permit.

26

REVISED VERSION: September 2011, based on public comment received.

This approach simplifies the above combination of gear fees and is our preferred alternative.

We would create a Hudson River Commercial Fish Gear Permit (HRCFGP): for individuals who want to harvest river herring or Atlantic menhaden; fee of $150. This would be instead of individual gear licenses.

a. Qualifications needed: proof of previous sale to a licensed retail bait shop; if a business (retail bait shop), proof of business incorporation (LLC)

b. If applicant holds a valid New York food fish or crab permit(s); cost of HRCFGP to be offset by valid permit fee(s)

c. To include all restrictions as listed in Table 6.

d. Gears to be used include anchored (fixed) and drifted gill nets, scap/lift nets, seines and cast nets (see Table 6 for size limitations)

Gear restrictions outlined above will still apply to any alternative chosen.

Closed FishingDays A 36-hour escapement period per week, from 6 AM prevailing time on Friday to 6 PM prevailing time on Saturday, is in effect for commercial gill nets from March 15 to June 15. We propose to expand this closure to include all commercial nets.

Reporting Current mandatory reports of daily catch and effort data are submitted annually. We will continue to require these reports, but decrease the time of report submission to monthly.

CharterBoat License In order to distinguish Charter Boat operators from recreational anglers, we propose to use the existing Marine & Coastal District Party & Charter Boat License (CPBL), as it exists for NY's Marine District. CPBL holders will follow all regulation as established for the Marine District with two exceptions: creel and size limit for striped bass will comply with limits set for the Hudson River above the G. Washington Bridge and the creel limit for a charter boat will be 20 river herring per day. Hudson valley charters can take up to three to six individuals per trip.

7 PROPOSED MEASURES OF SUSTAINABILITY 7.1 Targets Juvenile Indices We propose to set a sustainability target for juvenile indices using data from the time period of 27

REVISED VERSION: September 2011, based on public comment received.

1983 through 2010 for both species. We will use a more conservative definition of juvenile recruitment failure than described in section 3.1.1.2 of Amendment 2 to the ASMFC Interstate Fisheries Management Plan for Shad and River herring (ASMFC 2009). Amendment 2's definition is that recruitment failure occurs when three consecutive juvenile index values are lower than 90 % of all the values obtained in the base period. We will use a 75% cut off level.

The 75% level for alewife is 0.35 (instead of 0.19) and 11.14 (instead of 2.86) for blueback herring (Figure 16).

The fishery will close system-wide if recruitment failure, defined as three consecutive years below the recruitment failure limit, occurs in either species and will remain closed until we see three consecutive years of recruitment greater than the target values.

7.2 Sustainability Measures There are several measures of stock condition of Hudson River herring that can be used to monitor relative change among years. However, these measures have limitations (described below) that currently preclude their use as targets. These include mean length in fishery independent samples, catch per unit effort (CPUE) in the reported commercial harvest and age structure. We propose to monitor these measures during the fishery and use them in concert with the sustainability target to evaluate consequences of a continued fishery.

Mean Length Mean total length reflects age structure of the populations and thus some combination of recruitment and level of total mortality. Mean total lengths of both river herring species in the Hudson River system has declined over the last 20 years and the means are now the lowest of the time series. Since this has been a persistent change in the face of stable recruitment, we suggest that the reduction in length has been caused by excessive mortality of adults within the river and during their ocean residency (bycatch). The bycatch fishery is a large unknown and not solely controlled by New York State to effect a change. Current annual reproduction now relies on a few returning year classes making the populations vulnerable to impacts of poor environmental conditions during the spawning and nursery seasons. We propose to monitor mean total lengths during the proposed fishery.

Catch per Unit Effort in Report Commercial We suggest that CPUE values of the reported harvest reflect general trends in abundance.

However, annual values can be influenced by changes in reporting rate and thus we do not feel that CPUE should be used as a target. Rather, we will follow changes within gear types and fisheries for general trends.

Age structure and Total mortality We will monitor age structure, frequency of repeat spawning, and total mortality (Z) if we can 28

REVISED VERSION: September 2011, based on public comment received.

resolve uncertainties about aging methods and estimate methodology discussed in Status Section 4.2.2.

8 REFERENCES ASMFC. (Atlantic States Marine Fisheries Commission). 2009. Amendment 2 to the Interstate fishery management plan for shad and river herring. Washington, D.C. USA.

Caraco, N.F., J.J. Cole, P.A. Raymond, D. L Strayer, M.L. Pace, S.E.G. Findlay and D.T. Fischer. 1997, Zebra mussel invasion in a large turbid river: phytoplankton response to increased grazing.

Ecology 78:588-602.

Cating, J. P. 1954. Determining age of Atlantic shad for their scales. U.S. Fish and Wildlife Service Bulletin 54: 187-199.

Cieri, M., G. Nelson, and M. Armstrong. 2008. Estimate of river herring bycatch in the directed Atlantic herring fishery. Report prepared for the Atlantic States Marine Fisheries Commission.

Dufour, M. R. Adams, K. Hattala, and L. Abuza. 2009. 2008 volunteer river herring monitoring program.

NYS Department of Environmental Conservation, New Paltz, NY.

Gedamke, T, and J. M. Hoenig. 2006. Estimating mortality from mean length data in nonequilibrium Situations, with Application to the Assessment of Goosefish. Transactions of the American Fisheries Society 135:476-487.

Greeley J.R. 1937. Fishes of the area with annotated list IN A biological survey of the lower Hudson watershed. Suppplement to the twenty-sixth annual report, 1936, State of New York Conservation Department. J.B. Lyons Company Albany NY, USA.

Hattala, K., M. Dufour, R. Adams, K. McShane, J. Kinderd, and R. Lowenthal. 2011. Volunteer river herring monitoring program 2010 report. NT State Department of Environmental Conservation, New Paltz, NY.

Hattala, K. and A. Kahnle. 2007. Status of the Hudson River, New York, American shad stock. IN ASMFC Stock assessment Report No. 07-01 (supplement) of the Atlantic States Marine Fisheries Commission. American shad stock assessment report for peer review, Volume II. Washington, D.C.,USA.

Hughes, A, and C, O'Reilly. Monitoring Alewife Runs in the South Shore Estuary Reserve Report on the 2008 Volunteer Survey. July 2008. <http://www.estuary.cog.ny.us/council-priorities/living-resources/alewife_survey/Alewife%202008.pdf>. [accessed September 2008].

Kahnle, A., D. Stang, K. Hattala, and W. Mason. 1988. Haul seine study of American shad and striped bass spawning stocks in the Hudson River Estuary. Summary report for 1982-1986. New York State Dept. of Environmental Conservation, New Paltz, NY, USA.

29

REVISED VERSION: September 2011, based on public comment received.

Kritzer, J, A. Hughes and C, O'Reilly. 2007a. Monitoring Alewife Runs in the South Shore Estuary Reserve Report on the 2006 Volunteer Survey 06. <http://www.estuary.cog.ny.us/council-priorities/living-resources/alewife-survey/2006ý/%20Aewife%20Survey%20Report.pdf>.

[accessed September 2008].

Kritzer, J, A. Hughes and C, O'Reilly. 2007b. Monitoring Alewife Runs in the South Shore Estuary Reserve Report on the 2007 Volunteer Survey. <http://www.estuary.cog.ny.us/council-priorities/living-resources/alewife-survey/2007%20Alewife%2OSurvey%20Final.pdf>.

[accessed September 2008].

Lynch, R. and B. Kim. 2010. Sample size, the margin of error and the coefficient of variation.

http:i/interstat.statiournials.net/YEARI2OI 0/articles/1001 004.pdf Miller, D., and J. Ladd. 2004. Channel morphology in the Hudson River Estuary: past changes and opportunity for restoration. In Currents-newsletter of the Hudson River Environmental Society, Vol. XXXIV, No. 1. Available: http://www2.marist.edu/-en04/CUR34-1.pdf.

Normandeau Assoociates Inc. 2003. Assessment of Hudson River Recreational Fisheries. Final Report prepared for the New York State Dept. of Environmental Conservation, Albany NY, USA.

Normandeau Assoociates Inc. 2007. Assessment of Spring 2005 Hudson River Recreational Fisheries.

Final Report prepared for the New York State Dept. of Environmental Conservation, Albany NY, USA.

Normandeau Associates, Inc. 2008. Spawning stock characteristics of alewife (Alosapseudoharengus) and blueback herring (Alosa aestivalis) in the Hudson River Estuary and tributaries, including the Mohawk River. Final Report prepared for the New York State Dept. of Environmental Conservation, Albany NY, USA.

Nelson, G. A, P. Brady, J. Sheppard and M.P. Armstrong. 2010. An assessment of river herring stocks in Massachusetts. Massachusetts Division of Marine Fisheries Technical Report.

NEFSC (Northeast Fisheries Science Center). 2011. Part I - preliminary analyses for Amendment 14 to the Atlantic Mackerel, Squid, and Butterfish Fishery Management Plan. Prepared for the 10 May 2011 FMAT meeting, Woods Hole, MA.

NYCDEP http://home2.nyc.eov/hltnl/dep/html/news/hwqs.shtmii NYSDEC (New York State Department of Environmental Conservation). 2010. Volunteer river herring monitoring program 2009. NY State Department of Environmental Conservation, New Paltz, NY.

Schmidt, R. and S. Cooper. 1996. A catalog of barriers to upstream movement of migratory fishes in Hudson River tributaries. Final report to the Hudson River Foundation from Hudsonia, Annandale NY, USA..

Smith, C.L. 1985. Inland fishes of New York State. New York Department of Environmental Conservation, Albany, NY, USA.

Strayer, D.L., K.A. Hattala, and A.W. Kahnle. 2004. Effects of an invasive bivalve (Dreissena 30

REVISED VERSION: September 2011, based on public comment received.

polymorpha)on fish in the Hudson River estuary. Can. J. Fish. Aquat. Sci. 61:924-941.

Young, B. 2011. Report on Peconic River alewife run - 2010. Peconic River Fish Restoration Commission, Ridge, NY, youngb53@optimum.net.

31

REVISED VERSION: September 2011, based on public comment received.

UpperHudson (non-fdo4

... . . ... MohawkAver/Erie

. . .. ... . .. . . .. . Cana

. . . . . . . . .. .. 4 _ y 1

---

-_FedffmlDwanT ý Pesten R-il Alany (w 232)

Upperj, Es-vs ay Cuadeton 90 Bddge St*clorpc+/-Ceek

........ .......... ... ............ .. ( nWin.... ig Catskill reek RoeliffAnseanKill ESopUS Creek Kbtgston cn 146)

Mid Evrnwy,3 Rouadozt Cree Black Creek YougJufepsie (lwn 122)

WappingersCaeek Qcssaick Creek ' kill &eek Newburgh Bay (km 95)

Moodna Creek West Point (kmt 83)

- .arMo41tag n Ejd Annswtlle Creek Haverstraw Bay (kmi 55) Crcton Rwer TappanZee (kn 45)

.Lawr++. Esheaty Tappan Zee Bidge G. Washington BAdge Battery (In 0 New York City Verrazano Narrows Figure 1 Hudson River Estuary with major spawning tributaries for river herring. (see Appendix Table A for complete list) 32

REVISED VERSION: September 2011, based on public comment received.

33

REVISED VERSION: September 2011, based on public comment received.

ýCT Hudson River NY 6P Lon g Island Sound Montauk

,ton New York Atlantic Ocean Figure 2 Long Island, Bronx and Westchester Counties, New York, with some river herring (primarily alewife) spawning streams identified (See Appendix Table A for list) 34

REVISED VERSION: September 2011, based on public comment received.

Commercial landings of River herring in NY 250 200 150 100 50

-4~ ~-4 - 4 M4 - 4 M4 -M -4r -4 4 M4 .4r .4 .. 4 4 Figure 3 Commercial landings of river herring from all waters of New York State.

45 40 35 30 r 25 20 On 15 10 5

0 1995 1998 2001 2004 2007 2010 Figure 4 Commercial landings of river herring in the Hudson River and NY Ocean waters.

0.60 0.50 ..........

.................. I..

0.40 ...

-", Fixed bBMB 0.30 ----------

.. - Drift 0.20 -M-- Fixed aBMB CL

• Scap 0.10 0.00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Figure 5 Percent commercial catch by gear of river herring in the Hudson River (a/b BMB=above and below Bear Mountain Bridge).

35

REVISED VERSION: September 2011, based on public comment received.

L.00 Lower River Catch-Per-Unit-Effort 3.00 Mid & Upper River Catch-per-Unit-Effort : weighted mean 1.80 2.50 1.60 1.40 I

I I.OO I 040 0.20 0.00 0.00 2000 2001 2002 2003 2004 2005 2006 2007 2008 2000 2010 2000 2001 2202 2003 2004 2005 2006 2007 200R 20 2010 Figure 6 Catch per Unit Effort (number of fish per hours fished) by area of the river and gear. Lower estuary = below Bear Mountain Bridge

[rkm 75]; Mid & Upper estuary = above the Bear Mountain Bridge.

Alewife-M

- V-Alewlfe-F 300 ---- Blueback herring-M 290 -- Blueback herring-F E

E 280 270

ý,,s

00. ..............

260 00 2SO C

aJ 240 230 220 210 200 ........... . ........... .......--

!............... T .... . .... . ... . .. "

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Figure 7 Mean total length of river herring collected from commercial fishery monitoring trips in the Hudson River Estuary 36

REVISED VERSION: September 2011, based on public comment received.

340.0 Alewife +Male 320.0

'AFermale 2

2 300.0 4-280.0

.4- 260.0 0

I-.

240.0 zzU.u 200.U 1936 1976 1979 198Z 1985 1988 1991 1994 1997 2000 2003 2006 2009 340.00 Blueback herring 320.00 300.00 280.00 4-260.00

• Male S x.

Female

240.00 220.00 200.00 1936 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 Figure 8 Mean total length of river herring in the Hudson River Estuary. Symbols with an "X" indicate adequate sample size (N>34) to characterize the stock.

37

REVISED VERSION: September 2011, based on public comment received.

Blutback herring male Blueback herring female 200 200 4-0-HRmale89 IMamale89 150 ISO -"/*,

"100 i-s-Mdmale89 E 100 z z 50 50 50 I I- 0 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 Age Age 200 Blueback herring male Blueback herring female 150

-*-HKRfemale90 150 --*-HRmale90 --o<Maferna1e90

-~MAmale90 100 ",MDfemale9O 100 MdmaIe9O ED z 50 50 0 0 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 Age Age Figure 7 Hudson (HR) age structure and estimated age structure of Hudson River blueback herring based on length-at-age keys from Massachusetts (MA) and Maryland (MD) blueback herring.

38 I

REVISED VERSION: September 2011, based on public comment received.

Blueback herring 1992 1 60.0 50.

40.0

--M-Ma0

10 0 Z ki 2

20.0 10.0 O,0 2 3 4 S 6 7 8 9 10 Age Blueback herring 2009 Blueback herring 2010 80.0 35.0 70.0 1 30.0 60.0 M-Ma -- ,-M-Ma z

25.0 50.0 -M - M-Md IU.U 40.0 - F-Ma 30.0 -- # F-Md zz= 15.0 20.0 10.0 10.0 -f 5.0 0.0 0.0 2 3 4 5 6 7 8 9 10 2 3 4 5 6 7 8 9 10 Age Age Figure 10 Estimated age structure of Hudson River blueback herring based on length-at-age keys from Massachusetts (MA) and Maryland (MD).

Blueback h erring mean age 6.00 5.00 I ..... M da U be 4.00 3.00 2.00 2.00 i ..........

-TIn- -

M.0 N-Md-rn Ma-f Md-f EMdF 1.00 1.00 -- w U NY-ni NY-rn 0.00 7, NO-1989 1990 1991 1992 2009 2010 Figure 11 Mean age of Hudson River blueback herring based on length-at-age keys from Massachusetts (MA) and Maryland (MD).

39 I

REVISED VERSION: September 2011, based on public comment received.

Alewife Male 2001 Alewife Female 2001 90 MA 80 250 70 #MA 2O0 150 60 z 40 --*-ME z 30 100 20 so 10 0 5 i-s I A ---- 0 4 2 3 4 5 6 7 a 9 2 3 4 5 6 7 a 9 Alewife Male 2003 Alewife Female 2003 2000 - MA 100 90 MA 80 150 -MD 70

-- 1 ME 60 2 40 30 5o 20 10 0 4 -- .------ 0 2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 Age Alewife Male 2009 Alewife Female2009 300 70 MA 60 -,-MA 250 MD 50MD 200 40 E 150 z2 1O0 50 10 0 -

2 3 4 5 6 7 8 9 2 3 4 5 6 7 S 9 Age Ag.

Alewife Male 2010 Alewife Female 2010 150 - MA 60

-=,- MA 50 MD MD 100 40 4,E *~ME E 30 z

z 20 50 10 0

2 3 4 5 6 7 8 9 2 3 4 5 6 7 8 9 Age I Age Figure 12. Estimated age structure of Hudson River alewife based on length-at-age keys from Maine (ME), Massachusetts (MA) and Maryland (MD).

40

REVISED VERSION: September 2011, based on public comment received.

4.80 Alewife -male 4.60 EMA A MD U ME 4.40 4.20 4.00 3.80 3.60 3.40 3.20 3.00 1990 2001 2002 2003 2004 2005 2006 2007 2008 2 009 2010 5.00 Alewife -female 4.80 N MA EMD <ME 4.60 4.40 w 4.20 4.00 3.80 3.60 3.40 3.20 3.00 1990 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Figure 13. Mean age of Hudson River alewife, ages estimated from age-length keys from Maine (ME),

Massachusetts (MA) and Maryland (MD).

41

REVISED VERSION: September 2011, based on public comment received.

8.0 -

Male alewife 7.0 -'4-Lc=250 6.0 ..... -Lc=240

"- -Lc=230 65.0 0

E 4.0 3.0

.0

.* 2.0 0J A I-.U at  %

0.0 - 1 1 1 ý I I I 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Female alewife 4.0

-"* Lc= 250

• 3.5

-W- Lc=240 E

U 3.0 -r-Lc=230

" 2.5 0

E 2.0 1.0 1890.5 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Figure 14. Length-based mortality estimates for Hudson River alewife. Lc =minimum length of fish caught in the sample gear.

42

REVISED VERSION: September 2011, based on public comment received.

16.0 Male Blueback herring 14.0 Lc=250

= *--Lc=240 E

M~* 10.0V 0 6.0 8,

6.0 S4.0 2.0 N-0.09 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Female Blueback herring 14.0 1

-4"*Lc=250 12.0 Lc=240 E -,r- Lc=230 10.0 0

8.0 E

• 1 0 6.0 4.0 2.0 0.0 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Figure 15 Length-based mortality estimates for Hudson River blueback herring. Lc =minimum length of fish caught in the sample gear.

43

REVISED VERSION: September 2011, based on public comment received.

7.00 Alewife 6.00 6- - GeometricMean w CI 5.00 -5 25th percentile 1983-2010

- - 10th percentile 1983-2010 4.00 IMI 3.00 0

2.00 #i  !

1.00 1 11 0.00 1 98 19 9 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 70.00 Blueback herring . A- t p GeometrilcMean w C1 25th percentile 1983-2010 60.00 S10th percentile 1983-2010 50.00 3 40.00 E

30.00 20.00 10.00 0.00 I------------

1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 Figure 16. Annual young-of-the-year indices (with 95% CI) for alewife and blueback herring collected in the Hudson River Estuary.

44

REVISED VERSION: September 2011, based on public comment received.

Table 1. Summary of available fishery-dependent river herring data in Hudson River and Marine District of New York.

Data Type Time eriod/ Details Description Usefulness as index Fishey .Dependent.- Comtercial Harvest Historic data: Provide catch and effort data Gives historic perspective

-1904-1994: NMFS Not separated by area ( river v marine) Provides trend data for state as a whole, but does not

-1994-present: Hudson (see below)- River data reporting rate unknown separate river(s) from ocean until 1994.

NYSDEC; Marine waters- VTR/dealer report since 2002

-1994- present: transfer of historic NMFS data to ACCSP, data available in confidential and non-confidential form Marine monitoring River herring most likely occur as bycatch No port sampling in NY for 'herring' in variety of fisheries Hudson River Began in 1995 through the present Data from 2000 to present good Emigration area CPUE Mandatory reports Enforcement of reports in 2000 Reporting rate unknown -Fixed GN below BMB:

Catch and effort statistics Data separated by gear used: o Good indicator of abundance Licenses are open access with low fees, Fixed gill net below Bear Mountain Bridge (BMB); o increasing trend many recreational fishers purchase and use assive gear below spawning area; consistent manner of Spawning area CPUE ommercial gears to obtain bait ishing; weekly sum of CPUE approximating "area under o Drift GN - variable urve" method o Scap -Flat In spawning area above BMB o Fixed GN- increasing Drift gill (main-stem HR only) - active gear Fixed gill (main-stem HR only) - less effort than below MB

- Scap/lift net (main-stem HR and tributaries)

Hudson R. Fishery Began in 1999 through the present - Number of annual trips are low; co-occurs & conflicts Characterize catch Monitoring Onboard monitoring with FI sampling Catch and effort statistics Catch samples low I"Catch subsample NEED improved sample size to be useful Fishery Dependent - Recreational Harvest (primarily Creel surveys: 2001: provides point estimate of effort for striped bass, Combination of effort for striped bass and point sought as bait for 2001, river-wide, all year ancillary river herring (RH) data estimate of RH harvest; combine with below CAP striped bass; some 2005, spring only - 2005 provides point estimate of RH harvest & effort for data to estimate magnitude of recreational harvest for harvest for personal 2007, state-wide angler survey; effort for striped bass 2005 to the present.

consumption) striped bass Cooperative Angler Data 2006-present Diary program for striped bass anglers; includes data for Good RH use per trip- used above with rec. harvest Program RH catch or purchase, use by trip to estimate total recreational harvest 45

REVISED VERSION: September 2011, basedon public comment received.

Table 2. Summary of available fishery-independent river herring data in Hudson River, New York.

Data type Time peiod/Agency Description Usefulness as index Fishe*y Independent- H dson River Spawning stock 1936: Biological Survey Historic data, low sample size of 25 fish, species, Indication of size change to present sex, length & age 1975-1985: NYSDEC contaminant Sample size low and extremely variable by year Indication of size change to present sampling 1989-1990 NYSDEC Hudson-Mohawk Focused study, large sample size (1,100 fish): Primarily blueback herring River. species, sex, length & age 1999-2001 Normandeau Assoc. Inc. (NAI) Contract to assess gears for spawning stock survey Primary gear used was size selective gill nets; Developed own age key; not clear how compares to precludes use for length analyses; need method of other Atlantic coast states adjustment for ages 2001 to present: NYSDEC spawning stock Focused spawning stock survey; >300 fish Sample design precludes use for catch-per-unit-survey collected most years; species, sex, length & scales effort data (ageing not complete)

Overview of all above Problems Ok to use Spotty adequate sample size in most years (>34 Good sample size for data 1989-99, 2001,-03,-05, per species, sex) to provide trend for length and -08 to present weight Ageing technique varies greatly from 1936, 1980s, Used ME, MA & MD age-length keys to estimate NAI; techniques appear different from other Judson ages; Atlantic coast states Results: a slight non-consistent bias of age

- Mortality estimates from age structure (above) lifference, possibly attributed to ageing technique unusable as index k/or growth differences (MD fish grow faster than AA)

Suggest use trend in mean age Mortality estimates from age structure (above) rusable as index Beverton-Holt length based too dependent on Inputs (length at recruitment and age)

Volunteer River herring surveys 2006 to present; documents presence/absence of - Not yet useful as index; provide a mechanism to river herring in Hudson tributaries and in some Long mprove future sampling for adult runs Island streams Young-of-year Indices 1980 to present: annual yoy sampling July-Oct sampling within nursery area Both species index variable standardized since 1984; Geometric mean number per haul Alewife increasing Catchability may be affected by habitat change Blueback slight decreasing trend 46

REVISED VERSION: September 2011, based on public comment received.

I ý Selected conservative target of 25th percentile 47

REVISED VERSION: September 2011, based on public comment received.

Table 3. Commercial river herring fishery monitoring data for the Hudson River Estuary.

On-board Observations on Comnrwrcial Trips Alewife Blueback herring Unidentified "river herring" Nof Number Sexratio Number Sex ratio Number Sexratio Percent Year trips M F U M F M F U M F M F U M F Total Alewife Blueback 7 . f 1996 1 7 43 43 0%/0 100%/0 Fr F 7 7 1997 5 5 25 178 0.17 0.83 208 100% 0°/0 F 7r 1998 1 114 114 100%/0 00/0 7

• 1999 4 73 348 421 17% 00/0 7 7 7 7 2000 6 19 18 0.51 0.49 3 32 480 0.09 0.91 552 7%/o 93%

p. 7 7 7 2001 7 192 178 851 0.52 0.48 1221 1000/0 0°/0 7 7 2002 8 43 19 41 1225 0.32 0.68 1328 3% 97%

2003 2 171 171 100% 00/0 7 7 2004 11 124 168 8 0.42 0.58 5 6 0.45 0.55 500 796 297 0.39 0.61 1904 16% 1%

2005 1 428 28 456 94% 0(/0

• 7 2006 3 1 246 247 00/0 100%/0 2007 6 14 53 268 335 4% 16%

2008 1 44 0.50 0.50 44 0%/o 0°/%

2009 3 187 179 4 0.51 0.49 37 61 0.38 0.62 468 79% 21%

8 2 5 2010 1 80 42 2 0.66 4 0.34 33 70 6 0.32 0.68 233 53% 47%

48

REVISED VERSION: September 2011, basedon public comment received.

Table 4. Estimated recreational use and take of river herring by Hudson River anglers.

Herring Use*

% of all CAP Trips N-SB N Total Trips using Estimated using herring as Trips N bought caught / RH Estimated herring as Herring Year bait using RH / trip trip use/trip SB trips** bait** Use 2001 53,988 39,500 93,157**

2005 89% 2.36 72,568 64,500 152,117**

Cooperative Angler Program Data 2006 93% 263 1.47 2.57 4.04 2007 70% 331 1.66 1.80 3.46 90,742 69,700 241,318***

2008 71% 445 0.86 1.64 2.50 2009 77% 492 0.63 3.80 4.43 2010 74% 527 0.67 4.80 5.48

• Data from NYSDEC - HRFU Cooperative Angler Program (unpublished data)

• • Creel survey data: NAI 2003, NAI 2007; 2001 estimated use modified using 2005 RH use per trip* 2001 trips using herring as bait

• • • Estimate calculated from overall average RH/trip (CAP) and Estimated SB trips from NYSDEC statewide angler survey 49

REVISED VERSION: September 2011, based on public comment received.

Table 5. Current and proposed recreational fishery regulations for a river herring fishery in the Hudson River.

Regulation Current 2010 Recreational Proposed change- new Season All year March 15 to June 15 Creel/ catch limits None 10 per day per angler or a maximum boat (any size, any number) limit of 50 per day for a group of boat anglers(whichever is lower)

Charterboats: (see commercialfishing table)

Closed areas None below Troy Dam - the River HerringconservationArea: No

- Closure from Guard gate 2 to fishing within 825fi (250m) ofa man-made Lock 2 on the Mohawk River or naturalbarrier

- Closure from Guard gate 2 to Lock 2 on Mohawk River Gear restrictions -Angling All tributaries,including the Mohawk River

-Scap/lift net: 36 sq ft or above Troy: Angling only, no nets smaller Main river below Troy Dam: Angling or the Dip net: 14" round or 13"x 13" use of nets to obtain bait for personal use square only as follows:

Seine: 36 sq ft or smaller Scap/lift net 16 sqfi or less

-Cast net; 1Oft diameter Dip net: 14" round or 13"x 13" square Seine 36 sq ft or smaller Cast net 10 ft diameter Escapement (no fishing days) None None License Marine Registry Marine Registry Reporting None New York angler diary on ACCSP website 50

REVISED VERSION: September 2011, based on public comment received.

Table 6. Current and proposed commercial fishery regulations for a river herring fishery in the Hudson River.

Regulation Current 2010 Commercial Proposed change - new Season Mar 15 - Jun 15 Mar 15 - Jun 15 Creel/ catch limits None Charterboats: IOfish per day per paying customer or a maximum boat limit of 50 fish per day, (whichever is lower)

• Closed areas No gill nets above 190-Castleton Bridge No gill nets above 190 - Castleton Bridge No nets on Kingston Flats -No nets on Kingston Flats

- No nets in tributaries Gear restrictions Allowed gears Allowed gears for river herring

- Gill net - Gill net 0 600 ft or less 0 600 ft or less 0 3.5 in stretch mesh or smaller o 3.5 in stretch mesh or smaller D No fishing at night in HR 0 No fishing at night in HR above above Bear Mt Bridge Bear Mt Bridge Seine >36 sq ft 0 No fixed gill nets above the Bear No seine >100 ft allowed above 190 Mt Bridge bridge - Seine; no seine >100 ft allowed above Scap/lift net no size 190-Castleton Bridge Fyke or trap net Scap/lift net lOft by lOft maximum Cast net not exceeding ten ft diameter Cast net not exceeding ten ft diameter Escapement (no 36 hr lift (applies only to gill nets 36 hr lift fishing days) allowed in the main river) Applicable to all net gears Marine Permit Marine Permit Marinepermit only license to take

- Fees implemented in 1911 anadromousriver herring,the only net Gill net $0.05/foot gears allowed include drift and fixed gill Scap net <10 sq ft $1.00 net, scap/lift net,seine and cast net Scap net> 1Osq ft $2.00 Fees updated to include any of the Seine $0.05/foot following:

Trap nets $3 to $10 Ia. Gill or seine net - $115; scap net $25 Fyke net $1 to $2 1b.Gill or seine $1 per foot Bait license I c.Fishing vessel $350

- Cast net $10

2. Create Hudson River commercial fish permit; includes use of gillnets, scap/lift nets, seines and cast nets with all other restrictions as listed in this table; qualifications needed (see Sec 6.1.2, page 26)

Charter* Boat None for Hudson above the Tappan Require existing Maine &Coastal District License Zee Bridge Party boat/ Charter license for tidal Hudson and its tributaries- $250.00 Reporting Mandatory daily catch& effort; one Mandatory daily catch& effort; reports I annual report due monthly 51

REVISED VERSION: September 2011, based on public comment received.

Appendix A. River herring streams of New York including tributaries of the Hudson River Estuary, and the Mohawk River; streams in the Bronx and Westchester Counties and on Long Island. (This list may not be complete).

Hudson River River Mile County Prirary Tributary Secondary Tribl Secondary Trib2 M to barrierFt to barrier 18 Westchester Saw Mill 100 328 24 Rockland Sparkill Creek 1,620 5,315 25 Westchester Wicker's Creek 240 787 28 Westchester Pocantico River 950 3,117 33 Westchester Sing-Sing 450 1,476 34 Westchester Croton River 2,860 9,384 38 Westchester Furnace Brook 820 2,690 38 Rockland Minisceongo 2,100 6,890 39 Rockland Cedar Pond Brook 4,500 14,765 43 Westchester Dickey Brook 2,610 8,563 44 Westchester Annsville Creek Peekskill Hollow Sprout Brook 1,140 3,740 44 Westchester Annsville Creek Peekskill Hollow 2,310 7,579 44 Westchester Annsville Creek 3,000 9,843 46 Orange Popolopen Creek 840 2,756 52 Putnam Phillipse Brook 1,160 3,806 52 Putnam Indian Brook 1,240 4,068 53 Putnam Foundry Brook 880 2,887 55 Putnam Breakneck Brook 160 525 57 Orange Moodna Creek 4,740 15,552 58 Dutchess Malzingah Brook (Gordon's Brook) 100 328 59 Dutchess Fishkill Creek 980 3,215 67 Dutchess Hunters Brook 180 591 67 Dutchess Wappingers Creek Hunters Brook 3,380 11,090 69 Ulster Lattintown Creek S. Lattintown 550 1,805 69 Ulster South Lattintown 1,100 3,609 75 Dutchess Falkill 100 328 76 Ulster Twaalfskill Highland Brook 400 1,312 78 Dutchess Maritje Kill 190 623 81 Dutchess Crum Elbow 270 886 84 Dutchess Indian Kill 1,200 3,937 84 Ulster Black Creek 1,670 5,479 87 Dutchess Fallsburg Creek 2,000 6,562 87 Dutchess Lands man Kill 2,100 6,890 91 Ulster Roundout 3,820 12,533 98 Columbia South Bay Creek 890 2,920 98 Dutchess Saw Kill 970 3,183 100 Dutchess Stony Creek 2,290 7,513 101 Ulster Esopus Creek 1,850 6,070 105 Columbia Cheviot Creek 380 1,247 110 Columbia RoeliffJansen Kill 9,320 30,579 112 Greene Catskill Creek Kaaterskill Creek 4,940 16,208 118 Greene Murderers Creek 930 3,051 121 Columbia Stockport Creek Claverack Creek 1,250 4,101 121 Columbia Stockport Creek Claverack Creek Kinderhook Cree 1,780 5,840 126 Greene Co~sackie Sickles Creek (dry) 1,270 4,167 128 Columbia Mill Creek 1,870 6,135 131 Albany Hannacroix 1,650 5,414 132 Albany Coeymans 300 984 135 Renssalaer Schodack Muitzes Kill 10,900 35,763 136 Renssalaer Vlockie Kill 1,880 6,168 137 Albany Vloman Kill 1,130 3,708 137 Renssalaer Papscanee Moordener Kill 1,550 5,086 142 Albany Nomans Kill 2,970 9,745 144 Renssalaer Mill Creek 210 689 149.5 Renssalaer Wynants Kill 430 1,411 150 Renssalaer Poesten Kill 310 1,017 Above Troy Dam Mohawk River 183,000 600,423 52

REVISED VERSION: September 2011, based on public comment received.

Appendix Table A continued.

County Stream Bronx Bronx River Hutchinson River Westchester Beaver Swamp Brook Blind Brook Byram River Mamaroneck River New Rochelle Creek Otter Creek Long Island Shore Stream&.or Pond with outlet Tributary Alewife Present?

South Beaverdam Creek Unknown South Browns River Unknown South Carlls River Confirmed South Carmans River Confirmed South Connetquot River Westbrook, Rattlesnake Creek Unknown South Massapequa Creek Confirmed South Mud Creek Unknown South Patchogue River Unknown South Penataquit Creek Unknown South Swan River Unknown South Champlin Creek Unknown South Forge River Unknown South Pipes Creek Unknown North Beaver Brook Unknown North Cold Spring Brook Unknown North Fresh Pond/Baiting Hollow Confirmed North Mill River, Oyster Bay Unknown North Nissequogue River Confirmed North Setauket Mill pond Unknown North Stony Hollow Run, Ctrpt. Unknown North Sunken Meadow Creek Confirmed North Wading River Unknown East Fnd Alewife Brook Confirmed East End Alewife Creek/Big Fresh Pond Confirmed East End Big Reed Pond Confirmed East End Fly Pond Restoration stocking effort East End Gardiner Bay Creeks Unknown East End Georgica Pond Unknown East End Halsey's-Neck Pond Unknown East End Hog Creek Unknown East End Hook Pond Unknown East End Ligonee Brook Confirmed East End Mill Pond - MecoxBay Ext. Unknown East End Peconic River Confirmed East End Sagaponack Pond - Jeremy's Hole Unknown East End Scoy Pond Restoration stocking effort East End Silver Lake/Moore's Drain Unknown East End Stepping Stones Pond Unknown 53

REVISED VERSION: September 2011, based on public comment received.

Appendix Table B. Summary of current (2010) fishery regulations for alewife and blueback herring in New York State.

Fishery / Area CommercialHarvest:

Inland waters Hudson River Estuary: G. Washington Bridge north to Troy Dam (River kilometer 19-245)

- Season: 15 March through 15 June

- 36 hour4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> Escapement period (Friday 6 am to Saturday 6pm, prevailing time)

- Net size restriction: limit of 600 ft, mesh size restriction: mesh <3.5 inch stretch mesh

- Net deployment restrictions (distance between fishing gear > 1500 ft)

- Area restrictions (drifted gears allowed in certain portions of the river)

Long Island: No restrictions, except for some towns which have restricted fishing within their township Marine Waters: Hudson River - G. Washington Bridge south; and waters including NY Harbor and around Long Island

- No limits or season.

Delaware River: NY portion, north of Port Jervis

- No commercial fishery exists in this portion; no rules prohibit it Baitfish harvest: Take of bait fish (including alewife and blueback herring) are allowed with Bait License in the Inland water of New York State. Allowed gears are seines (all Inland waters) and cast nets in the Hudson River only.

RecreationalHarvest:

- No daily limit

- No season

- Harvest can be by hook and line, and some net gears: dip nets (14inches round), scoop nets (13 x 13 inches square), cast net (maximum of 10 feet in diameter) and seine and scap / lift nets 36 square feet or less. Anglers must be registered with the New York Recreational Marine Registry.

54

REVISED VERSION: September 2011, basedon public comment received.

Appendix C. Current regulations for river herring fisheries in the Hudson River watershed, and public suggestions for change summarized from meetings held in April, 2010. Published in the NYSDEC website: http://www.dec.nvy.gzov/animals/57672.littn Reguation Crrn 201.0 Comnmercial Public suggestions for change Season Mar 15 -Jun 15 Creel/ catch limits None - Possession limit of 24 fish for charter boats*

- Have a 100 fish daily limit

- Have some kind of quota Closed areas - No gill nets above 190 Bridge - Add: Close tributaries to nets

- No nets on Kingston Flats Gear restrictions - Gill net - Gill net o 600 ft or less o Shorten length to 100 or o 3.5 in stretch mesh or 200 ft smaller o Add mesh size restriction o No fishing at night in HR o Limit net size above Bear Mt Bridge - Allow no nets

- Seine >36 sq ft

- No seine >100 ft above 190 bridge Escapement (no fishing 36 hr lift (no gill nets allowed in - 36 to 72 hr closure days) the main river) - Stay away from the weekend

- does not apply to scap nets in higher demand for bait) tributaries License Marine Permit - *require a charter boat license Varies by gear $1 to $30 - Raise the price of a permit

- Increase fee to $75 to $200 Include cast nets as commercial Marine Permit (currently need a bait license)

Make a lottery for obtaining marine permit Reporting Mandatory daily catch& effort 55 I

REVISED VERSION: September 2011, based on public comment received.

Season All year - Be more restrictive

- Choose a season to protect alewife

- Choose closure (season) based on water temperature Creel/ catch limits None - 5 to 10 a day (any size, any number) - Allow a special limit for Charter boats: 24/day

- Need to know difference between creel and possession limit?

- Make a slot size &/orsize limit Closed areas None - Close all the tributaries to fishing

- Close the Mohawk to herring fishing

- Have rotating tributary closures (changes every 3 years)

- Close parts of tributaries Gear restrictions Angling - No nets, angling only 36 sq ft scap or smaller - No nets in tributaries 14" round or 13"x13" dip net - No nets or smaller gear 36 sq ft seine Maximum 10 ft diam. Cast net*

Escapement (no fishing days) None Close fishing 3 or 4 days a week Allow herring harvest either on odd or even days of the week Close the run during peak of spawning Time closures (hours during the day or night)

Opposed to day closures Make no-fishing days enough to protect spawning Have sliding closures during the week, i.e. "lure" days License Marine License $10 Reporting None - Have a call-in number for harvest like a HIP #) to get better information

- Create a website for anglers to input what they catch Other issues (other than a fishery) that are creating problems for river herring

- Chlorine discharge problems

- Ocean harvest is the problem- not the river fishery

- Increased silt (covers eggs)

Long Island streams: The lack of data means that no fishery will be allowed under the "sustainable" definition in the ASMFC Amendment 2. Information on habitat and passage issues will be gathered.

56

In a springtime ritual, adults and children went to their local streams and caught great quantities of the small fish. Prized as one Ice-fishing shacks (above) are evidence of of the best-tasting fried fish, smelt were brought home for dinner, New England's long tradition of fishing for rainbow smelt. Scientists (below) from three sold locally, and shipped to distant markets. Many animals-seals, states are studying causes of the smelt's striped bass, codfish, great blue herons, and others-feasted on recent decline, including loss of suitable stream habitats (bottom) for spawning.

rainbow smelt during the springtime bonanza. Although small in size, this fish played a big role in the ecosystem and economy.

Now rainbow smelt are declining, even in streams that once hosted abundant runs each spring. The diminishing numbers have become evident in the Gulf of Maine. Recognizing the plight of the rainbow smelt, the U.S. government listed it in 2004 as a federal Species of Concern.

T-he state governments of Maine, Massachusetts, and New Hampshire are working together to understand the rainbow smelt's status and threats, and to plan a regional conservation effort for the species. Scientific research by the three-state collab-orative focuses on the status of the smelt population and the condition of spawning areas in streams, which may be a key factor in the rainbow smelt's decline.

0I a

Rainbow Smelt -- 0

• 1, at a Glance

- Native to coastal waters of northeastern United States and Canadian Maritimes.

O Eats shrimp, marine worms, amphipods, euphausiids, mysids, 4I~

and smaller fish.

Eaten by porpoises, seals, A New England Tradition salmon, trout, bluefish, striped bass, Atlantic cod, and birds. Historically, people in New England valued rainbow smelt as an easy-to-catch, abundant source of fresh protein after the long winter.

Slender fish averaging 6 to 8 The commercial fishery for rainbow smelt is one of the oldest in 0o inches long.

New England, and for many years it was among the most valuable.

Can live up to 6 years, but more More recently, the catch along the Gulf of Maine coast has dwindled, typically lives 3 or 4 years. although parts of eastern Maine still have strong commercial fisher-Lives in estuaries, harbors, and ies. Recreational fishing for rainbow smelt continues to be a popular offshore waters during summer, pastime in Massachusetts, New Hampshire, and Maine.

fall, and winter.

Migrates into rivers and streams Fish in Peril to spawn beginning in late Rainbow smelt were so plentiful a hundred years ago that farmers winter (Massachusetts) to late spring (eastern Maine).

• caught them by the barrelful and had enough to eat, use as bait, and even spread on their fields as fertilizer. In many places now, it would be difficult to fill a single barrel with rainbow smelt. The species has largely disappeared from the southern part of its geographic range, and its numbers along the coast of the Gulf of Maine have dropped dramatically. In general, rainbow smelt are least abundant in Massa-

Red dots indicate streams chusetts and increase slightly toward eastern Maine. Reliable data on where rainbow smelt are population size are not available, but Maine fishery data show that I

,~known to spawn. E rainbow smelt landings have dropped tremendously since the 1800s.

4 While a decrease in fishing effort may contribute to the drop in land-ings, the overall trend is clear: rainbow smelt are in trouble.

0 Ix 4 D 10 OD

-1~

At present, rainbow smelt live only north of Long Island Sound (green area).

• ~; r*2*U..~~ 3/4~tMQ St r< 4 94*2 r

t" *2"~

t t 2<

t,~

-~

C Q*7t~

4 47* -i

~j' i~&r r,~ ~'t,.

7*

t1/4~ I I

.3 A 7

'(tLS

.~,pr

~;7

-~

>2

<'1 7*2 N

cc K~

itt U

- V

• 29~2.

7 .. v.**

A *i*¸ iF" 1~ i. wr4 3/4-s I

"A ii

-1

• 6" 1'

TI'

What Can You Do?

Individual citizens and towns can take important steps to help the rainbow smelt recover.

Local efforts are essential and can make a big difference in the survival of the species.

1. Use less fertilizer on your property.

Water carries fertilizer into streams, where the nutrients promote growth of algae on smelt eggs.

2. Fix dams and culverts blocking smelt from spawning areas.

Many dams and culverts prevent rainbow smelt and other fish from swimming upstream and downstream. In collaboration with owners and government agencies, dams can be removed, culverts reconfigured, and culverts replaced with bridges.

3. Plant shrubs and trees along stream banks and refrain from clearing existing vegetation.

Vegetated buffers help to filter out pollutants, sediment, and excess nutrients before they enter the waterway. Shrubs and trees also shade streams, keeping the water cool for fish.

4. Maintain natural stream channels and substrate; restore those altered with concrete walls or other structures.

Faster-flowing water in altered streams can lead to scouring or crowding of smelt eggs. Low water velocity and unnatural substrates can reduce egg attachment and incubation success.

5. Use less road salt and sand near streams.

When salt and sand are washed into streams, they can kill smelt eggs.

- 6. Clean storm drains annually.

Debris and infrequent maintenance can clog storm drains, forcing water to flow over ground. The water carries sediment into streams, which smothers smelt eggs.

7. Get to know your smelt runs.

Find out where smelt spawn in your town and insist that local officials protect these valuable habitats.

I

A Regional Conservation Plan for Anadromous Rainbow Smelt in the U.S. Gulf of Maine 0,---

By Claire L. Enterline Maine Department of Marine Resources 172 State House Station, Augusta, ME 04333 Bradford C. Chase Massachusetts Division of Marine Fisheries 1312 Purchase Street, 3rd Floor, New Bedford, MA 02740 Jessica M. Carloni New Hampshire Fish and Game Department 225 Main Street, Durham, NH 03824 Katherine E. Mills University of Maine, Gulf of Maine Research Institute 350 Commercial Street, Portland, ME 04101 STHE UNIVERtSITY OF IM"FF UMAINE

ACKNOWLEDGEMENTS This project would not have been possible without the leadership of John W Sowles and Seth L. Barker (Maine Department of Marine Resources)

And the many tireless hours of field, laboratory, and statistical work, and administration by:

Massachusetts Division of Marine Fisheries:

Matthew H. Ayer, Scott P Elzey, Christopher Wood, Katie L'Heureux, Kim Trull, Carolyn Woodhead, John Boardman, Mike Bednarski, Steve Correia, and Stephanie Cunningham New Hampshire Fish and Game Department:

Douglas Grout, Cheri Patterson, Simon Beirne, Joshua Borgeson, Eric Bruestle, Joshua T. Carloni, Jessica Devoid, Michael Dionne, Robert Eckert, Rebecca Heuss, Elizabeth Morrissey, Conor O'Donnell, Lon Robinson, Kevin Sullivan, Christopher Warner, Kristi Wellenberger, and Rene6 Zobel Maine Department of Marine Resources:

Linda Mercer, Ernie Atkinson, Tim Bennett, Denise Blanchette, Colby Bruchs, Amy Hamilton Vailea, Joseph Gattozzi, Jon Lewis, Marcy Lucas, Celeste Mosher, Anne Simpson, Peter Thayer, Chris Uraneck, Thomas Watson, and the Maine Marine Patrol University of Maine Sea Grant Extension:

Christopher Bartlett Downeast Salmon Federation:

Dwayne Shaw Submitted as part of:

A Multi-State Collaborative to Develop & Implement a Conservation Program for Three Anadromous Finfish Species of Concern in the Gulf of Maine NOAA Species of Concern Grant Program Award #NA06NMF4720249A Cover illustration by Victor Young

©2012

CONTENTS Introduction ............................................................................................................................................................ 3 1 - Species Status ..................................................................................................................................................... 6 1.1 - Basic Biology ................................................................................................................................................... 6 Life History ................................................................................................................................................ 6 Habitat Use ................................................................................................................................................ 6 Genetic Stock Structure in Gulf of M aine ............................................................................................... 8 1.2 - Historical Smelt Fisheries ............................................................................................................................... 12 M id-Atlantic ............................................................................................................................................. 12 New Jersey ................................................................................................................................................ 12 New York .................................................................................................................................................. 13 Connecticut .............................................................................................................................................. 14 Rhode Island ............................................................................................................................................. 14 M assachusetts ........................................................................................................................................... 15 H istorical Fisheries............................................................................................................................... 15 Recen t Trends ....................................................................................................................................... 15 New Hampshire ........................................................................................................................................ 16 HistoricalFisheries............................................................................................................................... 16 Recen t Trends ....................................................................................................................................... 16 M ai n e ....................................................................................................................................................... 17 HistoricalFisheries............................................................................................................................... 17 Recen t Trends ....................................................................................................................................... 17 Canadian Provinces ................................................................................................................................... 18 HistoricalFisheries............................................................................................................................... 18 Recen t Trends ....................................................................................................................................... 18 Summary .................................................................................................................................................. 19 1.3 - Population Status in the Gulf of M aine .......................................................................................................... 20 Previous Smelt Population Studies ............................................................................................................ 20 Current Fisheries Dependent M onitoring ............................................................................................ 21 New Hampshire Creel Survey ................................................................................................................ 21 M aine Creel Survey .............................................................................................................................. 22 Current Fisheries Independent Monitoring .......................................................................................... 22 State Inshore Trawl Surveys ................................................................................................................... 22 M aine and New HampshireJuvenile Abundance Surveys ................................................................... 23 New HampshireEgg DepositionM onitoring.................................................................................... 23 M aine Spawning Stream Use M onitoring........................................................................................ 24 RegionalFyke Net Sampling................................................................................................................. 24 Establishing Gulf of M aine Spawning Site Indices .................................................................... 24 2008-2011 Results ......................................................................................................................... 25 Study Area Summary ..................................................................................................................... 29 Conclusions About Regional Fyke Net Sampling ....................................................................... 30 2 - Threats to Rainbow Smelt Populations in the Gulf of M aine ...................................................................... 42 2.1 - Threats to Spawning Habitat Conditions and Spawning Success .............................................................. 42 Spawning Site Characteristics .................................................................................................................... 42 Obstructions ............................................................................................................................................. 43 Dams .................................................................................................................................................. 43 Roa d crossings ...................................................................................................................................... 44

Contents continued Channelization and Flow Disruptions ................................................................................................. 44 Dischargeand Velocity .......................................................................................................................... 44 Substrate and ChannelStability ............................................................................................................ 45 Watershed characteristics .......................................................................................................................... 45 2.2 - Threats to Embryonic Development and Survival ...................................................................................... 48 Water Chemistry ....................................................................................................................................... 49 Water Temperature............................................................................................................................... 49 Specific Conductivity............................................................................................................................ 49 Dissolved Oxygen ................................................................................................................................. 50 p H ...................................................................................................................................................... 50 Tu rbid ity ............................................................................................................................................. 50 Data An alysis ....................................................................................................................................... 51 Nutrient Concentrations ........................................................................................................................... 51 Total N itrogen ..................................................................................................................................... 51 Total Phosphorus.................................................................................................................................. 52 TN /TP R atio....................................................................................................................................... 52 DataAnalysis ....................................................................................................................................... 52 Periphyton ................................................................................................................................................ 52 Heavy M etal Concentrations .................................................................................................................... 53 Watershed characteristics .......................................................................................................................... 54 C on clu sions .............................................................................................................................................. 44 2.3 - Threats to Smelt in M arine Coastal Waters ................................................................................................. 59 Fish Health ............................................................................................................................................... 59 Fishing Mortality ...................................................................................................................................... 60 Overflshing in historicalfisheries ............................................................................................. ... 60 Incidentalcatch in small mesh fisheries............................................................................................ 61 Predator-prey relationships ........................................................................................................................ 62 P rey A vailability................................................................................................................................... 62 PredatorPopulationShifts .................................................................................................................... 62 Community shifts ................................................................................................................................. 63 Climate-driven environmental change ....................................................................................................... 63 3 - Conservation Strategies ................................................................................................................................... 65 3.1 - Regional Conservation Strategies ................................................................................................................... 65 3.2 - State M anagement Recommendations ........................................................................................................... 69 M assachusetts ........................................................................................................................................... 69 Smelt Stocking Efforts ........................................................................................................................... 70 HabitatRestoration.............................................................................................................................. 70 Recommendations................................................................................................................................. 70 New Hampshire ........................................................................................................................................ 71 Population monitoring......................................................................................................................... 71 HabitatRestoration.............................................................................................................................. 72 Recommendations................................................................................................................................. 73 Main e ....................................................................................................................................................... 73 Continue monitoringsmelt populationsat multiple life stages............................................................. 73 Improving connectivity andaccess to spawninggrounds..................................................................... 74 Assessing causesfor local decline............................................................................................................. 75 M arked larvalstocking at monitoredsites ......................................................................................... 75 Recommendations................................................................................................................................. 76 Literature Cited ........................................................................................................................................ 77 A p p en d ix .................................................................................................................................................. 85 2

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

INTRODUCTION The rainbow smelt (Osmerus mordax) is a small anadromous fish that over-winters in estuaries and bays prior to spawning each spring in coastal streams and rivers. Smelt have supported culturally important commercial and recre-ational fisheries throughout New England since at least the 1800s. However, in recent years, concerns have risen about the population status of rainbow smelt. The species has disappeared from the southern end of its geographic range, which once extended to the Chesapeake Bay and now may extend only as far south as Buzzards Bay, Massachusetts. High numbers of rainbow High numbers of smelt that once supported commercial fisheries in New England have declined rainbow smelt that precipitously since the late 1800s to mid-1900s. While recreational fisheries once supported for rainbow smelt continue, declining catches have also been noted by anglers, commercial fisheries particularly since the 1980s.

in New England have Based on these observations of range contraction and abundance declines, declined precipitously the National Oceanic and Atmospheric Administration (NOAA) listed rainbow smelt as a federal Species of Concern in 2004; New Hampshire also lists sea- since the late 1800s run rainbow smelt as a Species of Special Concern. Although rainbow smelt to mid-1900s.

population declines have been widely documented, the causes are not well understood. In listing the species, factors identified as potential contributors included structural impediments to their spawning migration (such as dams and blocked culverts) and chronic degradation of spawning habitat due to stormwater inputs that include toxic contaminants, nutrients, and sediment.

Following the designation of rainbow smelt as a species of concern, the Maine Department of Marine Resources received a 6-year grant from NOAAs Office of Protected Resources to work in collaboration with the Massachusetts Division of Marine Fisheries and New Hampshire Fish and Game Department to document the status of and develop conservation strategies for rainbow smelt (NA06NMF4720249). This conservation plan represents a summary of key elements of the project, which focused on several objectives:

1) Documenting range contraction and range-wide population declines based on historical data and accounts

2) Evaluating the status of rainbow smelt populations in the Gulf of Maine region

3) Developing a population index to track the strength of spawning runs

4) Assessing a range of potential threats to rainbow smelt populations

5) Proposing management actions to help conserve rainbow smelt throughout the Gulf of Maine region.

This study has significantly advanced our understanding of the biology, status, and threats to rainbow smelt in the Gulf of Maine. A major contribu-tion was the development of standardized procedures for indexing the abun-dance of spawning rainbow smelt. Four years of fyke net sampling of spawning ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 3

runs throughout the Gulf of Maine region have provided important baseline information about the status of the species. Observations of truncated age structures within the spawning run, high male to female ratios in some rivers, and lower survival rates and a higher portion of age-1 spawners than historically observed all indicate that Gulf of Maine rainbow smelt populations are cur-rently stressed.

Further evidence of the decline can be derived from a survey of historically active spawning sites throughout the state of Maine, using a study from the 1970s (Flagg 1974) as a valuable baseline for comparison. The recent survey found that 13% of the historically active spawning streams no longer support rainbow smelt spawning, and most of the streams that remain active now sup-port smaller runs than they did historically. The substantial decline in strong spawning runs merits concern and attention.

Many threats to rainbow smelt spawning habitat were identified as part of this study. Obstructions such as dams and improperly designed culverts may physically impede smelt migration to appropriate spawning sites. Further, extremely high or low flows can impede swimming ability or impair the cues smelt rely on to undertake this migration. Once on the spawning grounds, water quality conditions may affect the hatching and survival of smelt eggs. In many rivers studied as part of this project, pH, turbidity, nutrient levels, and dissolved contaminants warranted concern for water quality. Field observations also showed an association between nitrogen levels and periphyton growth at spawning grounds, and laboratory experiments demonstrated that high periph-yton growth significantly impaired the survival of smelt embryos.

Many of these threats-particularly flow patterns and water quality-are not driven by factors within the spawning rivers themselves, but rather by ac-tivities in the surrounding watersheds. Across a suite of water quality and heavy metal parameters, we found that high levels of development in the watershed were associated with poorer conditions for rainbow smelt, while high propor-tions of forest in the watershed supported high quality stream conditions.

In conjunction, watershed development was negatively associated with the strength of smelt spawning runs, while forested watersheds supported stronger runs in their receiving streams.

Our goal in assessing threats to rainbow smelt was to identify conditions that appear to negatively and positively affect smelt throughout their life cycle so that management actions can effectively target these factors. Based on our assessment of critical threats, management recommendations to protect and restore rainbow smelt populations include:

" Maintain the federal Species of Concern designation for rainbow smelt

" Continue monitoring population trends and biological characteristics in the extant range, and expand efforts towards estimating rainbow smelt population size

" Restore historical or degraded spawning habitat

" Maintain and, where necessary, improve fishery monitoring to ensure that fishing effort is compatible with sustainability of local and regional rainbow smelt populations

• Expand research initiatives to anticipate direct and indirect effects of 4

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

dimate change and variability on rainbow smelt

" Invest in research to further study environmental requirements, stress-ors, and drivers in order to effectively manage recovery

" Stock marked larvae to re-establish rainbow smelt runs at restored sites, as needed and as appropriate given considerations of genetic diversity and donor population viability This Conservation Plan provides: a description of the life history of anadromous rainbow smelt; an account of the historical fishing pressure on the species; a summary of the current population status and monitoring efforts; explanation of the threats to the species at different life stages, including the marine phase; and conservation and management strategies for the region and for each state in the Gulf of Maine. Our intent is that this information will provide important baseline information regarding the status of smelt popula-tions at the present time and that it will offer coastal and fishery managers guidance on appropriate actions and priorities to protect and restore rainbow smelt moving forward.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN e 5

1 - SPECIES STATUS Rainbow smelt (Osmerus mordax) are small anadromous fish that live in nearshore coastal waters and spawn in the spring in coastal rivers immediately above the head of tide in freshwater (Buckley 1989, Kendall 1926, Murawski et al. 1980). Landlocked populations of smelt also naturally occur in lakes in the Northeast U. S. and Canada and have been introduced to many freshwater sys-tems, including the Great Lakes. Anadromous smelt serve as an important prey species for commercially and culturally valuable species, such as Atlantic cod, Anadromous smelt serve Atlantic salmon, trout, Atlantic gray seals, striped bass (Clayton et al. 1978, as an importantprey O'Gorman et al. 1987, Kircheis and Stanley 1981, Kirn 1986, Stewart et al.

1981). Historically, the range of rainbow smelt extended from Chesapeake Bay species for commercially to Labrador (Buckley 1989, Kendall 1926), but over the last century, the range and culturally valuable has contracted and smelt are now only found east of Long Island Sound.

species, such as Atlantic cod, Atlantic salmon, 1.1 - BASIC BIOLOGY trout,Atlantic gray seals, striped bass.

Life History Smelt are small-bodied and short-lived, seldom exceeding 25 cm in length or five years of age in the Gulf of Maine region (Murawski and Cole 1978, Lawton et al. 1990). By age two, smelt are fully mature and recruited to local recreational fisheries and spawning runs. Life history appears to be influenced by latitude; few age-I smelt become mature and participate in Canadian smelt runs, however in Massachusetts, New Hampshire, and southern Maine, age-I individuals are present in the spawning runs (Collette and Klein-MacPhee 2002). Studies in Massachusetts found that the majority of age-1 spawners were male (Murawski and Cole 1978, Lawton et al. 1990). Our current spawn-ing surveys have found that runs in the Gulf of Maine are dominated by age-2 smelt, with few older smelt in Massachusetts, New Hampshire, and southern Maine; however the older ages are better represented in midcoast and eastern Maine. Fecundity estimates of approximately 33,000 eggs for age-2 smelt and 70,000 eggs for age-3 smelt were reported by Clayton (1976).

Habitat Use Annual movements and habitat use by adult rainbow smelt have been large-ly assumed based on discrete sampling or patterns in recreational and commer-cial fishing. Mark and recapture studies have focused on distinct phases of the life cycle, such as movements between spawning areas (Murawski et al. 1980),

composition of late and early populations of spawning adults (McKenzie 1964) and winter movements within a river system (Flagg 1983). Larger annual and regional migrations have been synthesized from anecdotal reports by anglers and commercial fishermen as well as from beach seine and spawning surveys.

6

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN I

Rainbow smelt overwinter in estuaries and bays and then spawn in early spring in pool and riffle areas above the head-of-tide in coastal streams and rivers. The spawning habitat characteristics are discussed in detail in sections 2.1 - Threats to Spawning Habitat Conditions, and 2.2 - Threats to Embry-onic Development and Survival. Because males have a longer physiological spawning period, they may return to spawning grounds multiple times within the same year (Marcotte and Tremblay 1948). Mark and recapture studies have observed the same male at different spawning sites within a given year, suggest-ing that males are able to spawn multiple times (Murawski et al. 1980, Rupp 1968). Murawski et al. (1980) hypothesized that spawning in different streams may be facilitated by passive tidal transport, however this has not been directly observed. Females, on the other hand, rarely ascend to the spawning grounds more than once in a season, based on recent mark-recapture surveys (C.

Enterline, unpublished data). Because female smelt are broadcast spawners, their spawning is expected to occur in a single event as most or all of their eggs are deposited in a single event.

Spawning females deposit demersal (sinking) adhesive eggs that attach to the substrate and hatch in 7-21 days, depending on temperature. Upon hatching, larvae are immediately transported downstream into the tidal zone, at which point the larvae begin feeding on zooplankton. Larval dispersion is mostly passive in response to river flow and coastal circulation patterns, but there is also an active (swimming) component (Bradbury et al. 2006b).

Although horizontal movements of smelt larvae appear passive, they actively migrate vertically in response to tidal flow in order to maintain their position in zooplankton rich water and minimize downstream movement (Laprise and Dodson 1989, Dauvin and Dodson 1990, Sirois and Dodson 2000). This active swimming behavior is overwhelmed by passive transport in local circulation patterns. The importance of local circulation on larvae dispersion is discussed more in the genetic stock structure section below.

Juvenile smelt remain in the estuary, bay, or sheltered coastal area through the summer, and sometimes through the early fall (NHF&G and ME DMR Juvenile Abundance Surveys, 1979-2011, analysis for current study). In Great Bay, NH, juvenile smelt are most abundant in August, while in the Kennebec and Merrymeeting Bay estuary complex in Maine, abundance is more evenly distributed between August, September, and October (Figure 1.1.1). In Maine, catches of juvenile smelt occur from July to October, while in New Hampshire, catches range from June to November.

Habitat use in marine waters is largely unknown but can be inferred through interviews with coastal fishermen and state trawl surveys. Smelt may migrate in search of optimum water temperatures, moving offshore during the summer months to greater depths with cooler water (Buckley 1989). Based on low catches by fishermen in freshwater and larger catches in brackish and saltwater in May, the presumed end of the spawning run, it has been assumed that adults return to estuaries and coastal waters immediately after spawning (Bigelow and Schroeder 1953). However, recent findings indicate that rainbow smelt may remain within estuaries and bays contiguous to their spawning sites for up to two months after spawning (C. Enterline, unpublished data).

Recent trawl surveys have found small schools of smelt as far from the coast as 60 km and in depths up to 77 m (data from the Maine-New Hampshire and ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 7

Massachusetts Trawl Surveys). Spring trawl surveys find smelt further from the coast and in deeper water (spring avg. depth = 29.7 m) than during fall trawl surveys (fall avg. depth = 19.9 m) (Figures 1.1.2 and 1.1.3; t-test comparing depth, p = 0.0338 < 0.05), however the average spring catch is smaller com-pared to the fall (spring average catch 2001-2012 = 31, fall average catch 2000-2011 = 129, Wilcoxon non-parametric test of means, p < 0.0001 < 0.05), likely because adult smelt are within coastal streams and rivers as part of the spawning event during the spring period. The smelt that are caught further offshore in the spring are smaller, with lengths associated with age- I fish; these are likely young fish that are not recruited to the spawning run.

As offshore water temperatures drop in the fall, smelt likely move towards the coast, eventually migrating into the upper estuaries where they overwin-ter (Buckley 1989; Clayton 1976; McKenzie 1964). Anecdotal reports from recreational hook-and-line ice-fishermen describe smelt moving in tidal rivers with the nighttime flood tide and out with the ebb tide, and some moving as far up as the head of tide each night. These foraging movements are the basis for robust recreational fisheries in the fall and winter at many locations in the Gulf of Maine.

Genetic Stock Structure in the Gulf of Maine Understanding the genetic structure of a species and the driving factors behind that structure is central to well-designed species management. A species may be comprised of one or more genetic stocks, separated by different spawning areas or physical barriers. Managing a species at too large a scale (i.e.,

assuming there is only one stock when there are multiple) may lead to the loss of genetic structure and the benefits of local adaptation. Managing at too small a scale (i.e., assuming stocks are isolated within individual rivers when in fact there is some mixing), neglects the important role of gene flow and results in loss of genetic variation (Kovach et al., in press).

From 2006-2010, we collected genetic samples at 18 spawning site index stations spanning the Gulf of Maine to understand if unique genetic stocks existed and the extent of gene flow between spawning populations. All informa-tion presented in this conservation plan was reported by the University of New Hampshire and in detail by Kovach et al. (in press). The three most genetically divergent populations were found in Cobscook Bay, Maine, Massachusetts Bay, and Buzzards Bay, Massachussetts. Penobscot and Casco bays in Maine also showed some differentiation. Gene flow was high between rivers from downeast coastal Maine, the Kennebec River, ME, and Great Bay, NH to northern Massachusetts; all were dominated by the same genetic signal. Midcoast Maine also seemed to be part of this large stock, but also showed distinct signals from Penobscot Bay and Casco Bay (Figure 1.1.4). These groupings can assist management decisions on stocking efforts, with the goals of maintaining distinct stocks where possible, while still preserving gene flow to maintain and replenish genetic diversity.

Although the study did not find evidence of genetic bottlenecking, genetic variation was significantly reduced in the two most distinct regions: Buzzards Bay (Weweantic River), and Cobscook Bay (East Bay Brook) (Kovach et al.,

in press). The reduced diversity in the Weweantic River is consistent with its 8 ° ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

location at the southern extent of the species range, where populations can have reduced gene flow and lower spawning population sizes (Schwartz et al. 2003).

The reduced variation in Cobscook Bay is more likely due to isolation by circulation patterns. The reduced diversity and distinctive nature of these smelt runs warrant further population monitoring and possibly updated protection measures.

The divergence patterns observed may be explained partly by coastal circu-lation patterns (Kovach et al., in press). Because the movement of smelt larvae is largely passive during the early development (Bradbury et al. 2006b), their dispersal is determined first by river flow and secondly by marine circulation.

The Gulf of Maine Coastal Current (GMCC) has a counter-clockwise pat-tern, which is strongest in the summer months when smelt larvae are present in coastal waters. The GMCC consists of two distinct portions. The Eastern Maine Coastal Current (EMCC) flows from the Bay of Fundy southwest along the coast and, in the area of Penobscot Bay, often splits southward and offshore.

The remaining portion of the EMCC combines with outflow from Penobscot Bay and continues southwestward towards coastal New Hampshire and Mas-sachusetts, creating the Western Maine Coastal Current (WMCC; Pettigrew et al., 1998, 2005). Backflow eddies are associated with large rivers (like the Penobscot) and to a lesser extent with Casco Bay, and as a result, larvae may be maintained within the nearshore area. Continuing further southwest along the coast, Massachusetts Bay maintains high larval retention as the strength of the WMCC pattern has largely diminished by this point (Incze et al. 2010).

Figure 1.1.1. Mean smelt catch 600 by month In the Maine and New HampshireJuvenile Abundance 500" Surveys 1979-2011 for all survey sites combined. Errorbarsrepre-400 sent one standarderrorfrom the mean.

300-200" 100 -j d 0" 500 400-300" 200-100 0 M I 5 7 8 9 10 11 Month ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 9

Figure 11.2. Smelt catches In the fall state nearshoretrawl surveys for Massachusetts,New Hampshire,and Maine 2000-2011.

Figure 1.1.3. Smelt catches In the spring state nearshoretrawl surveys for Massachu-setts (2000-2011),

New Hampshire,and Maine (2000-2012).

10

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Figure 1.i4. Genetic differentia-tion of smelt stocks In the Gulf of Maine from Kovach et aL,

("in press'). Divergencemay be explained by circulation patterns, where the Gulf of Maine Coastal Current carrieslarvae from downeast coastalMaine to New Hampshireand northern Mas-sachusetts,while other localized circulation patternsmaintain the distinctiveness of Penobscot Bay, Casco Bay, Massachusetts Bay, and Buzzards Bay. The color boxes display the 6 genetic signals Ca88o - boxes with the same colors EMCC Indicate the same signal. Length of boxes represents number of samples taken from the region.

WMCC

-Massachusetts Bay Buzzards Bay J 110 IKilometers ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 11 I

1.2 - HISTORICAL SMELT FISHERIES Smelt fishing is a longstanding tradition in many coastal communities of New England and the Canadian Maritimes. During winter and early spring, smelt schools enter estuaries and embayments and aggregate in preparation for the spring spawning run. During this period of migration, commercial, and recreational fisheries target smelt through the ice and from shore. Some shore fisheries also occur in fall, mainly with hook and line, during foraging move-ments that precede the spawning migration. Fishing methods for smelt vary by state; including weirs, hook and line, seines, dip nets, bag nets, and gill nets.

This section will describe the historical range of rainbow smelt and the fisheries that targeted them. We focus on the Gulf of Maine, but provide some background on populations throughout the range. We rely heavily on the classic work "The Smelts" by Kendall (1926) and the thorough recent literature review found in Fried and Schultz's (2006) investigation in Connecticut.

By the late 1800s, with The earliest record of smelt harvest in the U. S. was likely by Captain John the advancement of rail Smith in 1622; Smith noted the smelts were so plentiful that the Native Ameri-transport,smelt were an cans would harvest the fish by simply scooping them up in baskets (in Kendall important export product 1926). There is little additional information about early New England smelt shipped on ice from the harvests until the mid-1800s, although extensive subsistence and local com-mercial harvest occurred before this time, based on occasional references and Canadian Maritimes and town records. Early uses of smelt included livestock feed and fertilizer to enrich Maine to the Boston and farm fields. The abundance of smelt in the mid-1800s can be pictured from New York markets. the account of French settlers along the Buctouche River in New Brunswick harvesting 50 to 60 barrels (36 gallons/barrel) annually to serve as fertilizer for each homestead (Perley 1849 in Kendall 1926). About this time, food markets developed for smelt as human populations grew in coastal cities. By the late 1800s, with the advancement of rail transport, smelt were an important export product shipped on ice from the Canadian Maritimes and Maine to the Boston and New York markets (Kendall 1926).

Mid-Atlantic Smelt are considered a cold water fish, with a historical center of abundance north of Cape Cod but southerly populations ranging south to the Mid-Atlan-tic. Early references of smelt range include Virginia, Maryland and Delaware (Goode 1884, Kendall 1926, Bigelow and Schroeder 1953), but we found no information on smelt populations or harvests for these states. Later references on smelt range list New Jersey as the southern limit (Scott and Scott 1988, Collette and Klein-MacPhee 2002). Overall, references south of Delaware Bay are not well documented. The presence of smelt in states south of New Jersey may have been sparse, an indication of occupancy at the edge of the species' range, or alternatively the fisheries may have faded before the onset of recorded commercial harvest data in the early 20th century.

New Jersey In 1833, smelt were observed to be plentiful in New Jersey with "wagon-loads" of smelt harvested in Newark Bay, yet by 1849, smelt were reported as declining (New York Times 1881 in Fried and Schultz 2006). The Delaware 12 - ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN I

River had been listed as a southern smelt run, including an early observation in a tributary, the Schulykill River, of cast net fishing for smelt during late winter (Norris 1862). Spring runs of smelt, also called frost fish, were reported in the Delaware, Hackensack, Passaic and Raritan rivers during the late 1860s. By this time, only the Raritan River supported a lucrative commercial fishery, with annual catches nearing 10,000 lbs (NJCF 1872). The New Jersey Commis-sioners of Fisheries (NJCF) 1872 report also suggested that industrial water pollution in the rivers was severely impacting all anadromous fisheries. The last regular commercial catch in New Jersey was reported in 1921 (Fried and Schultz 2006).

Smelt were considered endangered in New Jersey by 1877 and the state launched an effort in the 1880s to study the reproductive biology of smelt and to stock smelt fry hatched from eggs collected in viable smelt runs to depleted smelt runs (NJCF 1886).

No evidence of stocking success has been located and by 1941 smelt were considered extirpated from New Jersey (Camp 1941 in Fried and Schultz 2006). The New Jersey Fish and Game Department has conducted trawl surveys throughout their coastal waters since the early 1980s, and no smelt have been detected during this time.

New York Historical references indicate that tributaries near the Hudson River and Long Island once supported prominent recreational and commercial fisheries but that overfishing and poor water quality likely caused declines be-fore the end of the 19th century (Kendall 1926). The smelt trade at the Fulton Market in New York City was reported to average 1,352,000 lbs annually in the 1870s (Scott 1875 in Kendall 1926). By 1887, the smelt fishery was no longer considered commercially viable (New York Times 1881, Mather 1887, Mather 1889; in Fried and Schultz 2006). State fishery agencies in New York became concerned about the declining status of smelt in the late 1800s and embarked on extensive stocking efforts that included placing 127 million eggs in Long Island streams during 1896-1898 (Kendall 1926). The stocking efforts faded when smelt eggs became scarce in the early 20th century (Kendall 1926).

Commercial catches declined and became sporadic in the 20th century. Rou-tine commercial harvests exceeding 1,000 lbs annually were last reported in the 1950s (Fried and Schultz 2006).

Since the 1970s, annual surveys in New York have detected rainbow smelt, but catches have become increasingly infrequent and have been rare since the 1990s. The Hudson River Estuary Monitoring Program has conducted ichthy-oplankton and juvenile fish surveys throughout the estuary since 1973, and the data show a dramatic decrease in smelt abundance since the mid-1990s, with only trace numbers detected today (ASA A&C 2010). Fish sampling efforts conducted by New York State Department of Environmental Conservation (NY DEC) have produced similar results, with very few adults detected since the 1980s. Today, smelt are considered extirpated or at extremely low numbers in the Hudson River system (C. Hoffman, NY DEC, pers. comm. Sept. 2010).

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN 9 13

Connecticut A synopsis of early fisheries records shows that smelt runs were present in most tidal rivers in coastal Connecticut, and economically important commer-cial fisheries targeted the seasonal occurrence of smelt (Visel and Savoy 1989, Fried and Schultz 2006). Smelt were targeted primarily with haul seines and gill nets in the Housatonic, Connecticut and Pawcatuck rivers (Visel and Savoy 1989). Hook and line angling was also common in the 19th century at numer-ous locations; smelt were described as an important export fish to New York City markets. Smelt landings were reported as peaking in Connecticut in the 1880s at 27,000 lbs and steadily declining with minor and intermittent land-ings since the 1930s (Fried and Schultz 2006). There was a modest increase in landings in the 196 0s when several thousand pounds were reported annually.

The last years with significant smelt runs in Horseneck Brook of Greenwich, were 1965 and 1966 (Visel and Savoy 1989).

By the 1980s, smelt were recognized as nearly absent from Connecticut's coastal rivers. Similar to regions south of New England, concern centered on the role of point and non-point pollution sources (Visel and Savoy 1989).

The decline of smelt in Connecticut prompted dedicated efforts to document their presence in the 2000s. The smelt fishery was formally closed to harvest in 2005, and smelt were listed as a state endangered species in 2008. Fried and Schultz (2006) carried out intensive surveys in five estuaries along the central and eastern Connecticut coast. They documented no evidence of smelt spawn-ing but did catch 9 adults while seining in the upper Mystic River during 2004.

State beach seine surveys infrequently encounter smelt, however there have been recent observations of a few adult smelt in 2007 (T. Wildman, CT DEP Inland Fisheries Division, pers. comm. Nov. 2010). The State of Connecticut is currently considering listing smelt as extirpated from the state.

Rhode Island Smelt landings first appear in Rhode Island records in 1880 with landings of 95,000 lbs, which remains the peak annual harvest for this state (Fried and Schultz 2006). Since that point, landings records steadily declined with minimal landings reported after 1932. Landings rebounded slightly during 1965-1970 when several thousand pounds were reported annually. Since this time, minimal commercial landings have been reported (Fried and Schultz 2006). In response to declining populations, the Rhode Island Division of Fish and Wildlife (RIDFW) began a smelt stocking and monitoring program in 1971 (RIDFW 1971). Over the next seven years, approximately 44 million smelt eggs were transferred from populations in Massachusetts and New Hampshire to four rivers in Rhode Island. Extensive monitoring was conducted at the four recipient rivers, and no evidence was found of successful recruitment following stocking (RIDFW 1978). The monitoring only found evidence of a viable smelt run in the Pawcatuck River where low densities of smelt eggs were observed in 1974. The stocking effort was considered unsuccessful and discontinued in 1977 (RIDFW 1978). In the last decade smelt were briefly listed as endangered in Rhode Island, then delisted and considered extirpated with a chance of a trace populations present. Adult smelt have been captured on rare occasions during coastal pond and bay surveys since the 1990s (A. Libby, RI DFW, pers. comm. Oct 2011).

14

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Massachusetts HistoricalFisheries Early accounts indicate that smelt populations in Massachusetts supported culturally important sustenance fisheries that evolved into small-scale commer-cial and recreational fisheries as coastal populations grew. The smelt fisheries prior to 1874 targeted fall and winter feeding aggregations with baited hooks and used dip nets and seine nets during the spring spawning runs (Kendall 1926). The local importance of these fisheries and the potential abundance of the populations is reflected in accounts that describe over nine million smelt taken from the Charles River at Watertown in 1853 (Storer 1858), and over 2,300 fishermen at Hough's Neck in Quincy in one day targeting smelt (Kend-all 1926). Overfishing concerns were raised in the 1860s that were attributed to with the use of nets during the spawning run. This concern led the Massachu-setts State Legislature to prohibit net fishing for smelt during the spawning run in 1868 (Kendall 1926). A quote the Massachu-In 1874, a law prohibited the taking of smelt by any method other than setts Commissioners on hook and line in all state waters with a few exempted rivers - most of these Fisheriesand Game in exemptions were revoked by the end of the century. Kendall (1926) relates 1917 expressed the accounts of rebounding smelt fisheries in the 1870s and praise for the net ban.

concern of the period, Catch records are sporadic and largely town or county specific during the latter "The smelt fishery of half of the 19th century. However, there was a general declining trend in this period, and by the 191 Os and 1920s there was growing concern about smelt Massachusettsis in a fisheries in Massachusetts and the influence of industrial pollution. A quote the depleted condition, and Massachusetts Commissioners on Fisheries and Game in 1917 expressed the strenuousand radical concern of the period, "The smelt fishery of Massachusetts is in a depleted measures will be required condition, and strenuous and radical measures will be required to save this spe-to save this species from cies from extinction" (MCFG 1917).

extinction."

Smelt fisheries are poorly documented in Massachusetts after Kendall's 1926 report. The annual reports of the state fisheries agency depict contrasting trends along a gradient. In southern Massachusetts, there was a sharp decline in commercial importance and the disappearance of smelt in some locations.

However, north of Cape Cod and in the greater Boston area, an active and popular fall and winter sportfishery persisted through the 1970s. Fried and Schultz (2006) summarized federal commercial catch records that show three time-series peaks in Massachusetts harvest: 1880 (82,034 lbs), 1919 (39,000 lbs), and 1938 (25,000 lbs). The early landings data were based on the available town and county records and are expected to be incomplete (Kendall 1926).

It is likely that no records adequately describe the true extent of smelt harvest at any time in Massachusetts's history. The view provided by the combined historical and anecdotal accounts suggests that smelt supported important sea-sonal fisheries that attracted large numbers of anglers and that smelt occurrence and abundance greatly exceeded the species' present status.

Recent Trends Striking changes appear to have occurred in smelt detection and abundance in Massachusetts since Kendall's report (1926). Contemporary studies began with river-specific work in the Jones and Parker rivers in the 1970s (Lawton et al. 1990, Murawski and Cole 1978, and Clayton 1976). These studies were the first to report biological characteristics of the spawning runs and timing of ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 15

movements in Massachusetts. Concerns over declines in smelt abundance grew after these studies, as sportfisheries' catches declined sharply in the late 1980s.

The MA DMF responded to concerns from the sportfishing community with a survey of all smelt spawning habitats on the Gulf of Maine coast within Massa-chusetts during the 1990s (Chase 2006) and the initiation of fyke net monitor-ing in 2004 to develop population indices.

Specific mention of Buzzards Bay is warranted because it is presently the southern limit of the documented spawning range. Buzzards Bay lies directly south of Cape Cod, which separates the Virginian marine ecoregion to the south from the Gulf of Maine/Bay of Fundy ecoregion to the north (Spalding et al. 2007). No historical records have been found of spawning runs on Cape Cod, a likely result of its glacial formation and flat gradient. Goode (1884) reported smelt harvest in coastal weir fisheries in Buzzards Bay in 1880. More recently, an anadromous fish survey from 1967 reported 10 rivers in Buzzards Bay with active smelt spawning runs (Reback and DiCarlo 1972). An estuarine survey of the Westport River in Buzzards Bay in 1966-1967 found smelt in seine and trawl surveys and reported a known spawning run and associated fish-ery in the river (Fiske et al. 1968). Smelt runs in the region have since quietly faded to low levels of detection. Fisheries monitoring during the last 10 years has documented the presence of smelt in only three Buzzards Bay rivers; with a lone viable spawning run in the Weweantic River.

New Hampshire HistoricalFisheries Significant smelt fisheries of commercial and cultural importance have occurred in the Great Bay estuary of New Hampshire since the 18th century or earlier. Hook and line fishing has mainly occurred in winter through ice on tidal waters. Additionally, bow nets were traditionally fished under the ice, and weirs were deployed during spring spawning runs (Warfel et al. 1943).

Historical fisheries in New Hampshire are poorly described relative to Maine and Massachusetts. Kendall (1926) provides very little information on coastal New Hampshire smelt runs, focusing more on landlocked populations. He does provide annual smelt harvest estimates for coastal fisheries as follows: 1888

- 36,000 lbs, 1908 - 2,600 lbs, and 1924 - 3,835 lbs. The reported peak of commercial catch in New Hampshire was between 1940-1945, with an estimated 150,000 lbs harvested per year (Figure 1.2.1; Fried and Schultz 2006). It is expected that the historical records substantially underreported actual harvest from the Great Bay fisheries.

Recent Trends The state of New Hampshire has monitored smelt fisheries in Great Bay since the 1970s, when concerns were voiced from fishery participants about declining catches. To this end, an angler creel survey was started in 1978 and a smelt egg deposition survey began in 1979. A project was also launched at that time to improve commercial harvest data by mandating bow net and weir net fishermen to record their catches in log books. In 1981, a statewide smelt fishery management plan was written by the New Hampshire Fish and Game Department (NHF&G) to maintain sea-run smelt populations and support commercial and recreational fisheries (NHF&G 1981).

16

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Data collected by the NHF&G indicate declining population trends in recent decades. The angler creel survey data depict a reduction in CPUE and total catch during the 2000s (Sullivan 2010). The smelt egg survey shows egg densities in the 2000s that are an order of magnitude lower than the 1980s (Sullivan 2007); the survey was discontinued in 2008 due to concerns over methodology and very low presence of smelt eggs. The commercial harvest re-cords in New Hampshire have also recorded declines since 1987 (Figure 1.2.1).

Commercial dip net and bow net permits remain active, but the fisheries have declined to low levels of catch and effort (J. Carloni, NHF&G, pers. comm.,

2011). Despite the apparent decreasing trends, recreational fishing for smelt in Great Bay still remains a popular winter fishery that attracts higher catch and effort than fisheries to the south in Massachusetts.

Maine HistoricalFisheries Commercial and sustenance smelt fisheries were important to Maine's colonial inhabitants as early as the 18th century, but are poorly documented.

Kendall (1926) provides detailed accounts of valuable commercial hook and line and net fisheries from the 1880s to 1920s. The opening of export markets to New York and Boston after the mid-i 800s, coupled with growing use of seine and bag nets, led to increases in harvest and the development of a signifi-cant commercial fishery. Goode (1884) provides the first reported commercial smelt harvest records for Maine, with landings exceeding a million pounds in the 1880s. In 1894 the smelt fishery was reported to support 1,100 fishermen with shore fishery landings that were the fourth most valuable behind lobster, clams, and sea herring (Whitten 1894). Statewide records are absent before this time, however subsequent catch data show a steep decline after the 1890s (Squires et al. 1976; Figure 1.2.1). The last year the Maine catch exceeded a million pounds was in 1903. As early as 1920, a report by the Maine Commis-sion of Sea and Shore Fisheries described the depleted status of smelt runs and the negative impacts of targeting spring spawning aggregations for commercial harvest (MECSSF 1920). An early management response to this decline was performing egg transfers from both landlocked and sea-run smelt populations to depleted runs (Kendall 1926); these were largely undocumented. While the commercial fishery continued to decline in the 20th century, the recreational fishery that targeted smelt both through the ice and during spawning runs increased in catch and effort starting in the 1940s. The rental ice shack fishery, in particular, grew in economic importance as out-of-state anglers were attract-ed to Maine's coastal rivers.

Recent Trends Recognizing the traditional importance of the smelt fishery and continued population declines, the Maine Department of Marine Resources (ME DMR) developed a Smelt Management Plan in 1976 (Squires et al. 1976). The plan outlined present conditions and made recommendations to improve fisheries and spawning habitat. It also attributed the dramatic decline observed in the mid 20th century to increased industrial pollution in Maine's rivers after World War II (Figure 1.2.1). The ME DMR also launched studies at this time to record the presence and distribution of smelt in coastal Maine and investigate ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 17

causes of the historic decline (Flagg 1974). Flagg's (1974) work on Maine's sea-run smelt documented catches at camp fisheries on the Kennebec River and Merrymeeting Bay, and catalogued spawning runs on 134 coastal streams.

As part of the present study, the ME DMR has reinstituted creel surveys and spawning habitat investigations so that current catch records can be compared to the 1970s monitoring. Maine continues to have important recreational fisheries featuring winter ice fishing on tidal rivers and spring dipnet fishing at spawning runs, although annual harvest is at historic lows. A modest commer-cial harvest continues in downeast Maine, largely centered on the Pleasant River in Columbia Falls, where gill and bag nets are allowed to fish in late winter.

CanadianProvinces HistoricalFisheries Anadromous smelt populations in Canada have long supported valuable Anadromous smelt commercial fisheries that greatly exceed the collective harvest from the United populationsin Canada States. Among provinces, New Brunswick has had the largest fishery, which have long supported historically targeted smelt for use as fertilizer and bait (Goode 1884). Growing valuable commercial export markets were driven by the Canadian harvests, which were, and continue fisheries that greatly to be, the largest commercial harvests in the species' range. Records are sparse exceed the collective before the 20th century, however Kendall (1926) cites accounts of fast develop-ing export markets to Boston and New York in the 1870s that created demand harvest from the United for large harvests - exceeding two million pounds by the 1880s. In 1901, the States. shipment records of one export company in New Brunswick approached eight million pounds. The highest aggregate landings reported for Canada was just over nine million pounds in 1914 (Kendall 1926). A report from the U. S.

Bureau of Fisheries in 1920 noted that while the Maine smelt fishery had de-clined in the early 1900s, the New Brunswick fisheries had undergone "remark-able" growth to support the market demands in the U.S. (USDOC 1920). The Miramichi River in New Brunswick was long a center of the province's smelt fishery. Shipments of smelt to U. S. markets from the Miramichi River region exceeded 4.3 million lbs for the winter fishery in 1924 (Kendall 1926), making the fishery one of the most valuable industries in the Province at that time.

Recent Trends New Brunswick and Nova Scotia continue to support important commer-cial fisheries. There is less evidence of population declines in these provinces than in the U. S. portion of the range. The capitalization of a Great Lakes fish-ery for smelt in the 196 0s and 1970s resulted in high landings that suppressed prices and may have reduced effort in the New Brunswick fishery (McKenzie 1964, DFO 2011). In spite of depressed prices, the eastern New Brunswick smelt fishery remained stable between 1988 and1998, with total reported landings between 1.5 and 2.5 million lbs, a sum that may under represent actual landings (DFO 2011).

The smelt fisheries of the St. Lawrence River have shown a decline com-parable to U. S. fisheries. Reduced commercial and recreational fisheries and spawning habitat abandonment in the St. Lawrence River tributaries triggered survey and restoration efforts in the 2000s (Trencia et al. 2005). The fisheries remain culturally important today while operating at historically low harvest 18 ° ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

levels with ongoing restoration efforts by Quebec's Ministry of Natural Resources (Verreault et al. 2012).

Summary Dramatic changes have occurred in both Gulf of Maine smelt fisheries and the distribution of smelt on the East Coast since the start of the 20th century.

Culturally and economically important smelt fisheries have disappeared or faded to historic lows. The trend is evident of wide-scale abandonment of the historic southern extent of the range, where commercial smelt fisheries were viable before the 20th century. Currently, the southern extent of the species range is likely in the Buzzards Bay, Massachusetts region, with higher popula-tion levels observed in more northern rivers.

Popular recreational fisheries remain in Maine and New Hampshire, although these fisheries also appear to be harvesting at historically low levels.

The traditional Massachusetts ice shack fisheries have been reduced to very low levels of participation and catch, and they are faced with warmer winters that bring insufficient ice to support shacks. The causes of this steep decline in smelt fisheries on the U. S. East Coast have not been defined, but have been discussed for over a century. Industrial pollution at spawning rivers, structural barriers, and overfishing have received the most attention as causal factors.

Watershed alterations, natural predation and climate change are potential fac-tors that have been implicated more recently.

Figure 1.2.1. Commercial smelt Commercial harvest of rainbow smelt in ME and NH landings for Maine (1887-2009) and New Hampshire (1950-2009).

1400 Datasources: U.S. Commissioners Report, U.S. Bureau of Fisheries, 1200 State of Maine landings data (as

~1. summarized by Squlers et al.

Zl 1976), and NMFS website.

~.800 v

-ME I~

C600-H c 400 0Z 200 I (,) 0) U) - I- C' 0) UZ) - t- C') 0) UZ) ,. C"- C') 0) U) ,-

rrrrr rrrrr rrrrN N Year ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 19 I

1.3 - POPULATION STATUS INTHE GULF OF MAINE Concerns have grown over the health of anadromous rainbow smelt popu-lations throughout much of their range. This concern has prompted interest in assessing smelt populations and developing restoration strategies. Limited information is available from both fisheries-dependent and independent sources on the present status of populations in New England. The Species of Concern (SOC) project reviewed existing smelt population data in New England to consider the potential for developing indices of abundance, and initiated field projects during 2008-2011 to establish new data series to provide information on the status of smelt runs.

Previous Smelt PopulationStudies The earliest smelt population studies occurred in northern portions of their range, likely in response to the commercial importance of smelt fisheries in Studies conducted in these regions. Kendall (1926) focused on smelt fisheries but did provide smelt the late 1950's described length data gathered from various sources during the 1850s to 1920s. Not several life history char- much information can be gleaned from these sparse data, except to say the max-acteristicsthat we also imum size of smelt from that time period of about 26-28 cm (total length) is observed in the present quite similar to the maximum size found in the present study (27 cm). Warfel study, such as declining et al. (1943) reported smelt age data for Great Bay, NH; this study provided some of the first age data for the area and perhaps the first documentation of averagelength as the age-I smelt participating in the spawning run. Summary statistics for Warfel et spawning run progresses al. (1943) and the following studies are presented in Table 1.3.1.

and few smelt over the McKenzie (1958 and 1964) followed the Great Bay study with a detailed age of three. study of the life history of smelt and their fisheries in the Miramichi River of New Brunswick during 1949-1953. McKenzie (1964) demonstrated several life history characteristics that have been confirmed in the present study, such as: declining average length of smelt as the run progresses, a more balanced sex ratio in the winter fishery than during the spawning run and few smelt older than age- 4 . The age composition in the Miramichi River during 1949-1953 had consistently higher representation of age-3 (22-49% annually) and age-4 (2-8% annually) than seen in the present study and had older fish present each year, although at low proportions (age-5 and age-6 at <0.5% and <0.1%,

respectively). Murawski and Cole (1978) calculated an annual survival rate (S) of 0.35 for the overall proportions in McKenzie's age composition data, a value found to be the highest among reported survival data for anadromous rainbow smelt (Chase et al. 2012).

The ME DMR devoted considerable time to the assessment of smelt fisher-ies in the 1970s and 1980s (Squiers et al. 1976, Flagg 1983). The majority of the effort was fishery-dependent assessments of the winter smelt fishery. The size composition data from these winter fishery studies may not be directly comparable to spawning run size composition. However, summary data on sampling proportion by age and mean length at age are included in Table 1.3.1 because the data document the size composition of smelt populations at the time and the relatively larger contribution of older smelt in the catch.

Murawski and Cole (1978) provided size, age and mortality data from the Parker River, Massachusetts spawning run and winter fishery during 1974-1975. This study sampled both the winter sport fishery catch and spring 20

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

spawning run with a fyke net, providing a valuable comparison to the Parker River data in the present study. Five age classes were represented in the fyke catches, with a majority at age-2. Murawski and Cole (1978) also provided one of the few estimates of smelt population mortality and survival rates. They reported mean values of the annual survival rate (S) of 0.28 and the instanta-neous total mortality (Z) of 1.27 for both sexes using three analysis methods for the spawning runs. They considered the estimated overall annual mortality rate of 72% of the adult population to be high and that increases in fishing pressure could limit reproductive success in the Parker River.

Lawton et al. (1990) investigated biological aspects of the Jones River (MA) smelt spawning run during 1979-1981. The study used a lift net at the upstream limit of smelt spawning habitat to collect mature smelt. All biologi-cal data collected by the lift net may not be directly comparable to the present study, wherein a fyke net was deployed downstream of the lowermost spawning habitat. However, the study did produce an age/length key based on length-stratified age subsamples that should be representative of the spawning run demographics and comparable to the fyke net age/length data. Five age classes were found in the Jones River with an age-2 majority for most years and very few age-5 smelt. For the three spawning seasons sampled, age-2 and age-3 smelt comprised 83-99% of the spawning smelt. Lawton et al. (1990) also estimated the Jones River spawning population by extrapolating smelt egg densities to to-tal spawning habitat area. The spawning stock abundance model calculated the spawning run of 1981 to exceed four million adult smelt. They also reported evidence of a strong 1978 year class with relative contributions of this cohort evident in the subsequent three spawning runs.

The smelt runs of the St. Lawrence River have supported culturally and ec-onomically important fisheries in Qu6bec for decades. Declining smelt fisher-ies landings attracted the interest of the Qu~bec Ministry of Natural Resources to conduct biological monitoring in the 1990s. Pouliot (2002) reported on size and age sampling of the spawning run in a St. Lawrence River tributary, the Fouquette River, during 1991-1996. A standardized dipnet sampling method was used at night at the spawning habitat. The results provide the first detailed population demographics and mortality estimates for smelt in the St. Lawrence River watershed. The Fouquette River smelt runs during the 1990s contained four or five cohorts in most years. Estimates of the annual rate of total mortal-ity were 74% for females and 73% for males.

CurrentFisheriesDependent Monitoring New Hampshire Creel Survey NHF&G has conducted winter creel surveys since 1978. The survey occurs from ice in to ice out, generally between the months of December and March. Four locations are sampled: the Lamprey, Oyster/Bellamy and Squamscott rivers as well as Great Bay. From 1983-1986 no survey was con-ducted due to lack of funding, and in 2002 and 2006 fishing, and subsequently surveys, were not possible due to lack of ice cover.

Biologists interview all anglers (or a sub-sample when large groups of an-glers are present) for catch and effort information during a two hour survey pe-riod per day, visiting locations on a rotating basis. The information collected is ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 21

expanded to provide estimates of catch, effort and catch per unit effort (CPUE) by month and location. Biological information from the smelt catch, including length, sex and scales for ageing, are taken from 150 fish weekly.

The average CPUE for 1987-2011 is 4.48 fish/hour over the entire sample period. High CPUEs have not been observed in the last ten year period (2000-2011, max CPUE 5.6), compared to the previous twenty year period (1980-1989 max CPUE = 10.3; 1990-1999 max CPUE = 10.6; Figure 1.3.1). In most recent years, the CPUE has been below the series average (4.48) until 2011 when it increased to 5 fish/angler hour. There has not been a peak in CPUE over 6 fish/angler hour since 1995. The CPUE shows large inter-annual variability, and seems to follow a 5-10 cyclical pattern (Figure 1.3.1).

Maine Creel Survey Adopting sampling methods currently used by NHF&G (Sullivan 2009) and methods used in a 1979-1982 study conducted by the ME DMR (Flagg 1983), ME DMR again began conducting creel surveys in 2009 in the Ken-nebec River and Merrymeeting Bay area. As part of this survey, ME DMR staff visited participating camps two or three times per week on a rotating basis to collect biological information about the recreational catch. Staff collected biological information from a subset of each angler's catch (up to 100 fish per angler), including length, sex, scale samples for ageing and fin clip samples for genetic sampling. The number of anglers, fishing hours, and the number of fishing lines used was also recorded.

CPUE was calculated as the total number of smelt caught per line-hour of fishing, as opposed to NHF&G calculation of CPUE as smelt caught per angler hour - ME DMR currently calculates CPUE using line-hour to remain consistent with surveys conducted by ME DMR 1979-1982. The recent survey found a slightly lower CPUE (0.48), compared to the 1979-1982 study CPUE (0.64), however inter-annual variability was significantly larger than the comparison between the two study periods (Figure 1.3.2, Flagg 1983). While annual fluctuations in CPUE occurred in both surveys, the recent survey had the lowest CPUE recorded (0.17) during the two time series.

Catch Card boxes were also posted at each camp for fishermen to voluntari-ly report information about their total smelt catch and any bycatch; responses varied widely between sites and between years. There were 122 responses in 2009, 6 in 2010, and 37 in 2011 for all camps combined. It is our hope that with continued interaction with anglers and camp owners that the number of responses will increase. Despite the low number of responses in 2010, the Catch Cards still reflected the catch patterns found in creel survey data.

Current FisheriesIndependent Monitoring State Inshore Trawl Surveys The three state fisheries agencies perform inshore small-mesh trawl sur-veys twice a year, in the spring (MA DMF in May, NH/ME in late May, early June) and fall (MA DMF in September, NH/ME in October, early November).

The MA DMF has been performing surveys since 1978, while the ME DMR began sampling the New Hampshire and Maine waters in fall 2000. These surveys provide information about marine habitat use and migration patterns 22 o ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

of rainbow smelt, as discussed in section 1.1 - Basic Biology. However, this survey is designed to monitor groundfish abundance, and has limited applica-tion for pelagic species like rainbow smelt. The data are helpful in determining the presence or absence of smelt in certain regions and depths, and can give a picture of inter-annual age cohort strength from size data, but are not powerful in showing trends in rainbow smelt abundance. However, trends in catches in both state surveys seem to have a 5-10 year cyclical pattern similar to the creel surveys and juvenile abundance surveys (Figure 1.3.3), although the causal factor behind these cycles is unknown.

Maine and New HampshireJuvenile Abundance Surveys In 1979, ME DMR established the Juvenile Alosine Survey for the Kenne-bec/Androscoggin estuary to monitor the abundance of juvenile alosines at 14 permanent sampling sites, sampled June through November. Four sites are on the upper Kennebec River, three on the Androscoggin River, four on Merry-meeting Bay, one each on the Cathance, Abadagasset, and Eastern rivers. These sites are in the tidal freshwater portion of the estuary. Since 1994, ME DMR added six additional sites in the lower salinity-stratified portion of the Ken-nebec River. The seine is made of 6.35 mm stretch mesh nylon, measures 17 m long and 1.8 m deep with a 1.8 m x 1.8 m bag at its center. The net samples an area of approximately 220 m2 .

Of all the river sections, the lower Kennebec catches considerably more juvenile smelt than all upstream sections; the average catch over the time period for the lower Kennebec was 92 smelt/haul/year, while all others were under 10 smelt/haul/year, and catches are sporadic. Though the highest average annual catch occurred in 2005 (316 smelt/haul) in the lower Kennebec, juvenile smelt abundance in this river segment has been low since 2007, with three of the four lowest average annual catches occurring in the past four years. Trends in abun-dance also seem to follow a 5-10 cyclical pattern similar to the other surveys (Figure 1.3.4).

The NHF&G has conducted an annual Juvenile Abundance Survey since 1997. It is designed as a fixed station survey, as opposed to a stratified random survey, because strong tidal currents, rocky shorelines, and various anthropo-genic structures limit the amount of suitable beach seining locations. A total of 15 fixed locations are sampled monthly from June to November. The stations are located within the Great Bay and Hampton-Seabrook estuaries. Seine hauls are conducted by boat using a 30.5 m long by 1.8 m high bag seine with 6.4 mm mesh deployed 10 - 15 m from the shore. Over the sampling period, the Piscataqua River has seen the highest CPUE (177 smelt/haul/year), however the highest annual average catch occurred in Great Bay in 2001 (225 smelt per haul). The lowest average catch over the entire sampling period was in the Hampton Beach/Seabrook area (11 smelt/haul/year). While these abundance data also seem to follow a cyclical pattern, there has been a decline in the juve-nile rainbow smelt being captured in recent years - excluding the first year of sampling, the four lowest average annual catches have occurred within the past 6 years (Figure 1.3.5).

New HampshireEgg DepositionMonitoring New Hampshire Fish and Game Department conducted egg deposition sampling from 1978-2007 using methodologies described by Rupp (1965). A ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 23

ring of known area (20.3 cm2 ) was tossed on natural substrate, and the number of eggs within the ring was counted. Egg counts were conducted weekly, from mid-March to mid-April, in the Oyster, Bellamy, Lamprey, Squamscott and Winnicut rivers. The mean number of eggs per square centimeter was used as an index of spawning stock abundance. Validation of the index was attempted by regressing the index with catch per unit effort (CPUE) of the creel survey but showed very poor correlation. The egg deposition sampling was discon-tinued in 2008 because comparisons between this dataset and other indices of smelt abundance (creel and juvenile surveys) did not correlate well, while trends in the other surveys did correlate well with each other.

Maine Spawning Stream Use Monitoring In 2005 and 2007-2009, biologists with the ME DMR worked with the Maine Marine Patrol to document coastal rivers and streams currently being used by rainbow smelt for spawning. The survey collected information about the spawning habitat (substrate, possible obstructions), and the strength of the run as characterized by the density of egg mats or number of spawning adults present. We compared the current use and strength of runs to information collected by ME DMR in the early 1970s (Flagg 1974) and to information compiled in 1984 by the U. S. Fish and Wildlife Service (USFWS 2012).

Of the 279 streams surveyed , the majority either supported smaller runs than they did historically or no longer support spawning, while only a small percentage (19%) seem to currently support strong runs (Table 1.3.2, Figure 1.3.6). Spawning decline was concentrated in southern Maine, lower Casco Bay, the Kennebec River, and the east side of Frenchman's Bay. Spawning runs remain strong in northern Casco Bay, the Medomak, St. Georges, and Penob-scot Rivers, and around Pleasant Bay and Cobscook Bay.

Regional Fyke Net Sampling Earlier research on anadromous smelt populations in the Gulf of Maine has primarily consisted of short-term efforts that monitor smelt size and age structure during spawning runs. These efforts have not produced long-term population indices of abundance for smelt, and presently, no indices exist in New England. The smelt SOC project targeted the spring spawning runs as a source of information on population status. The objective was to produce fishery-independent indices of abundance, with the understanding that only mature smelt participate in the spawning runs. The approach was to record biological data from spawning runs; to conduct analyses on size and age com-position, catch-per-unit-effort, and mortality; and to make comparisons as possible among rivers and to previous studies.

Establishing Gulf of Maine Spawning Site Indices Methods. As part of this project, fyke net stations were selected at coastal rivers in Maine, New Hampshire, and Massachusetts for monitoring during 2008-2011 (Figure 1.3.7, Table 2.1.1). The stations were chosen for suitability to maintain a fyke net in a known smelt run and to represent a range of run sizes and watershed conditions. The fyke net was set at mid-channel in the intertidal zone below the downstream limit of smelt egg deposition. The fyke net opening faced downstream, and nets were hauled after overnight sets. This approach was adopted to intercept the spawning movements of smelt that occur 24

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

at night during the flood tide. Fyke net catches were assumed to be representa-tive of the size and sex composition of the spawning run. With each haul, smelt were counted, sexed, measured (total length) and released. Scales were sampled weekly at some stations for ageing.

After pilot deployments in 2007-2008 to identify suitable stations, eight fyke net stations were monitored in Massachusetts, three stations in New Hampshire and six in Maine (Figure 1.3.7). The sampling period in Massachusetts targeted 11 weeks from the first week of March to the third week of May to cover the known smelt spawning period. The sampling dura-tion in New Hampshire and Maine varied due to a later ice-out and spawning season that occurs later with increasing latitude.

2008-2011 Results Smelt were captured at most fyke stations during the spring spawning runs of 2008-2011. The annual catches ranged from few individual smelt in some rivers to several thousand in the larger smelt runs. The following sections and Because smelt migrate graphics describe major findings in the fyke net catch data that portray popula- from marine to fresh-tion trends across the species' distribution on the Gulf of Maine coast. water habitatsto spawn Seasonality. Because smelt migrate from marine to freshwater habitats to during the spring freshet, spawn during the spring freshet, they are affected by a range of environmental they are affected by a factors most related to temperature and precipitation. Understanding how an range of environmental unpredictable environment can influence the timing, location and strength of a factors most related to smelt run is valuable for managing smelt populations. Accordingly, characteris-temperatureand tics of the onset, peak, and overall duration of a smelt run can provide measures of population health. In this study, the onset and ending of the spawning run precipitation, were based on the average date of first and last capture, respectively. Spawn-ing run peak was determined based on the average date of maximum catch. In several cases, the onset and the ending of the spawning run were inconclusive and had to be estimated using best professional judgment. Run duration was determined based on the average yearly duration of the run from 2008-2011.

Inter-system variability was noted in the timing of the spawning run (Figure 1.3.8). Within most systems in Massachusetts and New Hampshire, the spawning run had begun by mid-March. Within several Maine systems, how-ever, the spawning run was delayed and did not start until late-April. Similar patterns were observed in the peak and ending, with Massachusetts and New Hampshire systems having earlier peaks and earlier ending dates than those in Maine. Differences in run timing among states are presumably attributable to regional differences in climate, with cooler, more northerly systems displaying a delayed spawning run.

Run duration also varied with location. The longest run durations were observed for the Fore and Jones rivers, Massachusetts, and Tannery Brook, Maine. In these systems, average run duration appeared to exceed 70 days.

Conversely, the shortest runs were observed to occur in the North, Weweantic, and Saugus rivers, Massachusetts, where average run duration did not exceed 40 days. The causes for the differences in run duration are unknown, par-ticularly because previous studies have demonstrated shorter run durations in northern latitudes, with runs in individual tributaries often lasting less than two weeks in New Brunswick (McKenzie 1964) and Quebec (Pouliot 2002).

In the case of the U. S. Gulf of Maine surveys, population abundance and year ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 25

class strength may be influential, however the causal factors are not currently understood.

Catch Per Unit Effort (CPUE). The number of fish captured per a given amount of sampling, known as catch per unit effort (CPUE), is a measure used by fishery scientists to assess the relative abundance of a fish population, under the assumption that higher catches for a given amount of sampling effort (e.g., time, gear, habitat area, samplers) represents greater abundance. For the fyke net survey, the number of smelt caught per haul was used as a measure of CPUE. Yearly measures of CPUE were based on the geometric mean of weekly average CPUE.

The results of this study demonstrated that CPUE varied widely among rivers and years. For the entire region, the two highest overall CPUE were found in Maine (Deer Meadow Brook = 58.07, Schoppee Brook = 37.83),

while the two lowest were found in Massachusetts (Westport River = 1.01, North River = 1.37). There was an overall trend of higher CPUE in Maine compared to New Hampshire and Massachusetts - out of the 17 index sites, four out of the top five highest CPUE were found in Maine (Table 1.3.3).

Considering abundance by state, in Massachusetts, the Fore River had the highest overall CPUE (20.42), while the Westport River had the lowest (1.01).

In New Hampshire, the highest overall value was found at the Oyster River (5.62), while the lowest was at the Winnicut River (1.64). In Maine, the highest was found at Deer Meadow Brook (58.07), and the lowest at Long Creek (11.39, Table 1.3.3).

Yearly CPUE peaked in five of eight Massachusetts rivers in 2008, suggest-ing that in these systems, the largest smelt runs were observed at the beginning of the study (Table 1.3.3, Figure A. 1.1). In New Hampshire, the highest annual CPUE for all rivers was seen in 2011 (Table 1.3.3, Figure A. 1.2). In southern and midcoast Maine (Long Creek, Mast Landing, and Deer Meadow Brook),

the highest annual CPUE was seen in 2008 or 2009, while in eastern Maine (Tannery, Schoppee, and East Bay brooks), the highest annual geometric mean values were seen in 2010 (Table 1.3.3, Figure A. 1.3). It should be noted that when CPUE is calculated as simply the number of smelt per haul, the highest CPUE for East Bay Brook occurred in 2008 (Figure A. 1.3).

At this time, high levels of variability in CPUE and the limited duration of the study prohibit a statistical analysis of trends in relative abundance. How-ever, the CPUE data from 2008-2011 for some stations should be valuable as a reference point for future comparisons.

Length and Age Composition. Length and age information yields important insights into the health of a fish population. As a general rule, the presence of a variety of age classes is indicative of a healthy population. Further, populations containing older and larger individuals, which have a relatively high reproductive potential, are considered healthier than those containing only younger, smaller individuals. Smelt are fast growing fish that mature at small size and become fully recruited to the spawning stock at age-2 in the study area.

We measured total length of captured smelt to the nearest millimeter (mm).

Smelt ages were determined from scale samples.

The age class composition of the runs varied between sites, but displayed 26 - ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

geographical patterns. We found that runs in the southern portion of the Gulf of Maine (represented by the Fore River, Massachusetts, and Mast Landing, Maine) displayed two dominant age modes: one comprised mainly of age-I smelt and second mode comprised of mainly age-2 smelt (Figure 1.3.9 and 1.3.10). Age-1 smelt were common in Massachusetts and, in some years, were the dominant age class; yet this age class was present at much lower frequencies in spawning runs in the northern range of the study area (Table 1.3.4, Figures 1.3.9-1.3.14). In the mid-portion of the region (represented by Deer Meadow and Tannery brooks, Maine), age- I fish were encountered infrequently - the runs instead were dominated by age-2 fish, and the frequency of age-3 individ-uals was much higher than seen in more southern runs. Older ages (4-5) were also seen in these runs at higher rates than at all other runs, and these were the only sites to have age-6 fish represented in the runs (Table 1.3.4, Figures 1.3.11 and 1.3.12). In the northeastern portion of the Gulf of Maine (represented by Schoppee and East Bay brooks), runs were composed primarily of age-2 fish, with few to no age-I fish observed. Age-3 fish were observed, but at a lower frequency than the mid-portion of the region. The occurrence of older ages (4-5) was higher than the southern runs, but not as high as the mid-portion (Table 1.3.4, Figures 1.3.13 and 1.3.14). The fact that fish at age- 4 or older were unusual in Massachusetts, but relatively common in Maine samples, suggests higher levels of mortality in southerly systems.

Length at age also varied between sites, but again showed a geographic pattern. Because large sample sizes of age-2 males were present in each run, it is informative to compare the average lengths between sites using this category.

The largest length at age was observed in the southern portion of the region (Fore River avg. age-2 male = 184 mm, Mast Landing = 178 mm, Table 1.3.4),

indicating a faster growth rate at lower latitudes. Though the Oyster River geographically lies between these sites, age-2 males were comparatively smaller than the other southern sites (162 mm). This smaller age-at-length compared to surrounding sites may be evidence of a stressed population in the Oyster River, although further evidence would be needed to substantiate this idea.

Comparisons between previous studies show that length-at-age is observed to decline moving northward (Table 1.3.1). We observed a similar trend, how-ever the smallest length-at-age was observed in the mid-portion of the region (Deer Meadow Brook avg. age-2 male = 157 mm, Tannery Brook = 142 mm, Table 1.3.4). Sites at the most northeastern portion of the Gulf of Maine had larger age-2 males than in this mid-portion, but smaller than the southern sites (Schoppee Brook = 163 mm, East Bay Brook = 166 mm, Table 1.3.4). This pattern in age-at-length, as well as the pattern in run compositions discussed above, is coincident with the genetic stock structure of rainbow smelt reported by Kovach et al. (in press) and discussed in section 1.1 - Basic Biology, which found that the fish from Tannery Brook had a genetically differentiated signal that was also seen in fish from Deer Meadow Brook, but not in any other sites.

Because it was not possible to develop age-at-length keys for all sites due to low sample numbers at some sites, median length (calculated from all fish at a site) and length distributions are useful in understanding region-wide trends. Median lengths were lowest for males in the Massachusetts sites, and for females in the New Hampshire sites, and were generally higher for Maine sites (Table 1.3.5, Figure 1.3.15). The driving factor behind these patterns ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 27

seems to be the age composition of each of these runs rather than the length at age - runs in the southern portion of region are composed of a large proportion of age-1 fish, while runs in the mid- and northeastern portion have a higher proportion of age 3+ fish (Table 1.3.4). While not all fish were aged, modes corresponding to specific ages can help in affirming this idea. Length frequency figures for all sites with enough samples to produce relevant figures are included in the Appendix (Figures A. 1.4 - A. 1.14).

Sex Ratio. Although spawning runs of most anadromous fishes are male biased, those displaying a substantially higher proportion of males may be indicative of a stressed population. Because the limiting factor for population growth is often the abundance of females, populations dominated by males may be less robust than those containing a higher proportion of females. In this study, sex ratio was determined based on the ratio of the aggregate 2008-2011 catch of males to the catch of females.

The results of the fyke net survey demonstrated that each system contained a smelt population that was male biased. Overall, this survey observed an average sex ratio of 4:1. Of the systems sampled, the most heavily male biased were the Parker River, MA, and the Squamscott and Oyster rivers, NH, which all displayed a male to female ratio of greater than 8:1. The lowest male to female ratios were found in three systems in Maine: Tannery Brook, Schoppee Brook, and the East Bay River. In each of these systems, the sex ratio was less than 2: 1. We acknowledge that these sex ratios are biased themselves due to the behavior of male smelt spending more time on the spawning grounds than females (Murawski et al. 1980).

Mortality. Limited work has been done on population metrics for anadro-mous rainbow smelt throughout their range, but a few studies have calculated population mortality and survival rates based on age structure (Murawski and Cole 1978, Pouliot 2002). Survival and mortality analyses have potential biases that may limit their accuracy. Few age cohorts are available for the assessment:

the age-I cohort is excluded from mortality estimates because they are partially recruited to the spawning run, and age- 4 smelt are presently uncommon. Sec-ondly, the sampling method cannot distinguish the occurrence of repeat spawn-ing movements of individual smelt; this behavior could bias measurements of mortality and survival. Under the assumption that these biases were consistent among studies, we calculated mortality and survival estimates for sites that had sufficient age data for 2008-2011 and compared them to previous studies.

Within the study area, survival rates (S) and instantaneous total mortal-ity (Z) were calculated using the Chapman and Robson equation (Chapman and Robson 1960) at five stations in Maine and one each in Massachusetts and New Hampshire. However, the presence of some small sample sizes, few years of observations and the above discussed biases limit the reporting of these data to a relative comparison across the region and to past studies. Tannery Brook, Maine, had the highest average survival for 2008-2011 at S = 0.33, followed by S = 0.26 for 2009-2011 at Deer Meadow Brook in Maine. For sites that had at least three years of data, the Fore River, Massachusetts, had the lowest average survival at S = 0.17. The range of these spawning population survival estimates places the higher values in the present study among the highest reported by previous studies in the U.S. (Murawski and Cole 1978, Lawton et al. 1990) and 28 ° ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Canadian Provinces (McKenzie 1964, Pouliot 2002), and the sites at the lower range are the lowest survival values reported for anadromous rainbow smelt.

Study Area Summary Massachusetts. Of the eight fyke net stations monitored in Massachusetts, six caught enough smelt to allow summary comments on run demographics, but only the Fore River had a sufficient sample size to generate age composition data each year. The age and length data in Massachusetts suggest the pres-ence of a truncated age distribution, a sign of stressed populations due to high mortality and potentially poor recruitment. Male smelt in Massachusetts have similar median lengths compared to male smelt in New Hampshire and Maine.

However, female smelt in Massachusetts had higher median length than the other states; a statistic driven by larger age-2 to age- 4 females. Massachusetts stations are dominated by length modes that indicate age-I and age- 2 smelt, with very low presence of smelt older than age- 4 . The proportion of age-1 smelt in Parker River and Jones River spawning runs markedly exceeds that The age and length data found in previous studies. Changes in the contribution of age-1 smelt to the in Massachusettssuggest spawning run between previous studies and the present study, and the higher the presence of a trun-proportion of these small smelt in Massachusetts compared to New Hampshire cated age distribution, and Maine raises interesting questions on the significance of these apparent a sign of stressed differences. Smelt at the southern stations may experience faster growth in their first year and are reaching a body size that supports maturity sooner than populationsdue to high northern runs. mortality and potentially New Hampshire. The presence of mature smelt was documented in fyke poor recruitment.

catches in the Bellamy, Salmon Falls, Lamprey, Squamscott, Winnicut and Oyster rivers during 2008, and the standardized fyke net sampling protocol was followed in the Squamscott and Winnicut rivers from 2008-2011, and in the Oyster River from 2010-2011. Sufficient age samples were collected at the Oyster and Squamscott rivers in 2011 to prepare length frequency and age-length graphs. Two length modes are apparent in both rivers composed of age-I and age-2 smelt. However, more overlap is seen in these modes than is found in Massachusetts smelt age-length data. Few smelt reached age- 4 in New Hampshire rivers. For each available age key, age-4 comprised less than 2% of the annual age sample. Growth rates appear to be slower within New Hamp-shire runs, as age-3 smelt occur at smaller lengths than seen in Massachusetts and no age-2 smelt larger than 19 cm have been sampled.

Maine. All six Maine fyke net stations produced sample sizes large enough to summarize information on smelt run status. Median smelt length for the Maine stations was slightly larger than at the other states because these runs had a lower proportion of age-1 smelt, but higher proportion of age 3+ smelt; however, average length at age was smaller, indicating a slower growth rate compared to sites further south. The Maine smelt runs also averaged higher CPUE rates and showed more balanced age distributions and sex ratios than seen in southern runs. These patterns were most evident in catch data from the easternmost Maine stations. All these observations indicate relatively healthier smelt runs in Maine than in Massachusetts and New Hampshire. The age composition of smelt in Maine's spawning runs contributes to less separation between length modes and an extended age-2+ mode. These features could reflect interesting potential differences in growth rates, maturation, and survival ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 29

in Maine than at the southern runs.

Conclusions About Regional Fyke Net Sampling A common goal in fisheries management is to base decisions on a long-Table 1.3.1. Mean length at age term stock assessment that generates defensible biological benchmarks on the and proportion at age of anad- health of the fish stock. The present study does not achieve this goal, but it romous rainbowsmelt sampled starts the process of providing information on spawning run CPUE, temporal during spawn/ng runs In earlier characteristics, and size and age composition of rainbow smelt in three states.

studies in the study area and CanadianMaritimeProvinces. The sampling period from 2008-2011 is too brief for conclusions on Aft length data were converted to population trends. However, such baseline information is vital for all fish stock total length.

assessments. The task of assessing the status of rainbow smelt in the Gulf of Mean Length at Age Location Region Citation Year Sex N Age-1 Age-Z Age-3 Age-4 Age-S Age-6 Miramichi River NB McKenzie (1958) 1949-53 M NA 157 178 196 211 Miramichi River NB F NA 162 186 212 238 Fouquette River Quebec Pouliot (2002) 1991-96 M NA 133 166 198 215 227 Fouquette River Quebec F NA 135 173 213 237 245 Great Bay NH Warfel et al. (1943) 1942 both 287 86 145 171 220 264 Penobscott River ME Squiers et al (1976) 1974-75 both 260 165 196 226 264 Kennebec River ME Flagg (1984) 1980-82 M 1012 174 202 221 229 Kennebec River ME F 680 180 215 239 249 Parker River MA Murawski 1974-75 M 2097 141 188 208 236 242 Parker River MA and Cole (1978) F 584 140 197 219 245 249 Jones River MA Lawton et al. (1990) 1979-81 M 31394 132 184 208 221 242 Jones River MA F 5009 130 190 222 244 254 Proportion (%) at Age Location Region Citation Year Sex N Age-i Age-2 Age-3 Age-4 Age-5 Age-6 Miramichi River NB McKenzie (1964) 1949-53 both NA 66,2 29.3 4.1 0.4 Great Bay NH Warfel et al. (1943) 1942 both 287 3.5 65.9 29.6 1.0 0 Kennebec River ME Flagg (1984) 1979-82 both 1700 59.9 33.0 5.5 0.5 Penobscott River ME Squiers et al (1976) 1974 both 133 42.1 39.1 17.3 1.5 Penobscott River ME 1975 both 127 17.3 67.7 14.2 0.8 Parker River MA Murawski 1974 M 343 38.0 42.5 15.9 3.2 0.4 Parker River MA and Cole (1978) 1974 F 50 15.7 50.5 20.8 10.8 2.2 Parker River MA 1975 M 113 9.9 81.2 7.9 0.8 0.2 Parker River MA 1975 F 40 3.9 76.6 16.4 2.4 0.7 Jones River MA Lawton et al. (1990) 1979 M 364 15.0 64.6 19.7 0.7 <0.1 Jones River MA 1979 F 235 15.1 66.7 16.7 1.0 0.5 Jones River MA 1980 M 428 0.2 88.4 11.1 0.3 0 Jones River MA 1980 F 353 0 86.0 12.8 1.2 0 Jones River MA 1981 M 250 2.9 55.7 37.9 3.5 0 Jones River MA 1981 F 160 0.4 36.0 48.7 14 0.9 Notes

1. Lawton et al. (1990) and Murawski and Cole (1978) age composition is based on age key proportions assigned to total length sample.

2. Murawski and Cole (1978) mean length combines 1975 winter creel survey with 1974 and 1975 spawning run data.

3. McKenzie (195S and 1964) length data are converted to TL from SL Age-6 smelt were caught in most years at low frequency (<0.1%).

4. Pouliet (2002) fork length data were converted to total length using the conversion, TL= (FL-0.5584)/0.9142, from Chase et al. (20061.

5. Flagg (1983) and Squiers et al (1976) size and age data are both from winter smelt ice fishery, but included for comparative value.

30 , ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Maine is further complicated by the case of having distinct stock structure for some rivers, instead of a coast-wide stock complex. Finally, the assessment of anadromous fish is confounded by their migration between marine and freshwater habitats, where different factors influence their growth and survival.

Despite these challenges, the fyke net data from the present study show a gradi-ent of conditions with signs of stressed populations in southern Gulf of Maine and less evidence of stress moving north along the Maine coast, as evidenced by younger age distributions, smaller age-at-length, and lower CPUE rates.

Table 1.3.2. Currentstate of smelt Status N Percent spawning runs in Maine with re-Not historically listed, and currently do not support spawning 42 15%

Historical runs that do not currently support spawning spect to their historicalstatus.

35 13%

Currently support smaller runs than historically 95 34%

Currently support strong runs 53 19%

Historical runs that were not visited, current status Is unknown 54 19%

Table 1.3.3. Catch per unit effort Annual CPUE Overall Rim 5tQ 2008 2009 2010 2011 (CPUE) of rainbowsmelt at fyke Weweantlc R. MA 2.81 1.27 1.47 1.57 1.78 net spawning survey Index sites, Westport R. MA 1.00 1.00 1.00 1.02 1.01 by annual CPUEand overall CPUE Jones R. MA 9.13 5.58 7.56 5.13 6.85 Fore R. MA 33.55 10.41 22.00 15.70 20.42 for the entire samplingperiod, Saugus R. MA 6.30 1.19 1.07 2.49 2.76 2008-2011.

North R. MA 1.39 1.12 1.08 1.90 1.37 Crane R. MA 3.03 1.97 2.12 3.39 2.63 Parker R. MA 7.63 2.56 1.66 2.47 3.58 Squamscott R. NH 3.45 1.44 1.08 6.26 3.06 Winnicut R. NH 1.60 1.34 1.36 2.25 1.64 Oyster R. NH - - 5.45 5.79 5.62 Long Cr. ME - 18.69 5.56 9.93 11.39 Mast Landing ME 52.00 29.84 8.81 13.80 26.11 Deer Meadow Bk. ME 11.11 100.82 24.86 95.46 58.07 Tannery Bk. ME 15.28 28.26 41.87 14.03 24.86 Schoppee Bk. ME 38.42 37.25 37.83 East Bay R. ME 15.48 4.42 21.66 11.86 13.35 ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN o 31

Proportion (%) at Age Location Region Year Sex Length N Age N Age-1 Age-2 Age-3 Age-4 Age-S Age-6 East Bay Brook ME 2008 both 899 63 92.2 6.7 1.1 East Bay Brook ME 2009 both 236 68 0.8 62.3 33.9 3.0 East Bay Brook ME 2010 both 1387 261 2.0 80.7 13.7 3.6 East Bay Brook ME 2011 both 1211 268 72.0 26.7 1.2 0.1 Schoppee Brook ME 2010 both 2034 281 0.9 90.2 3.5 5.4 Schoppee Brook ME 2011 both 1831 245 2.2 90.7 7.1 Tannery Brook ME 2008 both 2001 74 60.0 34.2 5.8 Tannery Brook ME 2009 both 1778 72 3.9 78.6 7.9 6.4 3.2 Tannery Brook ME 2010 both 1892 344 2.5 49.6 45.4 1.4 1.0 0.1 Tannery Brook ME 2011 both 908 172 6.9 36.6 48.0 8.5 Deer Meadow ME 2008 both 179 85 5.0 77.1 17.9 Deer Meadow ME 2009 both 2016 135 0 90.2 5.7 3.4 0.7 Deer Meadow ME 2010 both 1366 320 2.8 26.0 64.7 5.0 1.5 Deer Meadow ME 2011 both 1946 108 1.5 83.6 6.9 6.7 0.9 Mast Landing ME 2008 both 1620 90 15.2 58.6 24.2 2.0 Mast Landing ME 2009 both 1106 128 0.6 85.6 13.9 2.9 Mast Landing ME 2010 both 355 268 75.5 8.7 13.8 1.7 0.3 Mast Landing ME 2011 both 1833 275 44.5 53.5 0.8 1.2 Oyster River NH 2010 both 421 185 65.8 29.0 4.5 0.7 Oyster River NH 2011 both 401 231 11.2 75.1 13.5 <0.1 Fore River MA 2008 both 1958 380 51.9 41.4 6.2 0.4 0.1 Fore River MA 2009 both 846 660 15.5 52.5 31.4 0.6 Fore River MA 2010 both 1441 493 89.6 7.9 2.4 0.1 <0.1 Fore River MA 2011 both 1241 486 48.3 48.7 2.6 0.4 <0.1 Mean Length at Age Location Region Year Sex N Age-1 Age-2 Age-3 Age-4 Age-5 Age-6 East Bay Brook ME 2008-11 M 322 145 166 197 215 241 East Bay Brook ME 2008-11 F 338 155 173 212 238 241 Schoppee Brook ME 2010-11 M 225 146 163 195 204 Schoppee Brook ME 2010-11 F 299 152 169 206 234 Tannery Brook ME 2008-11 M 339 135 142 166 183 190 Tannery Brook ME 2008-11 F 322 137 146 178 198 211 215 Deer Meadow ME 2008-11 M 397 138 157 185 209 220 226 Deer Meadow ME 2008-11 F 250 125 160 194 222 208 Mast Landing ME 2008-11 M 447 132 178 192 211 Mast Landing ME 2008-11 F 312 137 190 209 232 256 Oyster River NH 2008-11 M 344 117 162 179 209 Oyster River NH 2008-11 F 60 114 167 180 Fore River MA 2008-11 M 1113 141 184 202 215 Fore River MA 2008-11 F 507 142 194 217 249 251 266 Table L3.4. Mean length at age and proportionat age of anadro-mous rainbowsmelt sampled at fyke net stations for 2008-2011 for the presentstudy. Age keys were applied to length samples for proportion atage.

32 9 ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

MALE State River Code I Years N Mean SE Median Min Max MA Weweantie WW 4 188 151 2.01 145 104 238 MA Jones JR 4 1249 156 0.93 143 106 254 MA Fore FR 4 4396 166 0.43 157 109 241 MA Saugus SG 4 401 162 1.30 153 113 240 MA North NR 4 79 150 2.18 149 118 217 MA Crane CN 4 262 161 1.44 156 121 221 MA Parker PR 4 1217 167 0.98 156 86 255 NH Squamscott SQ 2 340 154 1.85 159 86 227 NH Oyster OY 2 344 149 1.74 156 88 22S ME Long Creek LC 4 1191 169 0.41 168 110 228 ME Mast Landing ML 4 3099 163 0A0 169 105 227 ME Deer Meadow DM 4 4367 166 0.33 163 83 241 ME Tannery Brook TB 4 4214 152 0.27 152 104 223 ME Schoppee SB 2 2303 164 0.24 163 125 222 ME East Bay EB I Total 4 2368 172 0.31 169 136 250 ME ME East Bay Schoppee SB EB 2 4 26018 2303 2368 172 164 0.24 0.31 125 136 250 222 I Total 26018 FEMALE State River Code Sex Ratio N Mean SE Median Min Max MA Weweantic WW 3.4 55 149 4.29 139 107 225 MA Jones JR 2.5 492 160 1.69 144 100 258 MA Fore FR 4.0 1090 168 1.06 154 111 270 Table 1.3.5. Rainbow MA Saugus SG 7.7 52 172 5.01 157 129 248 smelt length data from MA North NR 3.4 23 154 4.71 153 113 214 catches at fyke net MA Crane CN 2.8 94 169 3.31 162 114 257 stations, 2008-2011.

MA Parker PR 9.5 128 194 3.18 204 112 272 A few stations were NH Squamscott SQ 3.7 93 135 3.86 118 86 239 NH Oyster OY 5.7 60 151 4.80 166 88 224 excluded because of low ME Long Creek LC 3.3 360 178 0.99 176 118 251 sample sizes or poten-ME Mast Landing ML 2.7 1136 177 0.86 180 93 263 tially biased samples ME Deer Meadow DM 3.6 1209 165 0.71 159 83 258 from few hauls. Smelt ME Tannery Brook TB 1.8 2366 157 0.46 154 108 236 of unknown sex were ME Schoppee SB 1.5 1564 174 0.53 170 129 256 excluded from this table.

ME East Bay EB 1.7 1389 183 0.59 176 122 263 Sex ratio Is the ratio of Total 10111 males to females.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN e 33

4*

New Hampsshire Creel Survey 10 8

CL

-6

.C1

• 11

-E 4 I

ill 11111 WE 1Ii1 LU 2 Figure 1.3.1. New HampshireFish and Game Creel Survey catch per unit effort (CPUE) calculatedas U number of fish caught per hour of fishing 1978-2011 0.9 Maine Creel Survey 0.8 0.7 0.6 0.5 0.4 0.3 Figure1.3.2. Catch per unit effort 0.2 (CPUE) as smelt caught per line-0.1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of fishing observed during the rainbow smelt winter creel 0 survey In Maine during1979-1982 and 2009-2011.

200 W. . . . . . . . . .

180 160 140 1120 180 .MEfte4 00 Figure 1.3.3. Inshore Trawl Survey 40 average annualsmelt catches (in numbers of fish) from MA DMF 20 state survey (1978-2011)and ME DMR/NHF&G combined state survey (2000-2012).

34 a ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN I

350 300 250

- 200

~150 S100 50 Figure1.3.4. Average annual catch of rainbowsmelt YOY In ME 0 DMR Juvenile Abundance Survey W

In the lower Kennebec River.

Othersites are excluded due to low catches.

800 1 New Hampshire Juwenile Abundance Surey 700 6o00 S500 3 400

<~ 300 S200 Figure 1.3.5. Average annual catch of rainbowsmelt YOY in NHF&G Juvenile Abundance Sur-

__IL 100 vey. The 11 locations within the 0 Piscataqua River and Little/Great Bay were grouped Into two cohorts 1 Little and Great Bays to show annualtrends. The Hamp-NPiscataqua Riwr ton/Seabrook area was excluded due to low catches.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 35 I

Figure 1.3.6. Currentstatus of smelt spawningruns In Maine and historicalsites where the current status remains unknown.

36

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN I

SchW Bwook Dee Meadow Brook Oys erRi PaFwvrkolM i Janelan River

-A Rb~er Figure 1.3.7. Fyke net monitoring stations In Massachusetts,New 110 Hampshire, and Maine I,,, I OM 2008-2011.

Figure 1.3.8. Smelt runs progress in a bell-curve shape over the sea-son, where the beginningof the run sees few smelt and the num-ber steadily Increases to a peak In the run (red portion of the barsin the figure), after which point the run steadily declines (blue por-tion of the bars). These patterns are shown here, along with the average beginningand end date of each run 2008-201:. Stations are arrangedfrom south to north startingat the x-axis origin.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN a 37

S00 o0te-I 8Age,2 A4g-3 mAlel" 54 5e-S+ 048-1 A W40-2 IN4o-3 mAge4 IN48-5.

2009 ilS~ I1 N 2008 400 400 AgeN.380 AgFN: 458

'o0 3000 L/FN. 846 Fle 530/ No.

200 0

2 9 10 13 12 13 14 15 16 TOWl 17 18 Legt ml 19 2n 21 22 23 24 IS 26 27 28 100-8 9 10 11

--1F1--IIII 12 13 14 15 16 17 18 Total Length(-I8l 19 20 21 22 23 24 25 26 27 28 SOD 2 14g-I 56I42 l5483 maVe-4 mAge-5+

I 2011 0440-3 m504- 5A4e.3 5440-4 5440ta-S.

44eN - 486 L/FN = 1241 300 1/FN 1441 2o 00]

200 I00 9 10 11 l12 l 13 14 15 16 7 35 TotaLength(m) 19 20 21 2 23 24 2S 26 27 28 0o 100 8 9 00 11 H]

12 13 rv.IIll 14 15 16 17 18 Total le ngth (iM) 19 20 21 g

22 23 24 25 26 27 28 Figure 1.3.9. Age composition of Fore River, MA, fyke net catch In2008-2011. Both genders were combined with number of age samples reported as "Age N" and length frequency sample size reported as "i/F Ný Son .,40-I m44*2 m403 53404 O4-4 5 11e-2 5411-3 5Ape-4 SOD I

800 2008 4001 2009 L/FN. 1705 NO 200]

0 9 9 10 11 12 13 14 15 16 17 18 III.-19 20 01 22 23 24 25 26 27 28 8 9 10 12 12 13 14 IS 16 17 18 19 20 22 22 23 24 25 26 27 28 Totallength (s) Tota Leng4t h (1) 250 04A40. 5443-2 ii4e-3 mA504 44m.-S 5" 0440-1 54402 w440-3 a40. 53401.A-2011 201)A 2010 A40N 275 AS, N - 268 NO0 L/FN.1833 Iami.

L/FN - 3S7 p8e No.

100 200 So, 8 5 Ri R - 0..-

10 11 12 13 14 IS 16 17 18 15 20 2121 23 24 25 26 27 2S 0

8 9 10 11 12 13 14 16 16 17 18 Total LengthI-m) 19 20 21 22 11 24 IS 26 Z7 22(

Total e Length1 Bot )

Figure 1.3.10. Age composition of Mast Landing, ME, ilike net catch In 2008-2011. Both genders were combined with number of age samples reported as "Age N" and length frequency sample size reported as "L/F N"ý 38 toANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

0Ae-I S*Ap2 SIAV3 NAge4 OWL0- *60o2 3AP-3 MAI- 3440-50 2008 714 A1eN 135 AgeN: 175 400 L/FN 2016 so No1 ilium 200 25 0

8 9 10 11 12 13 14 IS 16 17 10 19 20 21 22 23 24 25 26 27 28 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Total Length(mn) Totl. Length 1Cm)

GM0 o0g4-1 MAg12 3mae-3 mIAe-4 3041-5 DA02-4 3A0e-2 34A4-3 3a4 4 44S 2010 2011 A0 N" 320 200 200 L/FN5-1367 200 8 9 10 01 12 13 14 15 16 17 IS Total Legt (-)

1 20 221 22 23 24 2S 26 27 20 10 21 12 14 15 53 16 17 18 19 20 22 22 23 24 25 2S 27 28 Tot."140 I-0)

Figure 1.3.11. Age composition of Deer Meadow Brook, ME, fyke net catch In 2008-2011. Both genders were combined with number of age samples reported as "Age N" and length frequency sample size reportedas "L/F N".

Figure1.3.12. Age composition of Tannery Brook, ME, fyke net catch in 2008-2011. Both genders were combined with number of age samples reported as 'Age N" and length frequency sample size reportedas "L/F N".

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN a 39 I

2011 2010 AgeN 281 0e0N= 245 L/FN 2036 L/FN= 1831 No, No.

400 400 200 200

- IO I . . .

8 9 11 12 13 14 15 16 20 i8 17 19 20 21 22 23 24 25 26 27 28 a1 8 9 10 12 13 14 15 16 17 18 19 20 22 22 23 24 25 26 27 28 Total Length (nmj Total Length (-)

Figure 1.3.13. Age composition of Schoppee Brook, ME, fyke net catch In 2010-2011. Both gen-ders were combined with number of age samples reported as "Age N" and length frequency sample size reportedas "L/F N'.

200 0A.I iB4.-Z R88.8 UA,04 *E0.-S 000e-2 m008-2 §Age-3 30A0-4 2008 2009 40eN.63 40eN068 V, -.92.

/FN9 237 200 500

.1,.'..-

08 9 20 21 12 13 14 15 26 17 IS 19 20 21 22 23 24 25 26 27 28 8 9 10 11 12 13 14 15 16 17 18 19 20 22 22 D 24 25 26 27 28 Tow Lengthio) Total Length (rm) 500 C0e8- 11A9.2 384.3 IA0-4 3Age-5 2010 2011 400 40 0 AgeN 268 AV N 261 1SF N

• 1211 1 9N.1308 300-No, No.

200-100-0 8 9 10 11 12 13 14 IS 16 17 19 19 20 21 22 23 24 25 26 27 28 6 9 10 11 12 1 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 TOWaLength.(M)j Tool Length c-)

Figure1.3.14. Age composition of East Bay Brook, ME, fyke net catch In 2008-2011. Both gen-ders were combined with number of age samples reported as "Age N" and length frequency sample size reported as "L/F N.

40

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

MALE 250 200-I I 10 1 1

• 0 0 0

• i 100-50 I I I I I I WW JR FR SG CN PR SQ OY LC ML DM TB SB EB River FEMALE Figure 1.3.15. Median total length 250 of smelt caught at 14 fyke net stations In the study area,2008-2011 The top of the box plots 200- Is the 75th percentile and the bottom Is the 25th percentile.The line in the box Is the median and E 10 150 I I I i1i jj i the errorbars mark the 10th and 90th percentiles.The stations are arrangedon the x-axis from the southernmost MA station to the 100- northernmost ME station. Station medians for females and males were found to be significantly different with Kruskal-Wallis test,

-so LC ML DM TB SB EB KW = 1324.94, df - 13, p <0.001; WWV JR FR SG CN PR SQ OY and KW - 2000.77, df - 13, p River <0.001,respectively.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 41

2 - THREATS TO RAINBOW SMELT POPULATIONS IN THE GULF OF MAINE Rainbow smelt encounter a variety of potential threats during their fresh-water and marine life stages. Dams, overfishing, and pollution have typically been considered the most important factors affecting diadromous fish, includ-ing rainbow smelt (Saunders et al. 2006, Limburg and Waldman 2009). While these factors may have played major roles in the declines of rainbow smelt, other factors may also be responsible for recent declines. Changes in trophic interactions, community shifts, watershed land use, and climate-driven Dams, overfishing, and environmental conditions may all need to be considered when evaluating factors that affect rainbow smelt populations.

pollution have typically been consideredthe most important factors 2.1 - THREATS TO SPAWNING HABITAT CONDITIONS AND SPAWNING affecting diadromous SUCCESS fish, including rainbow smelt. Spawning Site Characteristics Across their distribution range, smelt spawning runs are variable in regard to habitat use, spawning substrate, spawning period, and water temperature range (Bigelow and Schroeder 1953, Hurlbert 1974, Kendall 1926, Pettigrew 1997, Rupp 1959). Investigations of Massachusetts smelt runs have found that spawning begins between late February and mid-March when water tempera-tures reach 4-6 'C and concludes in May (Chase 1990, 2006; Chase and Childs 2001; Crestin 1973; Lawton et al. 1990). In New Hampshire, spring runs begin in early to mid-March when the water temperatures reach 3-6 'C and conclude in May (NHF&G, current study). In Maine, the timing of the run varies geographically, beginning in late March in waters west of the Kennebec River, in mid-April in waters between the Kennebec River and the Penobscot River, in late April to early May in the Penobscot River and advancing to mid-May in most waters in downeast Maine. Water temperature at the beginning of runs varies from 1.5-9 'C, and most runs in Maine last four to five weeks (ME DMR, current study). There is also some evidence that rainbow smelt may spawn in the main stem of large rivers in Maine earlier than runs begin in smaller streams close to these rivers. In rivers such as the Kennebec, Penobscot, Union, and Pleasant, spawning may occur under the ice or directly following ice-out in mid-March to early April (ME DMR, current study).

The best documentation of the physical characteristics of smelt spawning habitats in the Gulf of Maine is provided by a detailed assessment of Massachusetts rivers that was conducted between 1988 and 1995 (Chase 2006).

This study identified both stream attributes and water chemistry conditions that were suitable for smelt spawning. Chase (2006) documented and mapped smelt spawning habitat at 45 locations in 30 rivers on the Gulf of Maine coast of Massachusetts. Rainbow smelt egg deposition was documented to take place over stream sections ranging from 16 m to 1,111 m in length, with an average 42 ° ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

of 261 m. In most cases, the downstream limit of egg deposition occurred near the interface of salt and fresh water, while the upstream limits were typically delimited by physical impediments that prevented further passage. When passage allowed, smelt would continue spawning in freshwater riffles beyond tidal influence. The average patch size of substrate where smelt eggs were observed was 2,336 M 2 , with a range of 16 m 2 to 13,989 m2 .

Smelt were found to spawn in shallow riffles where water velocity increased in stream channels. Within the streams where smelt eggs were found, channel width averaged 6.8 m. Depth transects conducted in 16 of these streams found that the average depth of spawning riffles was 0.28 m, and the range of average depths was 0.1 - 0.5 m under baseflow conditions. However, smelt eggs were found in depths up to 1.5 m in three surveyed rivers. The average water veloc-ity at the riffle transects was 0.39 m/s, with a range of 0.1 to 0.9 m/s. These measurements and observations of associated egg deposition led Chase (2006) to hypothesize that 0.5 - 0.8 m/s was an optimal range for adult attraction and egg survival.

Observations in smelt spawning rivers in Massachusetts led Chase (2006) to conclude that the ideal channel configuration for spawning habitat may be-gin with a deep channel estuary where the salt wedge rises to meet a moderate gradient riffle at the tidal interface and follows into the freshwater zone with ample vegetative buffer and canopy and an extended pool-riffle complex that spreads out egg deposition and provides resting pools. However, this scenario was not common in Massachusetts spawning rivers, and likely is not in many other rivers and streams in the Gulf of Maine. Many of the spawning streams and rivers were altered by: (1) a range of passage obstructions (undersized cul-verts, dams, etc.) that limited or completely blocked the smelts' ability to reach their spawning grounds, (2) channelization and flow alterations that changed water velocity and substrate conditions, and (3) removal of riparian vegeta-tion, leading to increased amounts of polluted runoff flowing directly into the stream, as well as reduced canopy cover leading to increased water temperature.

These three categories represent major threats to spawning habitat and to smelt spawning success, and they are described further in the following sections. In many cases, these threats are present simultaneously in more developed water-sheds, compounding the threats to successful smelt spawning.

Obstructions Damns Industrial development depended on rivers for power, and over 500 dams remain on rivers in Maine, New Hampshire, and Massachusetts that may have a large impact on diadromous species (Martin and Aspe 2011). Dams block access to spawning habitats for many anadromous species, but their effect on rainbow smelt is particularly acute. The small body size of rainbow smelt makes them unable to jump to heights necessary to migrate through fish ladders, which pass other diadromous fish over dams. In Maine, at least 13 out of 275 (5%) historical and current spawning sites are either reduced in area or the spawning habitat is blocked by coastal dams (Abbott, USFWS, pers.

comm., 2012). In New Hampshire, although smelt spawning occurs in most of the coastal rivers, head-of-tide dams exist on all of these rivers (with the ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 43

exception of the Winnicut River), reducing habitat and forcing smelt to spawn within areas subject to tidal influence. Although the exact number has not been documented, the same situation exists in Massachusetts, where head-of-tide dams limit spawning habitat.

Road crossings The majority of smelt spawning streams in the Gulf of Maine are small coastal streams that are not dammed. More frequently, barriers are road-stream crossings. Undersized, improperly installed, or poorly maintained culverts at road-stream crossings can severely impair smelt migration. This can occur when culverts have become perched, where the downstream side stream height is well below the culvert height, or when culverts are undersized to such an extent that they create velocity barriers or reduce freshwater flow to levels that impede environmental cues for smelt. Reducing stream habitat fragmentation is critical for increasing access to smelt spawning habitat. In Maine, there is an ongo-ing effort to ground-survey all stream barriers. At the time of this report, 35%

of the state has been surveyed. Of the 88 smelt historical or current spawning sites falling within this surveyed portion, 34 (39%) sites have potential barri-ers to passage. Extending the scope to the entire state, 127 historical or cur-rent spawning sites out of a total of 275 are crossed by roads at least once, and multiple times in many cases. While some of these crossings may have adequate passage, it is estimated that two-thirds of these crossings are undersized and may present passage problems for smelt (A. Abbott, USFWS, pers. comm..,

2012). The frequency of the problem is magnified in Massachusetts where only I of 45 mapped smelt spawning habitats were unaltered by road crossings or impediments (Chase 2006).

ChannelizationandFlow Disruptions Discharge and Velocity In Massachusetts, New Hampshire, and Maine, most smelt runs occur in small coastal rivers or streams with low seasonal baseflows where spring stream discharge is sufficiently high to attract adults and support egg incubation. In the Northeast United States, early spring flows are typically enhanced by snow melt and precipitation, but discharge may decline progressively later in the season. In a survey of 45 spawning rivers in Massachusetts, aside from the Merrimack River, only nine had average spring discharges over I m3/s (35 cubic feet per second (cfs)), and only four exceeded a spring average of 10 m3/s (353 cfs) (Chase 2006).

During the current study, when USGS gauge stations were present, we re-corded river discharge weekly at our smelt spawning sampling sites. None of the survey stations in Maine were located on rivers with gauge stations; however, measurements were available for two New Hampshire sites and four Massa-chusetts sites (Table 2.1. 1). Over a two year period (2008-2009), we found an average discharge of 1.83 m3/s (65 cfs) across all sites, with most values (75%)

under 1.99 m3/s (70 cfs) (Table 2.1.2). Discharge varied significantly between the sites, and was directly correlated to watershed size (Spearman's rank correla-tion = 0.78).

Although high discharge is not a threat to smelt spawning, if it results in sharp increases in velocity it impairs smelts' ability to reach their 44 a ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

spawning grounds. In watersheds with large amounts of impervious surface and not managed for stormwater, infiltration of runoff is reduced and the smoother impervious surfaces allow water to run off the surface and into streams faster.

The combined result is a rapid increase in both volume and velocity (Cooper 1996, Klein 1979). Substantial variability in velocity may be found within a coastal stream depending on specific location (e.g. pool versus riffle), and tim-ing (precipitation events and tidal stage will affect daily velocities). However, as part of the current study we found that velocities at all spawning index sites fell within a fairly narrow range (0.32 m/s - 0.58 m/s) when measurements were taken within riffles when no tidal influence was present (Table 2.1.2). Velocity exceeded 0.79 m/s only 10% of the time, and generally the catch per unit effort of spawning adult smelt was lower during those high velocity events.

Conversely, low discharge may also threaten successful spawning. Suffi-cient freshwater flows are necessary for other anadromous species to cue their migrations and enable them to successfully locate their spawning site (Yako et al. 2002). Low discharge associated with urbanization may also lead to insuf-ficient water mixing, resulting in higher water temperatures, lower dissolved oxygen, increased sedimentation, and increased concentrations of pollutants and contaminants (Klein 1979). Reductions in baseflow can be caused by water withdrawals and impounding as well as increases in impervious surface (Klein 1979, Simmons and Reynolds 1982). In many cases, withdrawals during the spring months may be expected to remove a small proportion of available spring flows. However, concerns are growing in urban areas where human population growth has increased water demands. Furthermore, a gradual but measured loss in snow pack over the last century has led to a reduction of spring baseflow in coastal streams, a situation that could compound concerns over water withdrawals.

Substrate and Channel Stability Natural stream and river channels that are vegetated and dynamic can absorb the impacts of flooding by accommodating changes in discharge and water levels. However, in urbanized areas with extensive impervious surface or where streams have been channelized by fixed walls, the runoff from large rain events flows directly into streams, leading to increases in the frequency and severity of flooding. In turn, these events can cause channel erosion and alteration of the stream bed (Klein 1979). The timing of flood events can cause positive responses to smelt spawning substrata by scouring sediment and periphyton before spawning occurs or negative responses by scouring away large egg sets (Chase 2006). Booth and Reinelt (1993) report that pool and riffle habitat may be altered and channel stability may be degraded when impervious surface exceeds 10-15% of the watershed area These impacts can be mitigated by restoring riparian buffers along stream and river banks.

Watershedcharacteristics Watershed activities can have a substantial influence on many of the condi-tions identified above as potentially affecting rainbow smelt spawning habitat.

Land cover in a watershed affects habitat conditions and biological communi-ties in receiving waters in a variety of ways (Burcher et al. 2007, Allan 2004).

Urbanization and agricultural activities can contribute to erratic flow levels, Warmer water temperatures, channel alterations, sedimentation, chemical and ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 45

bacterial pollution, and nutrient loading (Wang et al. 200 la, Allan 2004).

In addition, barriers to spawning passage are more likely to exist due to road networks in more urbanized watersheds than in less developed areas. These watershed-associated factors can all influence the suitability of streams for rainbow smelt spawning.

Associations between watershed characteristics and spawning site use have been observed for other anadromous species. Limburg and Schmidt (1990) noted that spawning activity of anadromous fishes (mostly alewife) in tributar-ies to the Hudson estuary was inversely related to the proportion of urban land use in the surrounding watershed. In the Pacific Northwest, Pess et al. (2002) found that median densities of spawning coho salmon were 1.5-3.5 times higher in forest-dominated areas than in urban or agricultural areas. These examples indicate that there may be linkages between spawning success and watershed characteristics. While the causal factors have not been identified, Our analysis found that urbanization may influence in-stream habitat by increasing water velocities as-sociated with flood events, changing substrate, removing canopy cover and thus weak spawning runs ex-increasing water temperature, and other habitat changes.

isted in rivers surrounded In this study, we evaluated correlations between rainbow smelt catch per by urbanized watersheds, unit effort at the spawning index sites and land use in the adjacent watersheds while rivers draining at two spatial scales: (1) the full drainage basin and (2) the 210 meter buffer forested watersheds sup- immediately adjacent to the stream. Watersheds within which rainbow smelt ported strongersmelt spawning runs were sampled represented a wide variety of conditions (Table spawning populations. 2.1.1). A principal components and cluster analysis suggests that the smelt spawning watersheds can be classified into three distinct types: (1) urbanized, (2) forested, and (3) wetlands/agricultural (Figure 2.1.1). Correlations be-tween the aggregate mean CPUE of spawning rainbow smelt over 2008-2011 (standardized based on net coverage of the stream width) indicate that weak spawning runs exist in rivers surrounded by urbanized watersheds, while rivers draining forested watersheds support strong smelt spawning populations.

Interestingly, the negative association between development and CPUE was substantially stronger at the scale of the full drainage basin than when only the riparian buffer zone was considered (Table 2.1.3). This appears to be because many rivers within urbanized watersheds have extensive riparian wetlands in their buffer zones. The presence of these wetlands at the 21 0-m scale weakens the influence of urbanization on smelt spawning. Other land cover types and the number of downstream crossings, at either the scale of the watershed or riparian buffer zone, were not significantly correlated to the strength of rainbow smelt spawning populations.

46 e ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Fyke Net Location Hydrologic Information Watershed Intormation Average Average Channel Discharge Velocity Drainage River Latitude Longitude Town State Width (m) (m*ls) (mIs) Watershed (HUC 10) Area (kmn) Land Cover (11/20)

Westport River 41.6209 -71.0598 Westport MA 11.3 Buzzards Bay 26.5 Forest I Agriculture Weweantic River 41.7662 -70.7461 Wareham MA 35,7 Buzzards Bay 148.2 Forest I Agriculture Jones River 41.9960 -70,7233 Kingston MA 27.3 1.92 0.492 South Coastal Basin 69.3 Forest I Wetland Fore River 42.2225 -70.9732 Braintree MA 13.7 1.92 0.623 Boston Harbor 74.7 Development I Forest Saugus River 42.4680 .71.0077 Saugus MA 55.4 Boston Harbor 55.8 Development I Forest North River 42.5221 -70.9116 Salem MA 9.1 0.49 0.454 North Coastal Basin 12.6 Development I Forest Crane River 42.5966 -70.9364 Danvers MA 8.2 0.17 0.497 North Coastal Basin 14.0 Development I Forest Parker River 42.7505 .70.9282 Newbury MA 54.8 0.516 Plum Island Sound 66.0 Forest I Wetland Squamscott River 42.9824 -70.9461 Exeter NH 101.0 5.65 0.384 Exeter River 276.9 Forest I Wetland Winnicut River 43.0389 -70.8455 Greenland NH 36.6 1.05 0.3 Great Bay 45.5 Forest I Wetland Oyster River 43.1310 .70.1310 Durham NH 32.0 Great Bay 59.9 Forest I Development Long Creek 43.6332 3133 S. Portland ME 24.3 0.64 Fore River 17.5 Development I Forest Mast Landing 43.8587 -70.0842 Freeport ME 15.2 0.468 Casco Bay Basin 20.7 Forest / Wetland Deer Meadow Brook 44.0369 -69.5874 Newcastle ME 24.0 0.489 Sheepscot River 27.6 ForestlWetland Tannery Brook 44.5706 .68.7888 Bucksport ME 07.7 0.402 Penobscot River and Bay 13.2 Forest I Agriculture Schoppee Brook 44.6627 -67.5533 Jonesboro ME 16.0 0.583 Roques Bluffs Frontal Drainages 9.3 Forest / Wetland East Bay Brook 44.9547 -67.1041 Perry ME 21.9 0.217 Cobscook Bay 3.0 Forest I Wetland Table 2.1.1. Rainbow smelt spawning habitatstation loca-tions for water quality monitoring.

Drainageareasare GIS calcula-tions set from the location of fyke net placement.

Table 2.1.2. Dischargeand velocity Discharge (ms Velocity (m/s)

Minimum Value 0.04 0.050 measurements from spawning Lower Quantile (25%) 0.35 0.323 survey Index sites. Discharge Mean 1.83 0.478 measurements taken from Upper Quantile (75%) 1.99 0.579 USGS gauge stations upstream Maximum Value 12.81 :1.483 of spawningsites and velocity measurements taken by state biologists at the spawningsites (dischargen = 6, velocity n m 13) in active riffle areas.

Correlation with smelt spawning CPUE Land Cover Watershed Level Stream Buffer Zone (210m)

% developed -0.62 -0.48 Table 2.1.3. Spearman's rank

% developed open space (parks, golf courses) -0.47 -0.32 correlationbetween rainbowsmelt

% forest 0.60 0.60

% wetland spawning CPUE and land cover

-0.29 -0.28

% agriculture -0.06 0 at two spatial scales. Correlation number of downstream crossings -0.46 -0.46 coefficients In bold type indicate significanceat the p = 0.5 level ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 47

Figure2.1.1. Cluster analysis (Ward's method) of study water- _ __,

sheds based on dominant land uses (as indicated by the propor-tion of developed, developed open, forest, agriculture,and wetland areas) and watershed characteristics (i.e., population density, stream crossings,and 0 proportion of Impervious sur- 0 * ,_ ,

face). Station codes: NR - North E -, - _ _

River, LC = Long Creek, CR=

Crane River, FR = Fore River, SR

= Saugus River, WE = Weweantic River, WN = Winnicut River, SQ

= Squamscott River, JR = Jones 0 River, PR = ParkerRiver, EB X 0 W x W. ix W* a

_Z __ U~LL. W U) (L U4 ~ l East Bay Brook, OY = Oyster River, TB = Tannery Brook, SB = Schop- Developed Agriculture Forestedl pee Brook, DM = Deer Meadow Wetland Brook, ML = Mast Landing.

2.2 - THREATS TO EMBRYONIC DEVELOPMENT AND SURVIVAL Smelt deposit demersal (sinking), adhesive eggs at fast-flowing riffles, where they attach to the substrate or aquatic vegetation. The duration of egg incubation is related to water temperature (McKenzie 1964), and in the Gulf of Maine, eggs hatch 7-21 days after fertilization (Chase et al. 2008, McKenzie 1964). The success of this reproductive strategy depends on access from ma-rine waters, low predation, and suitable water and habitat quality for successful recruitment. In many watersheds, the tidal interface is the physical location favored for the development of commerce and community centers. This change in landscape can lead to hydrologic alterations, particularly in urban areas, leaving streams vulnerable to point and non-point source pollutants; nutrient enrichment; and reduced streamflow, shading and riparian buffer.

Changes in spawning habitat may be a major factor in the decline of smelt populations. However, up to this point, the degree to which water quality impairment may be impacting smelt populations in the Gulf of Maine has not been described. With this concern in mind, we developed monitoring pro-grams to assess baseline water and habitat conditions at smelt spawning habitat index sites spanning the entire Gulf of Maine and explored possible impacts on spawning success resulting from changing habitat conditions. This informa-tion is applied to support recommendations for conserving and restoring smelt populations and habitats.

Four indicators were measured to assess water quality at smelt spawning index sites: basic water chemistry, nutrient concentrations, periphyton growth and heavy metal concentrations. The sampling was guided by a Quality Assur-ance Program Plan (QAPP) for monitoring water and habitat quality at smelt spawning habitats in coastal rivers on the Gulf of Maine coast (Chase 2010).

The QAPP integrates smelt life history with existing state and federal water quality criteria, with the objective of developing a standardized process to classify the suitability of smelt spawning habitat. Beyond characterizing smelt 48 - ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

habitat, it is our hope these data will contribute to water quality and habitat restoration efforts at coastal rivers in New England.

Summary statistics were generated for water quality data by site and then compared to thresholds assembled from existing water quality criteria (Table 2.2.1). The U.S. Environmental Protection Agency (EPA) developed criteria for turbidity, total nitrogen (TN) and total phosphorus (TP) based on the 25th percentile of the distribution of observed values in an ecoregion (US EPA 2000). The 25th percentile is the value of a given parameter where 25% of all observations are below and 75% are above. The 25th percentile was adopted by EPA as the threshold between degraded conditions and minimally impacted locations. Additionally, the Massachusetts Department of Environmental Protection (MassDEP) established Surface Water Quality Standards (SWQS) for temperature, pH and dissolved oxygen (DO) as part of their Clean Water Act waterbody assessment process (MassDEP 2007). These thresholds were selected to protect designated categories of aquatic life, including fish habitat.

Stations were classified as Suitable (minimally impacted) or Impaired for each parameter. Water quality data were also evaluated to explore the potential of establishing new thresholds specifically derived from smelt spawning habitat measurements.

Water Chemistry Basic water chemistry parameters were measured during smelt spawn-ing runs at 19 index station stations: the 16 fyke survey sites and 3 additional spawning sites of interest in Massachusetts (Figure 1.3.7 and Table 2.1. 1) following the QAPP protocol. Yellow Springs Incorporated (YSI) water chem-istry sondes were used to measure water temperature (*C), DO (mg/L and %

saturation), specific conductivity (mS/cm), pH and turbidity (NTU, Neph-elometric Turbidity Units) in freshwater at the spawning grounds. At most stations, discrete water chemistry measurements were recorded three times per week. The seasonality of water chemistry monitoring was not synchronized for all stations due to the later onset of the spawning season at the northern end of the study area. For this reason, detailed comparisons of some parameters, such as temperature, should be made cautiously.

Water Temperature Water temperature has an important influence on smelt metabolism, the onset of smelt spawning and the duration of egg incubation. Median water temperatures during the spawning period were fairly consistent across the study area, with a range of 8.8 - 12.9 'C (Table 2.2.2, Figure A.2.1). No measure-ments exceeded the water temperature criterion of 28.3 0C adopted from Mass-DEP SWQS to protect aquatic life. The relatively high temperature threshold has little relevance for smelt that spawn in the cool water of the spring freshet; however, the temperature data have value for documenting baseline conditions and may have future application for monitoring reference values, such as station medians or 75th percentiles.

Specific Conductivity Specific conductivity is proportional to the concentration of major ions in solution corrected to the international standard of 25 °C. High conductance in ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN a 49

freshwater can indicate high watershed contributions of natural alkaline com-pounds or ionic contributions from pollution sources. For this reason, conduc-tivity has been discussed as a potential proxy for pollution sources, urbaniza-tion, and eutrophication. Median specific conductivity during the spawning period ranged from 0.031 - 0.997 uS/cm (Table 2.2.2, Figure A.2.2). The four highest medians occurred at urban sites near the Boston metropolitan area.

Dissolved Oxygen Adequate dissolved oxygen (DO) concentrations are necessary for embry-onic survival and normal development. The QAPP provides a DO criterion of > 6.0 mg/L to protect aquatic life. Median DO concentrations during the spawning period ranged from 9.5 - 12.5 mg/L (Table 2.2.2, Figure A.2.3), and median DO saturation levels ranged from 91.0 - 107.8% (Table 2.2.2, Figure A.2.4). All individual DO measurements were well above the DO threshold.

Similar to water temperature, the DO threshold may have limited relevance because of the high concentrations of DO found in turbulent riffles during the spring freshet. The distribution of DO saturation data does show increasing supersaturation in urban Massachusetts and a declining DO saturation mov-ing north in the study area. Supersaturation of oxygen can indicate eutrophic conditions, where due to the photosynthetic cycle of the algal communities, supersaturation is observed during the daylight hours, but anoxic conditions are present during darkness (Carlton and Wetzel 1987).

pH Increased acidification of water bodies in New England is a widely recog-nized threat to fish populations, as low water pH can increase the impact of alu-minum toxicity and disrupt fish respiration. Geffen (1990) conducted laborato-ry experiments to examine the influence of pH on smelt embryo survival; trials found that survival was most influenced by the duration of low pH exposure and embryo developmental stage. For example, high mortality occurred to early stage smelt eggs (4-6 days post-fertilization) at 5.5 pH when exposure ranged from 6-11 days. Fuda et al. (2007) conducted similar experiments and found survival was not affected until pH was < 5.0. The QAPP adopted the water pH criterion of > 6.5 to _<8.3 from MassDEP (2007) to protect aquatic life. Most stations had pH measurements in a range that was not a concern for rainbow smelt. Median pH during the spawning period ranged from 5.92 - 7.67 (Table 2.2.2, Figure A.2.5). Of the 19 rivers sampled, seven were classified as Impaired

(>10% of individual measurements below pH 6.5). Among the stations classi-fied as Impaired, only four had routine measurements below 6.0 pH: the three southernmost Massachusetts stations and Schoppee Brook in Maine.

Turbidity Turbidity in water is the result of suspended inorganic and organic matter; it can be caused by natural fluctuations in sediment transport or by changes in productivity. The QAPP adopted the turbidity criterion of : 1.7 (NTU) from the EPA Northeast Coastal Zone ecoregion (US EPA 2000). Most rivers had median turbidity values >1.7 NTU, and all were classified as Impaired for having at least 10% of measurements > 1.7 NTU (Table 2.2.2, Figure A.2.6).

Several stations in New Hampshire and southern Maine had median values well above the threshold. However, this elevated turbidity may result from the natural suspension of sediments, either due to soil type or the naturally high 50

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

turbidity in the spring associated with snow melt and higher runoff. Adopting the study's 25th percentile of 1.9 NTU would still result in all stations being classified as Impaired. The turbidity data will be further evaluated to determine if a more appropriate turbidity threshold can be established by removing pre-cipitation effects through an analysis of baseflow data.

DataAnalysis Median values of water temperature, DO, specific conductivity, pH and turbidity were compared among sampling stations (Kruskal-Wallis, p < 0.001),

and a multiple comparison test was used to determine which stations were significantly different from others (Siegal and Castellan, 1988; R code, krus-kalmc; p = 0.05; Figures A.2.1 - A.2.6). Significant differences were found for all parameters; trends between parameters were common among rivers and regions. Conductivity was especially variable among sites and may be related to watershed characteristics; in the most urban sites (Crane and North rivers, Massachusetts) conductivity was significantly higher than most other sites, whereas at the forested sites (Deer Meadow and East Bay Brooks, Maine),

conductivity was significantly lower than most other sites, The relation of these variables to spawning smelt populations is discussed in the Watershed Charac-teristics Section.

Nutrient Concentrations Nitrogen and phosphorus are vital nutrients for plants but can cause exces-sive growth and degrade the health of aquatic life at high concentrations. The influence of nutrient pollution on water and habitat quality in rivers and lakes is a growing concern in the United States (Mitchell et al. 2003). The health or trophic state of aquatic habitat is influenced most by light, carbon sources, nutrients, hydrology and food web structure (Dodds 2007). Among these influences in developed watersheds, nutrient enrichment is most dependent on human activity and may be most amenable to remediation efforts. Total nitrogen and total phosphorus were recorded weekly at index stations in the freshwater portion of the streams on the spawning grounds from 2008-2011.

Field sampling procedures are documented in the QAPP (Chase 2010), and the laboratory analysis followed EPA-approved Quality Assurance /Quality Control (QA/QC) protocols.

Nutrient concentrations for smelt spawning habitat were classified us-ing EPA recommended thresholds for freshwater streams and rivers that were developed from the distribution of available water quality data (US EPA 2000).

These EPA thresholds for Suitable habitat for the study area are 0.57 mg/L for total nitrogen (TN) and 23.75 ug/L for total phosphorus (TP). The EPA also recommends that states develop their own nutrient water quality criteria for protecting specific designated uses of aquatic habitat under Clean Water Act assessment and remediation processes (US EPA 2000). In this light, the TN and TP data recorded for this study were compared to the EPA nutrient criteria and the data distributions were evaluated for potential smelt habitat-specific thresholds (Table 2.2.3)

TotalNitrogen Measurements of TN at 20 stations during 2008-2011 showed a trend of ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 51

higher concentrations in urban areas (Table 2.2.3, Figure A.2.7). The range of median concentrations for all stations was 0.216 - 1.395 mg/L. Only five stations were classified as Suitable for TN (_<10% of measurements below 0.57 mg/L; EPA 2000), with four of these stations at the northeastern end of the study area. All others were classified as Impaired. The TN 25th percentile generated from the study sites was 0.340 mg/L, which was 40% lower than the EPA ecoregion threshold.

Total Phosphorus Measurements of TP displayed a more stable trend across the study area (Table 2.2.3, Figure A.2.8). The range of median concentrations for all stations was 12.18 ug/L to 36.72 ug/L. Only 4 stations were classified as Suitable for TP

(< 10% of measurements below 23.75 ug/L; EPA 2000). All others were classi-fied as Impaired. The TP 25th percentile generated from the study stations was 17.56 ug/L; 26% lower than the EPA ecoregion threshold.

TN/TP Ratio While total concentrations of nitrogen and phosphorus are important for plant production, the balance or ratio of TN to TP can also influence growth and species composition. Most TN:TP ratios were in a range expected for freshwater systems in New England (15:1-30:1). Higher ratios indicating high nitrogen and possible phosphorus limitation were found at the most urbanized stations, and low ratios most influenced by high phosphorus were only found at a few stations where watershed development was low.

Data Analysis Comparisons of median TN, TP and TN:TP ratios among sampling sta-tions found significant differences for all three parameters (Kruskal-Wallis, p <

0.001). A multiple comparison test was used to determine which stations were significantly different from others (Siegal and Castellan, 1988; R code, krus-kalmc; p = 0.05). The box plots in Figures A.2.7 - A.2.8 represent a graphic display of the multiple comparisons. The high TN concentrations at Crane River and North River (> 1.0 mg/L) in Massachusetts were significantly dif-ferent from all stations except the Saugus River. The four stations with median TN < 0.3 mg/L were significantly lower than most the remaining stations, all but one found in urban areas of Massachusetts and New Hampshire.

Periphyton Periphyton is the complex of benthic algae, detritus and other microorgan-isms that attaches to the river bed and is an important indicator of primary pro-duction and environmental disturbances in aquatic habitats. Periphyton growth responds to nutrient enrichment and can reach excessive or nuisance growth in eutrophied systems (Biggs 1996). Eutrophication has been identified as a major concern for smelt spawning habitat due to the potential impact of excessive pe-riphyton growth on smelt embryo survival at spawning riffles in Massachusetts (Chase 2006). These concerns have also been raised for smelt runs in tributar-ies to the St. Lawrence River in less urban regions of Qu~bec (Lapierre et al.

1999). Periphyton monitoring was conducted to provide a biological response variable for nutrient concentrations that may be directly related to successful embryonic survival. Laboratory experiments studying the effect of periphyton 52 , ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

growth on smelt embryo survival complimented the field monitoring, The lab results demonstrated that embryo survival was significantly lower on substrata with high periphyton growth/concentrations than on clean surfaces (Wyatt et al. 2010).

Field monitoring measured the growth of periphyton on spawning ground substrate at the index sites during the spawning period to determine how growth may differ between sites. Ceramic tiles were deployed to collect pe-riphyton during the 2008-2009 spawning period at riffle habitat where smelt deposit eggs. Periphyton growth on the tiles was collected biweekly to quan-tify daily growth and describe algal species composition. Ash-free dry weight (AFDW, g/m 2/day) was calculated as a measure of periphyton biomass. Average periphyton growth ranged from 0.006 to 0:120 g/m 2/day at 12 smelt spawning habitat stations (Table 2.2.3). The range of periphyton growth included very low growth at the easternmost Maine stations to high growth at urban centers in Massachusetts.

No algal biomass thresholds are available specifically for smelt spawning habitat. In the absence of published thresholds, the 25th percentile of 0.0143 g/m 2/day was calculated from the AFDW medians observed during this study and compared to all values. All river stations exceeded this threshold and were classified as Impaired for periphyton, except for Deer Meadow Brook, Chan-dler River and East Bay Brook, Maine. The periphyton data suffer from high variability and low sample sizes at some sites. However, there appears to be potential value in using the 50th percentile (0.0533 g/m 2/day) as a threshold for moderately impacted rivers. At the stations with medians above the 50th percentile (Figure 2.2.1), the periphyton could be characterized as excessive growth that could impede egg incubation and appears to be associated with higher TN and urbanization. However, more work is needed to understand the range of periphyton growth at different spawning streams, how this var-ies annually in response to environmental conditions, and the point at which periphyton growth impairs embryo survival.

Heavy Metal Concentrations Heavy metals such as cadmium, chromium, copper, iron, lead, manganese, mercury, silver and zinc can be absorbed by both fish embryos and larvae and lead to developmental abnormalities and reduced survival (Finn 2007, Jezierska et al. 2009, Wegwu and Akaninwor 2006). Short-term, high-intensity con-tamination mostly occurs in the spring months during snowmelt periods, when mild water acidification that is associated with snow melt leads to free metal ions being leached from sediments (Jezierska et al. 2009). Long term exposure to lower concentrations of heavy metals may be of equal concern. The toxic ef-fects of aluminum on salmonid embryos are seen when pH is below 6.5; at this level, pH can inhibit the swelling of the egg shell, reducing the amount of space for the embryo to develop and move, and leading to stunted growth or physical abnormalities (Finn 2007). Cadmium, lead and copper at low levels can exac-erbate these effects at any pH (Jezierska et al. 2009). Above critical thresholds, mercury, lead, cadmium, chromium, iron, and zinc have all been shown to reduce the number of embryos successfully hatching (Wegwu and Akaninwor 2006), as well as to disturb skeletal growth, impair hemoglobin (red blood cell)

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 53

formation, cause osmoregulatory failure, and limit overall growth because the organism's energy is spent ridding the body of the toxic contaminants (Finn 2007; Jezierska et al. 2009).

We sampled heavy metal concentrations and other minerals (calcium and magnesium) at all index sites during baseflow conditions over the course of the spawning period in 2010 and 2011 to describe the range of concentrations to which smelt embryos are chronically exposed. Although not part of this study, corollary laboratory experiments should be performed to ascertain which metals and what concentrations reduce survival and impair normal development in smelt embryos and larvae.

Of the heavy metals, silver, cadmium, and mercury concentrations were below detection levels for all sites during all sampling periods (detection levels 0.002 mg/l, 0.5 ug/l, 0.5 ug/l, respectively). Chromium was detected only once during the sampling period, in the Oyster River, New Hampshire (0.003 mg/l; detection level 0.002 mg/I). Although these metals were not detected, or de-tected only once, it should not be assumed that they are not present. They may in fact be present either at concentrations below the detection levels or during runoff or precipitation events neither of which our sampling captured. All other metal concentrations were detected at most sites, and the range of values fol-lowed a log distribution. As log distributions are typical of metal concentrations in many regions, the values we measured likely represent much of the range of metal concentrations present in the region during the smelt spawning season (Table 2.2.4).

A principal components analysis (PCA) was performed using the 2010-2011 average concentrations (log transformed to produce normal distributions) to determine which metal and mineral concentrations trended together, and which seemed to vary on their own. From this analysis, we find that lead (Pb; abbreviations refer to labels in associated figure, and are not the full elemental symbols with ionic sign), copper (Cu), and zinc (Zn) are highly related and trend opposite from aluminum (Al). This pattern indicates that when high values of lead, copper, and zinc were present, aluminum values were low, and vice versa. Being drivers of water hardness, calcium (Ca) and magnesium (Mg) were highly related to hardness and alkalinity, but notably nickel (Ni) was also highly related to these variables (Figure 2.2.2).

The relationship between metal concentrations and watershed characteris-tics is explored in the following section.

Watershed characteristics As suggested throughout the preceding sections, watershed land use can affect water quality in receiving streams and rivers in a variety of ways. The development of wetlands, agricultural fields, or forested areas replaces porous soils with impervious surfaces, which increases the velocity of water flowing off the land and the supply of suspended sediments, nutrients, and contaminants to adjacent streams (Brenner and Mondok 1995, Corbett et al. 1997, Strayer et al. 2003, US EPA 2004). In addition, agricultural areas contribute nutrients-both nitrogen and phosphorus-to receiving streams. In aquatic ecosystems, these nutrients can promote algal blooms, deplete oxygen, and degrade fish habitat (Carpenter et al. 1998, Howarth et al. 2000).

54

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Understanding how water quality, nutrient levels, and heavy metal concen-trations are related to watershed land use is important for developing manage-ment strategies to minimize impacts to rainbow smelt eggs and larvae.

Correlations between watershed land use and water quality parameters, nutrient levels, periphyton growth, and heavy metal concentrations were evalu-ated using Spearman's rank correlation statistic. Results are presented in Table 2.2.5 at the scale of the full drainage basin and riparian buffer zone. Several key patterns emerge from these correlation results that are relevant to rain-bow smelt conservation. First, patterns are very similar at full watershed and riparian buffer scales, indicating that land use in the broader watershed exerts a similar influence on water quality as land use immediately adjacent to the receiving stream. Second, the percent of development and forest in the water-shed show the strongest associations with water quality, with the direction of influence occurring in opposition to one another. For example, higher percent-ages of developed areas are associated with higher stream dissolved (available) Understandinghow water nitrogen and heavy metals concentrations; conversely, highly forested water-quality, nutrientlevels, sheds are associated with lower concentrations of nitrogen and metals (Craw-ford and Lenat 1994). Because periphyton growth is dependent on available and heavy metal con-nutrients (like dissolved nitrogen), and because heavy metals can negatively centrationsare related affect embryo development and survival, this pattern suggests that protecting to watershedland use is forested areas is important for maintaining water quality conditions that are importantfor developing beneficial to rainbow smelt. managementstrategies to minimize impacts to Conclusions rainbow smelt eggs When compared to the established EPA thresholds, the water quality data and larvae.

collected during 2008-2011 show widespread impairment due to elevated TN, TP, and turbidity and more localized impairment from acidification and excessive periphyton growth. More work is needed to evaluate existing criteria and to establish new thresholds that are specific to smelt spawning habitat.

For example, the turbidity criterion is likely too low to be relevant for stream riffles during spring; conversely, the water temperature and DO criteria may be too high, as smelt embryos require a lower temperature than the current EPA threshold. The highest median values for TN, conductivity and periphyton were associated with urban sites. Most sites with few identified impairment were at the northern end of the study area.

These results provide a range of water quality conditions that affect successful embryonic survival. From high impairment in urban settings to suit-able water quality in rural settings, these sites are examples of both conditions requiring remediation and demonstrating restoration targets. We encourage resource managers to use these baseline conditions to consider potential reme-diation measures (e.g., riparian buffers, stormwater improvements, point source reductions) to improve impairments and to plan for protecting locations with suitable conditions for supporting smelt spawning success.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 55

Table 2.2.1. Water chemistry Existing Water Ouallty Criteria criteriarelated to smelt spawn-Ing habitat.The water chemistry Suitable Minimally Minimally Moderately parameterswere adopted to Impacted Impacted Impacted protect Aquatic Life at Class B 25th Percentile 25th Percentile 50th Percentile Inland Waters (MassDEP2007), Parameters (MassDEP 2007) (US EPA 2000) (2008-2011 data) (2008-2011 data) and US EPA reference conditions Temperature (°C) 5 28.3 (25th percentile) for the Northeast Sp. Conductivity (mS/cm) 5 0.131 CoastalZone sub-Ecoreglon (US pH > 6.5 to 5 8.3 EPA 2000). Potential criteriaare DO (mg/L) a 6.0 Turbidity (NTU) S 1.7 5 1.9 5 2.1 presentedbased on 25th and TN (mg/L) 5 0.570 5 0.340 5 0.452 50th percentiles from 2008-2011 TP (ug/L) 5 23.75 5 17.56 5 20.43 Perlphyton Biomass (g/m2/d) 5 0.0143 5 0.0533 projectdata. Blank cells Indicate either that no criterionexists or the derived percentile has limited relevance for smelt habitat.

_ _mTI_ Cond. DO % SN NTU Rie MWia Cods e in Exceed Median Mjn Exceed Meia Exceed Median Excee Medfaia Exceed ECW MA Weatpoft W~p 95S5 0% 0.130 96.1 0.6 0 .92 1.4 -3 MA Woantlic WW 11.05 0% 0.092 95.9 10.55 0% 6.23 W% 2.2 40 MA Jones JR 9.71 0% 0.200 100.0 11.74 0% 639 M% 2.8 OW MA Fore FR 10,26 0% 0.558 105.1 12D6 0% 7.09 2% 2.1 A16 MA Saugus SG 8-89 0% 0.663 102.3 11.98 0% 7.28 0% 2.9 of MA North NR 9,57 0% 0.962 105.0 12.45 0% 7,24 0% 2.0 74%

MA Crone CN 9.22 0% 0.997 99.1 11.89 0% 7.18 1% 3A 90 MA Essex ER 9,83 0% 0200 1Q5.2 12,32 0% 6,71 28 13 293 MA parker PR 9.11 0% 0.2.2 105.1 11.88 0% 7.02 1% 1,8 NH Squamsoott So 11t89 0% 0.152 100.4 10.93 0% 6.93 2% 1.8 57%

NH WiMIcut WR 11.50 0% 0.315 97.6 1121 0% 7.43 0% 4.3 0%

NH Oyster OY 10,48 0% 0.195 101.0 11.31 0% 7.38 0% 4.4 10*%

ME Long Creek LC 10.36 0% 0.526 97.0 11.07 0% 7.25 0% 6.9 1OO%

ME Mast Landing ML 8.79 0% 0.134 98.1 11,52 0% 7.11 8% 8.8 ME DeerMeadow DM 10.99 0% 0,031 9810 11.14 0% 6.84 1.. 24 ME Tannery Brook TB 12O58 0% 0.157 98.1 10.43 0% 7.67 4% 1,8 ME Sdchppe SB 9.48 0% 0.089 92.5 10.2 0% 6.27 2.1  %

ME Chandler River CR 12,86 0% 92.8 9.47 0% 6.72 21 2.0 10, ME EtBa Es 9.80 0% 0.044 95.1 10.78 0% 7.31 7% 2.0 2Mth Periet 9.51 0.131 96.6 10.85 6.72 1.9 50th Percentile 9.83 0.197 98.1 11.21 7.09 2.1 Table 2.2.2. Basic water chemistry measuredat 19 smelt fyke net index stationsin the U. S. Gulf of Maine and Buzzards Bay, Mas-sachusetts.Median values were calculated from all available data from 2008-2011. The percent-age of samples at each station that exceed the QAPP (Chase 2010) thresholds are presented In shaded cells, indicatingan Impaired classificationfor the pa-rameter.No waterquality criteria are available for conductivity or DO saturation.

56

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

TP TN N:P AFDW (alelday)

Sbft Rhr Code N Median Exceed N Median Exceed N Medan N Medin MA Westport WP 25 19.20 ýo 25 0... W 25 332 0 MA Weweenbc WW 26 37.80 4% 23 0.283 '11 23 7.8 0 MA Jones JR 48 16.70 13* 47 0.659 9 47 34.1 8 0.0169 MA Fore FR 47 21.10 M% 48 0.530 l% 47 23.1 8 0.0154 MA Saugus SG 10 26.95 70% 11 0.917 1*% 10 35.1 0 MA North NR 47 21.06  % 49 1,395 10% 47 88.0 6 0.0828 MA Csrne CN 48 21.89 A 48 1.285 10ft 48 58.9 8 0.1198 MA Essex ER 11 12.80 9% 11 0.411 9% 11 31,3 0 MA MIN MR 45 21.80 3 46 0.844 45 28.0 8 0.0685 MA Parker PR 11 17.80 0% 11 023 411 31.0 0 NH Squamscott SO 37 17.44 9 37 OA20  % 37 22.7 9 0.0598 NHi Winnlcut WR 37 20.10 36 0.516 M 36 25.3 9 0.08867 NH Oyster OY 15 22.70 15 0.387 20 15 18.3 0 ME Long Creek LC 30 20375 29 0.425 ... 29 23.8 4 0.0625 ME Mat Landing ML 37 18.81 22% 37 0.258 0% 37 11.9 0 ME DeerMeadow DM 37 17.0 . 35 0.253 0% 35 16.5 4 0.0068 ME Tannery Brook TB 32 23.840 32 0&332 0% 32 13.9 18 0.479 11% 18 15.5 0 ME Scloppe. SB 18 27.00 81%

ME Chandler River CR 10 14.95 0% 9 0.342 111% 9 24.8 4 0.0111 ME East Say Es 34 11,15 6% 33 0.216 0% 33 17.7 4 0.0065 25th Pevetl 17.56 0.340 17.4 0.0143 5lih Percentle 20.43 0.452 24.1 0M0533 Table 2.2.3. Nutrient and periphy-ton measurementsfor all Index stationsIn the U. S. Gulf of Maine and Buzzards Bay, Massachu-setts. The percentage of samples at each station that exceed the QAPP (Chase2010) thresholds are presented in shaded cells, Indicatingan Impaired classifica-tion for the parameter.No criteria are availablefor the N:P ratio or perlphyton.

Analyte Unit 2010 2011 2010-2011 2010-2011 2010-2011 Detection Detection Mean Low High Limit Limit Value Value Value Aluminum mg/L 0.005 0.01 0.1347 0.0059 1.0000 Arsenic ug/L 0.5 0.5 1.30 0.51 4.00 Cadmium ug/L 0.5 0.5 BDL BDL BDL Calcium mg/L 0.05 0.05 13.78 0.55 52.00 Chromium mg/L 0.002 0.002 0.003 0.003 0.003 Alkalinity mg/L 1 1 29.14 3.26 100.00 Copper mg./L 0.0005 0.0005 0.0013 0.0005 0.0077 Iron mg/L 0,05 0.05 0.62 0.16 2.70 Lead ug/L 0.5 0.5 1.05 0.38 3.10 Magnesium mg/L 0.05 0.05 4.27 0.27 39.00 Table 2.2.4. Analytes measured in Nickel mg/L 0.0005 0.0005 0.0016 0.0005 0.0050 Silver mg/L 0.002 0.0005 BDL BDL BDL water samples taken at baseflow Zinc mg/L 0.002 0.002 0.006 0.002 0.021 atsmelt spawning Index sites Total Hardness mg/L 0.35 0.33 54.6 2.5 430.0 2010-2011. Detection limits and Not Sampled Mercury ug/L 0.5 in 2011 BDL BDL BDL mean, low, and high concentra-tions are shown for each analyte.

BDL - below detection limit.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 57 I

Full watershed Stream buffer

%dev %devopen %forest %wetiand %a I %dev %devooen %forest %wetland %aa 4

water quality conductivity 0.S5 0.9 -0.83 -0.16 -0.12 0.94 0.92 -0.79 -0.01 -0.24 DO conc. 0.67 0.65 -0.38 -0.18 -0.19 0.51 0.56 -0.34 -0.05 -0.2 pH 0.36 0.39 -0.25 -0.42 0.01 0.42 0.43 -0.3 -0.33 -0.14 turbidity 0.32 0.47 -0.22 -0.14 -0.04 0,28 0.51 -0.12 -0.21 -0.18 TP 0.26 0.34 -0.46 -0.21 0.04 0.36 0.31 -0.48 -0.11 -0.02 TN 0.87 0.77 -0.81 0.1 -0.16 0.85 0.74 -0.74 0.24 -0.19 AFOW 0.62 0.49 -0.57 -0.1 0.23 069 0.55 -0.58 0.04 0.02 alkalinity 0.83 0.77 -0.66 -0.23 -0.14 0.8 0.76 -0.66 .0.05 -0.25 hardness 0.83 0.78 -0.7 -0.24 -0.11 0.88 0.88 -0.68 -0.16 -0.33 Metals Ad -0.s3 -0.44 0.46 0.2 0.22 -0.39 -0.28 0.56 0.02 0.13 As 0.S4 0A5 -0.44 0.13 0.22 0.61 0.57 -0.67 0.15 -0.04 Ca 0.83 0.75 -0.68 -0.3 -0.16 0.86 0.82 -0.19 -0.36 Cu 0.58 0.45 -0.37 -0.41 0 0.42 0.35 -0.41 -0.25 -0.09 Fe 0.26 0.43 -0.32 0.22 0,42 0.19 0.41 -0.3 0,24 0,34 Mi 0,86 0.84 -0.74 -0.05 -0.06 0.89 0.92 -0.67 -0.01 -0.26 NI 0.89 0.89 -0.74 -0.21 -0.04 O.8i 0.83 -0.69 -0.08 -0.2 Pb 0.64 0.63 -0.72 -0.45 -0.81 0.81 0.59 -0.64 -0.36 -0.8 Zn 0.7 0.74 -0.87 -0.29 -0.35 0.74 0.71 -0.82 -0.25 -0.44 Table 2.2.5. Spearman's rank correlationbetween water quality metrics and land cover at two spatialscales (e.g., full water-shed and riparianbuffer zone).

Correlationcoefficients In bold type indicatesignificance at the p-O.05 level.

0.35 0,30 92008 32009 0,25 0.20 AFDW 0.15 Figure2.2.1. Annual median periphyton growth (ash-freedry weight, g/M 2/day) displayed by 0.00 -

sample stationwith 50th per-centile of station median values JR FR NR CM MR SQ WR IC DM TB CR EB marked by green line. Refer to Table 2.2.2 for river codes. River 58 e ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

oMg Alk Ca Figure2.2.2. Principalcompo-

. . nents analysis (PCA) performed 0 on 2010-2011 average metal and mineral concentrations(log

-2C transformed). The first compo-nent Is driven most by hardness Zn (a variable which representsthe total mineral concentrationof water,driven by calcium and mag-neslum), magnesium, calcium, alkalinity,and nickel. The second component Is driven most In the positive direction by aluminum and arsenic and less so by iron, Component 1 (56.6 %) and In the negative direction by zinc, copper, and lead.

2.3 - THREATS TO SMELT IN MARINE COASTAL WATERS Smelt spend at least half the year in marine coastal waters during the summer and fall months. As adults and juveniles they are a schooling fish that attract a wide range of predators. While monitoring this life phase can be more difficult than monitoring discrete spawning runs, it is no less important when considering the species decline. During this period, smelt are susceptible to environmental influences on survival, shifts in natural mortality and to capture in small mesh fisheries targeting other species. These topics are discussed below, using the best available information to discuss how each issue may be affecting smelt populations; however, to fully understand the implications, each requires further study.

Fish Health Improving understanding of fish health status as well as the abundance, geographic distribution, and vectors of areas of study necessary to support the development and implementation of conservation strategies designed to protect and restore rainbow smelt populations. Pathogens can adversely affect both juveniles and adults in both general and acute ways, including organ failure, energy loss, interruption of hormonal pathways and reproductive weakness (D.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 59

Bouchard, University of Maine, pers. comm., 2011).

We characterized pathogen presence endemic to smelt at fourteen spawn-ing index sites spanning the Gulf of Maine over a two-year period, 2009-2010 (Bouchard 2010). Sampling did not detect bacterial pathogens of regulatory concern but did detect endemic parasites that are well documented for similar anadromous species. Parasitological results were typical of wild fish populations, with various trematodes (e.g., black grub), cestodes, nematodes and protozoa observed at all sites. A microsporidian parasite detected in various tissues of many individuals in this study was not identified as to species, but is consis-tent with (Glugea hertwigi), which was confirmed at one site: the Fore River, Massachusetts. This parasite has been documented extensively in freshwater smelt can be detrimental to successful spawning because this parasite infests the gonads of smelt (Jimenez et al. 1982, Nsembukya-Katuramu et al. 1981). The observation of large numbers of (Philometraspp.)-like nematodes in the gonads Fish from a majority of of the majority of female fish in the study is also consistent with reports of this parasite as an opportunistic pathogen of spawning female fish in other species the sites spanning the (Moravec and de Buron 2009).

entire Gulf of Maine Virology results revealed a viral agent from adults from Casco Bay, Maine; region showed evidence however, it is difficult to place any significance to this agent at the present time of erythrocytic disease, or because the virus is not similar to currently catalogued agents (IPNV, IHNV, degradationof red blood ISAV, and VHSV have been ruled out by PCR techniques). More analysis on cells, leading to anemic this agent is needed to fully understand the physiological effects it may be hav-effects. ing. Fish from a majority of the sites spanning the entire Gulf of Maine region showed evidence of erythrocytic disease, or degradation of red blood cells, leading to anemic effects (Bouchard 2010). This last point may be of specific concern and warrants further investigation to understand the extent of disease and causal factors.

FishingMortality Overfishing in historicalfisheries While historical fisheries for rainbow smelt landed thousands (and in Maine millions) of pounds annually in the 1800s, because the relative size of the entire population was unknown, it is not possible to quantify the effect of these targeted fisheries on smelt populations.

As populations declined in the 20th century, and as regulations limited fishing gear and take in response to this decline, targeted fishing effort has also been reduced. Today, few targeted commercial fisheries exist: a dip and bow net fishery is open to permitted individuals in Great Bay, New Hampshire; and a gill and bag net fishery are allowed during a regulated time period to per-mitted individuals on five rivers in downeast Maine. Large-scale recreational hook-and-line ice fisheries also exist in Great Bay, New Hampshire, and on many rivers and embayments in Maine (most notably the Kennebec River and Merrymeeting Bay area). While these fisheries are not thought to contribute high mortality for the smelt populations they target, the current extraction rates are unknown. Studies by the ME DMR in the late 1970s estimated that the ice-fishery on the Kennebec River extracted less than 5% of the total smelt popula-tion in the river (Flagg 1983). In Maine there is also a large recreational dip net fishery that targets adult smelt on the spawning grounds during the spring runs.

60

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

While there is a limit of 2 quarts of smelt per person per day in this spring fishery, the contribution to mortality is unknown.

Incidentalcatch in small mesh fisheries Five small mesh fisheries operate in the Gulf of Maine, all capable of en-countering rainbow smelt. Because smelt is not a regulated species for federally permitted fisheries, incidental catch (bycatch) is not required to be reported, although it is in some cases. Thus, it is difficult to determine the total amount of smelt bycatch; however, the relative impact on the species can be assessed based on reports from the Northeast Fisheries Science Center Observer (NEF-SC) Program, which monitors catch from a representative sample of each fleet (NEFCS 2012). The following analyses represent all Gulf of Maine states.

The Northern shrimp fishery operates in nearshore coastal waters during the winter and early spring months. Since 1992, the fishery has been required to install a finfish excluder device in their nets, the Nordmore grate. Prior to 1992, total bycatch in this fishery comprised almost two-thirds of the catch (Howell and Langan 1992). Subsequent surveys have found that the grate is extremely effective in limiting bycatch; Eayrs et al. (2009) observed reductions to 4-8% of the total catch over a two-year period.

Using NEFSC observer records, the effect of the Nordmore grate on reduc-ing smelt bycatch can specifically be seen. In the period directly preceding the requirement of the excluder device (1989-1992), there were 197 observed trips on vessels targeting Northern shrimp, and smelt were caught on 38 (19%) of these trips. A total of 201 lbs of smelt were caught during these trips combined, for an average of 5.3 lbs per trip. The highest was 46 lbs of smelt bycatch, although 87% of these trips caught less than 10 lbs. In the period directly following the excluder panel requirement (1993-2006), the amount of smelt bycatch on observed trips decreased, although not significantly (Wilcoxon ranked sum test: p = 0.129 > 0.05). During this period, smelt were observed on 74 (24%) out of 303 observed trips. A total of 289 lbs of smelt bycatch were caught during these trips, with an average weight per trip of 3.1 lbs. The high-est smelt catch was 31 lbs, and 92% of these trips had less than 10 lbs. Recent data (2007-2011) show that smelt bycatch has decreased significantly from the last two time periods (Wilcoxon ranked sum test: p < 0.0001 < 0.05). During this most recent period, smelt bycatch was observed on only 22 162 (14%)

observed trips, all of which saw less than 10 lbs. The average smelt bycatch for this recent period was 0.5 lbs, with a maximum catch of 2 lbs.

Vessel Trip Reports (VTRs) were implemented in 1996, at which point it became mandatory for vessels to report all catch. From the VTR reports, smelt were only reported in the shrimp fishery post-2006, but reported annually since then. From 2006-2011, smelt were reported in 35 trips out of 14,339 trips (0.2%). Of the trips that did report smelt, the average catch was 5.3 lbs, the highest 100 lbs (one occurrence), and 94% of trips reported less than 10 lbs. Further work is needed to estimate the total amount of smelt taken in the shrimp fishery using both observer and VTR data.

The mackerel, whiting (silver hake), Atlantic herring, and loligo squid fisheries are all also capable of encountering smelt as bycatch. These fisher-ies operate on multiple scales with various gear types, including pound (trap) nets at fixed locations close to shore, offshore trawling, and bag netting. Smelt ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 61

bycatch has been reported on VTRs in the Atlantic herring and whiting fisher-ies, however too few reports have been given from the mackerel fishery to draw any inferences, and no smelt bycatch has been reported from the loligo squid fishery.

In the Atlantic herring fishery, some smelt bycatch was reported in each year 1996-2011, although was reported on fewer than five reports in 1997, 2002, and 2008-2011. For the total period, smelt were reported in 135 trips out of 5463 total Atlantic herring trips (2.4%). The average reported catch was 5.1 lbs, the highest was 100 lbs (one occurrence), and 84% of these trips reported less than 10 lbs.

In the whiting (silver hake) fishery, smelt bycatch was reported for 71 trips out of a total of 20,204 trips (0.3%) for 1996-2011. In seven of these years, fewer than 5 VTRs reported smelt (1999, 2004, 2005, 2008-2011). The aver-age reported catch was 6.4 lbs, the highest was 42 lbs, and 73% of these trips reported less than 1Olbs.

If these data are representative of smelt bycatch in these fisheries, it is likely that they are not having a large effect on smelt populations at this time.

However, because we do not have a population estimate for smelt, it is not possible to ascertain the mortality rate due to bycatch in these fisheries. Further, the effect of small-mesh fisheries in the past cannot be determined. To fully understand the effect of small-mesh fisheries on smelt populations, more work is necessary to ensure that the observer and VTR programs are accurately capturing the extent of smelt bycatch.

Predator-preyrelationships Prey Availability Rainbow smelt are voracious feeders on amhipods, euphausiids, mysids, shrimps, marine worms, and any available small fishes (e.g., silverside, mummi-chog, herring) (Scott and Scott 1988). We do not know of existing broad-scale data to evaluate changes in the prey of rainbow smelt over time, however, the prey base was likely affected by changes in primary production and zooplank-ton community composition during the'1990s (Greene et al. 2012), and such variability should be expected as a result of oceanographic and climate variabil-ity. In addition, the balance between small prey species and larger fishes may shift as a result of ocean acidification (Wootton et al. 2008), which will likely affect calcifying organisms such as zooplankton and shrimp.

PredatorPopulationShifts Predators of rainbow smelt include a variety of aquatic birds (e.g.,

mergansers, cormorants, gulls, terns), fish (e.g., Atlantic cod, Atlantic salmon, striped bass, bluefish), and seals (Collette and Klein-MacPhee 2002). While the abundance of some of these predators has declined since the 1990s, others have increased. For example, striped bass populations have increased dramati-cally over the past 20 years, although the recovery has not been seen consis-tently along the coast. Maine striped bass populations have actually declined or remained at low levels compared to other regions (ASMFC 2011). Striped bass predation has been shown to have a significant impact on blueback herring populations in Connecticut River, and has been attributed as one of the factors 62 , ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

limiting blueback herring restoration in this river (Davis et al. 2009). Similarly, populations of grey seals in the Gulf of Maine have increased dramatically over the past few decades (NEFSC 2010). Like striped bass, grey seals are capable of ingesting large amounts of forage fish, and are found feeding in nearshore coastal waters in late spring when smelt are present in large schools. Although not as closely documented, cormorant populations have also sharply increased in recent years and are known to prey heavily on smelt. Striped bass, cormo-rants, and grey seals have received protections as managed species that have increased their populations sharply in short periods of time. Although these are natural predators that smelt have coexisted with while adapting to Gulf of Maine environments, it is possible that the impact of increasing predation on declining smelt populations results in proportionally higher natural mortality than in the past.

Recent shifts in predator range may also increase the exposure of smelt to predators. Friedland et al. (2012) suggested that the survival post-smolt Atlantic salmon may be affected by increasing predator abundance in the Gulf of Maine; increasing predator abundance that is due not necessarily to increas-ing population size, but to northward shifts in range due to recent changes in climatic and oceanic conditions. Because many of these species prey on a wide range of forage fish, this increasing predator abundance may affect smelt populations as well, although more research would be necessary to assess this relationship.

Community shifts Dramatic declines of diadromous fish populations have been observed across North America (Limburg and Waldman 2009; Hall et al. 2012).

Saunders et al. (2006) proposed that coherent declines within a co-evolved diadromous community could negatively affect individual species. While Saunders et al. (2006) focused on benefits that may have been lost for Atlantic salmon through community-level shifts, several of these could also affect rain-bow smelt. In particular, the decline of species such as alewives, blueback her-ring, and American shad-which are present in rivers and estuaries as juveniles during the same time as rainbow smelt-could have resulted in the loss of a prey buffer for rainbow smelt juveniles, making them more vulnerable to predation.

Climate-driven environmental change It is anticipated that climate change will influence temperature and pre-cipitation patterns in New England, and some of these effects may already be evident in recent environmental trends. Surface water temperature has been monitored monthly nearly continuously since 1905 (ME DMR 2011). This temperature series shows periods of warming during the 194 0s-1950s and again from the 1990s to mid-2000s, with the warmest water on record observed in 2006 (Figure 2.3.1). Because smelt are a cold water species, their geographic distribution shift northward may be influenced by the trend in warmer waters.

In addition to warmer coastal waters, freshwater conditions have changed in recent years as well. During the 1980s and 1990s, the Northeast experi-enced an increase in heavy precipitation events, and warmer temperatures have reduced ice cover and prompted earlier spring flows (Hodgkins et al. 2003, ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 63

Frumhoff et al. 2007). On New England streams that are substantially affected by snowmelt, the winter/spring center of volume dates and peak flow dates advanced by 1-2 weeks between 1970 and 2000 (Hodgkins et al. 2003). Water temperature and flow changes may affect spawning migration timing (Juanes et al. 2004, Ellis and Vokoun 2009), development rates, and early life stage survival in rainbow smelt. More research is needed to understand how climate-related environmental changes influence smelt abundance and distribution changes and to anticipate future implications for rainbow smelt.

With concern to species communities and shifts that are due to climate change, evidence suggests that the balance between small prey species and larger fishes may shift as a result of ocean acidification (Wootton et al. 2008). As the amount of atmospheric carbon increases, the amount of dissolved carbon in oceanic water also increases, in turn decreasing the pH of seawater. At lower pH values, the development and survival of calcifying marine organisms like coralline algae and phytoplankton are inhibited. Because these organisms are the base of the marine food chain and the direct diet of many of smelts' prey species, a decline in these organisms may also negatively affect smelts' prey base.

This hypothesis has been examined on the Pacific coast, but with no conclu-sive results, and has only begun to be considered in the Gulf of Maine. More research is needed to fully understand the effect of climate change on species composition changes in this region.

13 12

• 10i 10 E

7 Figure 2.3.1. Mean annual 6 surface water temperatureat U) to to U) to L U) Ul) U) U') to 0 +_- (" i~t U) to 1'- 03 0) 0 Boothbay Harbor,Maine, from O 0) 0) 0) 0) 0) 0) 0) 0) 0) 0 1905-2010.

64

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

3 - CONSERVATION STRATEGIES We recommend that rainbow smelt remain federally listed as a Species of Concern. Populations have disappeared from their southern range in a short period of time and are also declining in their present distribution in the Gulf of Maine. The species should continue to be monitored, and factors contributing to its decline should continue to be assessed.

3.1 - REGIONAL CONSERVATION STRATEGIES We recommend that rainbow smelt remain Recommendation 1: Continue monitoringprograms federally listed as a Each state within the present distribution of rainbow smelt in the Gulf of Species of Concern and Maine currently monitors populations through inshore trawl, juvenile abun-dance, fyke net, and/or creel surveys. that currentpopulation monitoringefforts In states at the extreme southern limit of the range where spawning populations have not been documented within the past ten years, inshore trawl continue in the Gulf of surveys are likely the most effective way to monitor the remnant populations. Maine.

In the Gulf of Maine states, trawl surveys provide the only source of data on the marine life phase of smelt. It is necessary that these surveys continue to document smelt presence and quantify abundance, and it is recommended that biological information is collected from a sub-sample of catches.

The regionally standardized fyke net survey developed for this study should be continued in the Gulf of Maine. A standardized survey is necessary to provide long-term data that can track inter-annual variability across distinct spawning stocks. This information is critical for detecting whether populations are declining or showing signs of stress, as may be characterized by truncated age distributions, decreases in length at age, and decreases in CPUE over time.

The juvenile abundance surveys should also be continued in New Hampshire and Maine as the only surveys targeting this life stage. Further, creel surveys should be maintained at recreational fishing sites to provide a measure of the impact of the fishery as well as information about changes in population size and biological characteristics over time.

Because some pathological concerns were found as part of this project (see section 2.3 -Threats to Smelt in Marine Coastal Waters), Gulf of Maine states should periodically monitor rainbow smelt from multiple spawning stocks for pathology, including parasite occurrence, viral agents, and systemic physi-ological problems. Further, states should cooperate with Canadian provinces to compare parasite and disease prevalence in the entirety of the species' range.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 65

Recommendation 2: Restore historicalor degradedspawning habitat Spawning habitat degradation and obstructions to access have been identi-fied as two important factors that have reduced successful spawning. Restoring in-stream habitat (e. g. substrate, water volume and velocity, pool and riffle areas), riparian buffer, improving and preserving watershed functions, and restoring access are important management strategies to improve local smelt populations.

Where possible, head-of-tide dams should be removed. Eggs deposited be-low dams are subject to periods of salinity during high tide and may be exposed to air at low tide if freshwater flows coming over the dam are low. Perched culverts and small water control barriers can also have this effect. When these obstructions are removed, smelt are able to ascend into freshwater, where water chemistry is more stable over time and water level is relatively constant. While undersized culverts (less than 1.2x bank-full width) may not completely block Restoring in-stream access, they can limit the number of smelt that reach the spawning grounds by habitat(e. g. substrate, creating velocity barriers. Restoration projects to improve road-stream cross-water volume and ings should design replacement culverts that target minimum water depth of 6 inches with average velocities in the culvert of 0.5 m/s or less, and flood veloci-velocity, pool and riffle ties below 1.5 m/s (see section 2.1 - Threats to Spawning Habitat Conditions areas),riparianbuffer, and Adult Spawning).

improving and preserv-Additionally, water quality at the spawning grounds must support healthy ing watershed functions, embryonic development and survival. We found that diminished rainbow smelt and restoringaccess are spawning runs existed in rivers surrounded by urbanized watersheds, while importantmanagement rivers draining forested watersheds supported strong smelt spawning popula-strategiesto improve tions. Comparing watershed conditions to water quality, higher concentrations of nutrients and toxic contaminants were associated with developed areas, local smelt populations.

while highly forested watersheds were associated with lower concentrations of nutrients and metals. This pattern suggests that protecting forested areas is important for maintaining water quality conditions that are beneficial to rainbow smelt. Furthermore, regional efforts to purchase conservation lands should consider parcels in watersheds that support smelt spawning habitats.

When development does occur in watersheds with smelt spawning habitat, the amount of impervious surface should be minimized, and stormwater mitigation techniques should be implemented to curtail the impacts on water quality (e. g.

riparian buffers, vegetated stormwater retention pools, underground filtration systems, etc.).

Recommendation 3: Smelt Fishery ManagementActions The results of the present study documented evidence of high population mortality (truncated age distribution) and poor recruitment (low abundance) in smelt populations in the southern portion of the study area. The time series of population data collected among the fishery dependent and independent surveys is too brief to determine the causes of these stressors on smelt popula-tions. However, overfishing was consistently identified as a significant concern in the latter half of the 19th century and the early 20th century in the southern portion of smelt's distribution.

The sustainabiliry of current smelt fisheries, both recreational and commer-cial, will require management strategies to quantify natural mortality and fish-66 - ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

ing mortality. We recommend that each state in the study area review current smelt fishery regulations and identify locations where present management may not be sufficient to protect distinct populations that display evidence of stress.

We recommend that states estimate fishing mortality from all targeted smelt fisheries and review bag limits on both commercial and recreational fisheries that target smelt.

Recommendation 4: Expand research to estimate populationsize and assess the potentialimpacts of ecosystem and climate changes The surveys carried out as part of this project did not enable us to develop a population estimate for rainbow smelt. However, the standardized fyke net survey established by the study should be continued with additional research in order to assess smelt population status in the region, understand the im-pact of targeted fishing and incidental bycatch, and to understand the relative contributions of each spawning stock to the regional population. This may be accomplished through a large-scale mark and recapture effort that targets each genetic stock (Kovach et al., in press; section 1.1 - Basic Biology). Tagging studies carried out as part of this project to understand habitat use and within-season repeat spawning behavior documented few inter-annual returns (less than 1%), although approximately 200 smelt per year were tagged (assumed to be less than 10% of the entire run based on estimated fyke net catch efficien-cies). Future tagging studies should tag a representatively larger sample of the spawning population to effectively monitor inter-annual repeat spawning and estimate population size. Additionally, improved and validated age structure data are needed to support future estimates of population size. Efforts should be made to maintain sufficient age structure sample sizes in each state.

Further research is needed to understand how changes in prey availability and predator abundance affect smelt populations. Other studies have found connections between increasing predator populations and depressed forage fish populations (see section 2.3 - Threats to Smelt in Marine Coastal Waters).

Because these studies looked at predators that also feed on anadromous smelt, the impact on smelt populations should also be examined.

Species that are important prey of rainbow smelt may be particularly af-fected by changes in the chemistry of marine waters. Increases in the amount of carbon in the atmosphere are associated with increases in the amount of carbon in salt water, which leads to a reduction in oceanic pH that may negatively im-pact small prey species, such as calcareous plankton (Wooton et al. 2008). This relationship needs to be better quantified to understand the effect of a smaller prey base on smelt populations. Conversely, predator populations that have shifted in their range in response to climate conditions may be preying upon forage fish populations more than in previous times (Friedland et al. 2012).

Further studies are necessary to understand how rainbow smelt will be affected by changes to their prey and predators as a consequence of climate change.

Climate change may also impact smelt populations by changing the extent of available spawning areas. Smelt spawn directly above the head of tide, and the upstream extent of the freshwater spawning area is typically either a natu-ral barrier or road crossing. Thus, a rise in sea level that extends the tidal limit to these barriers may greatly reduce the number of spawning sites or the area ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 67

within sites that is suitable for spawning. Conversely, a rise in sea level could increase habitat by raising tidewater above natural barriers allowing access to new reaches. Future research should model the potential effects for various sea level rise projections.

Expanded research to understand reasons for systemic health issues and reduced survival is needed to effectively guide management actions. While it is helpful to understand overall relationships such as watershed composition and smelt population responses, it is only a starting point. For example, research into dose responses to specific water quality constituents at all life stages would enable managers to develop smelt specific water quality criteria. These criteria may then be used to guide water treatment goals around which non-point or point source controls can be designed. This would be especially important in those already developed watersheds that are impractical to restore to forest.

Controlled studies in both laboratory and field settings are critical to improve Expanded research to our understanding of cause and effect, not just correlations, and to develop measureable relationships. Lastly, post-restoration monitoring is necessary to understandreasons for evaluate the success of any prescribed restoration technique.

systemic health issues and reduced survival is Recommendation 5: Implement stocking of marked larvae, with needed to effectively continued monitoring and genetic considerations guide management Rainbow smelt are currently extirpated or have severely declined in many actions.

coastal rivers and streams that once supported robust spawning populations.

Historical fishing pressure at the spawning grounds and degraded habitat and water quality may be causal factors. When improvements are made to water quality and habitat in these streams, restoration practices, such as stocking, may be appropriate to re-establish rainbow smelt runs at these sites.

Successful stocking efforts must include marking and subsequent recapture of hatchery stocked smelt to quantify effectiveness of restoration efforts. Utiliz-ing recent advances in smelt culture techniques, Ayer et al. (2012) developed methods for marking otoliths in larval rainbow smelt with oxytetracycline (OTC) for monitoring returns. Using these methods, the Massachusetts Divi-sion of Marine Fisheries began a pilot program to stock OTC-marked smelt larvae in the Crane River, MA, after water quality suitability was confirmed and passage improvements were made to upstream spawning habitat (Chase et al 2008). Over 10 million marked smelt larvae have been stocked into the Crane River since 2007, and spawning adult smelt with OTC-marked otoliths have been recaptured, providing a positive response for the project to continue stock-ing and monitoring.

New restoration sites for rainbow smelt are being examined in both Mas-sachusetts and Maine. In many situations, the protection and enhancement of existing habitat and water quality at both donor smelt runs and potential stocking sites will be preferential to initiating a stocking effort. Before any stocking begins, these sites will be sampled for baseline population data, and a site suitability assessment will be conducted, which will include water quality monitoring, streambed characterization, and flow measurements. Further, the genetic information presented in this plan (section 1.1 - Basic Biology) must be used in determining the appropriate parent stock. Managing at too fine a scale can lead to reduced allelic diversity and ignores the natural occurrence of gene 68

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

flow, while managing at too large a scale can reduce genetic diversity and ignore local adaptations. Another important consideration is the status of donor popu-lations to support stocking efforts. Careful planning should be made to remove a minimal proportion of a donor smelt run's productivity for stocking. Finally, long-term post-stocking monitoring should be performed to demonstrate stocking success.

3.2 - STATE MANAGEMENT RECOMMENDATIONS Massachusetts Massachusetts has a long history of implementing management measures to ensure sustainable smelt fisheries. Concern over the capability of net fisher-ies during smelt spawning runs to negatively impact the long-term viability of smelt runs was documented in the 1860s (Kendall 1926). In 1874, the Massa-chusetts state legislature banned harvest using nets during the spawning period and limited harvest to hook and line for most coastal rivers in Massachusetts.

By the start of the 20th century, nearly all smelt runs had this protection, and local smelt fisheries continued mainly as sportfisheries with little change until recent decades.

The only location in Massachusetts that presently allows net fishing for smelt during the spawning run is the Weweantic River in Wareham. This fish-ery is conducted under authority of M.G.L 67 of 1931 that gives the Town of Wareham the responsibility to manage a smelt fishery from March 1 to March

31. This recreational fishery continues today with a 36 smelt/day bag limit for each permitted fisherman and limits the net size to 5 square feet. This location was monitored as a smelt fyke net station during the present study. The smelt catch at the Weweantic River station had low CPUE for Massachusetts rivers and a size composition dominated by the age-I mode. MA DMF intends to initiate cooperative efforts with the Town of Wareham to ensure this unique southern smelt run can be sustained.

Following the net bans of the 19th and early 20th centuries, no smelt laws or regulations were made in Massachusetts until 1941 when three provisions were added to M.G.L. Chapter 130 that focused specifically on smelt fisher-ies. Section 34 of Chapter 130 standardized the spawning run ban for harvest during March 15 to June 16. Section 35 standardized the method of harvest to hook and line only in Massachusetts. Section 36 gave the Division of Ma-rine Fisheries authority to close smelt spawning river beds to entry during the spawning season. Following these three laws, no changes to smelt regulations were made until 2009 when a daily bag limit of 50 smelt per angler was adopt-ed. Unlike Maine and New Hampshire that drafted smelt management plans in the 1970s and 1980s, no such plan has been prepared in Massachusetts.

Declining recreational smelt catches in the 1980s prompted a review of the status of smelt fisheries and spawning runs by the MA DMF A survey of all coastal drainages on the Gulf of Maine coast of Massachusetts was conducted from 1988-1995, during which 45 smelt spawning locations were documented and mapped in 30 coastal rivers (Chase 2006). The report for this survey in-cluded specific habitat and water quality recommendations for each smelt run.

Following the survey, effort was directed toward acquiring smelt population ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 69

data. A grant was received from NOAA's Office of Protected Resources to de-velop fyke net indices at six smelt runs during 2004-2005 (Chase et al. 2006).

This approach and the six fyke net stations were adopted for the present study.

These contemporary efforts, when compared to the historical records and fish-ery accounts from the 1960s and 1970s, present evidence of a sharp decline in Massachusetts smelt populations in the past 2-3 decades. Locations that once supported popular winter ice fisheries for smelt no longer have fisheries, and some known spawning runs have had no recent evidence of spawning activity.

Smelt Stocking Efforts The transfer of smelt eggs from larger donor smelt runs to smaller runs or rivers with no smelt spawning was a common practice late in the 19th century in Massachusetts, followed by a large dedicated effort during 1910 to 1920 (Kendall 1926). The ease with which smelt eggs could be collected and the appearance of large numbers of excess eggs in some settings contributed to the zeal behind decades of stocking. Unfortunately, documentation of responses to stocking is essentially absent, other than brief narratives in annual agency re-ports. Short-term increases in smelt spawning run size appear to have occurred in some systems, especially for coastal to inland lake transfers. However, no evidence can be found of long-term benefits of coastal to coastal river transfers.

Smelt egg transfers continued periodically through the 1980s with strong sport-fishing constituency support. Recent requests to stock smelt eggs led to a MA DMF evaluation that attempted to quantify the number of eggs transferred, egg survival and returning adult smelt (Chase et al. 2008). Returning spawning adults were documented in a pilot river with no smelt run during the first year of possible returns, but low egg survival and expected low recruitment conclud-ed with MA DMF discouraging the use of smelt egg transfers and prioritizing passage, water quality, and habitat quality improvements over stocking as meth-ods for restoring smelt populations. MA DMF presently does not support the use of egg transfers but is conducting a pilot study on the stocking of oxytetra-cycline marked larvae as a potential substitute for egg stocking in specific cases where population enhancement can be coupled with habitat improvements and monitoring.

HabitatRestoration The survey of smelt spawning habitat provided recommendations for specific habitat improvement projects (Chase 2006), four of which have since been conducted. Each of these projects has focused on improving spawning substrate. Two of these projects were able to take advantage of planned culvert replacements to add substrate improvements as part of the scope of work, while the other projects specifically targeted grant and mitigation funds to augment spawning substrate. The experience gained from these projects will assist future efforts in the region.

Recommendations

1) Apply the information gained from the present study and recent smelt habitat improvement projects to identify potential restoration sites and design smelt spawning habitat improvements that meet the life history requirements of smelt. Projects that can remove barriers and extend habitat connectivity for smelt and other diadromous fish should be prioritized 70 e ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

2) Continue monitoring smelt fyke net stations from the present study that have been identified as having promise to support long-term indices of abundance (i.e., Weweantic River, Jones River, Fore River and Parker River).

Improve and maintain data collection at fyke net stations to support future development of biological population benchmarks

3) Develop water quality criteria that relate to designated uses within the Massachusetts Wetlands Protection Act in order to protect the specific habitats of anadromous fish, including smelt spawning habitat

4) Conduct a smelt habitat survey of the Buzzards Bay region of Massachusetts that was not mapped during the previous Gulf of Maine survey in Massachusetts

5) Develop a state smelt conservation plan similar those completed for Maine (1976) and New Hampshire (1981)

New Hampshire The recreational smelt fishery in New Hampshire has been monitored and regulated for decades, and current fishing pressure is not believed to pose a ma-jor threat to the smelt population in the state. Ensuring that fishing pressure is compatible with a sustainable smelt population requires continuing monitoring efforts that are already underway, including creel surveys, spring spawning run surveys, and biological sampling during the ice fishery and young-of-the-year seine surveys. Current monitoring of the fishery does not capture recreational fishing for smelt that occurs in the fall prior to the onset of ice. There is also a limited hook and line commercial fishery for smelt in New Hampshire with local markets that is not well recorded. Developing surveys that obtain data from these portions of the fishery would be helpful for appropriately charac-terizing fishing related mortality. Currently, the daily limit for recreational smelt fishing is 10 liquid quarts, which is approximately equivalent to half of a 5 gallon bucket. Given that smelt is a species of concern, this limit would be re-evaluated if in the future fishing pressure is believed to pose a major threat to the population. Neighboring states of Maine and Massachusetts, which have larger smelt runs, have a daily limit of 2 quarts and 50 fish, respectively.

Populationmonitoring The most current statewide fisheries management plan for rainbow smelt was written in 1981, but it predominately focuses on lake smelt populations.

The objectives for smelt management were to maintain or increase the popula-tion of smelt and to provide for commercial and recreational fisheries. Man-agement measures implemented following development of the plan included closure of the fishery to net or weir fishermen from March 1 to December 15, a 10 quart daily possession limit, and implementation of a smelt egg transfer program that occurred intermittently until 1991.

To evaluate the effectiveness of the management measures and detect trends in smelt abundance, an annual creel survey of the recreational ice fishery was implemented, and a smelt egg deposition index was developed. Data have been collected for the smelt egg index from 1979-2006. The intent of the index was to provide a fisheries independent relative measurement of spawning stock abundance. Validation of the index was attempted in 1993 by regressing ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 71

it with catch per unit effort of the winter fishery, but results showed very poor correlation between the two. The Department also compared data from the creel survey with the abundance of young of the year (YOY) rainbow smelt col-lected via a seine survey that was initiated in 1997. This comparison resulted in a much stronger correlation with age-2 smelt CPUE from the creel survey. The Department discontinued egg deposition surveys in 2006 as a result of poor data correlation with other surveys, but will continue to monitor rainbow smelt through juvenile abundance surveys, creel surveys, as well as spawning surveys at the fyke net index stations that were implemented for this project.

HabitatRestoration Improving water quality in the Great Bay Estuary is expected to benefit smelt using New Hampshire waters. An increase in the concentration of dis-solved nutrients and substantial increases in nutrient loading have been de-tected in the estuary in recent years. These observations prompted the New Hampshire Department of Environmental Services (NH DES) and the U. S.

Environmental Protection Agency to develop nutrient criteria for the estuary.

Applying these criteria will result in water quality being classified as impaired in the entire estuary, including all of its tributaries. These noted nutrient increases have the potential to spur periphyton growth, which may reduce the viability and hatching of smelt eggs, as discussed in section 2.2 - Threats to Embryonic Survival and Development. The current nutrient criteria assessment is motivat-ing local action to reduce nutrient loading, which should result in improved water quality and reduced periphyton during the smelt spawning season.

Habitat assessment and restoration are key conservation strategies that will be pursued in New Hampshire to enhance spawning conditions for smelt.

While main stem spawning habitats are well known in the major tributaries to Great Bay, a comprehensive assessment of other potential spawning locations in smaller tributaries would be beneficial. Habitat improvement projects that would benefit smelt include mitigating siltation and removing head-of-tide dams to increase the amount of freshwater area available for spawning. Cur-rently most spawning in New Hampshire occurs in intertidal areas. Intertidal bars have developed in some tributaries following recent flood events; smelt eggs are deposited on these rocky bars and are then exposed to air at low tide.

Grading of these bars to minimize their intertidal exposure would reduce egg mortality.

In addition, head-of-tide dams currently block smelt migration on most of the major tributary rivers to Great Bay. One of these obstructions has recently been removed; the dam in place for 55 years on the Winnicut River in Green-land, NH, was recently demolished, restoring spawning habitat for smelt.

Following the dam's construction in 1957, there was a steady decline of a once well-known large smelt run. Other head-of-tide dams in the Great Bay Estuary are under consideration for removal. The potential benefits to smelt will be a key factor in deliberations about the future options for these dams.

Finally, siltation in some rivers has reduced smelt spawning habitat. Dam removal should increase stream flows and help remove accumulated sediments, and actions to reduce nutrient inputs will also reduce sediment inputs to the Great Bay Estuary and its tributaries. These actions should improve smelt spawning habitat conditions in the tributaries.

72 e ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Recommendations:

1) Continue monitoring efforts in place including: winter creel survey, juvenile abundance seine survey, spring spawning run fyke net sampling

2) Improve water quality and support NH DES in developing nutrient criteria for Great Bay Estuary

3) Identify habitat restoration projects to enhance smelt spawning conditions.

4) Continue to support dam removal projects to connect smelt to historical spawning habitats

5) Conduct a smelt spawning habitat assessment of coastal areas in New Hampshire.

Maine Through this project, we have found that while rainbow smelt populations are contracting rapidly in range, there are still strong populations in Maine.

However, our surveys have also shown that smelt populations in the state are not as strong as previous Department studies have found. Comparing the num-ber and strength of spawning runs currently to that of the late 1970's, we have found that many runs have declined, while others are extirpated (see section 1.3 - Population Status). Data collected during our fyke net survey and creel surveys has also shown that length at age has declined compared to historical records in upper Casco Bay and Kennebec River populations. Because smelt continue to support an economically important and sizable recreational fishery in Maine, as well as a locally economically important commercial fishery in Washington County, it is imperative to pursue management measures that will sustain and restore this species.

Continue monitoringsmelt populations at multiple life stages The state surveys that are currently in place target four important life history stages for rainbow smelt. The annual fyke net survey, which began in 2008, monitors the adult spawning runs at six index sites spanning the Maine coast. From this survey, we collect information about the inter-annual variabil-ity of the spawning stock, the strength of age classes, and mortality rates. The genetic information combined with movement and habitat studies show that while adult smelt may not home to the same stream each year, they do show fidelity to larger bay and estuary areas. Thus, by monitoring adult smelt during the spawning season, we can observe changes in a specific stock over time. The other surveys do not have this ability. While the inshore trawl survey can track relative population abundance over time, it likely catches mixed genetic stocks and annual CPUEs may be skewed by stock variability.

The creel survey that targeted the Kennebec River and Merrymeeting Bay beginning in 2009 was expanded with the help of the Downeast Salmon Federation in 2010 to survey anglers on the Pleasant and Narraguagus rivers.

Flagg (1984) estimated an extraction rate of less than 5% on the Kennebec River in the late 1970s. However, the population during that time period was likely larger than at present (see section 1.3 - Population Status in the Gulf of Maine); the fishery may have a more significant effect when population levels are low. Given the cultural and economic value of these fisheries, the creel ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN - 73

survey should be expanded to target aggregations of fishing camps in other locations (e.g., Great Salt Bay on the Damariscotta River), and efforts should be made to repeat the mark-recapture survey performed by Flagg (1984) to deter-mine a current extraction rate.

The juvenile abundance survey is extremely important in understanding the reproductive success and early life stage survival in the Kennebec River and Merrymeeting Bay. Because we also monitor adult populations in this river system through creel surveys, it may be possible to compare data from the two surveys to quantitatively link adult winter catches to late summer juvenile abundance as NHF&G has been able to do. Additionally, by further under-standing how juvenile abundance varies between river segments, we may be able to identify important juvenile habitat.

Improving connectivity and access to spawning grounds In many locations where smelt runs have historically declined or disap-Local smelt runs may be peared on the Maine coast, the decline is due to the inability of smelt to reach affected by a combina- the spawning grounds. Road crossings on small coastal streams are often tion of factors, including provided by undersized or hanging culverts or by small historic water control habitatdegradation, dams that no longer have purpose. Undersized culverts present problems when access problems, and velocities increase during rain events because the water is constricted to a width smaller than the natural streambed. Because smelt are not strong swimmers, current fishing practices.

high water velocities can impede their ability to swim through the culvert, and thus to reach their spawning grounds. Hanging culverts (those where the downstream water level is lower than the culvert height) and dams that are downstream of the spawning grounds completely block access. Unlike other anadromous fishes (e.g., alewife and salmon) that can ascend fish ladders or jump vertical obstructions, smelt are unable to pass vertical obstructions over six inches.

State agencies in Maine, including ME DMR, are currently working to catalogue such obstructions and prioritize which should be removed or rede-signed to allow for anadromous fish passage. As part of this effort, a web-based tool will be publicly available so that municipalities and land trust organizations can identify road crossings in their area where improvements could re-establish smelt habitat access. In many cases, removing these barriers can have immedi-ate effects in opening smelt spawning passage into a stream when strong runs exist nearby. If this is not the case, stock enhancement may be considered in the absence of other habitat degradation. The ME DMR will continue to work with other state agencies, municipalities, and non-governmental organizations to identify barriers to historical smelt habitat and restore access.

Assessing causesfor local decline Some smelt populations in Maine have declined or become extirpated, while others remain strong. In some cases, local declines can be attributed to historical overfishing; however, habitat degradation, access problems, and cur-rent fishing practices may also be impacting smelt populations in the state.

Effective stormwater management techniques can reduce the impact of de-velopment on water quality in urbanized watersheds in the state. As an exam-ple, the Maine Department of Environmental Protection has worked with the South Portland Water District and businesses within the Long Creek watershed 74 - ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

to build stormwater retention areas that reduce the amount of nutrients and contaminants flowing directly into the stream. While the stream quality still shows the effects of development, impairment is reduced and the stream is able to support a limited smelt spawning run. Because this regional smelt project has found that development within a watershed can impact water quality to the point where smelt embryonic health and survival are impaired, watershed management efforts that reduce runoff into receiving streams are recommended in urbanized or developing watersheds.

Current fishing regulations regarding anadromous rainbow smelt limit take by season and location. Recreational fishing is allowed July 1 through March 14; there is no catch limit, but the gear is restricted to hook and line or dip net. During the spawning season (March 15 through June 30), take is limited to two quarts per person per day, and it is predominantly a dip net fishery.

While the state Marine Patrol does actively enforce this regulation regarding gear and catch limitations, the number of violations that go without reprimand is unknown. Further, it is currently unknown what impact the recreational fishery may have on smelt populations. With the creel survey of the ice fishery beginning again in 2009, the ME DMR now has the opportunity to assess the extraction rate of the winter fishery and determine if a limit on take is neces-sary. However, at this point there is no survey of the spring dip net fishery; the effect of fishing mortality during the spawning season and the subsequent loss of possible embryos is unknown. Future work should include an effort to quan-tify fishing mortality due to both the recreational winter and spring fishery. In locations where there is evidence of stressed smelt runs, management action should be considered to limit mortality during spawning runs.

Commercial fishing for smelt is allowed in only six tidal rivers in the state, all in Washington County: the East Machias, Pleasant, and Narraguagus rivers from January 1 through April 10, without any limit on quantity; and the Indi-an, Harrington, and Chandler rivers with no limit on quantity or time period.

Anyone fishing commercially for smelt must possess a Pelagic License from the ME DMR. With possession of this license, the fisherman is required to submit landings data to the ME DMR. The ME DMR is working with Downeast Salmon Federation to survey the biological composition of the catches to determine if the fishery may be impacting life history or age structure. This collaboration is necessary to monitor the fishery, and should continue in the fu-ture. If over time there is evidence of smelt population decline in this region or evidence that the commercial fishery may be contributing to a high mortality, management actions should address the fishing effort possibly by limiting take or further gear restrictions.

Marked larvalstocking at monitored sites As part of this project, the ME DMR revisited historical spawning runs to document their current status and found that many sites no longer support spawning or support only limited runs (see section 1.3 - Population Status in the Gulf of Maine). When the decline at these sites can be attributed to historical fishing pressure that no longer exists or to habitat degradation or pas-sage constraints that have been addressed, larval stocking may be an option to reintroduce smelt.

Adapting methods by Ayer et al. (2012), the ME DMR began a project ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN a 75

to restore rainbow smelt populations to North Haven, Maine, an island in the center of Penobscot Bay that supported robust smelt populations up until the 1950s. After visits by ME DMR to identify the most appropriate stream for the project, the North Haven Community School completed pre-monitoring and found no water quality impairments that would affect smelt embryo survival.

In spring 2012, the ME DMR and school worked together to mark larvae with oxytetracycline (OTC) for release at the stream. The school and ME DMR will continue to monitor adult returns in subsequent years to determine the suc-cess of the project. Following this model, the ME DMR hopes to continue to re-establish smelt populations at sites where restoration projects have improved habitat quality or connectivity. However, habitat restoration must always pre-cede any stocking efforts.

Recommendations With continued population monitoring and threat assessment in collabora-With continued tion with fisheries managers, university scientists, recreational and commercial populationmonitoring fishermen, and interested citizens, the rainbow smelt populations in Maine and threatassessment in I could be maintained or possibly expanded. To this end, the ME DMR has begun to implement restoration efforts, including a stocking project in North collaboration with Haven and assessment of culvert replacements that would provide access to fisheries managers, historical habitat. Future work in the state of Maine to protect this species of university scientists, concern should include:

recreationaland 1) Continuing monitoring of smelt populations through fyke net sampling, commercialfishermen, creel surveys, the inshore trawl survey, and the juvenile abundance survey and interestedcitizens, 2) Developing a mark-recapture study to estimate the current extraction rate rainbow smelt popula- of recreational ice fishing on the Kennebec River and Merrymeeting Bay tions could be maintained and other rivers and embayments that support recreational ice fishing or possibly expanded. 3) Restoring stream connectivity and access to historical spawning grounds with monitoring to assess pre- and post-construction conditions and smelt populations

4) Assessing threats to smelt habitat and evaluating connections between degraded habitat and local smelt population decline

5) Stocking rainbow smelt larvae marked with oxytetracycline into historical smelt spawning streams that maintain good habitat, while maintaining the genetic structure as identified by this project and annually monitoring stocking success.

76 ° ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

LITERATURE CITED Allan, J.D. 2004. Landscapes and riverscapes: the influence of land use on stream ecosystems. Annual Review of Ecology, Evolution, and Systematics 35:257-284.

ASA A&C (ASA Analysis and Communication). 2010. 2008 year class report for the Hudson River Estuary Monitoring Program. Prepared for Dynegy Roseton L.L.C.

by ASA Analysis and Communication, Washingtonville, NY ASMFC (Atlantic States Marine Fisheries Commission). 2011. Striped Bass Stock Assessment Update. 132 pp. www.asmfc.org Ayer, M.A., S.P. Elzey, and B.C. Chase. 2012. Laboratory marking of anadromous rainbow smelt embryos and larvae and the implications for restoration. Pages 60-62 in The Fourth North American Workshop on Rainbow Smelt: Extended Abstract Proceedings. Massachusetts Division of Marine Fisheries, Technical Report TR-5 1.

Bigelow, H.B., and W.C. Schroeder. 1953. Fishes of the Gulf of Maine. U.S. Fish and Wildlife Service Fishery Bulletin 53:1-577.

Biggs, B.J.E 1996. Patterns in benthic algae of streams. Pages 21-56 in Stevenson, R.J., M.L. Bothwell, and R.L. Lowe, editors. Algal Ecology: Freshwater Benthic Ecosystems. Academic Press, San Diego, California.

Booth, D., and L. Reinelt. 1993. Consequences of urbanization on aquatic systems:

measured effects, degradation thresholds, and corrective strategies. Pages 545-550 in Proceedings Watershed '93 A National Conference on Watershed Management.

Alexandria, Virginia.

Bouchard, D. 2010. August 2010 Executive Summary for Atlantic Sturgeon and Rainbow Smelt Pathology Analysis. Maine Department of Marine Resources Cooperative Agreement # CT-I 3A-20090126*4283.

Bradbury, I.R., K. Gardiner, P.V.R. Snelgrove, S.E. Campana, P. Bentzen, and L.

Guan, (2006b). Larval transport, vertical distribution, and localized recruitment in anadromous rainbow smelt (Osmerus mordax). Canadian Journal of Fisheries and Aquatic Sciences 63:2822-2836.

Brenner, EJ., and J.J. Mondok. 1995. Nonpoint-source pollution potential in an agricultural watershed in northwestern Pennsylvania. Water Resources Bulletin 31(6):1 101-1112.

Buckley, J.L. 1989. Species Profiles: Life Histories and Environmental Requirements of Coastal Fishes and Invertebrates (North Atlantic); Rainbow Smelt. U.S. Fish and Wildlife Service Biological Report 82(11.106). U.S. Army Corps of Engineers, TR EL-82-4. I1pp.

Burcher, C.L., H.M. Valett, and E.E Benfield. 2007. The land-cover cascade:

relationships coupling land and water. Ecology 88(l):228-242.

Carpenter, S.R., N.E Caraco, D.L. Correll, R.W. Howarth, A.N. Sharpley, and V.H.

Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications 8(3):559-568.

Chase, B.C. 1990. Summary report for the North Coastal Basin. Massachusetts Division of Marine Fisheries, Massachusetts Bay Smelt Spawning Habitat Program, STAP Doc.90-102.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 77

Chase, B.C. 2006. Rainbow smelt (Osmerus mordax) spawning habitat on the Gulf of Maine Coast of Massachusetts. Massachusetts Division of Marine Fisheries, Technical Report TR-30.

Chase, B.C. 2010. Quality Assurance Program Plan (QAPP) for Water Quality Measurements Conducted for Diadromous Fish Habitat Monitoring, Version 1.0, 2008-2012. Massachusetts Division of Marine Fisheries, Technical Report TR-42.

Chase, B.C., M.H. Ayer, K.J. MacGowan, and K. Taylor. 2006. Population Indices of Rainbow Smelt Spawning Runs in Massachusetts. Massachusetts Division of Marine Fisheries, Completion Report for NMFS Project No: O-2003-NER4, Gloucester, MA.

Chase, B.C., and A.R. Childs. 2001. Rainbow smelt (Osmerus mordax) spawning habitat in the Weymouth-Fore River. Massachusetts Division of Marine Fisheries, Technical Report TR-5.

Chase, B.C., C. Enterline, K. Mills, M.H. Ayer, S.P Elzey, J. Fischer, S.J. Correia, and C.H. Wood. 2012. Rainbow smelt (Osmerus mordax) spawning population moni-toring on the Gulf of Maine coast of New England. Pages 13-15 in The Fourth North American Workshop on Rainbow Smelt: Extended Abstract Proceedings. Massachusetts Division of Marine Fisheries, Technical Report TR-51.

Chase, B.C., J.H. Plouff, and M. Gabriel. 2008. An evaluation of the use of egg trans-fers and habitat restoration to establish an anadromous rainbow smelt (Osmerus mordax) spawning population. Massachusetts Division of Marine Fisheries, Techni-cal Report TR-33.

Chapman, D.G., and D.S. Robson. 1960. The analysis of a catch curve. Biometrics 16:354368.

Clayton, G.R. 1976. Reproduction, first year growth, and distribution of anadro-mous rainbow smelt, Osmerus mordax, in the Parker River and Plum Island Sound estuary, Massachusetts. Master's Thesis. University of Massachusetts, Amherst, Massachusetts.

Clayton, G.R., C.E Cole, S.A. Murawski, and J.D. Parrish. 1978. Common marine fishes of coastal Massachusetts. University of Massachusetts, Amherst, Massachusetts.

Collete, B.B., and G. Klein-MacPhee. 2002. Bigelow and Schroeder's Fishes of the Gulf of Maine. Smithsonian Institution Press, Washington, D.C.

Cooper, C. 1996. Hydrologic effects of urbanization on Puget Sound lowland streams.

Master's Thesis. University of Washington, Seattle.

Corbett, C.W, M. Wahl, D.E. Porter, D. Edwards, and C. Moise. 1997. Nonpoint source runoff modeling: a comparison of a forested watershed and an urban watershed on the South Carolina coast. Journal of Experimental Marine Biology and Ecology 213:133-149.

Crawford, J.K., and D.R. Lenat. 1994. Effects of Land Use on the Water Quality and Biota of Three Streams in the Piedmont Province of North Carolina. Hydrobiolo-gia. 294:185-199.

Crestin, D.S. 1973. Some aspects of the biology of adults and early life stages of the rainbow smelt, Osmerus mordax (Mitchill), from the Weweantic River Estuary, Wareham-Marion, Massachusetts, 1968. Master's Thesis. Boston University, Boston, Massachusetts.

Dauvin, J.C., and J.J. Dodson. 1990. Relationship between feeding incidence and vertical and 644 longitudinal distribution of rainbow smelt larvae (Osmerus mordax) in a turbid well-mixed estuary. Marine Ecology Progress Series 60:1-12.

Davis, J.P., E.T. Schultz, and J. Vokoun. 2009. Assessment of river herring and striped 78 e ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

bass in the Connecticut River: abundance, population structure, and predator/prey interactions. Ecology & Evolutionary Biology Articles Paper 26.

http://digitalcommons.uconn.edu/eebarticles/26 DFO (Department of Fisheries and Oceans, Canada). 2011. Integrated Smelt Fishery Management Plan, Eastern New Brunswick Area, Gulf Region, 2007-2011. DFO IFMP, New Brunswick, Canada. http://www.glf.dfo-mpo.gc.ca/e0008079 Dodds, W.K. 2007. Trophic state, eutrophication and nutrient criteria in streams.

TRENDS in Ecology and Evolution Vol.22 No. 12.

Eayrs, S., N. Stevens, D. Salerno, and M. Raymond. 2009. A contemporary assessment of the bycatch of regulated species and the Nordmore Grate in the northern shrimp fishery. Northeast Consortium Final Report PZ09020.

http://www.northeastconsortium.org/pdfs/awards 2007/Eayrs%2007/Eayrs%20 07%20Final%2OReport.pdf Ellis, D., and J.C. Vokoun. 2009. Earlier spring warming of coastal streams and implications for alewife migration timing. North American Journal of Fisheries Management 29:1584-1589.

Finn, R.N. 2007. Review. The physiology and toxicology of salmonid eggs and larvae in relation to water quality criteria. Aquatic Toxicology 81:337-354.

Fiske, J.D., J.R. Curley, and R.P. Lawton. 1968. A study of the marine resources of the Westport River. Massachusetts Division of Marine Fisheries, Monograph Series No. 7.

Flagg, L.N. 1974. Distribution, food habits, and management of striped bass and smelt in Maine coastal waters. Completion report AFS-4, Maine Department of Marine Resources.

Flagg, L.N. 1983. Evaluation of anadromous fish resources. Final Report AFS-2 l-R, Maine Department of Marine Resources.

Fried, H.A., and E.T. Schultz. 2006. Anadromous rainbow smelt and tomcod in Connecticut: Assessment of populations, conservation status, and need for restoration plan. Master's thesis. University of Connecticut, Storrs.

Friedland, K.D., J.P. Manning, J. Link, J.R. Gilbert, A.T. Gilbert, and A.E O'Connell, Jr. 2012. Variation in wind and piscivorous predator fields affecting the survival of Atlantic salmon, Salmo salar,in the Gulf of Maine. Fisheries Man-agement and Ecology 19:20-25.

Frumhoff, P.C., J.J. McCarthy, J.M. Melillo, S.C. Moser, and D.J. Wuebbles. 2007.

Confronting Climate Change in the U.S. Northeast: Science, Impacts, and Solutions. Synthesis report of the Northeast Climate Impacts Assessment (NE-CIA). Cambridge, MA: Union of Concerned Scientists (UCS). http://www.north-eastclimateimpacts.org/pdf/confronting-climate-change-in-the-u-s-northeast.pdf Fuda, K.M., B.M. Smith, M.P. Lesser, B.J. Legare, H.R. Breig, R.B. Stack, and D.L.

Berlinsky. 2007. The effects of environmental factors on rainbow smelt Osnierus mordax embryos and larvae. Journal of Fish Biology 71:539-549.

Geffen, A.J. 1990. Response of rainbow smelt, Osmerus mordax (Mitchill), eggs to low pH. Journal of Fish Biology 37:865-871.

Goode, G.B. 1884. The Fisheries and Fishery Industries of the United States. U.S.

Commission of Fish and Fisheries, Washington D.C.

Greene C.H., B.C. Monger, L.P. McGarry, M.D. Connelly, N.R. Schnepf, A.J. Persh-ing, L.M Belkin, P.S. Fratantoni, D.G. Mountain, R.S. Pickart, R. Ji, J.J. Bisagni, C. Chen., S.M.A. Hakkinen, D.B. Haidvogel, J. Wang, E. Head, P. Smith, and A. Conversi. 2012. Recent Arctic Climate Change and its Remote Forcing of Northwest Atlantic Shelf Ecosystems. Oceanography 25:2.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 79

Hall, C.J., A. Jordaan, and M.G. Frisk. 2012. Centuries of anadromous forage fish loss: consequences for ecosystem connectivity and productivity. BioScience 62(8):723-731.

Hodgkins, G.A., R.W. Dudley, and T.G. Huntington. 2003. Changes in the timing of high river flows in New England over the 20th century. Journal of Hydrology 278:244-252.

Howarth, R., D. Anderson, J. Cloern, C. Elfring, C. Hopkinson, B. Lapointe, T.

Malone, N. Marcus, K. McGlathery, A. Sharpley, and D. Walker. 2000. Nutrient pollution of coastal rivers, bays, and seas. Issues in Ecology 7:1-15.

Howell, W.H., and R. Langan. 1992. Discarding of commercial groundfish species in the Gulf of Maine shrimp fishery. North American Journal of Fisheries Manage-ment 12:568-580.

Hurlbert, P.J. 1974. Factors affecting spawning site selection and hatching success in anadromous rainbow smelt (Osmerus mordax), Mitchell. Master's Thesis. University of Maine, Orono.

Incze, L., H. Xue, N. Wolff, D. Xu, C. Wilson, R. Steneck, R. Wahle, P. Lawton, N.

Pettigrew, and Y Chen. 2010. Connectivity of lobster (Homarus americanus) populations in the coastal Gulf of Maine: part II. Coupled biophysical dynamics.

Fisheries Oceanography 19:(1) 1-20.

Jezierska, B., K. Lugowska, and M. Witeska. 2009. The effects of heavy metals on embryonic development of fish (a review). Fish Physiology and Biochemistry 35:625-640.

Jimenez, D., J.E. Pelczarski, and H.R. lwanowicz. 1982. Incidence of piscine erythro-cytic necrosis (PEN) and Glugea hertwigi (Weissenberg) in rainbow smelt (Osmerus mordax) in selected Massachusetts streams. Estuaries 5(2):145-149.

Juanes, E, S. Gephard, and K.E Beland. 2004. Long-term changes in migration timing of adult Atlantic salmon (Salmo salar)at the southern edge of the species distribution. Canadian Journal of Fisheries and Aquatic Sciences 61:2392-2400.

Kendall, WC. 1926. The smelts. Bulletin of the Bureau of Fisheries, Vol. XLII, Doc.

No. 1015.

Kircheis, EW, and J.G. Stanley. 1981. Theory and practice of forage-fish management in New England. Transactions of the American Fisheries Society 110:729-737.

Kirn, R.A. 1986. Assessment of rainbow smelt (Osmerus mordax) as forage in Lake Champlain. Master's thesis. University of Vermont, Burlington.

Klein, R.D. 1979. Urbanization and stream quality impairment. Water Resources Bulletin 15(4):948-962.

Kovach, A.I., T.S. Breton, C.L. Enterline, and D.L. Berlinsky. In press. Identifying the spatial scale of population structure in anadromous rainbow smelt (Osmerus mordax). Fisheries Research.

Lapierre, L., C. Cormier, G. Trencia, and G. Verreault. 1999. Periphyton of four St.

Lawrence estuary tributaries used for rainbow smelt spawning. Pages 33-38 in First North American Workshop on Rainbow Smelt, Quebec.

Laprise, R., and J.J. Dodson. 1989. Ontogeny and importance of tidal vertical migra-tions in the retention of larval smelt Osmerus mordax in a well-mixed estuary.

Marine Ecology Progress Series 55:101-111.

Lawton, R., P. Brady, C. Sheehan, S. Correia, and M. Borgatti. 1990. Final report on spawning sea-run rainbow smelt (Osmerus mordax) in the Jones River and impact assessment of Pilgrim Station on the population, 1979-1981. Pilgrim Nuclear Power Station Marine Environmental Monitoring Program, Report Series No. 4, 72 pp.

80 ° ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Limburg, K.E., and R.E. Schmidt. 1990. Patterns of fish spawning in Hudson River tributaries: response to an urban gradient? Ecology 71(4):1238-1245.

Limburg, K.E., and J.R. Waldman. 2009. Dramatic declines in North Atlantic diadromous fishes. BioScience 59:955-965.

Marcotte, A., and J.L. Tremblay. 1948. Notes sur la biologie de l'6perlan (Osmerus mordax Mitchell) de la Province de Qu~bec. Sta. Biol. Saint-Laurent, Que.

Contrib. 18:107 pp.

Martin, E.H., and C.D. Apse. 2011. Northeast Aquatic Connectivity: An Assessment of Dams on Northeastern Rivers. The Nature Conservancy, Eastern Freshwater Program. http://cw-environment.usace.army.mil/restore/fishpassage/pdfs/NEA-quaticConnectivityReport.pdf MassDEP (Massachusetts Department of Environmental Protection). 2007.

Massachusetts Surface Water Quality Standards. Massachusetts Department of Environmental Protection, Division of Water Pollution Control, Technical Services Branch, Westborough, MA (Revision of 314 CMR 4.00, January 2007).

MCFG (Massachusetts Commissioners on Fisheries and Game). 1917. Fifty-first an-nual report, for the year 1916. Boston, MvA.

McKenzie, R.A. 1958. Age and growth of smelt, Osmerus mordax (Mitchill), of the Miramichi River, New Brunswick. Journal of the Fisheries Research Board of Canada 15(6):1313-1327.

McKenzie, R.A. 1964. Smelt life history and fishery of the Miramichi River, New Brunswick. Journal of the Fisheries Research Board of Canada, Bulletin 144.

MECSSF (Maine Commission of Sea and Shore Fisheries). 1920. Second Biennial Report of the Commission of Sea and Shore Fisheries of the State of Maine. Auburn, Maine.

ME DMR (Maine Department of Marine Resources). 2011. Boothbay Harbor Environmental Data. http://www.maine.gov/dmr/rm/environmentaldata.html Mitchell, D.F, M. Liebman, L. Ramseyer, and B. Card. 2003. Riffles and Reservoirs

- nutrient criteria and downstream effects. Proceedings of the Water Environment Federation, National TMDL Science and Policy, 22:572-592.

Moravec, F, and I. De Buron. 2009. New data on three gonad-infecting species of Philometra (Nematoda, Philometridae)from estuarine fishes in South Carolina, USA. Acta Parasitologica 54(3):244-252.

Murawski, S.A., and C.F Cole. 1978. Population dynamics of anadromous rainbow smelt Osmerus mordax, in a Massachusetts river system. Transactions of the American Fisheries Society 107:535-542.

Murawski, S.A., G.R. Clayton, R.J. Reed, and C.H. Cole. 1980. Movements of spawning rainbow smelt, Osmerus mordax, in a Massachusetts Estuary. Estuaries 3(4):308-314.

NEFSC (Northeast Fisheries Science Center). 2010. Gray Seal (Halichoerusgrypus grypus): Western North Atlantic Stock in U.S. Atlantic and Gulf of Mexico Marine Mammal Stock Assessments. NOAA Technical Memorandum NMFS-NE-219.

NEFSC (Northeast Fisheries Science Center). 2012. Northeast Fisheries Observer Program. http://www.nefsc.noaa.gov/fsb/program.html NHF&G (New Hampshire Fish and Game Department). 1981. Fisheries manage-ment plan for rainbow smelt (Osmerus mordax). New Hampshire Fish and Game Department, Concord, NH.

NJCF (New Jersey Commissioner of Fisheries). 1872. Second Annual Report of the Commissioners of Fisheries of the State of New Jersey. Trenton, New Jersey.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 81

NJCF (New Jersey Commissioner of Fisheries). 1886. Report of the New Jersey Fish Commissioners to the One Hundred and Tenth Legislature of the State of New Jersey. Vol. III, Camden, New Jersey.

NOAA. 2012. http://www.nefsc.noaa.gov/fsb/program.html Norris, T. 1862. Remarks on a species of Osmerus taken in the Schuykill, below Fairmont Dam. Proceedings, Academy of Natural Sciences of Philadelphia, 1861 (1862), pp. 58-59, Philadelphia, PA.

Nsembukya-Katuramu, S., E.K. Balon, and R. Mahon. 1981. A comparison of spawning, harvested, and die-off rainbow smelt, Osmerus mordax, in eastern Lake Erie. Journal of Great Lakes Research 7(2):144-154.

O'Gorman, R., R. Bergstedt, and T.H. Eckert. 1987. Prey fish dynamics and salmo-nine predator growth in Lake Ontario 1974-1984. Canadian Journal of Fisheries and Aquatic Sciences 44(Supplement 2):390-403.

Pess, G.R., D.R. Montgomery, E.A. Steel, R.E. Bilby, B.E. Feist, and H.M.

Greenberg. 2002. Landscape characteristics, land use, and coho salmon (Oncorhynchus kisutch) abundance, Snohomish River, Washington, USA.

Canadian Journal of Fisheries and Aquatic Sciences 59:613-623.

Pettigrew, P. 1997. Suivi de la reproductio de L'eperlan arc-en-ciel de la rive sud de r'Estuaire du Saint-Laurent en 1994, 1995, et 1996. Ministere de l'Env. et de la Faune, Direction regionale du Bas-Saint-Laurent, Riviere-du-Loup, Fevrier, 52 pp.

Pettigrew, N.R., D.W. Townsend, H. Xue, J.P.Wallinsa, P.J. Brickley, and R.D.

Hedand. 1998. Observations of the Eastern Maine Coastal Current and its offshore extensions in 1994. Journal of Geophysical Research 103:30623-30640.

Pettigrew, N.R., J.H. Churchill, C.D. Janzen, L.J. Mangum, R.P. Signell, A.C.

Thomas, D.W. Townsend, J.P. Wallinga, and H. Xue. 2005. The kinematic and hydrographic structure of the Gulf of Maine Coastal Current. Deep Sea Research 1152:2369-2391.

Pouliot, G. 2002. Dynamique de la population d'eperlans arc-en-ciel (Osmerus mordax) du sud de l'estuaire du Saint-Laurent par l'analyse des cohortes de reproducteurs frequentant la riviere Fouquette entre 1994 et 2001. Societe de la faune et des parcs du Quebec, Direction de l'amenagement de la faune de la region du Bas-Saint-Laurent. 47 p.

Reback, K.E., and J.S. DiCarlo. 1972. Completion Report on the anadromous fish project. Massachusetts Division of Marine Fisheries, Publication No. 6496:113 p.

RIDFW (Rhode Island Division of Fish and Wildlife). 1971. Restoration and Estab-lishment of Sea Run Fisheries. Rhode Island Division of Fish and Wildlife, Federal Aid in Fish Restoration, Project F-26-R-6, Job Completion Report.

RIDFW (Rhode Island Division of Fish and Wildlife). 1978. Restoration and Estab-lishment of Sea Run Fisheries. Rhode Island Division of Fish and Wildlife, Federal Aid in Fish Restoration, Project F-26-R-13, Job Completion Report.

Rupp, R.S. 1959. Variation in the life history of the American smelt in inland waters of Maine. Transactions of the American Fisheries Society 88:241-252.

Rupp, R.S. 1965. Shore-spawning and survival of eggs of the American smelt, Osmerus mordax (Mitchill), in Maine. Transactions of American Fisheries Society 90 (1):42-48.

Rupp, R.S. 1968. Life history and ecology of the smelt (Osmerus mordax) in the inland waters of Maine. Maine Department of Inland Fish Game, Federal Aid in Fish Restoration, Project F- 10-R, Final Report.

82

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

Saunders, R., M.A. Hachey, and C.W Fay. 2006. Maine's diadromous fish community: past, present, and implications for Atlantic salmon recovery. Fisheries 31(11):537-547.

Schwartz, M.K., L.S. Mills, Y. Ortega, L.E Ruggiero, and EW. Allendorf. 2003.

Landscape location affects genetic variation of Canada lynx. Molecular Ecology 12:1807-1816.

Scott, W, and M. Scott. 1988. Atlantic Fishes of Canada. Canadian Bulletin of Fisheries and Aquatic Sciences 219:731 pp.

Siegel, S. and N.J. Castellan. 1988. Nonparametric Statistics for the Behavioural Sciences, 2nd edition. McGraw-Hill, New York.

Simmons, D. and R. Reynolds. 1982. Effects of urbanization on baseflow of selected South-Shore streams, Long Island, NY Water Resources Bulletin 18(5):797-805.

Sirois, P., and J.J. Dodson. 2000. Critical periods and growth-dependent survival of larvae of an estuarine fish, the rainbow smelt Osmerus mordax. Marine Ecology Progress Series 203:233-245.

Spalding, M.D., H. Fox, G. Allen, N. Davidson, Z. Ferdafia, M. Finlayson, B.

Halpern, M. Jorge, A. Lombana, S. Lourie, K. Martin, E. Mcmanus, J. Molnar, C. Recchia, and J. Robertson. 2007. Marine Ecoregions of the World: A Biore-gionalization of Coastal and Shelf Areas. BioScience 57(7):573-583.

Squires, T.S., L. Flagg, and L. Austin. 1976. Smelt Management Plan. Maine Depart-ment of Marine Resources, Completion Report Project #AFSC-13/FWAC-2.

Stewart, D.J., J.E Kitchell, and L.B. Crowder. 1981. Forage Fishes and Their Salmonid Predators in Lake Michigan. Transactions of the American Fisheries Society 110:751-763.

Storer, H.R. 1858. Fishes of Massachusetts. Memoirs of the American Academy of Sciences United States Government Printing Office.

Strayer, D.L., R.E. Beighley, L.C. Thompson, S. Brooks, C. Nilsson, G. Pinay, and R.J. Naiman. 2003. Effects of land cover on stream ecosystems: roles of empirical models and scaling issues. Ecosystems 6:407-423.

Sullivan, K. 2007. Monitoring of the rainbow smelt resource and winter ice fishery.

New Hampshire Fish and Game Department, Federal Aid in Fish Restoration, Project F-61-R, Progress Report, Durham.

Sullivan, K. 2009. Rainbow smelt creel survey design and calculation methodologies.

New Hampshire Fish and Game Department, Federal Aid in Fish Restoration, Project F-61-R, Progress Report Appendix 1, Durham.

Sullivan, K. 2010. Monitoring of the rainbow smelt resource and winter ice fishery.

New Hampshire Fish and Game Department, Federal Aid in Fish Restoration, Project F-61-R, Progress Report, Durham.

Trencia, G., G. Verreault, M. Legault, and V. Tremblay. 2005. L' eperlan arc-en-ciel (Osmerus mordax) anadrome du sud de l'estuaire du Saint-Laurent, une population en voie de designation comme espece vulnerable. Le Naturaliste Canadien, Vol.

129 No. 1.

US DOC (U.S. Department of Commerce). 1920. Progress in Biological Inquiries.

Report of the Division of Scientific Inquiry for the Fiscal Year of 1920. Depart-ment of Commerce, Bureau of Fisheries, Doc. No. 896, Washington, D.C.

US EPA (U.S. Environmental Protection Agency). 2000. Nutrient Criteria Technical Guidance Manual. Rivers and Streams. Office of Water, USEPA, Document EPA 822-B-00-OO2Washington, D.C.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN a 83

US EPA (U.S. Environmental Protection Agency). 2004. National water quality inventory. USEPA, Report to Congress, 2004 reporting cycle, Washington, DC.

USFWS (U.S. Fish & Wildlife Service). 2012. http://www.fws.gov/GOMCP/identify.

html#diafish Verreault, G., G. Trencia, and M. Legault. 2012. The rainbow smelt restoration plan for the St. Lawrence estuary: where are we eight years later? Pages 62-64 in Fourth North American Workshop on Rainbow Smelt: Extended Abstract Proceedings, Massachusetts Division of Marine Fisheries, Technical Report TR-5 1.

Visel, T., and T. Savoy. 1989. The historical decline of the rainbow smelt "Osmerus mordax"in Connecticut's coastal waters. Proposal to the University of Connecticut Center for Environmental Health, University of Connecticut Cooperative Exten-sion Service. Reprinted in 2005.

Wang, L., J. Lyons, P. Kanehl, and R. Bannerman. 2001 a. Impacts of urbanization on stream habitat and fish across multiple spatial scales. Environmental Management 28(2):255-266.

Warfel, H.E., T.P. Frost, and WH. Jones. 1943. The smelt, Osmerus mordax, in Great Bay, New Hampshire. Transactions of the American Fisheries Society 72:257-262.

Wegwu, M., and J. Akaninwor. 2006. Assessment of heavy-metal profile of the New Calabar River and its impact on juvenile Clarias gariepinus. Chemistry &

Biodiversity 3:79-87.

Wootton, J.T., C.A. Pfister, and J.D. Forester. 2008. Dynamic Patterns and Ecological Impacts of Declining Ocean pH in a High-Resolution Multi-Year Dataset. Proceedings of the National Academy of Sciences of the United States of America 105(48):18848-18853.

Whitten, D.R. 1894. Commissioner of Sea and Shore Fisheries of the State of Maine, 1894, Augusta, ME, 38 pp.

Wyatt, L.H., A.L. Baker, and D.L. Berlinsky. 2010. Effects of sedimentation and periphyton communities on embryonic Rainbow Smelt, Osmerus mordax. Aquatic Sciences 72:361-369.

Yako, L.A., M.E. Mather, and E Juanes. 2002. Mechanisms for migration of anadro-mous herring: an ecological basis for effective conservation. Ecological Applications 12(2):521-534.

84 - ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

APPENDIX 25 :Weweantic River 20 0- 2008

-dr-2010 0

-"-2011 3,1 3/21 4/10 4/30 5/20 6/9 150 Jones River 100

• 2008

-,-2010 0 4

• 2011 3/1 321 4/10 4/30 5S20 6/9 300 Fore River 4150 16-2009 100 so -d-21 0 - - -- 2011 3J1 3/21 4!10 4(30 5/20 6,9 5oo Parker River 400 o- 20oo 300 200 -2009 100 -2010 FigureA.1.1. Catch-per-unit-effort 0 ..... -X- 2011 (number of smelt per haul) at V'1 3121 4/10 4/30 5i20 6/9 selected Massachusetts fyke net stations,2008-2011.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 85

200 Squamscott River 1*0

• ý 200$

100 0-5-2009 so r-2010 hi4-2011.

0 311 31 3/21 4,10 4130 S"20 6/9 3:

10 M Winnicut River 6i *+ 200$

2009 43

-- 2010 042011 0

121 4/10 4130 5/20 6/9 100 60 Oyster River

-","2010 40

--0-2011 20 FigureA.1.2. Catch-per-unit-effort /

0 (number of smelt perhaul) at 3;21 4;10 4,30 5,,20 6i9 New Hampshirefyke net stations, 2008-2011.

86 0 ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

600 500 Mast Landing 4I+ 2008 400 4 300 -.- 2009 200 AA --- 2010 100

-)- 2011 0 A'.3 3/1 3V21 4/10 4/30 5/20 6,9 70 1500 Deer Meadow Brook 0M X 1000 2008

- ,-2009 500

4. -,-,,-2010 (0

-- 2011 3/" 3/21 4/10 4/30 5/20 619 1500 Tannery Brook 1000 4,* 2003

-- 2009 500

-,"-2010 0- 2011 3!1 3/21 4110 4/30 5/20 6i9 400 East Bay River 300

• - 2008 200 - 2009 100 i S2010 N&AW 2011 FigureA.1.3. Catch-per-unit-effort 3/1 (number of smelt per haul) at 3121 4/10 4/30 5/20 6/9 selected Maine fyke net stations, 2008-2011.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 87

so I 2009 - 2.01, 10 20"noo" FigureA.3-4. Length frequency of rainbowsmelt caught In the Weweantic River, MA, fyke net, 94 0 U 12L 1 ID 2S 2008-2011.

go 2006 2009 .W096 IN- 013 90 NVION 374210)

N4, 40 40 20 0

a 9 Ji 1i.

20 U1 a ll o6 1617 TONI1 nulL..

0 tol U a 0 a 2121 2 L'"lv1 ao) 2 002 27 2 0

1 10 1IO U iii U Ii111 upijilMi -

1

" I Towj 1440 1iM) 1 0 1 22 1 4a 2026 2.7 W0 f Ia -1714 IN 2011 .~fqlO4 01N291 amok1 (N. fil am*41 IN-109 10.

a a~l I111.9

• 6 0I U 12 II14 20 16 7 1* I 1021 22 2 24 2526 27 30 0 9 10 II 12 U* 14 10 1I 17 10 09 1 21 22 21 2 20 2 27 Figure A.1.5. Length frequency of rainbowsmelt caught In the Jones River, MA, fyke net, 2008-2011.

2006 f.m98 041.114271P 2009 ,Fo.1-- (11.14711 am*.J Im41000)

_Jill...Iii 394.6 IN-6"0 No.

.u.Ijj W40 I m I0i _. .. -. AM

64. .

11 9 W10 12 U It 14 16 27 A1 6. Z) 11 UZ 23 24 JS 26 21, a *9 I 1* 12 U!It 14 A2 17V 3 12 20 U1 2*a 24 25 A4 27 2 TotjidLeU, ton) low 'Ifoh tw 2010 UFO-*. IN-am0 WF0odw (0.2271, X00 hu0. tN1.111*2* 31043 IN. 1010 Nd.

100 No.

am A L ~

9 a 1r U IJl Ua 1 lJ-

- U-14 15 U 17 NTOW R

nI OALth

-Um.

20 21 22 2214 a CO A 26 27 R.

N we0

• 1011 U 14 18 s1 1 a0 39 02 22 50 72 Figure A.1.6 Length frequency of rainbow smeit caught In the Fore River, MA, fyke net 2008-2011.

88 a ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

ll40 2*011 zn"Fel (,$

40 2010 40 42 20 20 9 0 WU 2U 14 IS U17 IS It20 Z12 23 Z4 2 5247 n1 ol2* dI11729012 34557Z TMW La~t AM 0411110"II Figure A.1. 7. Length frequency of rainbowsmelt caught at the Oyster River, NH, fyke net, 2010-2011.

so Na.

to Figure A.1.8. Length frequency of 9 S IOU U U 14 m3 is 10 712111920 21 2223 24 252 all7 2 rainbowsmelt caught at the Lam-prey River, NH, fyke net, 2008.

40S 30 No.

22 Figure A.1.9. Length frequency of rainbowsmelt caught at the I R 30 U 12 13 14 IS l6 17 IN 192D221 22 X3 M4 2S 26 22 .9 Squamscott River, NH, fyke net 2011 ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 89 I

40.

200 UFWm. tmN SZ7) 2009 .Ivm*2 jfl-m M0 am0w N4 132) ample 04-5.221 No.

1gW 10 1 9 10 U 12 U 14 15 16 17 IN 2 20 W 2 22 23 24 IS 21 Z? 28 101 0,

9 la 2 U21122

.I1 4 is

. --iM-.A 26 1? 0 is 20 21 U2 32 14 2S 20 27 Is

%WLS2 A04ao) TA LWOOlaiR 20 4W0 2010 2021 82fi...* tN - 76 ý.402. (N.""

220 DIMS 0~211l 3W ilL0 t No.

110 3 2W 50 0e S

.0,1 20 11 12 U M

1.4 IS It muI mo A MA"U U

17 26 19 20 21 nl 2l l4 25 26 27 28 we.

a 9 9 20 94 U It 104 IS1 117IIS 21 20 72 TOWiL-0 w M8[- fm FigureA.1.10. Length frequency of rainbowsmelt caught at Mast Landing, ME, fyke net, 2008-2011.

2100, SOD0 200=

~J~Sa#

fN~7 all 2009

'Fe..* (N.14A 79~ISO E~il 25, 9 IN241 U Ing ai" 12 4 IS 16 12 19 If 29 21 22 23 24 25 26 27 n6 a

1 9 10 22 Ulf0 14122 ly 2)119301 U. 22 22 7224 26 2? 7 Tow Lii(e) 24 2010 Wffnaie IN-1774 ,fem.0I (N-3347 44

• Mast (N-2l21) UM*IO IN26227 4 3W No. x9.

24 200

'I SOD

• 9 20 I 12 13 14 IS 16 1? A1 V 29 0 21 Z2 23 24 2!5 26 27 25 I SOIt1 2 23 14 U5 15 17 26 26 A0 1 27 71 24 S0 2i "? A TOWb"w d~ TOW1040 (CM)

FigureA.1.11. Length frequency of rainbowsmelt caught at Deer Meadow Brook, ME fyke net 2008-2011.

90

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN I

1.0 zoOn W1FOi3 N.43 30 * * (N

• 11Sgl 10 Ii..

4W 4.

• 9 U) I 1.2

,JJil '.

U 14 15 16 11 18 19 20 21 V VAWLaN"I1-U 25 4 25 26, 27 IS 30 10 30 11 12 U 14 15 16 V7 IS 1.9 Z0 U1 U22 Taw3£0340 lIwo 25 24 I 26Z 20 sm 2010 ýWsr4*! p4'?772) 40 *Am.. j".3121 WI.

0ý lIi] 03

.iJi.IL~. -

9 to0U 12 U 14 I"Ewo*0L 1 9 10 11 12 U 14 1 16 10 19 It N0 U1 a 23 14 as 16 22 28 UWUW40"W, FigureA.1.12. Length frequency of rainbowsmelt caught at Tannery Brook, ME, fyke net, 2008-2011.

2W0 NI.(?.Laa

• 1110W11.113222254W222 8f Fft am~ aow Figure A.1.13. Length frequency of rainbowsmelt caught at Schoppee Brook, ME, tyke net, 2010-2011.

20" Z009 5,m311 INAZ1I 1.501 DOOM.(W, S1O) 9M.

no a 9 2.0 1.1 U I 111d.

HA 15 A 1.7 A* " 20 1 U0 Z) 24 25 20 IF 28

-- I 8 20 3 11 U 13

.IA. 1.

N4 15 16 I1 18 U1 20 2a 22 is 24 27 a 21 24 lo efat ".6

50. 40 2011 JFOXVU 00-0#1I no. a".6 (N-691) OW 613440 M-22113 0N Io.

a I, 0 9 to2 U U2 13 14 is 1 1.7 U to3a 2011 22 53 24 25 x6 2? 25 S$ 8 11 12 U3 14 is 16 1?1*Uigu 20r1 AS .1 24 20 20 IV 2o Bok f")20 M, Figure A.1.14. Length frequency of rainbow smelt caught at East Bay Brook, ME, tyk net 2008&2011.

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 91 I

25 20 -

FigureA.2.1. Water temperature data distributionsfor 19 smelt samplingstations In study area.

alil ggIi t 15-The top of the box plots is the 75th percentile and the bottom C.

E Is the 25th percentile. The line within the box Is the median and IT 10 -

the error bars represent the 10th and 90th percentiles. The stations 5

are arrangedon the x-axis from the southernmost MA station to the northernmostME station.

0 Station medians were found to WP WWNJR FR SG NR CN ER PR SQ WR OY LC ML DM TB SB CR EB be significantlydifferent with Kruskai-Wallistest (KW - 93.21, River df - 18, p < 0.001).

1.4-FigureA.2.2. Specific conductiv- 1.2 -

ity data distributionsfor 18 smelt samplingstations in study area.

1.0-The top of the box plots Is the 75th percentile and the bottom U 0.8 is the 25th percentile. The line 09 within the box Is the median and C-o) 0.6-the errorbars representthe loth and 90th percentiles.The stations 0.4-are arrangedon the x-axis from the southernmostMA stationto the northernmostME station. 0.2- T Station medians were found to 0.0 be significantlydifferent with WP WW JR FR SG NR CN ER PR SQ WR OY LC ML DM TB SB EB KruskaI-Wallls test (KW -1374.4, df - 17, p < 0.001). River 92 9 ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

16 FigureA.2.3. Dissolved oxygen (mg/L) data distributionsfor 19 14-T T T T smelt samplingstations In study area. The top of the box plots is 12 - T T T TTTT T T the 75th percentile and the bot-tom Is the 25th percentile. The

-J qHTT T line within the box Is the median 10- and the errorbars representthe E

0 10th and 90th percentiles.The E0 8- stationsare arrangedon the x-axis from the southernmost MA station to the northernmostME station.

6-Station medians were found to be significantlydifferent with 4- KruskaI-Wallis test (KW - 439.51, df - 18, p < 0.001). The green line WP WW JR FR SG NR CN ER PR SQ WR OY LC ML DM TB SB CR EB marks the MassDEPDO criterion River (6.0 mg/L) for protectingAquatic Life.

14U FigureA.2.4. Dissolved oxygen 120 t -- (% saturation)data distributions for 19 smelt samplingstations In study area. The top of the box T T T T plots is the 75th percentile and 0 100 a the bottom Is the 25th percentile.

The line within the box is the me-dian and the errorbars represent 80 the 10th and 90th percentiles.The stationsare arrangedon the x-axis from the southernmost MA station to the northernmostME station.

tjU Station medians were found to WP WI JR FR SG NR CN ER PR SQ WR OY LC ML DM TB SB CR EB be significantlydifferent with River Kruskal-Wallis test (KW - 439.51, df - 18, p < 0.001).

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

• 93 I

Figure A.2.5. Water pH data 9.0 distributionsfor 19 smelt sam-8.5' pling stationsin study area.The top of the box plots is the 75th 8.0-percentile and the bottom Is the 7.5. T T 25th percentile.The line within 7T-the box Is the median and the 7.0-error bars representthe 10th and *. 6.5-90th percentiles.The stations are arrangedon the x-axis from the 6.0-southernmost MA station to the 5.5-northernmostME station.Station medians were found to be sig- 5.0-nflcantly different with Kruskal-Wallis test (KW - 1041.3, df - 18, 4,5 p < 0.001). The green lines mark 40 the lower MassDEPpH criterion PVWV JR FR SG NR CN ER PR SQWROY LO ML DM TB SB CR EB (a6.5 and s 8.3) for protecting River Aquatic Life.

FigureA.2.6. Turbidity (NTU) data distributions for 19 smelt 24 samplingstations in study area. 22 The top of the box plots is the 20 75th percentileand the bottom 18 is the 25th percentile. The line within the box is the median and 16 the errorbarsrepresent the 10th 14 and 90th percentiles. The stations 12 z

are arrangedon the x-axis from 10 the southernmost MA station to the northernmostME station. 8 Station medians were found to 6 be significantlydifferent with 4 Kruskai-Waills test (KW - 660.8, 2

df - 18, p < 0.001). The green line marks the EPA turbiditycriterion 0 WP WV JR FR SG NR CN ER PR SQ WR OY LC ML DM TB SB CR EB for minimally Impacted water quality (:5 1. 7 NTU). River 94

• ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN

2.5 FigureA.2. 7. Total nitrogen (TN) data distributionsfor 20 smelt samplingstationsin study area.

2.0 The top of the box plots is the 75th percentileand the bottom Is the 25th percentile. The line within

-J 1.5 the box Is the median and the

76) errorbars representthe 10th and E

zI-- 90th percentiles. The stations are 1.0 arrangedon the x-axis from the southernmostMA station to the T T northernmostME station. Station 0.5- medians were found to be sig-nificantly different with Kruskal-Wallis test (KW - 408.4, df - 19, 0.0 p < 0.001). The green line marks WPWWW JR FR SG NR CN ER MR PR SQ WR OY LC ML DM TB SB CR EB the EPA total nitrogen criterion for minimally Impacted water quality River

(< 0.57 mg/L).

FigureA.2.8. Total phosphorus (TP) data distributionsfor 20 60 smelt sampling stationsIn study area. The top of the box plots Is 50, the 75th percentileand the bot-tom Is the 25th percentile. The a.

40, tt line within the box Is the median and the errorbars representthe 10th and 90th percentiles. The 30-stationsare arrangedon the x-axis from the southernmostMA station 201 to the northernmost ME station.

Station medians were found to

10. be significantlydifferent with Kruskal-Wallis test (KW - 174.7, 0- df - 19, p < 0.001). The green line WPWWAJR FR SG NR CN ER MR PR SQ WR OY LC ML DM TB SB CR EB marks the EPA total phosphorus criterionfor minimally Impacted River water quality (< 23.75 ugIL).

ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN e 95

96 o ANADROMOUS RAINBOW SMELT REGIONAL CONSERVATION PLAN I

Fora copy of this report online, please visit www.restorerainbowsmelt.com and click on the "Learn More" tab Fora printedcopy, please contactyour state marine agency:

Massachusetts Division of Marine Fisheries Website: http://www.mass.gov/dfrele/dmfl Boston Offices: (617) 626-1520 Gloucester Regional Office: (978) 282-0308 New Bedford Regional Office: (508) 990-2860 New Hampshire Fish and Game Department Website: http://www.wildlife.state.nh.usl Durham Marine Fisheries Division: (603) 868-1095 Maine Department of Marine Resources Website: http://www.maine.gov/dmr/index.htm Sea Run Fisheries Division: (207) 287-9972 Bureau of Marine Sciences: (207) 633-9500 MAR12003.INDD/NHFG 2012

New York State Department of Environmental Conservation Office of General Counsel, 1 4 tb Floor 625 Broadway, Albany, New York 12233-1500 Fax: (518) 402-9018 or (518) 402-9019 Website: www.dec.ny.gov Joe Martens Acting Commissioner January 28, 2011 VIA ELECTRONIC MAIL AND HAND DELIVERY Hon. Maria E. Villa Hon. Daniel P. O'Connell Administrative Law Judges New York State Department of Environmental Conservation Office of Hearings and Mediation Services 625 Broadway, 1st Floor Albany, New York 12233-1550 Re: Enterzy Nuclear Indian Point.Units 2 and 3 CWA Section 401 WQC Application Proceeding NRC--Atomic Safety and Licensing Board'sDec. 3,2010 FSEIS

Dear ALJs Villa and O'Connell:

This letter constitutes Department staff's filing in compliance with the Ruling on Proposed Issues for Adjudication and Petitions for Party Status dated December 13, 2010, issued in the Entergy Indian Point §401 WQC proceeding ("Issues Ruling"), and with item 3 of the Scheduling Order attached to the Issues Ruling. Specifically, page 9 of the Issues Ruling and item "3" of the Scheduling Order directed Department staff to:

" advise the ALJs and the parties as to whether the Nuclear

• Regulatory Commission, Atomic Safety and Licensing Board's December 3, .2010 Final Supplemental Environmental Impact Statement ('FSEIS') is sufficient for Department Staff to -make the findings required by Section 617.11 of 6 NYCRR.'

It is Department staff's position after due deliberation that, in conjunction with or as otherwise supplemented by the Final Environmental Impact Statement by the Department concerning the Applications to Renew SPDES Permits for Three Hudson River Power Plants accepted June 25, 2003, along with the Department's records of proceedings (administrative hearing records) for both the Entergy Indian Point SPDES permit (DEC No.: 2-5522-00011/00004) and Entergy Indian 'Point §401 WQC application (DEC Nos.: 3-5522-00011/0030 and 3-5522-00105/00031), as well as with the NRC's record of proceeding (hearing file and record) for Entergy's license renewal for Indian Point Units 2 and 3 (Docket Department staff notes that the December 3,2010 FSEIS was prepared by staff of the NRC, not by.the Atomic Safety and Licensing Board, and that such FSEIS is not yet actually "final."

Nos. 50-247-LR and 50-286-LR; ASLBP No. 07-858-03-LR-BDO1), including but not limited to any contentions, -attachments, reports, declarations, comments; and administrative hearings relating to or arising from the publication by the NRC Staff on. December 3, 2010, of the FSEIS for the renewal of the Indian Point nuclear operating licenses, the NRC Staffs FSEIS (insofar as it may be further supplemented or amended by future proceedings noted herein) would be sufficient .for the purpose of making findings as required by 6 NYCRR §617.11.

Department staff notes that, consistent with the provisions of 6 NYCRR §617.15(c), a final decision by a Federal agency is not controlling on any agency decision on the proposed action, but may be considered by the agency. In addition, consistent with the provisions of 6 NYCRR

§617.11 (e), Department staff further notes that, because the Indian Point nuclear facilities are located in the coastal area (as defined in 6 NYCRR §6i7.2[f]), the agency' cannot make a final determination on the proposed action until there has been a written finding that the action is consistent with applicable policies set forth in 19 NYCRR §600.5.

Thank you for your courtesies and attention to this matt6r.

Very truly yours, Mark D. Sanza Assistant Counsel Via U.S. Mail and E-Mail:

Elise N. Zoli, Esq. ezoli@goodwinprocter.com John C. Englander, Esq. jenglander@goodwinprocter.com Goodwin Procter, LLP rfitzgerald@goodwinprocter.com Exchange Place Boston, Massachusetts 02109 Rebecca Troutman, Esq. rtroutman@riverkeeper.org Riverkeeper, Inc.

20 Secor Road Ossining, New York 10562 Melissa-Jean Rotini, Esq. mjrl @westchestergov.corn Assistant County Attorney County of Westchester Room 600, 148 Martine Avenue White Plains, New York 10601 Richard L. Brodsky, Esq. richardbrodsky@gmail.com 2121 Saw Mill River Road White Plains, New York 10607

Daniel Riesel, Esq. driesel@sprlaw.com Sive, Paget & Riesel, P.C.

460 Park Avenue,. 10't Floor New York, New York 10022 Steven Blow, Esq. stevenblow@dps.state.ny.us Assistant General Counsel New York State Department of Public Service Agency Building Three Empire State Plaza Albany, New York 12233-1350 Sam M. Laniado, Esq. sml@readlaniado.com David B. Johnson, Esq. dbj@readlaniado.com Read and Laniado, LLP 25 Eagle Street Albany, New York 12207-1901 Michael J. Delaney, Esq. mdelaney@dep.nyc.gov Director, Energy Regulatory Affairs New York City Department of Environmental Protection 59 17 Junction Boulevard, 109t Floor Flushing, New York 11373-5108 Robert J. Glasser, Esq. Bob.glasser~robertjg~asserpc.com Robert J. Glasser, P.C.

284 South Avenue Poughkeepsie, New York 12601 Via E-Mail Only:

Ned Sullivan, President nsullivan@scenichudson.org Hayley Mauskapf, Esq. hMauskapf@scenichudson.org Paul Schwartzberg Schwartzberg@scenichudson.org Scenic Hudson, Inc.

Karl S. Coplan, Esq. kcoplan@law.pace.edu Daniel E. Estrin, Esq. destrin@law.pace.edu Pace Environmental Litigation Clinic, Inc.

Deborah Brancato, Esq. dbrancato@riverkeeper.org Phillip H. Musegaas phillip@,riverkeeper.org Riverkeeper, Inc.

,I Geoffrey H. Pettus, Esq. gfettus@nrdc.org Natural Resources Defense Council Frank V. Bifera, Esq. fbifera@hblaw.com Hiscock & Barclay, LLP Kelli M. Dowell, Esq. kdowell@entergy-com Entergy Services, Inc.

EDMSI390924vI.