ML071430367
ML071430367 | |
Person / Time | |
---|---|
Site: | Wolf Creek ![]() |
Issue date: | 05/09/2007 |
From: | Wolf Creek |
To: | Document Control Desk, Office of Nuclear Reactor Regulation |
References | |
ET 07-0017 | |
Download: ML071430367 (157) | |
Text
11. Section 2.5 of the ER (Wolf Creek Generating Station (WCGS), 1980)states that U.S. Fish and Wildlife Service (USFWS) and U.S.Geological Survey (USGS)recommended that additional data be obtained on the habitat requirements of the Neosho River madtoms. Please provide any new information related to these habitat requirements.
Aquatic Ecology Page 1 of 3-Available water or sediment quality data for John Redmond Reservoir (JRR), Coffey County Lake (CCL), and Neosho River.* Available information regarding local, state, or federal management measures for the JRR, CCL, and the Neosho River. This may include fisheries management, watershed management, flow regulation, etc.* Any available documentation.
regarding minimum flows in the Neosho River.* Available records regarding the operation of the intake screens at either the Neosho River or CCL screen houses as well as information on the ongoing and periodic maintenance that occurs on the screens.-Available information on invasive or nuisance species observed in the facility's intake, JRR, CCL, or the Neosho River and available information on Wolf Creek Nuclear Operating Corporation (WCNOC) efforts to address this issue.* Documentation regarding the WCNOC or Coffey County Sheriff's office management of the CCL access program.* Examples of the fishery regulations developed by the Kansas Department of Wildlife &Parks for CCL.-If available, information on any occurrences of the Topeka Shiner in the JRR, CCL, Neosho River, or any water bodies crossed by the facility's transmission lines.Section 2.5 of the ER (Wolf Creek Generating Station (WCGS), 1980) states that U.S.Fish and Wildlife Service (USFWS) and U.S. Geological Survey (USGS) recommended that additional data be obtained on the habitat requirements of the Neosho River madtoms. Please provide any new information related to these habitat requirements.
-Section 2.5 of the ER (WCGS, 1980) states that USFWS and USGS recommended that flows below the John Redmond Dam be increased.
during critical periods for the Neosho River madtoms. Please provide any new information related to this issue.* If available, information on any occurrences of the Neosho river madtom in the JRR, CCL or Neosho river.-If available, information on any occurrences of the Neosho mucket mussel in the JRR, CCL or Neosho river.-Section 3.1.2 of the ER (WCGS, 1980) states that water is released to Wolf Creek infrequently.
Please provide available records documenting these releases and/or information regarding the frequency of these releases.-Details on the anti-scalants, dispersants, biocides, and corrosion inhibitors which are released into the circulating water system. Specifically, names of additives used, concentrations used, and frequency of application.
-System operating procedures for the circulating water system traveling screens.
Aquatic Ecology Audit Needs request #38"Section 2.5 of the ER (WCGS, 1980) states that USFWS and USGS recommended that additional data be obtained on the habitat requirements of the Neosho River madtoms. Is the applicant aware of any new information related to this?" Reference to "(WCGS, 1980)" is unclear. Wildhaber, et al (2000) was cited in Section 2.5 in the Environmental Report Operating License Renewal Stage as recommending USFWS and USGS obtain additional data to "assess if fish populations and habitat respond to changes in flow as predicted by this study." WCNOC is aware of some research since the Wildhaber et. al. (2000a) study, however, WCNOC is not aware of any increases from John Redmond Dam to assess fish population responses as suggested in that study. Known research are listed below and attached.Reference (attached)
Wildhaber, M.L., V.M. Tabor, J.E. Whitaker, A.L. Allert, D. Mulhern, P.J. Lamberson, and K.L. Powell. 2000a. Ictalurid Populations in Relation to the Presence of a Main-stem Reservoir in a Midwestern Warmwater Stream with Emphasis on the Threatened Neosho Madtom. Transactions of the American Fisheries Society 129:1264-1280.
Known research since Wildhaber et al.(2000a) are listed below and attached.Bryan, J. L., M. L. Wildhaber, and D. B. Noltie. 2004. Threatened Fishes of the World: Noturus placidus Taylor, 1969 (Ictaluridae).
Enviromental Biology of Fishes 70: 80.Bryan, J. L., M. L. Wildhaber, and D. B. Noltie. 2005. Examining Neosho Madtom Reproductive Biology Using Ultrasound and Artificial Photothermal Cycles. North American Journal of Aquaculture 67:221-230.
Bryan, J. L., M. L. Wildhaber and D. B. Noltie. 2006. Influence of Water Flow on Neosho Madtom (Noturus placidus)
Reproductive Behavior.
The American Midland Naturalist 156:305-318.
Gillette, D.P., J. Tiemann, D. R. Edds, and M. L. Wildhaber.
2006. Habitat Use by a Midwestern U.S.A. Fish Assemblage:
Effects of Season, Water Temperature and River Discharge.
Journal of Fish Biology 68:1494-1512.
Gillette, D.P., J. Tiemann, D. R. Edds, and M. L. Wildhaber.
2005. Spatiotemporal Patterns of Fish Assemblage Structure in a River Impounded by Low-head Dams.Copeia 2205(3):539-549.
Tiemann, J.S., D.P. Gillette, M.L. Wildhaber, and D.R. Edds. 2004. Effects of Lowhead Dams on Riffle-Dwelling Fishes and Macroinvertebrates in a Midwestern River. Transactions of the American Fisheries Society 133:705-717.
Tiemann, J.S., D.P. Gillette, M.L. Wildhaber, and D.R. Edds. 2004. Correlations
- Among Densities of Stream Fishes in the Upper Neosho River, With Focus on the Federally Threatened Neosho Madtom Noturus placidus.
Transactions of the Kansas Academy of Science 107:17-24.
Bulger, A.G., C.D. Wilkinson, D.R. Edds, and M.L. Wildhaber.
2002. Breeding Behavior and Reproductive Life History of the Neosho Madtom, Noturus placidus (Teleostei:
Ictaluridae).
Transactions of Kansas Academy of Science 105(3-4):
106-124.Bulger, A.G., M.L. Wildhaber, and D. Edds. 2002. Effects of Photoperiod on Behavior and Courtship of the Neosho Madtom (Noturus placidus).
Journal of Freshwater Ecology 17:141-150.
Wildhaber, M. L., A. L. Allert, C. J. Schmitt, V. M Tabor, D. Mulhern, K. L. Powell, and S. P. Sowa. 2000b. Natural and Anthropogenic Influences on the Distribution of the Threatened Neosho Madtom in a Midwestern Warmwater Stream. Transactions of the American Fisheries Society 129: 243-261.Wildhaber, M.L. 2006. The Role of Reproductive Behavior in the Conservation of Fishes: Examples from the Great Plains Riverine Fishes. The Conservation Behaviorist 4(1): 15-19.
Copyrighted Material Protect accordingly Transactions qolihe Anerican Fisheries Societ.v 129:1264-1280, 2000© Copyright by the American Fisheries Society 2000 Ictalurid Populations in Relation to the Presence of a Main-Stem Reservoir in a Midwestern Warmwater Stream with Emphasis on the Threatened Neosho Madtom MARK L. WILDHABER*
U.S. Geological Survey, Columbia Environmental Research Centel;4200 New Haven Road, Columbia.
Missouri 65201. USA VERNON M. TABOR U.S. Fish and Wildlife Service, 315 Houston Street, Suite E, Manhattan, Kansas 66502, USA JOANNE E. WHITAKER AND ANN L. ALLERT U.S. Geological Survey, Columbia Environmental Research Center, 4200 .New Haven Road, Columbia, Missouri 65201, USA DANIEL W MULHERN U.S. Fish and Wildlife Service, 315 Houston Street, Suite E, Manhattan, Kansas 66502, USA PETER J. LAMBERSON U.S. Geological Survey, Columbia Environmental Research Center, 4200 New Hat'en Road, Columbia, Missouri 65201, USA KENNETH L. POWELL Kjolhaug Environmental Services Company 4767 Richmond Road, Mound, Minnesota 55364, USA Abstract.-Ictalurid populations, including those of the Neosho madtom Noturus placidus, have been monitored in the Neosho River basin since the U.S. Fish and Wildlife Service listed the Neosho madtom as threatened in 1991. The Neosho madtom presently occurs only in the Neosho River basin, whose hydrologic regime, physical habitat, and water quality have been altered by the construction and operation of reservoirs.
Our objective was to assess changes in ictalurid densities, habitat, water quality, and hydrology in relation to the presence of a main-stem reservoir in the Neosho River basin. Study sites were characterized using habitat quality as measured by substrate size, water quality as measured by standard physicochemical measures, and indicators of hydrologic alteration (IHA) as calculated from stream gauge information from the U.S. Geological Survey. Site estimates of ictalurid densities were collected by the U.S. Fish and Wildlife Service annually from 1991 to 1998, with the exception of 1993. Water quality and habitat measurements documented reduced turbidity and altered substrate composition in the Neosho River basin below John Redmond Dam.The effccts of the dam on flow were indicated by changes in the short- and long-term minimum and maximum flows. Positive correlations between observed Neosho madtom densities and increases in minimum flow suggest that increased minimum flows could be used to enhance Neosho madtom populations.
Positive correlations between Neosho madtom densities and increased flows in the winter and spring months as well as the date ofthe I -d annual minimum flow indicate the potential importance of the timing of increased flows to Neosho madtoms. Because of the positive relationships that we found between the densities of Neosho madtoms and those of channel catfish Ictalurus puncratus.
stonecats Noturusftavus.
and other catfishes, alterations in flow that benefit Neosho madtom popu-lations will probably benefit other members of the benthic fish community of the Neosho River.Reservoirs decrease the flow of sediments and water quality downstream (Obeng 1981; Stanford alter the hydrologic regime, physical habitat, and et al. 1996). Hesse and Mestl (1993) showed that a reservoir system can substantially alter the hy-drograph of a river system. Patton and Hubert* Corresponding author: markwildhaber@usgs.gov (1993) found that the presence of a reservoir on a Received July 2, 1999; accepted April 17, 2000 Great Plains stream eliminated braided channels 1264 ICTALURID POPULATIONS AND STREAM HABITAT 1265 that had historically existed downstream.
After the construction of an extensive reservoir system with-in the lower Kansas River, the range of flow was reduced, turbidity declined, phytoplankton in-creased, and substrate changed from loose and"quick" to firm and stable (Cross and Moss 1987).Cross and Moss (1987) suggested that the changes in hydrology and habitat that result from water control by dams and reservoirs have detri-mental impacts on fish populations.
Subsequent research has demonstrated this (Bain et al. 1988;Scheidegger and Bain 1995). Cross and Moss com-pared predam and postdam data to show the impact of the reservoirs in the Arkansas River system on fish communities, flows, and turbidity.
Frequent extreme variations in hourly water flow caused by a hydroelectric dam have been shown to reduce the abundance of habitat specialists (i.e., fish spe-cies that strongly prefer shallow, slow water along stream margins; Bain et at. 1988), lower abun-dance in the larval fish community (Scheidegger and Bain 1995), and result in a downstream re-covery gradient of the fish community (Niemi et al. 1990; Yount and Niemi 1990; Kinsolving and Bain 1993). One might expect the impacts that flood control dams have on downstream riverine ecosystems to be similar to those of hydroelectric dams because of the regulated nature of both sys-tems. However, the effects of hydroelectric dams should be larger and more immediate than those of flood control dams owing to the higher fre-quency of changes in flow associated with the for-mer (Baxter 1985). Our study focuses on a dam-and-reservoir system designed primarily for flood control.One of the most recently developed methods for evaluating how the hydrology of a river system has changed over time is known as indicators of hydrologic alteration (IHA; Richter et al. 1996;The Nature Conservancy 1997). 1HA is a series of different annual measures calculated from the stream gauge data for each water year (October 1 of the year before to September 30 of the current year) that are available from the U.S. Geological Survey (USGS). The lHA measures include the average daily discharge for each of the 12 months;the average minimum and maximum daily dis-charges over 1, 3, 7, 30, and 90 d; the days of the year of the 1-d minimum and maximum discharg-es; the number of reversals between rising and falling discharges and the mean rate of rising and falling discharges; and the number of low pulses and high pulses and their average length in days.These measures help to characterize the hydrologic variation of a stream system within each year. The goal of Richter et al. (1996) was to provide an analytical tool for describing complex hydrologic variation to facilitate investigation of the ecosys-tem effects of hydrologic alterations.
The hydro-logic measures developed by Richter et al. were guided by the paradigm that the most effective way to manage riverine ecosystems is to protect or re-store "natural" hydrologic regimes (Sparks 1992;Poff et al. 1997). By trying to maintain or restore the natural hydrologic regime, resource managers should be able to maintain or restore the entire group of species in a riverine ecosystem (Sparks 1992).The Neosho madtom Noturus placidus (Taylor 1969) is a small (<75 mm total length) ictalurid that is native to the main stems of the Neosho and Cottonwood rivers in Kansas and Oklahoma and the Spring River in Kansas and Missouri (Luttrell et al. 1992; Cross and Collins 1995; Wilkinson et al. 1996). This species occupies portions of riffles with mean flows of 79 cm/s, mean depths of 0.23 m, and unconsolidated pieces of pebble and gravel 2-64 mm in diameter (Moss 1983). Neosho mad-toms feed at night on larval insects found among the gravel (Cross and Collins 1995). Based on sam-ples collected throughout the year (both day and night), the highest numbers of Neosho madtoms occur in riffles during daylight hours in late sum-mer and early fall after young of year are believed to have recruited to the population (Moss 1983;Luttrell et al. 1992; Fuselier and Edds 1994). Pre-vious research suggests that Neosho madtoms have a life cycle that is annual in nature, with recruit-ment of young of year into adult collection gear about the time the adults begin to disappear from collections (Fuselier and Edds 1994).Downstream of its confluence with the Cotton-wood River, the Neosho River is regulated by John Redmond Dam, which forms John Redmond Res-ervoir, a body of water that is used for flood con-trol, water supply, maintenance of downstream wa-ter quality, and recreation (Figure 1). The Neosho madtom was federally listed as threatened by the U.S. Fish and Wildlife Service (USFWS) in May 1990, and a recovery plan was approved in Sep-tember 1991 (USFWS 1991). The plan suggested the need for maintaining springtime flows for spawning as well as minimum flows during low-water times for the survival of Neosho madtom populations.
Our objective was to assess the impact of the presence and operation of John Redmond Dam and Reservoir on ictalurid population trends (in par-1266 WILDHABER ET AL.W+ E 0 20 40 Kilometers I ~~Missouri
-Arkansas FIGURE I.-Sampling sites (triangles) on the Neosho and Cottonwood rivers from 1991 to 1998. U.S. Geological Survey gauging stations are represented by half-filled circles. Numbers represent locations as follows: I = Plymouth, Kansas; 2 = Americus, Kansas; 3 = Burlington, Kansas; 4 = lola, Kansas; 5 = Parsons, Kansas; and 6 = Commerce, Oklahoma.ticular those of the Neosho madtom), as well as on the hydrology, habitat, and water quality in the Neosho River. Specifically, our hypothesis was that the presence of John Redmond Dam has sig-nificantly changed the hydrology, habitat, and wa-ter quality in the Neosho River below the dam and thus is detrimentally affecting ictalurid popula-tions, particularly those of the Neosho madtom.We tested this hypothesis by comparing the fol-lowing: (I) current ictalurid densities, habitat, and water quality above the reservoir and those below the dam, (2) predam and postdam hydrographs for the Neosho River below the dam; (3) postdam hy-drographs above the reservoir and those below the dam; and (4) the relationships between annual ic-talurid densities and the current and preceding wa-ter year values of each IBA measure.' Study Area The study area included the main stems of the Neosho and Cottonwood rivers in Kansas and Oklahoma, which we will refer to as the Neosho River basin (Figure 1). The Neosho and Cotton-wood rivers are part of the Arkansas River drain-age and drain mainly tallgrass and mixed-grass prairie, with mature riparian vegetation along some sections (Moss 1983). The Cottonwood Riv-er joins the Neosho River near Emporia, Kansas.The Cottonwood River and the Neosho River up-stream of its confluence with the Cottonwood Riv-er are fifth-order streams (Moss 1983). with mean annual discharges from 1990 to 1998 of 25.0 m 3/s at Plymouth, Kansas, and 12.7 m 3/s at Americus, Kansas, respectively (Figure 1). The headwaters of the Neosho and Cottonwood rivers are regulated by reservoirs.
Construction of Marion Reservoir at the headwaters of the Cottonwood River began in 1964, with embankment closure in 1967 and full flood control operation in 1968 (Tulsa District USACE 1993). Construction of Council Grove Lake at the headwaters of the Neosho River began ICTALURID POPULATIONS AND STREAM HABITAT 1267 in 1960, and the lake was placed in full flood con-trol operation in 1964 (Tulsa District USACE 1993). Downstream of its confluence with the Cot-tonwood River, the Neosho River is regulated by John Redmond Dam, construction of which began in 1959, with embankment closure in 1963, full flood control operation in 1964, and final comple-tion in 1965 (Tulsa District USACE 1993). All three reservoirs were constructed with water stor-age capacity that could be used for flood control, water supply, maintenance of downstream water quality, and recreation (Tulsa District USACE 1993).Methods Gravel bar sampling.-From 1991 to 1998, 12 gravel bars (shoreline collections of pebbles up to 38 mm in diameter that extend out into the river and are believed to provide suitable habitat for Neosho madtoms) were selected by USFWS for monitoring Neosho madtom populations (USFWS 1991). Generally, the same gravel bars, or ones in the same river reaches, were sampled each year (Figure 1). The number of sites sampled above John Redmond Reservoir and below John Red-mond Dam was similar but varied somewhat owing to high-water conditions and shifting gravel bars.Sampling at all sites occurred during daylight hours between August and October, after Neosho madtom young-of-year recruitment was expected to have occurred.
In 1993, no sampling was pos-sible owing to extreme flooding.
Data from 1996 were excluded from analyses because only four sites, all above the reservoir, were sampled owing to inaccessibility resulting from high waters.Before sampling, three to five transects perpen-dicular to the river channel were spaced equally to span the length of the bar. In most instances, five stations were spaced equally along each tran-sect, with a minimum distance of 2 in between adjacent stations.
Fewer than five stations were established if the river channel was less than 10 m wide or if a station occurred at a depth too great to seine (> 1.25 m). Transects on a gravel bar were sampled in order from downstream to upstream.On each transect, stations were sampled in order from nearest to most distant from the gravel bar.At each station, sampling proceeded in the follow-ing order to minimize the impacts of samples on each other: ictalurids (Neosho madtom, channel catfish ictalurus punctatus, slender madtom No-turus exilis, stonecat N. flavus, brindled madtom N. miurus, freckled madtom N. noclurnus, and flat-head catfish Pylodictis olivaris), substrate, water depth, water velocity, and surface water. Mea-surements of substrate and surface water quality were collected only in 1991. Fishes were collected from a 4.5-M 2 area by disturbing the substrate starting 3 m upstream of a stationary seine (3.0-mm 2 mesh) and proceeding downstream to the seine (i.e., kick-seining).
All ictalurids were iden-tified, measured for total length, and released back into the river. Substrate was collected from an un-disturbed area adjacent to the fish sampling lo-cation with a 13-cm deep X 10-cm-diameter cy-lindrical grab sampler. Substrate samples were placed in plastic bags, transported to the labora-tory, dried for 10-15 min at 100-110*C, and sieved into five size-classes
(<2 mm, 2-9 mm, 9-19 mm, 19-38 mm, and > 38 mm) that were then weighed. Water depth was recorded, and water ve-locity at 60% of water depth was measured with a Marsh-McBirney model 201 current meter. After all station samples were collected at a site, a sur-face water sample was analyzed with a Hach model DREL/IC portable colorimeter for pH, alkalinity, hardness, conductivity, turbidity, un-ionized am-monia (NH 3), nitrite plus nitrate (NO 2 + NO 3), sulfate (SO 4), phosphate (PO 4), and chloride (CI-).Hydrologic data.-We obtained USGS hydro-logic data in the form of daily mean flows from gauges on the Neosho and Cottonwood rivers be-fore and after the construction of John Redmond Dam. The 1960s was the decade of ongoing res-ervoir construction; full flood control operation was not in effect for all three primary reservoirs combined (Council Groves, John Redmond, and Marion) until 1968 (Tulsa District USACE 1993).Therefore, the postconstruction hydrograph for the period prior to the initiation of Neosho madtom population monitoring was represented by data from water years 1970.to 1989 (1970 is the water year that started in October 1969). This supplied the 20-year minimum data set suggested by Richter et a]. (1997) for representation of the postconstruc-tion period. We used hydrologic data from 1940 to 1959 to represent the 20-year preconstruction hydrograph needed for comparison with the post-construction hydrograph.
We used 1990-1998 hy-drographic data from the gauges at Americus, Plymouth, Burlington, lola, and Parsons, Kansas, and Commerce, Oklahoma (Figure 1). to directly test relationships among IHA measures and Neo-sho madtom densities.
We used predam and postdam data from the gauge near lola, Kansas, to compare the hydro-graph of the Neosho River below John Redmond Dam before and after the construction of the three 1268 WILDHABER ET AL.primary reservoirs in the Neosho River basin (Fig-ure 1). The gauge at Iola was the closest one down-stream of John Redmond Dam (89 kin) that was in operation before construction of the three res-ervoirs (Putnam et al. 1996).We used postdam hydrologic data from the gauges near Americus and Burlington, Kansas, on the Neosho River and Plymouth, Kansas, on the Cottonwood River to more directly assess the ef-fects of John Redmond Dam on the hydrograph of the Neosho River. The gauge near Americus is 38.6 kin upstream of the mouth of the Cottonwood Riv-er. The gauge near Plymouth is 63.1 km upstream of the confluence with the Neosho River. The Americus and Plymouth gauges are the nearest ones upstream of John Redmond Reservoir.
There are no gauges between the confluence of the Ne-osho and Cottonwood rivers and the reservoir.
Be-cause the hydrologic data used consisted of daily means, the data from the Americus and Plymouth gauges could be considered synchronized, There-fore, as a conservative estimate, we combined the hydrologic data from these two gauges to represent the postdam Neosho River hydrograph above the reservoir.
Data from the gauge near Burlington were used to represent the postdam Neosho River hydrograph below John Redmond Dam because that gauge is much closer to the dam (8.5 km downstream) than the one near lola.Statistical analyses.-We analyzed the physical, chemical, and fish population data to assess dif-ferences between sites above John Redmond Res-ervoir and sites below John Redmond Dam as well as differences across years. Arithmetic means were calculated for sites' depth and velocity.
We cal-culated site densities for four categories of fish: Neosho madtoms, channel catfish, stonecats, and all non-Neosho madtoms ictalurids combined.
The last grouping was established because some spe-cies of ictalurids were not found often enough to allow separate species-level analyses.
We calcu-lated fish densities by dividing the total number of Neosho madtoms., all other catfishes combined, channel catfish, stonecats, and other catfishes col-lected at a site by the total area sampled by kick-seining. We calculated site means for substrate size categories by dividing the total weight of a size category by the total weight of all size categories for the site. We also calculated the geometric mean and fredle index (geometric mean adjusted for dis-tribution of particle sizes) for the substrate at each site, as suggested by McMahon et al. (1996), to characterize substrate suitability for Neosho mad-toms. When the geometric means of substrate sam-pies are similar, a high fredle index indicates sub-strate consisting largely of sizes near the mean, while a low fredle index indicates more evenly distributed substrate sizes. The fredle index is pos-itively correlated with the potential permeability of sediment to water, and hence to dissolved ox-ygen transport within the sediment, and it has been shown to be positively conrelated with the emer-gence of salmonid alevins (Platts et al. 1983 and references therein).
Mean daily flow data were summarized before analysis by means of1HA mea-sure estimates produced with the IHA software (Nature Conservancy 1997).All statistical tests were conducted using SAS software (SAS Institute 1990). Site means were checked for normality and tested for homogeneity of variance using a chi-square test (Steel and Torrie 1980). Most nonnormal variables were log 1 o trans-formed; variables that were proportions or ratios were transformed using the arcsine of the square root. Various statistical methods were used to make primary comparisons between the situation before and after dam construction; between the situation above John Redmond Reservoir and that below John Redmond Dam; among Neosho mad-tom densities and ictalurid densities, depth, and velocity; and among the IHA measures and Neosho madtom densities.
These methods included anal-ysis of variance (ANOVA), Levene's test for dif-ferences in variance, correlation analysis, and analysis of covariance (ANCOVA) (SAS Institute 1990). Generally, relationships for which P < 0.05 were considered significant and those for which 0.05 !5 P < 0.10 were considered marginally sig-nificant.
Separate two-way ANOVAs with years and location above the reservoir or below the dam were performed on site means for fish densities, depth, and velocity.
Separate one-way ANOVAs were performed on the annual IHA values for com-parisons between the predam and postdam situa-tions and those above the reservoir and below the dam. Our ability to normalize variables through transformations and equal sample sizes allowed us to use the power and robustness of ANOVA to test hydrologic differences in mean values relative to location above the reservoir or below the dam de-spite the differences in variance that may have existed (Milliken and Johnson 1984). Because Richter et al. (1997) emphasized the importance of the interannual variation in IHA measures to a riverine ecosystem-along with the central ten-dencies of those measures-we also used Levene's test to assess the effects of the presence of the danm on interannual variation in the IHA measures.
ICTALURID POPULATIONS AND STREAM HABITAT 1269 Correlation analyses were used to assess rela-tionships between the densities of Neosho mad-toms, all other catfishes combined, channel catfish, and stonecats and depth and velocity.
Separate two-way ANCOVAs were performed on Neosho madtom densities for each IHA measure to assess the relationships between Neosho madtom densi-ties and IHA measures after adjusting for effects attributable to year and position relative to the dam and reservoir.
The above-the-reservoir versus below-the-dam effect in the ANCOVA models accounted for relationships that were the result of upstream-downstream patterns resulting from potentially im-portant factors such as stream size and the presence of the dam and reservoir.
The ultimate goal of the ANCOVAs was to determine if Neosho madtom population trends were related to IHA measures independent of the two main factors present within the design of the study (i.e., year and presence of the dam and reservoir).
For the ANCOVAs, we used current and preceding water years for the gauge closest to the gravel bar sampled. Because there were fewer gauges than there were gravel bars sampled, we grouped gravel bars by nearest gauge (i.e., Plymouth, Americus, Plymouth and Americus combined, Burlington, lola, Parsons, or Commerce).
We then averaged Neosho madtom site densities in each year by gauge grouping be-fore any ANCOVAs were conducted.
The annual nature of the life cycle of the Neosho madtom made it possible for us to compare density rela-tionships with IHA measures in the current and preceding water years to assess whether population trends were the result of survival to reproductive age (or reproductive success) and survival of young of year to recruitment to the population, respectively.
Before we present our results, two important points need to be made concerning the validity and strength of those results as they relate to the anal-yses chosen. First, if it had been possible, this study should have been done in the context of a before-after-control-impact-pairs (BACIP) design (Stewart-Oaten et al. 1992). In the BACIP design, the system is studied before and after the impact (in this case dam construction) simultaneously with a control system to ensure that observed changes are the result of the impact and not some unknown environmental factor. Because the Ne-osho madtom was not identified as a species until after construction of the dam (Taylor 1969; Tulsa District USACE 1993), a before-after comparison of fish populations was not possible.
Owing to the regional nature of the distribution of Neosho mad-tom populations (Luttrell et al. 1992; Cross and Collins 1995; Wilkinson et al. 1996), a control system for comparison is not available for the spe-cies. Even so, the validity of the information gained from this study is supported by the length of time covered, which provided data over an 8-year period for a fish that generally lives 2 years or less (Fuselier and Edds 1994). In essence, the multiyear nature of the study and the Neosho mad-tom's annual life cycle allow year to be used in ANOVA and ANCOVA models as the control and before-after comparison by accounting for year-to-year variability stemming from unknown en-vironmental factors. Second, the Neosho madtom's annual life cycle results in recruitment of young of year into adult collection gear about the time adults begin to disappear from collections (Fuse-lier and Edds 1994); as a result, we could not ef-fectively compare young of year and adult Neosho madtoms.Results The densities of Neosho madtoms and channel catfish were greater above John Redmond Reser-voir than below John Redmond Dam; the density of the all catfishes other than Neosho madtoms was marginally greater above the reservoir; and the densities of stonecats did not differ significantly between the two locations (Table 1). None of the measured catfish densities differed among years.Neosho madtom densities were positively corre-lated with the densities of all other catfishes com-bined (r = 0.41, P = 0.001), channel catfish (r =0.34, P = 0.007), and stonecats (r = 0.35, P =0.006). Neosho madtom and stonecat densities were positively correlated with water velocity (r= 0.33, P = 0.008; and r = 0.50, P = 0.0001, respectively).
The densities of all other catfishes combined, channel catfish, and stonecats were neg-atively correlated with depth (r = -0.47, P =0.00]; r = -0.33, P = 0.001; and r = -0.51, P= 0.0001). The negative correlation between Ne-osho madtom densities and depth was marginally significant (r = -0.22, P = 0.08). In all of the above correlations, N = 61.Physical habitat and water quality characteris-tics differed above John Redmond Reservoir and below John Redmond Dam (Table 2). Turbidity was higher, water temperature lower, and the fredle index marginally lower above the reservoir (Table 2). Water depth was marginally lower above the reservoir (Table 1). Water velocity, individual sub-strate size-classes, and the geometric mean of sub-strate size did not differ between the two locations 1270 WILDHABER ET AL.TABLE 1.--Two-way analysis of variance (ANOVA) results for 1991-1998 ictalurid densities, water depth, and velocity among years and location above John Redmond Reservoir or below John Redmond Dam.Density (fish per 100 M 2)Catfishes other than Neosho Neosho Channel Depth Velocity Variable madtoms madtoms catfish Stonecats (m) -(nm's)Two-way ANOVA (P-value [F])Year 0.15 0.45 0.28 0.33 0.36 0.013[1.69] [0.973 [1.291 [1.18] [1.13] [3.24]Location 0.0015 0.066 0.051 0.44 0.070 0.71[11.31] 13.53] [4.01]] [0.60] 13.44] [0.14]Year X location 0.29 0.089 0.071 0.53 0.93 0.95[1.28] [2.05] [2.19] [0.83] [0.26] [0.21]Means Neosho and Cottonwood rivers above John Red-mond Reservoir 19.82 45.40 34.31 4.61 0.33 0.34 Neosho River below John Redmond Dam 5.64 25.66 18.73 2.83 0.38 0.35 (Tables I and 2). Dissolved oxygen concentrations, temperature, and P0 4 concentrations were lower (the last marginally), while alkalinity and NH 3 were higher above the reservoir (Table 2). Con-ductivity, pH, hardness, SO 4 , and Cl- did not differ.The concentration of NO, + NO 3 was 1.5 mg/L at one site below the dam but 0.23 mg/L or less at all' other sites, with six of these sites having nondetectable levels. For this reason, NO 2 + NO 3 data were not analyzed statistically and are not reported in Table 2. Water velocity was different among years, but depth was not (Table 1).Hydrographic data (IHA values) were related to the presence of John Redmond Dam and John Red-mond Reservoir (Figures 2-5), and Neosho mad-tom densities were related to IHA values (Table TABLE 2.-Means, standard deviations (SD), and one-way analysis of variance (ANOVA) results for 1991 habitat and water quality measurements for locations above John Redmond Reservoir and below John Redmond Dam. For all variables except water quality measures above John Redmond Reservoir N = 6; for those measures, N = 5.Above John Below John Redmond Reservoir Redmond Dam Variable Mean SD Mean SD P-value (F)Substrate (percent by weight)B38 mm 4.35 0.11, 7.5 0.15, 0.38 (0.83)<38 and ---19 mm 26.63 12.26 23.26 6.57 0.57 (0.35)<19 and -9 mm 31.47 5.64 36.32 11.00 0.36 (0.92)<9 and ->2 mm 21.54 5.89 19.52 6.03 0.57 (0.35)<2 mm 15.14 6.75 11.95 5.61 0.39 (0.79)Geometric mean 10.40 0.061 11.74 0.04a 0.50 (0.49)Fredle index 5.52 1.85 7.82 2.12 0.073 (4.02)Water quality Water temperature
(°C) 24.74 1.80 27.58 1.30 0.014 (9.26)Turbidity (NTU) 57.0 17.89 27.17 7.36 0.0045 (14.09)pH 8.37 0.07 8.47 0.21 0.37 (0.90)Dissolved oxygen (mg/L) 4.66 0.67 5.62 0.59 0.033 (6.31)Conductivity 548.00 189.12 433.33 32.04 0.17 (2.18)Alkalinity (mg/L) 161.20 9.31 145.00 3.46 0.0032 (15.84)Hardness (mg;L) 220.92 0.171 180.22 0.02b 0.24 (1.59)NH 3 (mg/L) 0.40 0.05 0.15 0.09 0.0003 (33.15)S0 4 lmg/L) 65.0 57.12 48.83 6.15 0.50 (0.48)P04 (mg/L) 1.38 0.71 2.43 1.07 0.094 (3.50)C1- (ntg/L) 14.20 9.31 13.67 2.88 0.90 (0.02)Standard deviation of arcsin(proportion t 1 2 l-transformed data.Standard deviation of Ioglt-transformed data.
ICTALURID POPULATIONS AND STREAM HABITAT 1271 10001 E 0 U-C>M 1001 101 11 O -b b b b b b b b b Indicator of Hydrologic Alteration FIGURE 2.-Means and standard deviations for 22 flow indicators before and after construction of John Redmond Dam and John Redmond Reservoir (Richter et al. 1996). The U.S. Geological Survey hydrologic data used are from the gauge near lola, Kansas. The period 1940-1959 represents the predam period, the period 1970-1989 the postdam period (N = 20 in each period). Asterisks indicate significant (P S 0.05) differences between flows before and after construction of the dam and reservoir; pound signs indicate significant differences in variance. "Min" and "max" refer to minimum and maximum, respectively.
3). The mean February, November, and December flow rates, date of annual maximum flow, and num-ber of reversals between rising and falling water were lower before the construction of the dam (Figures 2 and 3). The variability of the minimum and maximum flows at 1, 3, 7, and 30 d was greater before the construction of the dam, as was that of the 90-d maximum flow and July flow rate (Figure 2); the variability in annual maximum flow date was greater after construction of the dam. The mean minimum flows at 1, 3, 7, and 30 d, along with the l-d and 3-d maximum flows and the flow rise rate values were all greater above the reservoir (Figures 4 and 5); the 30-d and 90-d maximum flow rates, low-pulse count, and high-pulse dura-tion were lower above the reservoir.
The variabil-ities of the l-d and 3-d maximum flow rates, low-pulse count, and rise rate of flow were greater above the reservoir (Figures 4 and 5); variability in high-pulse length was lower above the reservoir.
The effects of the dam on the minimum and max-imum flows of the Neosho River tended to de-crease with increasing distance downstream (Fig-ure 6). There was a dramatic decrease in Neosho madtom densities just downstream of the dam near the Burlington gauge (Figure 6). Further down-stream, near the lola gauge, Neosho madtom pop-ulation densities increased to levels near those found above John Redmond Reservoir.
Densities began to decrease again from lola downstream to Parsons; however, they were still greater than those at Burlington.
The ANCOVA of Neosho madtom densities and IHA measures for the current and preceding water years showed significant positive relationships between Neosho madtom density and (1) the minimum flows at 1, 3, 7, 30, and 90 d;(2) the mean daily flows in October, December, January, May, June, and August; and (3) the date 1272'WILDHABER ET AL.400 300> 200 100 0 32.6 a, M co 0 0 0 Rise Fall Min Max High High Reversal Low Low Rate Rate Flow Flow Pulse Pulse Count Pulse Pulse Date Date Count Length Count Length Indicator of Hydrologic Alteration FIGURE 3.-Means and standard deviations for the frequency, duration, and timing of extreme flows as well as the frequency and magnitude of changes in flow rate before and after construction of John Redmond Dam and John Redmond Reservoir (Richter ct al. 1996). The U.S. Geological Survey hydrologic data used are from the gauge near lola, Kansas. The period 1940-1959 represents the predam period, the period 1970-1989 the postdam period (N = 20 in each period). Asterisks indicate significant (P -- 0.05) differences between flow indicators before and after construction of the dam and reservoir; pound signs indicate significant differences in variance. "Min" and "Max" refer to minimum and maximum, respectively.
of the annual minimum daily flow (Table 3). The annual minimum flow occurred around the middle of September.
Discussion We documented important relationships be-tween location in relation to a main-stem dam and reservoir and ictalurid densities, water quality, substrate size composition, and hydrology.
These findings all suggest potential impacts on the ic-talurid populations, particularly populations of Neosho madtoms, in the Neosho River as a result of alterations in water quality, the availability of suitable habitat as measured by substrate, and hy-drology below John Redmond Dam. Our results are consistent with previous surveys that impli-cated the construction and operation of John Red-mond Dam and John Redmond Reservoir as causes of the changes in the structure of the fish com-munity that have occurred within the upper Neosho River basin (Cross and Braasch 1969). This study helps to define more specifically how and to what extent water quality, habitat as measured by sub-strate size composition, and hydrology as. mea-sured by IHA factors have been altered in the Ne-osho River basin as an aid to formulating river management strategies to improve the chances for recovery.
of the ictalurid community within that basin.The differences in turbidity, the fredle index, water temperature, dissolved oxygen, alkalinity, NH 3 , and PO 4 above John Redmond Reservoir ver-sus below John Redmond Dam suggest reservoir impacts on the water and habitat quality of the Neosho River. Changes in turbidity and substrate in the Neosho River below the dam parallel ob-ICTALURID POPULATIONS AND STREAM HABITAT 1273 10001 Above John Redmond Reservoir[ Below John Redmond Dam P < 0.05* Mean# Variance 0 CU 1001 101 11 r 1 Indicator-of Hydrologic Alteration FIGURE 4.-Means and standard deviations for 22 flow indicators above John Redmond Reservoir and below John Redmond Dam (Richter et al. 1996). The U.S. Geological Survey hydrologic data used are from 1970 to 1989 (N = 20). The Neosho River hydrograph above the reservoir is represented by combined data from the gauges near Americus and Plymouth, Kansas; that below is represented by the gauge near Burlington, Kansas (see Figure 1). Asterisks indicates significant (P -- 0.05) differences between indicators above the reservoir and the corre-sponding indicators below the dam; pound signs indicate significant differences in variance. "Min" and "max" refer to minimum and maximum, respectively.
servations of previous researchers (e.g., Cross and Moss 1987). As suspended solids are deposited on the reservoir bottom due to decreases in stream flow, turbidity levels decline and the composition of the substrate changes below the dam (Baxter 1985). The higher fredle index below the dam than above the reservoir represents a shift to the pre-dominance of larger gravel below the dam, a sta-tistic that is supported by the higher (though not significant) geometric mean and percent makeup for two of the three upper size-classes of substrate (Table 2). This increased coarseness of the sub-strate is a common effect of reservoirs (Baxter 1985). Decreased turbidity and increased substrate size were identified as habitat factors limiting Ne-osho madtom populations in the Spring River (Wildhaber et al. 2000). This suggests that similar limiting effects on Neosho madtom populations may be occurring below John Redmond Dam as a result of changes in turbidity and substrate size.Although water temperature, dissolved oxygen, alkalinity, NH 3 , and P0 4 differed above John Red-mond Reservoir and below John Redmond Dam, it is unclear whether these factors could be limiting ictalurid populations, especially Neosho madtom populations, because nothing is known about the effects of these factors on Neosho madtoms. Mean temperatures above the reservoir and below the dam differed by nearly 3°C; however, nothing is known about the preferred temperature or optimal temperature for growth for Neosho madtoms.Mean dissolved oxygen levels differed by less than I mg/L and were greater than levels shown to af-fect the behavior of other plains fishes (e.g., Bryan et al. 1984). The mean alkalinity level was only 10% less below the dam than above the reservoir 1274 WILDHABER ET AL.400 300 200 100 0 32.6-0 0 Rise Fall Min Max High High Reversal Low Low Rate Rate Flow Flow Pulse Pulse Count Pulse Pulse Date Date Count Length Count Length Indicator of Hydrologic Alteration FIGURE 5.-Means and standard deviations for the frequency, duration, and timing of extreme flows as well as the frequency and magnitude of changes in flow rate above John Redmond Reservoir and below John Redmond Dam (Richter et al. 1996). The U.S. Geological Survey hydrologic data used are from 1970 to 1989 (N = 20).The Neosho River hydrograph above the reservoir is represented by combined data from the gauges near Americus and Plymouth, Kansas; that below is represented by the gauge near Burlington, Kansas (see Figure 1). Asterisks indicate significant (P <: 0.05) differences between indicators above the reservoir and the corresponding indicators below the dam; pound signs indicate significant differences in variance. "Min" and "Max" refer to minimum and maximum, respectively.
The right axis only applies to low pulse count and low pulse length.and in both cases was well within acceptable limits for aquatic life (USEPA 1986). Mean NH 3 levels were below the USEPA water quality criteria for chronic toxicity to warmwater species (USEPA 1986). The phosphorus levels that would have been present as a result of the observed mean P0 4 levels above the reservoir and below the dam were greater than recommended levels (USEPA 1986).Along with removal of particulates, the presence of John Redmond Dam and John Redmond Res-ervoir has resulted in lower minimum flows, lower short-term (l-d and 3-d) maximum flows, in-creased occurrence of low-flow events, much less variability in flow rates, increased winter flows, increased long-term (30-d and 90-d) maximum flows, increased length and variability in length of high-flow events, and a later and more variable date of maximum annual flow below the dam. In essence, the Neosho River below John Redmond Dam has become a river with lower minimum flows and lower short-term and higher long-term maximum flows. The lower minimum flows are probably the result of a need to maintain water levels in the reservoir.
The lower short-term and higher long-term maximum flows are probably the result of managing the reservoir to minimize downstream flooding.Lower minimum flows below John Redmond Dam may directly constrain the available area of riffle habitat suitable for Neosho madtoms and therefore represent a limiting factor for such pop-ulations in the Neosho River. Neosho madtoms are almost entirely collected on gravel riffles with moderate flows (Fuselier and Edds 1994). Our ob-served negative correlations between depth and the densities of Neosho madtoms, all other catfishes ICTALURID POPULATIONS AND STREAM HABITAT 1275 TABLE 3.-Coefficients and P-values from two-way analysis of covariance (ANCOVA) ofNeosho madtom densities and flow-rate (IHA) measures.
The two-way model included year, location relative to John Redmond Reservoir, and their interaction.
The ANCOVAs were done with indicators of hydrological alteration (lIHA) data from both the current and the preceding water years in which Neosho madtom densities were collected.
Each water year extended from 1 October of the previous calendar year to 30 September of the current calendar year. Above John Redmond Reservoir and below the confluence of the Neosho and Cottonwood rivers, IHA values were based on the combined data from the gauges at Plymouth and Americus, Kansas (Figure 1). Neosho madtom densities were grouped and averaged for each year by the nearest U.S. Geological Survey stream gauge (N = 28).Coefficient from Coefficient from preceding water current water Measure year (P) year (P)Monthly discharge (m 3/sec)Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Magnitude and duration of annual extremes (m 3/sec)1-d minimum 1-d maximum 3-d minimum 3-d maximum 7-d minimum 7-d maximum 30-d minimum 30-d maximum 90-d minimum 90-d maximum Timinj of annual extremes (day of the year)Annual minimum Annual maximum Rate and frequency of change in conditions Fall rate (m 3 isec)Rise rate (m 3/sec)Number of reversals between rising and falling discharges Frequency and duration of low and high pulses Low-pulse length (d)Low-pulse count High-pulse length (d)High-pulse count 0.75 0.61 0.93 0.84 0.69 0.99 0.44 0.75 0.60 0.56 1.03 0.75 0.61 0.69 0.79 0.45 0.84 0.38 0.70 0.74 0.95 0.80 (0.053)(0.13)(0.034)(0.048)(0.092)(0.069)(0.39)(0.16)(0.17)(0.24)(0.031)(0.081)(0.048)(0.25)(0.023)(0.40)(0.023)(0.51)(0.066)(0.25)(0.064)(0.21)1.66 0.85 0.59 0.60 0.65 0.64 0.44 0.83 0.94 0.91 0.89 0.60 0.76 0.50 0.98 0.46 1.12 0.42 1.41 0.57 0.94 0.68 (0.0081)(0.14)(0.22)(0.17)(0.12)(0.15)(0.24)(0.049)(0.0093)(0.06)(0.047)(0.073)(0.0 16)(0.14)(0.0057)(0.17)(0.0031)(0.24)(0.0022)(0.17)(0.010)(0.12)0.01 (0.011) 0.00092 (0.78)0.0057 (0.16) 0.002 (0.64)-0.00012 (0.85)0.46 (0.44)0.013 (0.23)0.08 (0.86)-0.55 (0.32)-0.02 (0.75)0.02 (0.60)-0.00018 (0.80)0.18 (0.39)0.00096 (0.88)-0.14 (0.74)-0.48 (0.36)-0.20 (0.86)0.01 (0.86)combined, channel catfish, and stonecats and the positive correlations between velocity and Neosho madtom and stonecat densities support the obser-vation that depth and velocity are important de-terminants of habitat quality for Neosho madtoms and other catfishes in the Neosho River basin (Moss 1983; Fuselier and Edds 1994). Based on these observations, we suggest that the lower post-dam low flows downstream from the dam limit availability of suitable habitat for Neosho mad-toms, a phenomenon that has been observed for fishes in other systems that have been altered by dams (e.g., Aadland 1993). Further support for the need to increase minimum flows downstream of John Redmond Dam is given by Kinsolving and Bain (1993), who documented recovery ofriverine fishes along a disturbance gradient downstream of a hydroelectric dam, and Travnichek et al. (1995), who found that initiation of a minimum-flow re-gime below the same dam increased the number 1276 WILDHABER ET AL.10001 a)U, 4~~E a)0, 1~U U, 0 (U a)1001 101 11 M M V E I 0 M M V V V K 101 CD 0 CO)0 CL 0-.0 2 lu C)Qi V V*A M V*1_L_1 2 1+2 3 4 5 6 Gauge FIGURE 6.-Mean values for minimum (solid symbols) and maximum (open symbols) flow measures and mean annual density of Neosho madtoms (M) at gravel bars nearest a given gauge for each U.S. Geological Survey gauging station for water years 1990-1998.
The symbols represent the following time intervals:
squares, I d: upward triangles, 3 d; circles, 7 d; downward triangles, 30 d; and stars, 90 d. The gauges are ordered from upstream to downstream as follows (see Figure 1): 1 = Plymouth.
Kansas; 2 = Americus, Kansas; 3 = Burlington, Kansas;4 = lola, Kansas; 5 = Parsons, Kansas; and 6 = Commerce, Oklahoma.of riverine fish species collected 3 km downstream of the dam. Travnichek et al. (1995) found that the impact on the fish community was greater at the most upstream site below the dam.The combination of lower minimum flows and lower short-term maximum flow is also likely to indirectly limit the available area of riffle habitat that is suitable for Neosho madtoms by increasing consolidation of gravel bars. Though the substrate is coarser below John Redmond Dam, as indicated by the fredle index and substrate size distribution, it still contains a moderate amount of fine materials (<2 mm; Table 2). It is these finer sediments that allow for the "cementation" of gravel bars during low-water periods that has been hypothesized as a potential limiting factor for Neosho madtoms (Deacon 1961; Moss 1981). Deacon (1961) dem-onstrated a negative relationship between Neosho madtom abundance and drought. Deacon (1961)and Moss (1981) hypothesized that the drying of organic matter during drought results in the ce-mnenting together of the gravel that composes bars normally occupied by Neosho madtoms. In support of the cementation argument, Cross and Moss (1987) documented a change in the substrate from loose and "quick" to firm and stable in the lower Kansas River after construction of an extensive reservoir system. Moss (1981) further suggests that the compaction and cementation of gravel bars, which occurs during periods of low water, hinders recovery of the gravel substrate to the loose consistency preferred by Neosho madtoms that was present before drying. Bulger (1999) hy-pothesized that the periods of low water resulting in cementation of gravel bar sediments would tend to exclude Neosho madtoms from the interstitial spaces and the habitat in which they are predom-inantly found (Wenke et al. 1992; Fuselier and Edds 1994). Furthermore, this cementation of gravel during low-water periods would increase bar stability and thus increase the flows necessary to unconsolidate the gravel (Gordon et al. 1992).
ICTALURID POPULATIONS AND STREAM HABITAT 1277 Therefore, the lower short-term maximum flows vations further support the contention that Neosho that now occur in the Neosho River below John madtom densities below John Redmond Dam are Redmond Dam probably limit the ability of the at least partially the result of direct and indirect river to reverse the consolidation of gravel that effects of dam operation that occur far down-occurs as a result of drying during low-water stream.flows. a& The alterations in water quality and habitat that Our results suggest that minimum flows and occur simultaneously with changes in hydrology their timing are critical for the reproductive suc- as the result of damming may be as important as cess ofNeosho madtoms. The USFWS (1991) sug- the changes in hydrology themselves in impacting gested that certain minimum flows and the timing fish communities (Cross and Moss 1987; Patton of the spring rise may be critical to successful and Hubert 1993). Our results suggest that, along reproduction of the Neosho madtom. The positive with the changes in water flow, changes in water relationship that we observed between Neosho quality and habitat may be negatively affecting madtom densities and average daily flows in May Neosho madtom populations below John Redmond and June supports the hypothesis that timing of Dam. Unfortunately, our water quality and habitat the spring rise in water may be critical for suc- measurements were limited to the first year of fish cessful reproduction of the Neosho madtom. _ collection and thus precluded more complex anal-The relationships between Neosho madtom den- yses that would have allowed us to better deter-sity and average daily flow during August, Octo- mine the effect of water quality and habitat alter-ber, December, and January suggest that minimum ations on Neosho River ictalurid populations.
The flows may be critical to overwinter survival of importance of water quality and habitat to Neosho Neosho madtoms. Previous research has indicated madtom populations has been previously docu-that a larger body size (Miranda and Hubbard mented (Moss 1983; Fuselier and Edds 1995;1994a; Smith and Griffith 1994) and the presence Wildhaber et al. 2000).of shelter (Miranda and Hubbard 1994b; Quinn Changes in hydrology, habitat, or water quality and Peterson 1996) increased overwinter survival that increase Neosho madtom populations will for fish species such as largemouth bass Microp- probably increase the populations of other riffle-terus salmoides, rainbow trout Oncorhynchus my- dwelling benthic fishes. The strong positive rela-kiss, and coho salmon Oncorhynchus kisutch. It tionships between Neosho madtom densities and may be that inadequate flows just prior to and dur- those of all other catfishes combined, channel cat-ing the overwintering period affect Neosho mad- fish, and stonecats suggest that the status of the tom populations by limiting habitat resources and Neosho madtom may be an indicator of environ-thereby decreasing their chance for overwinter surT mental impacts on all the benthic fish communities vival. The positive relationship between annual of the Neosho and Cottonwood rivers. Previous minimum flow date and Neosho madtom densities studies of environmental impacts on Neosho mad-suggests that delaying minimum flows until after tom populations (Wildhaber et al. 1999, 2000)population recruitment in the fall may also en- have shown positive relationships between the hance Neosho madtom population numbers. In- densities of Neosho madtoms, channel catfish, and creased minimum flows would provide more hab- other riffle-dwelling benthic fishes, such as darters.itat for the Neosho madtom (a fluvial specialist), The Neosho madtom is a better indicator species 47 especially in the form of shallow edge habitat over for the Neosho River benthic fish community than gravel bars. other, more abundant species because it seems to The lower Neosho madtom densities that were be more sensitive to environmental impacts (Fu-observed downstream from John Redmond Dam selier and Edds 1994; Wildhaber et al. 2000) as a do not reflect a natural longitudinal gradient in the result of its habitat specificity and short life cycle species' abundance.
Even though the Neosho Riv- (Moss 1983; Fuselier and Edds 1994). This sen-er increases in size and the hydrologic patterns sitivity and the ease with which Neosho madtoms recover to levels more like those above the res- can be collected in reasonable numbers (Figure 6)ervoir, Neosho madtom densities do not exhibit suggest that the status of Neosho madtom popu-parallel recovery (Figure 6). Past (Moss 1983) and lations would provide an effective early-warning present data appear to show that densities of Ne- system for environmental impacts that may affect osho madtorns can be as high below the dam as the rest of the Neosho River fish community.
above the reservoir, with no evident longitudinal Further research is needed to directly test the gradient below the dam (Figure 6). These obser- implications of our findings for habitat and water;k 1278 WILDHABER ET AL management strategies within the Neosho River basin. Monitoring of Neosho madtom and other ictalurid populations and physical habitat should continue for the next 3-5 years. Concurrently, flows should be increased below John Redmond Dam so that the flows at Burlington, Kansas, mir-ror the combined flows at Plymouth and Americus to assess if fish populations and habitat respond to changes in flow as predicted by this study. Fur-ther, age and growth information on the widely distributed stonecat should also be collected as a surrogate to Neosho madtom age and growth in-formation owing to the positive relationship be-tween Neosho madtom and stonecat densities that we observed.
Because turbidity limits visual ob-servations in the field, laboratory tests are needed to determine if and how flow affects the repro-ductive behavior and success of Neosho madtoms.Because substrate composition seems to be a crit-ical limiting factor for Neosho madtoms (Moss 1983; Fuselier and Edds 1995; Wildhaber et al.2000), field studies are needed that test the impacts of habitat alterations, such as gravel mining, on Neosho madtom populations.
Implementation of these recommendations, along with those resulting from further research, should lead to a high prob-ability of recovery for Neosho madtom popula-tions in the Neosho River to the levels identified as necessary for delisting of the species as threat-ened (USFWS 1991).Acknowledgments This study was jointly funded and undertaken by the U.S. Geological Survey (USGS), through its Columbia Environmental Research Center (CERC), and the U.S. Fish and Wildlife Service, through its Ecological Services Field Offices in Manhattan, Kansas, and Tulsa, Oklahoma.
We thank T. Mosher, G. Horak, J. Stephen, R. Schultz, J. Silovsky, E. Miller, and S. Lynott of the Kansas Department of Wildlife and Parks; W. Busby of the Kansas Biological Survey; and K. Edgecomb and L. Cory of the Tulsa District of the U.S. Army Corps of Engineers for assistance in collection of field data. We thank D. Lacock of the USGS in Lawrence, Kansas, for supplying us with the hy-drologic data from USGS gauging stations that was used in the IHA analyses.
We also thank L. Sap-pington of CERC for computing assistance throughout the IHA analyses.
We gratefully ac-knowledge the cooperation of the many private landowners in Kansas, Missouri, and Oklahoma who granted us permission to sample on their prop-erty. This manuscript was greatly improved by comments from Z. Bowen of the USGS Midcon-tinent Ecological Science Center. G. R. Luttrell, and two anonymous reviewers.
References Aadland, L. P. 1993. Stream habitat types: their fish assemblages and relationship to flow. North Amer: ican Journal of Fisheries Management 13:790-806.
Bain, M. B., J. T. Finn, and H. E. Booke. 1988. Stream-flow regulation and fish community structure.
Ecol-ogy 69:382-392.
Baxter, R. M. 1985. Environmental effects of reservoirs.
Pages 1-26 in D. Gunnison, editor. Microbial pro-cesses in reservoirs.
Dr. W. Junk, Dordrecht, The Netherlands.
Bryan, J. D., L. G. Hill, and W. H. Neill. 1984. Inter-dependence of acute temperature preference and respiration in the plains minnow. Transactions of the American Fisheries Society 113:557-562.
Bulger, A. G. 1999. Population structure, habitat use, and breeding behavior of the Neosho madtom, No-turus placidus.
Master's thesis. Emporia State Uni-versity, Emporia, Kansas.Cross, F. B., and M. Braasch. 1969. Qualitative changes in the fish-fauna of the upper Neosho River system, 1952-1967.
Transactions of the Kansas Academy of Science 71:350-360.
Cross, F. B., and J. T. Collins. 1995. Fishes in Kansas, 2nd edition. University Press of Kansas, Public Ed-ucation Series, Lawrence.Cross, F B., and.R. E. Moss. 1987. Historic changes in fish communities and aquatic habitats in plains streams of Kansas. Pages 155-165 in W. J. Mat-thews and D. C. Heins, editors. Community and evolutionary ecology of North American stream fishes, 1st edition. University of Oklahoma, Nor-man.Deacon, J. E. 1961. Fish populations, following a drought, in the Neosho and Marais des Cynges riv-ers in Kansas. University of Kansas publications of the Museum of Natural History 13:359-427.
Fusclier, L., and D. R. Edds. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleostei:
Ictaluridae:
Noturus placidus).
Southwestern Natu-ralist 39:217-223.
Fuselier, L., and D. R. Edds. 1995. An artificial riffle as restored habitat for the threatcned Neosho mad-tom. North American Journal of Fisheries Manage-ment 15:499-503.
Gordon, N. D., T. A. McMahon, and B. L. Finlayson.
1992. Stream hydrology:
an introduction for ecol-ogists. Wiley, Chichester, West Sussex, UK.Hesse, L. W., and G. E. Mestl. 1993. An alternative hydrograph for the Missouri River based on the pre-control condition.
North American Journal of Fish-eries Management 13:360-366.
Kinsolving, A. D., and M. B. Bain. 1993. Fish assem-blage recovery along a riverine disturbance gradi-ent. Ecological Applications 3:531-544.
Luttrell, G. R.. R. D. Larson, W. J. Stark, N. A. Ash-baugh, A. A. Echelle, and A. V. Zale. 1992. Status ICTALURID POPULATIONS AND STREAM HABITAT 1279 and distribution of the Neosho madtom (Noturus placidus) in Oklahoma.
Proceedings of the Oklahoma Academy of Science 1992 72:5-6.McMahon, T. E., A. V. Zale, and D. J. Orth. 1996.Aquatic habitat measurements.
Pages83-120 in B.R. Murphy and D. W. Willis, editors. Fisheries tech-niques 2nd edition. American Fisheries Society, Bethesda, Maryland.Milliken, G. G., and D. E. Johnson. 1984. Analysis of messy data, volume 1: designed experiments.
Wads-worth, Belmont, California.
Miranda, L. E., and W. D. Hubbard. 1994a. Length-dependent winter survival and lipid consumption of age-0 largemouth bass in Bay Springs Reservoir, Mississippi.
Transactions of the American Fisheries Society 123:80-87.
Miranda, L. E., and W. D. Hubbard. 1994b. Winter sur-vival of age-0 largemouth bass relative to size, pred-ators, and shelter. North American Journal of Fish-eries Management 14:790-796.
Moss, R. E. 1981. Life history information for the Ne-osho madtom (Noturus placidus).
Kansas Depart-ment of Wildlife and Parks, Final Report 38, Pratt.Moss, R. E. 1983. Microhabitat selection in Neosho River riffles. Doctoral dissertation.
University of Kansas, Lawrence.Nature Conservancy.
1997. Indicators of hydrologic al-teration.
Nature Conservancy and Smythe Scientific Software, Haydon, Colorado.Niemi, G. J., P. DeVore, N. Detenbeck, D. Taylor, A.Lima, J. Pastor, J. D. Yount, and R. J. Naiman. 1990.Overview of case studies on recovery of aquatic systems from disturbance.
Environmental Manage-ment 14:571-587.
Obeng, L. E. 1981. Man's impact on tropical rivers.Pages 265-288 in M. A. Lock and D. D. Williams, editors. Perspectives in running water ecology. Ple-num, New York.Patton, T. M., and W. A. Hubert. 1993. Reservoirs on a Great Plains stream affect downstream habitat and fish assemblages.
Journal of Freshwater Ecology 8: 279-285.Platts, W. S., W. F. Megahan, and G. W. Minshall.
1983.Methods for evaluating stream, riparian, and biotic conditions.
U.S. Forest Service, Intermountain For-est and Range Experiment Station, U.S. Forest Ser-vice Technical Report INT-138, Denver.Poff, N. L., D. Allan, M. B. Bain, J. R. Karr, K. L.Prestegaard, B. D. Richter, R. E. Sparks, and J. C.Stromberg.
1997. The natural flow regime: a par-adigm for river conservation and restoration.
BioScience 47:769-784.
Putnam, J. E., D. L. Laccock, D. R. Schneider, M. D.Carlson, and B. J. Dague. 1996. Water resource data, Kansas, Water Year 1996. U.S. Geological Survey, Water Resources Division, Water-Data Re-port KS-96-1. Lawrence.
Kansas.Quinn, T. P., and N. P. Peterson.
1996. The influence of habitat complexity and fish size on over-winter sur-vival and growth of individually marked juvenile coho salmon (Oncorhynchus kisutch) in Big Beef Creek, Washington.
Canadian Journal of Fisheries and Aquatic Sciences 53:1555-1564.
Richter, B. D., J. V. Baumgartner, J. Powell, and D. P.Braun. 1996. A method for assessing hydrologic alteration within ecosystems.
Conservation Biology 10:1163-1174.
Richter, B. D.. J. V. Baumgartner, R. Wigington, and D.P. Braun. 1997. How much water does a river need?Freshwater Biology 37:231-249.
SAS Institute.
1990. SAS procedures guide, volume 6, 2nd edition. SAS Institute, Cary, North Carolina.Scheidegger, K. J., and M. B. Bain. 1995. Larval fish distribution and microhabitat use in free-flowing and regulated rivers. Copeia 1995:125-135.
Smith, R. W., and J. S. Griffith.
1994. Survival of rain-bow trout during their first winter in the Henrys Fork of the Snake River, Idaho. Transactions of the American Fisheries Society 123:747-756.
Sparks, R. E. 1992. Risks of altering the hydrologic regime of large rivers. Pages 119-152 in J. Carins, Jr., B. R. Niederlehner, and D. R. Orvos, editors.Predicting ecosystem risk: advances in modern en-vironmental toxicology, volume 20. Princeton Sci-entific Publishing, Princeton, New Jersey.Stanford, J. A., J. V. Ward, W. J. Liss, C. A. Fissell, R.N. Williams, J. A. Lichatowich, and C. C. Coutant.1996. A general protocol for restoration of regu-lated rivers. Regulated Rivers Research and Man-agement 12:391-413.
Steel, R. G. D., and J. H. Torrie. 1980. Principles and procedures of statistics-a biometrical approach, 2nd edition. McGraw-Hill, New York.Stewart-Oaten, A., J. R. Bence, and C. W. Osenberg.1992. Assessing effects of unreplicated perturba-tions: no simple solution.
Ecology 73:1396-1404.
Taylor, W. R. 1969. A revision of the catfish genus No-tutus Rafinesque, with an analysis of higher groups in the Ictaluridae.
Smithsonian Institution Press, U.S. Museum Bulletin 282, Washington, D.C.Travnichek, V. H., M. B. Bain, and M. J. Maceina. 1995.Recovery of a warmwater fish assemblage after the initiation of a minimum-flow release downstream from a hydroelectric dam. Transactions of the Amer-ican Fisheries Society 124:836-844.
Tulsa District USACE (U.S. Army Corps of Engineers).
1993. Tulsa District civil works projects pertinent data. Tulsa District USACE, Tulsa, Oklahoma.USEPA (U.S. Environmental Protection Agency). 1986.Ambient water quality criteria.
USEPA 440/5-86-001, Washington, D.C.USFWS (U.S. Fish and Wildlife Service).
1991. Neosho madtom recovery plan. USFWS, Denver.Wenke, T. L., M. E. Eberle, G. W. Ernsting, and W. J.Stark. 1992. Winter collections of the Neosho mad-tom (Noturusplaciduis).
Southwestern Naturalist 37: 330-333.Wildhaber, M. L., A. L. Allert, and C. J. Schmitt. 1999.Potential effects of interspecific competition on Ne-osho madtoms (Noturusplacidus) populations.
Jour-nal of Freshwater Ecology 14:19-30.Wildhaber, M. L., A. L. Allert, C. J. Schmitt, V. M.Tabor, D. Mulhern, K. L. Powell, and S. P. Sowa.
1280 WILDHABER ET AL.2000. Natural and anthropogenic influences on the distribution of the threatened Neosho madtom in a Midwestern warmwater stream. Transactions of the American Fisheries Society 129:243-261.
Wilkinson, C., D. R. Edds, J. Dorlac, M. L. Wildhaber.
C. J. Schmitt, and A. Allert. 1996. Neosho madtom distribution and abundance in the Spring River.Southwestern Naturalist 41:78-81.Yount, J. D., and G. J. Niemi. 1990. Recovery of lotic communities and ecosystem from disturbance-a narrative review of case studies. Environmental Management 14:547-569.
Copyrighted Material Protect accordingly Environmental Biology of Fishes 70: 80, 2004.C©2004 Kluwer Academic Publishers.
Printed in the Netherlands.
Threatened fishes of the world: Noturus placidus Taylor, 1969 (Ictaluridae)
Janice L. Bryan 5 , Mark L. Wildhaberb
& Douglas B. Noltie 5 2Department of Fisheries and Wildlife Sciences, 302 ABNR, University of Missouri-Columbia, Columbia, MO 65211, U.S.A (e-mail: jbryanC@usgs.gov) bU.S. Geological Survey, Columbia Environmental Research Center, 4200 New Haven Road, Columbia, MO 65201, U.S.A Common names: Neosho madtom (T). Conservation status: '-'listed federally as threatened 22 May 1990 (USEWS 1991).Identification:
One of 25 madtom species, distinguished from four madtoms in its range by two distinct crescent-shaped bands of pig-ment on caudal fin and lack of dark pigment extending to edge of adipose fin (Taylor 1969). Pectoral spines have poorly developed saw-like teeth on front margin. Fin ray counts: anal rays 13-16 (14.72); pelvic rays 8-12 (9.06), soft pelvic rays 7-9 (7.99); caudal rays 49-59 (54.32); vertebrae:
32-36 (33.62). Adults typically greater than 50 mm TL (Bulger & Edds 2001). Males in spawning condition exhibit swollen cephalic epaxial muscles and elongated genital papil-lae; both sexes exhibit reddened tooth patches during spawning season. Photograph by Janice L. Bryan. Distribution:
Endemic to Neosho, River basin in Kansas, Missouri, and Oklahoma (Taylor 1969). Species' range historically extended south to Illinois River in Oklahoma;currently restricted by reservoirs to approximately two-thirds of original range (Moss 1981). Abundance:
Large population fluctuations occur seasonally and annually (Moss 1981). In I year, Wilkinson et al. (1996) reported a mean summer density of occurrence in the Spring River of 11.3 fish versus 30.0 per 100 im in the autumn when generational overlap ocurrs. Higher mean autumn densities found in the Neosho River compared to the Spring River (Wildhaber et al. 2000a). Habitat and ecology: Occur over gravel bars and riffles in fourth to sixth order streams having moderate current, permanent flow, and unconsolidated gravel (Fuselier
& Edds 1994; Bulger & Edds 2001).Nocturnal benthic insectivores gathering prey from gravel interstices (Moss 1981). In nature, young-of-the-year (YOY) 15-49 mm TL;mature individuals reach 33-82 mm TL by end of first year (Fuselier
& Edds 1994; Bulger & Edds 2001). Suspected life span 1-2 years.Reproduction:
Sexual maturity reached during first year (Bulger & Edds 2001). Spawning occurs mid-summer, YOY found July-August.
Nests excavated beneath large rocks in unconsolidated gravel. Male parental care lasts 18-19 days (Bulger et al. 2002). Clutch size 60;mean chorion diameter 3.1 mnm. Threats: Impoundments restrict migration, alter natural hydrograph, and with gravel mining, eliminate preferred habitat (USFWS 1991). Water quality and quantity impacted by municipalities, agriculture, zinc-lead mining, urbanization, and industrialization (Wildhaber et al. 2000a). Conservation actions: Federally listed as threatened in 1990; threatened status in Kansas and endangered in Missouri and Oklahoma.
Recovery plan published in 1991 (USFWS 1991). Five-year moratorium on gravel mining in Neosho River instituted spring 1991. Annual population monitoring conducted since 1991 (Wildhaber et al. 2000b). Recent and ongoing research efforts focused on Neosho madtom ecology. Conservation recommendations:
Continue to limit gravel mining, prohibit dam construction, and encourage removal of unused dams. Further research needed on life history. Remarks: Cooperation of government agencies and private landowners crucial to recovery.Bulger, A.G. & D.R. Edds. 2001. Population structure and habitat use in Neosho madtomn (Noturus placidus).
Southwest Nat. 46: 8-15.Bulger, A.G., C.D. Wilkinson, D.R. Edds & M.L. Wildhaber.
2002. Breeding behavior and reproductive life history of the Neosho madtom, Nonnrusptacidus (Teleostei:
Ictaluridae).
Trans. Kans. Acad. Sci. 105: 106-124.Fuselier, L.C. & D.R. Edds. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleostei:
Ictaluridae:
Noturus placidus).
Southwest Nat. 39: 217-223. .Moss, R.E. 1981. Life history information for the Neosho madtom, Noturus placidus.
Kansas Department Wildlife Parks. Final Report. Contract No. 38. Pratt, Kansas. 33 pp.Taylor, W.R. 1969. A revision of the catfish genus Noturus Rafinesque, with an analysis of higher groups in the Ictaluridae.
U.S. National Museum Bulletin 282.Smithsonian Institution Press, Washington D.C. 259 pp.USFWS. 1991. Neosho madtom recovery plan. U.S. Fish and Wildlife Service, Denver, Colorado.
42 pp.Wildhaber, M.L., A.L. Allert, CJ. Schmitt, V.M. Tabor, D. Mulhem, K.L. Powell & S.P. Sowa. 2000a. Natural and anthropogenic influences on the distribution of the threatened Neosho madtom in a midwestern warmwater stream. Trans. Amer. Fish. Soc. 129: 243-261.Wildhaber, M.L, W.M. Tabor, J.E. Whitaker.
A.L. Allen, D.W. Mulhem, PJ. Ia.mberson
& ILL. Powell. 2000b. Ictalurid populations in relation to the presence of a main-stem reservoir in a midwestern warmwater stream with emphasis on the threatened Neosho madtom. Trans. Amer. Fish. Soc. 129: 1264-1280.
Wilkinson, C.D., D.R. Edds, J. Dorlac, M.L. Wildhaber, CJ. Schmitt & A.L. Allen. 1996. Neosho madtom distribution and abundance in the Spring River.Southwest.
Nat. 41: 78-41.
79 Reznick, D.N., H.A. Bryga & J.A. Endler. 1990. Experimentally induced life-history evolution in a natural population.
Nature 346: 357-359.Reznick, D.N., D.B. Miles & S. Winslow. 1992. Life history of Poecilia picta (Poeciliidae) from the island of Trinidad.
Copeia 1992: 782-790.Reznick, D.N., F.H. Rodd & M. Cardenas.
1996. Life-history evolution in guppies (Poecilia reticulata:
Poeciliidae).
IV.Parallelism in life-history phenotypes.
Am. Nat. 147: 319-338.Rodd, F.H., D.N. Remnick & M.B. Sokolowski.
1997. Phenotypic plasticity in the life history traits of guppies: Responses to social environment.
Ecology 78: 419-433.Roff, D.A. 2001. Life History Evolution.
Sinauer Assoc.Sunderland, MA. 465 pp.Sokal, R.R. & FJ. Rohlf. 1981. Biometry, 2nd edition. Freeman, San Francisco.
859 pp.Steams, S.C. 1976. Life history tactics: A review of ideas. Q. Rev.Biol. 51: 3-47.Steams, S.C. 1983. The evolution of life-history traits in mosquitofish since their introduction to Hawaii in 1905: Rates of evolution, heritabilities and developmental plasticity.
Am.Zool. 23: 65-75.Steams, S.C. 1992. The Evolution of Life Histories.
Oxford University Press, New York. 320 pp.Steams, S.C. & R.D. Sage. 1980. Maladaptation in a marginal population of the mosquitofish Gambusia affinis. Evolution 34: 65-75.Todd, R.G. 1979. Selected aspects of the biological control of mosquitoes in the Cayman Islands. Ph.D. Thesis, University of Reading, U.K. 274 pp.Trexler, J.C. 1985. Variation in the degree of viviparity in the sailfin molly, Poecilia latipinna.
Copeia 1985: 999-1004.Trexler, J.C. 1997. Resource availability and plasticity in off-spring provisioning:
Embryo nourishment in sailfin mollies.Ecology 78: 1370-1381.
Turner, C.L. 1941. Morphogenesis of the gonopodium in Gambusia affmis affinis. J. Morph. 69: 161-185.Vondracek, B.W., W.A. Wurtsbaugh
& J.J. Cech, Jr. 1988. Growth and reproduction of the mosquitofish, Gambusia affmis, in rela-tion to temperature and ration level: Consequences for life history. 6nv. Biol. Fish. 21: 45-57.Williams, G.C. 1966. Adaptation and Natural Selection:.
A Critique of Some Current Evolutionary Thought, Princeton University Press, Princeton, NJ. 320 pp.Winemiller, K.O. & K.A. Rose. 1992. Patterns of life-history diversification in North American fishes: Implica-tions for population regulation.
Can. J. Fish. Aquat. Sci. 49: 2196-2218.
Copyrighted Material Protect accordingly
[Article)North American Journal of Aquaculure 67:221-230, 2005 0 Copyright by the American Fisheries Society 2005 DOI: 10.1577/A04-020.1 0 Examining Neosho Madtom Reproductive Biology Using Ultrasound and Artificial Photothermal Cycles JANICE L. BRYAN*I Department of Fisheries and Wildlife Sciences, School of Natural Resources, 302 Anheuser-Busch Natural Resources Building, University of Missouri-Columbia, Columbia, Missouri 65211-7240, USA MARK L. WILDHABER U.S. Geological Survey, Columbia Environmental Research Center, 4200 New Haven Road, Columbia, Missouri 65201, USA DOUGLAS B..NOLTIE Department of Fisheries and Wildlife Sciences, School of Natural Resources, 302 Anheuser-Busch Natural Resources Building, University of Missouri-Columbia, Columbia, Missouri 65211-7240, USA Abstract.-We examined whether extended laboratory simulation of natural photothermal con-ditions could stimulate reproduction in the Neosho madtom Noturus placidus, a federally threatened species. For 3 years, a captive population of Neosho madtoms was maintained under simulated natural conditions and monitored routinely with ultrasound for reproductive condition.
Female Neosho madtoms cycled in and out of spawning condition, producing and absorbing oocytes annually.
Internal measurements made by means of ultrasound indicated the summer mean oocyte size remained consistent over the years, although estimated fecundity increased with increasing fish length. In the summer of 2001, after 3 years in the simulated natural environment, 13 out of 41 fish participated in 10 spawnings.
Simulation of the natural photothermal environment, coupled with within-day temperature fluctuations during the spring rise, seemed important for the spawning of captive Neosho madtoms. The use of ultrasound to assess the reproductive status in Neosho madtoms was effective and resulted in negligible stress or injury to the fish. These procedures may facilitate future culture of this species and other madtoms Noturus spp., especially when species are rare, threatened, or endangered.
Captive breeding programs provide knowledge of the reproductive biology of rare species and assist in the recovery of endangered and threatened fish species (e.g., bonytail chub Gila elegans, humpback chub Gila cypha, razorback sucker Xy-rauchen texanus, and Colorado pikeminnow Ptych-ocheilus lucius; see Haniman 1982a, 1982b, 1985, 1986). To increase the knowledge base, we inves-tigated the effects of natural photothermal cycles on the reproductive cycles of a captive population of Neosho madtoms Noturus placidus.The Neosho madtom is one of five madtoms Noturus spp. included on the U.S. list of threatened and endangered species (USFWS 1991; Burr and Corresponding author: jbryan@usgs.gov I Present address: Johnson Controls World Services, Inc., U.S. Geological Survey, Columbia Environmental Research Center, 4200 New Haven Road, Columbia, Missouri 65201, USA.Received April 14, 2004; accepted February 11, 2005 Published online June 17, 2005 Stoeckel 1999). They are endemic to the Neosho River basin in Kansas, Missouri, and Oklahoma, and are mostly found at main-stem gravel bars (Taylor 1969; Moss 1983; Fuselier and Edds 1994;Bulger and Edds 2001).As in other madtoms, spawning occurs May through July as temperatures approach 25"C (Pfingsten and Edds 1994; Bulger and Edds 2001;Bulger et al. 2002b). Nests are constructed under large objects in the gravel (Bulger et al. 2002a, 2002b). Observations of female parental care in madtoms are rare (Burr and Stoeckel 1999); only male parental care has been observed in Neosho madtoms and it lasts for 8-9 d after spawning (Bul-ger et al. 2002b). Short life spans of Neosho mad-toms restrict reproduction to one or two spawning seasons (Fuselier and Edds 1994; Bulger and Edds 2001).The use of ultrasound for sexing large common fish species is not new (Martin et al. 1983) and has recently been used on threatened and endan-gered species (Moghim et al. 2002; Columbo et 221 222 BRYAN ET AL.al. 2004; Wildhaber et al. 2005). Ultrasound has been especially useful in identifying gravid fe-males in fish species with nondescript sexual char-acteristics (Mattson 1991; Blythe et al. 1994; Kar-lsen and Holm 1994; Martin-Robichaud and Rom-mens 2001). Even though madtoms are sexually dimorphic during part of the year (Pfingsten and Edds 1994; Bulger and Edds 2001), researchers have had difficulty distinguishing madtom gender (Bulger et al. 2002a). In addition to gender iden-tification, ultrasound techniques have been used to measure internal reproductive structures, such as oocyte diameter and gonad size (Mattson 1991;Blythe et al. 1994; Martin-Robichaud and Rom-mens 2001). Standard oocyte collection techniques disrupt female maturation in margined madtoms N. insignis (Stoeckel and Neves 2001), so alter-native techniques such as ultrasound need to be examined for their potential use on madtoms. Be'-cause madtoms are difficult to sex, obtain oocyte samples, and many species are threatened or. en-dangered (Burr and Stoeckel 1999; Bulger et al.2002a), we employed ultrasound techniques to help us better understand the Neosho madtom re-productive cycle.Photothermal cycles influence the reproductive chronology of madtoms (Stoeckel 1993; Burr and Stoeckel 1999). Bulger et al. (2002a) showed that greater photoperiod length increases the amount of Neosho madtom reproductive behavior.
Stoeck-el (1993) also found that margined madtoms ex-posed 1 year to a natural photothermal cycle in the laboratory had a similar gonadosomatic index pat-tern as wild fish. Not known is whether exposure to successive years of artificial photothermal cy-cles will induce yearly gonadal maturation and spawning of captive populations of Neosho mad-toms. For the Neosho madtom and other endan-gered madtom species, knowing how environmen-tal cues influence reproduction (and having a tech-nique to assess reproductive status) will help prop-agate .and manage these populations.
The objectives of this study were to examine (1) the use and applicability of ultrasound to assess the reproductive status of Neosho madtoms, and (2)whether simulated natural photothermal cycles can trigger gonadal development and spawning of the Neosho madtom in successive years.Methods Laboratory conditions.-Neosho madtoms were collected from Kansas sections of the Neosho (n= 24) and Cottonwood rivers (n = 34) in the sum-mer of 1998 (Bulger et al. 1998). These fish were assumed to be 1-year-olds because of their small size. They were maintained by Bulger et al. (1998)in 1998 and by us from January 1999 until August 2001. Fish were kept in a 720-L Living Stream (Frigid Units, Inc., Toledo, Ohio, Model LS-900)in a photoperiod-and temperature-controlled room. Males and females were initially sexed based on their external secondary sexual charac-teristics (Bulger et al. 2002b) and were separated in the Living Stream. Fish were fed to satiation 3 times/week with frozen brine shrimp Anemia spp., frozen bloodworms Chironomus spp., or Hikari Sinking Carnivore Pellets (Kyorin Company, Ltd., Himeji, Japan). Excess food -was removed rou-tinely to prevent water fouling.Photothermal cycles.-Laboratory photoperiods mimicked the times of sunrise and sunset for Em-poria, Kansas (38 0 26'N, 96"12'W), near the lo-cation where the fish were collected (Figure 1; U.S.Naval Observatory 1999-2001).
Illumination was provided by eight General Electric (GE) Cool White 34-W fluorescent bulbs suspended 0.6-1.4 m from the water surface. Lights were controlled by a Precision time switch (Precision Multiple Controls, Inc., Midland Park, New Jersey, Model CD103).To simulate natural thermal cycles, water tem-peratures in the Living Stream were regulated with a 0.746-kW water heater-chiller (Frigid Units, Model DI-100, 3000-W heater). In the individual aquaria, we used 300-W Visi-Therm.
aquarium heaters (Aquarium Systems, Inc., Mentor, Ohio)to regulate water temperatures.
Water temperatures mimicked monthly Neosho River temperatures re-corded from August 1996 to October 1997 and from April to. July 1998 (Figure 1; Bulger et al.1998), except the minimum winter temperature of 9 0 C (Figure 1) was warmer than the river winter low of 0*C because of chiller limitations.
In 1999 and 2000, once water temperatures reached 25"C, we varied the laboratory within-day water tem-peratures to simulate natural daily fluctuations (Figure 1, inset). They were varied daily in 1999 between an evening low of 22"C and a daily high of 25 0 C (morning increase at 1100 hours0.0127 days <br />0.306 hours <br />0.00182 weeks <br />4.1855e-4 months <br /> and even-ing decrease at 2300 hours0.0266 days <br />0.639 hours <br />0.0038 weeks <br />8.7515e-4 months <br />), and in 2000 they were varied between 23"C and 25"C. We also varied temperatures in 2001 during the spring rise starting at 22"C (Figure 1). During the 4-week spring rise, the respective daily temperature ranges were 21-23°C, 21-24 0 C, 21-25,C, and 21-26"C. After the 4th week, the temperature ranged between 21*C and 25"C. Daily temperature fluctuations were continued until fish ceased reproductive activity.*'i NEOSHO MADTOM REPRODUCTIVE CYCLE 223 35.30O 0 25-20.15.E 1.-5-0 20 15 0 2 5 CL 0 Fish spawning.-At the beginning of each lab-oratory spawning season, males and females that exhibited prominent secondary sexual character-istics were paired and placed into separate aquaria.Each pair was provided with an inverted half-sec-tion of polyvinyl chloride pipe (1999: 13.3 cm long x 10.2 cm wide X 5.0 cm high; 2000: 12.7 cm long X 14.0 cm wide X 6.4 cm high). Pairings were maintained until all reproductive activity ceased. If a pair spawned, the female was replaced with another gravid female. Because of the noc-turnal nature of Neosho madtoms, we recorded all activities with video cameras. For each videotaped spawning, the number of eggs per clutch was de- -termined by measuring the chorion diameters of all visible eggs in the clutch (approximately 50-75) with Optimas image analysis software (Media Cybernetics 1999). We then calculated each egg's area (nr[diameter/2]
2). Clutch size was estimated by calculating the total area occupied by all layers that comprised the egg mass and dividing the total area by the mean egg area for that clutch. After spawning, females were examined with ultrasound (see Ultrasound Examinations) to determine whether or not they had expelled all eggs.Length and weight measurements.-We assessed the size and reproductive status of the fish in our captive population of 2-year-old and older Neosho madtoms. During the summer and winter of 2000 and 2001, we measured the lengths and weights of either all fish or a random subsample (Table 1).We did not include fish measured in summer 1999 because this subsample consisted of only the larg-est fish (Table 1).N -N./ N. N. /Month FIGURE 1.-Laboratory (solid lines) and correspond-ing natural (dashed lines) temperatures and photoperiods for the years 1999 through 2001 during a laboratory assessment of the reproductive behavior of Neosho mad-toms from the Neosho and Cottonwood rivers in Kansas.The inset gives an example of a daily temperature fluc-tuation during the spawning season.TABLE 1.-Number of Neosho madtoms during each length-weight examination during a laboratory assessment of their reproductive behavior.
Numbers in parentheses represent numbers of fish for which data were not available.
Ovaries Contained eggs Length and and Date Sex Live fish Mortalities weight fecundity n %Winter 1999 V 36 (start of study) 6 22 Summer 1999 V 28 8 23& (5) 21 (7) 21 100 (17 Aug) 6 22 0 13a (9) 16(6) 0 Winter 2000 9 27 1 26 (1) 27 1 4 (7 Jan) 6 21 1 19 (2) 12(9) 0 Summer 2000 9 27 0 15 (12) 25 (2 b) 27 100 (24 May) a 20 1 14 (6) 17 (3) 0 Winter 2001 9 27 0 27 27 7 26 (26 Feb) 6 15 5 14 (1) 15 0 Summer 2001 9 27 0 27 25 (2c) 25 93 (I May) 6 14 1 13 (1) 14 0 Total 228 17 Body measurements not used because of a size-biased sample of fish.b Unable to acquire ovary depth or length because ovary periphery was outside ultrasound image.c Females did not contain eggs.
224 BRYAN ET AL.Fish to be measured were anesthetized with tri-caine methanesulfonate (MS-222 ; Argent Chem-ical Laboratories, Inc., Redmond, Washington) and placed into a transparent, water-filled pan. Total length (TL) was measured to the nearest millimeter with a ruler placed under the pan. Fish wet weight was measured to the nearest 0. .g with a Mettler top-loading balance (Mettler Toledo, Inc., Colum-bus, Ohio, Model P120). We calculated the logio weight: log 1 0 length relationship by means of lin-ear regression (SAS 1992).Ultrasound examinations.-We examined fish with ultrasound to monitor fish gonadal develop-ment by means of either a GE LOGIQ 700 Expert (GE Medical Systems, Waukesha, Wisconsin) with an 8- or 13-MHz probe (summer 1999, winter and summer 2000) or a Shimadzu SDU-400 Plus (Shi[madzu Corporation, Kyoto, Japan) with a 7.5-MHz probe (winter and summer 2001). We referred to Zweibel and Sohaey (1998) for correct ultrasound technique.
We examined each surviving fish each summer and winter between summer 1999 and summer 2001 (Table 1). To acquire gonad images, each fish was first anesthetized with MS-222 and then rolled belly-up in a pan of water so that the ultrasound probe could be placed against the fish's abdomen.Each ultrasound image depicted a two-dimensional longitudinal section through the sagittal plane of the body cavity (Figure 2).Ovary and fecundity measurements.-We mea-sured the diameter of each clearly defined oocyte in the ultrasound image with Optimas image anal-ysis software (Media Cybernetics 1999). By mea-suring only clearly defined oocytes, we minimized the use of oocytes that were not scanned through their equators.
Oocyte volume was estimated by means of a cubic volume equation (volume =mean diameter 3) to incorporate the interstitial spaces among the oocytes.The estimate of a fish's ovary volume was based on the ovary, pair's being ellipsoidal in shape (el-lipsoid volume = [47rr3]abc, where a, b, and c are the three respective orthogonal axes. [length, width, and depth]; Beyer 1981). Length and depth of the ovary were measured from the ultrasound images each year. Because the ultrasound image showed the thickness of the body wall to be neg-ligible (Figure 2), we measured ovary width ex-ternally in summer 2001. For summers 1999.and 2000, we estimated ovary width from ovary depth with the equation, ovary width = 1.862 + 0.346 X ovary depth (n = 24; r 2 = 0.16; P = 0.0561).The ovary width of one fish from summer 2001 FIGURE 2.-Ultrasound images of captive female Ne-osho madtoms during a laboratory assessment of the reproductive status of Neosho madtoms, 1999-2001.
Image A is of a gravid female during summer and image B is of a female without visible eggs during winter. The images are of the sagittal plane of the fish (anterior to the left, ventral on the top, posterior to the right, and dorsal on the bottom). Annotations A, B, and C in image A indicate oocyte diameter measurements of 2.3, 2.0, and 2.2 mm, respectively.
was also estimated this way to compensate for an erroneous body width measurement.
The fecundity estimate was calculated by dividing each fish's ovary volume by its mean oocyte volume.To test for yearly differences in mean oocyte diameter and estimated fecundity, we used a mul-tivariate analysis of covariance (MANCOVA;model df = 3, error df = 67; SAS 1992) and ex-amined the resulting value of Wilk's lambda, where the dependent Variables were oocyte di-ameter and fecundity.
and the independent vari-ables were year and TL. Previous research has found a positive relationship between madtom length and fecundity (Burr and Stoeckel 1999);consequently, we used length as a covariate.
Fe-cundity data were square-root transformed to ho-mogenize variance and normalize the distribution.
Results Each year as spring temperatures reached 15-17°C (April-May), captive males and females de-veloped secondary sexual characteristics typical of wild fish (Bulger et al. 2002b). Average mortality among years was similar (Table 1). Female mor-NEOSHO MADTOM REPRODUCTIVE CYCLE 225 1- 150 E 100 4-'"50 0)0 4--0 FIGURE 1998 throu their repro bars repres males, soli oratory. Th in early su the numbe.15 14 range = 4.0-12.0 mi, mean CV = 13.1%, range 20015 = 6.8-39.3%;
2000: mean = 15.6 mm, range =5.0-40.0 mm, mean CV = 13.5%, range = 6.5-227.2%; 2001: mean = 14.7 rmm, range = 5.0-25.0 22 7 26 mm, mean CV = 10.0%, range = 5.2-19.8%).
41 Females exhibited a cyclic pattern of oocyte de-velopment in spring and resorption in winter (Ta-ble 1). Only 4-26% of the females contained dis-cernible oocytes during the winter. Oocytes visible in the winter were either very large oocytes being resorbed (1 female in 2000) or very small oocytes 0 1 2 3 4 5 being produced for the following summer (7 fe-Age males in 2001; Table 1). Overall, 93-100% of the females developed oocytes during the summers,.-Growth of captive Neosho madtoms from and each summer's average fecundity estimate in-ugh 2001 during a laboratory assessment of ductive behavior.
Points represent means and creased (1999, 121 oocytes; 2000, 210 oocytes;sent minima and maxima. Circles (open = 2001, 234 oocytes).
For all 3 years, the fecundity d = females) represent fish used in the lab- estimates greatly exceeded those reported previ-he triangle represents fish at the time of capture ously for Neosho madtoms, 79 oocytes being the immer 1998. The numbers on the graph are largest number of oocytes found in a gravid Ne-rs of fish in each sample. osho madtom (Bulger et al. 2002b).In summer 1999 we obtained ultrasound images from 28 females, 9 of which lacked a TL mea-summer 1999 resulted from a bacterial surement.
To estimate the TL of these fish for sub-whereas male mortalities in summer sequent length: fecundity analysis, we used esti-the result of male: male aggression (Ta- mates derived from the summer 2000 and 2001 regression relationship:
female madtom TL =80.75 + 7.551 X ovary length (n = 48; r 2 = 0.48;l Weight Measurements P < 0.0001). The MANCOVA model, which in-ivity, the Neosho madtoms grew to un- cluded mean oocyte diameter and fecundity as de-ed sizes and ages (Figure 3). The one pendent variables, detected significant effects of was a single male that failed to grow year and fish TL (Wilk's lambda, P < 0.0001 and emaciated throughout the 3-year study; P = 0.0004, respectively).
Mean oocyte diameter ntly, it was eliminated from the analysis.
did not vary among years or among fish TL (?I =Ils reached sizes almost twice that seen 71; r 2 = 0.10; P = 0.2747 and P = 0.3459, re-(117.0 mm TL; range, 100.0-135.0 mm spectively; overall mean oocyte diameter = 2.5 e 3; Fuselier and Edds 1994; Bulger and mm, range = 2.0-3.3 mm). The square root of 1). The resulting overall weight: length fecundity significantly increased with year and fish n relationship for the captive Neosho TL (n = 71; r 2 = 0.54; P < 0.0001 and P = 0.0039, was significant (malesand females com- respectively; Figure 4).talities in infection, 2000 wern ble 1).Length ai In capt precedent exception and was consequer Individua in nature TL; Figur Edds 200 regression madtoms bined across seasons and years; logl 0 weight =-4.033 + 2.613.1ogi 0 TL; n = 155; r 2 = 0.82; P< 0.0001).Ultrasound Examinations
- For Neosho madtom males, we could not dis-tinguish testes from other organs in ultrasound im-ages. In contrast, the images from females clearly showed oocytes; consequently females could be distinguished from males during the spawning sea-son (Figure 2). The number of oocyte diameters measured per fish and the within-fish oocyte di-ameter coefficients of variation (CV) were com-parable among summers (1999: mean = 6.8 mm, Spawnings Spawning only occurred during summer 2001 among 10 of the 17 pairs of Neosho madtoms (Ta-ble 2). Three different males (out of 10) spawned with 10 different females (out of 30). One male spawned with 5 successive females (male ID 3), another male with 3 females (male ID 5), and an-other with 2 females (male ID 4; Table 2). For the first six spawnings, the females expelled all of their eggs, whereas the females in the last four spawn-ings did not (Table 3).The clutch-size average of 230 eggs (range,20-418 eggs) falls within the range of the prespawn 226 BRYAN ET AL.:Z'25~20 U-I15 0)0*ci0 Go 0 Big W 0;IQ A 0 20001 80 100 120 Total Length (mm)FIGURE 4.-Estimated fecundity versus total length of captive Neosho madtoms during a laboratory assessment of their reproductive behavior, 1999-2001.
All years are combined in the length: fecundity equation:
square root of fecundity
= -6.22 + 0.1947 total length (n = 71; r 2= 0.37; P < 0.0001).fecundities we estimated for these females by means of the summer 2001 length: fecundity re-lationship (range, 208-271 eggs; mean, 246 eggs).The mean chorion diameter of the spawned eggs was 3.5 mm, which was higher than the mean pre-spawn ooctye diameter obtained from the ultra-sound images of all fish (summer 2001, mean =2.6 mm). Only the eggs of one spawn developed to the neurula stage, indicating that fertilization had occurred.
These eggs died when we attempted to culture them further. For four spawnings, the eggs stopped developing at the morula stage, sug-gesting that false activation had occurred.
For the five remaining clutches, one or both of the parents consumed the eggs the day after spawning.Discussion Ultrasound proved to be an effective, nonin-vasive tool for determining the sex, oocyte di-ameter, and fecundity of Neosho madtoms. The noninvasive ultrasound allowed us to perform re-peated fish examinations without causing injury or appreciable stress. In addition, the technique was easy to learn. Females were easily distinguished from males during the summer, and developing oocytes could be discerned, even during late-win-ter periods. The application of the ultrasound is limited to females because testes cannot be distin-guished from other organs. This outcome is not surprising because madtom testes are small, elon-gate, and closely associated with the intestines (Clugston and Cooper 1960; Stoeckel 1993;Stoeckel and Burr 1999). Sonography is a viable, noninvasive technique for ascertaining the repro-ductive condition of female Neosho madtoms, an important consideration for a rare species. New portable ultrasound models allow this technique to be used in the field.Madtoms are notorious for not spawning in the laboratory (Burr and Stoeckel 1999); thus, use of daily temperature fluctuations during the pre-spawning and spawning seasons appears to be im-portant for the induction of spawning.
The first 2 years of study did not yield any Neosho madtom spawnings from mature gravid individuals, even when daily temperature fluctuations were incor-porated into the summer spawning season. Spawn-ing occurred only during the third year, when daily temperature fluctuations accompanied the spring and summer periods..TABLE 2.-Summary of Neosho madtom spawning conditions in summer 2001. The temperatures given are the high and low water temperatures on the day of spawning.Date T ID 8 ID Time (hours) Temperature (0 C) Photoperiod (h) Female TL (num)16 Jun 1 4 -2000 21-22 15.0 116 26 Jun 2 3 0031-0127 21-24 15.8 113 30 Jun 3 3 Daylight before 2059 21-24 15.8 105 5 Jul 4 4 -1030 21-25 16.0 113 9 Jul 5 3 Night 21-26 16.0 114 16 Jul 6 3 Afternoon 21-25 16.0 112 22 Jul 7 3 Daylight before 2100 21-25 16.0 114 26 Jul 8 5 1300-1500 21-25 16.0 113 30 Jul 9 5 1300-1430 21-25
- 16.0 112 6 Aug .10 5 -1430 21-25 16.0 105 Mean 112 8 Visual estimate.A NEOSHO MADTOM REPRODUCTIVE CYCLE 227 TABLE 3.-Summary of Neosho madtom spawning results in summer 2001. Mean egg diameter was measured after egg deposition.
Estimated fecundity was calculated from the fecundity
- length relationship for the summer of 2001; nd= not determined.
Mean egg chorion Estimated clutch Estimated All eggs, Date diameter (mm) size fecundity expelled Fertile Clutch outcome 16 Jun 4.0 313 271 Yes nd Consumed next day 26 Jun 3.4 418 253 Yes Yes Dead by 29 Jun 30 Jun 4.1 388 208 Yes nd Female consumed next day 5 Jul 3.7 285 253 Yes No Male parental care Stopped 7 Jul 9 Jul nd nd 259 Yes No Dead on 13 Jul 16 Jul nd 1500 247 Yes No Dead on 18 Jul 22 Jul 3.1 276 259 No No Dead on 26 Jul 26 Jul 3.1 95 253 No nd Consumed next day 30 Jul 3.6 125 247 No nd Male consumed next day 6 Aug 3.2 208 208 No nd Male consumed next day Mean .3.5 230 .246 Laboratory fish exceeded the length, weight, and age maxima recorded for wild Neosho madtoms (Fuselier and Edds 1994; Bulger and Edds 2001).This was probably a result of the ideal captive conditions (i.e., abundance of food, warmer winter temperatures, lack of competitors, and absence of predators).
Relative to data collected on other madtoms following similar methods, the Neosho madtom logl 0 length-weight relationship was most like that of the margined madtom (loglo weight =-4.75 + 2.89.1ogl 0 length; Clugston and Cooper 1960).Even though the ultrasound technique yielded conservative fecundity estimates when compared with actual clutch sizes of females that expelled all eggs, fish fecundities were similar to those of other like-aged madtom species (Burr and Stoeckel 1999). The summer 2001 average fecundity was high for madtoms; only the stonecat N. flavus yielded higher mean fecundity estimates (Langlois 1954, 973 eggs; Walsh and Burr 1985, 278 eggs).This increase in fecundity is attributable to both the larger size and larger abdominal cavity volume of the Neosho madtom females. An increase in fecundity with increasing body size is common among ictalurid species that live longer than 2 years (Burr and Stoeckel 1999). Since wild Neosho madtoms rarely live beyond 2 years (Fuselier and Edds 1994; Bulger and Edds 2001), future prop-agation efforts would benefit by using conditions in this study that resulted in both an increased life span and a higher fecundity of captive Neosho madtoms.The largest Neosho madtom clutch in this study was significantly greater than the previously re-ported maximum (Bulger et al. 2002b). Only the stonecat clutch size exceeds the maximum clutch size estimated in this study (500 eggs; Greeley 1929). Clutch sizes were atypically large because of two factors: the large sizes of the females in-volved and the time at which clutch sizes were estimated.
The clutch size counts from the vid-eotape were made almost immediately after spawning and included infertile eggs as well as eggs that died later or were ingested by the parents.For example, Chan (1.995) found up to a 50% re-duction in brown madtom N. phaeus clutch size in the first 48 h after spawning and attributed this to natural causes.The photothermal conditions of the laboratory spawnings were comparable to the environmental conditions reported in previous Neosho madtom spawnings.
The range of maximum daily temper-atures on the days the fish spawned (22-26°C) was slightly lower than that recorded for previous spawnings (25-28 0 C; Pfingsten and Edds 1994;Bulger et al. 2002b), but was well within the range of other madtom species (Burr and Stoeckel 1999).The photoperiod during the spawnings (15-16 h)was consistent with that of the natural spawning season (Bulger and Edds 2001; Bulger et al. 2002a)and previous Neosho madtom spawnings in the laboratory (Bulger et al. 2002b).Egg diameters of the pre- and postspawn eggs were also comparable to those found in previous studies.;
The prespawn oocyte diameters obtained with ultrasound were the same as those measured from a preserved prespawn Neosho madtom fe-male (mean diameter = 2.5 mm; Bulger et al.2002b) and were comparable to those for other madtom species (Burr and Stoeckel 1999). The postspawn mean egg chorion diameters were sim-228 BRYAN ET AL.ilar to those obtained from previous clutches (3.1 mm, Pfingsten and Edds 1994; 3.1 and 3.7 mm, Bulger et al. 2002b). The apparent increase in egg size after spawning is probably attributable to the water absorption that occurs during water hard-ening (Saunders 1982; Jaffe 1985).As in other madtom studies (Stoeckel and Burr 1999), we noted the high frequency at which the parents consumed their egg clutches after spawn-ing. Human disturbance of the nest has been the predominant rationale offered on why the parents consume the eggs (Stoeckel and Burr 1999). How-ever, some of the egg masses in this study were consumed even when they were not disturbed.
This suggests that other reasons could cause this oop-hagy, including egg infertility or the postspawning confinement of parents. Use of larger tanks has been suggested as a possible solution (Stoeckel and Neves 2001) because it would allow females to leave the area of the nest.Partial spawning is a recognized occurrence in madtoms, as several researchers have reported in-complete egg depositions for some females (May-den and Burr 1981; Walsh and Burr 1985; Dinkins and Shute 1996; Bulger et al. 2002b). The occur-rence of partially spent females, spawned clutch sizes smaller than ovarian egg numbers, or both have been used as evidence of potential polyandry in madtoms (Menzel and Raney 1973; Burr and Stoeckel 1999). However,, in this study partial spawnings during the later part of the 2001 spawn-ing season were probably a consequence of dete-riorating fish condition and suboptimal spawning conditions.
The last four spawnings involved male 3 (that had completed three previous spawnings) and male 5 (that had participated in incomplete spawnings with three different females).
In addi-tion, the photothermal increases had reached a pla-teau by this time (16 h light: 8 h darkness with daily temperature fluctuations between 21°C and 25°C).In conclusion, this study has identified several aspects of Neosho madtom reproductive biology that could facilitate future research and propaga-tion efforts (i.e., serial polygyny, repeated attain-ment of reproductive maturity in sequential years, and sex identification with ultrasound).
Combined, their application may facilitate further life history research and the culture of this and other threat-ened and endangered madtoms.Acknowledgments We would like to thank David R. Edds of Em-poria State University in Kansas and Charles F Rabeni of the University of Missouri-Columbia for their advice and review comments.
Special thanks go to Jimmy Latimer and Kristi Cook of the University of Missouri-Columbia Veterinary Teaching Hospital, who contributed their exper-tise, time, and guidance to the ultrasound assess-ments. Thanks also go to Dan W. Mulhern and Vernon M. Tabor of the U.S. Fish and Wildlife Service, Office of Ecological Services, Manhattan, Kansas, who helped collect the Neosho madtom survey data. Angela Bulger provided help and as-sistance, and researchers and staff at the U.S. Geo-logical Survey Columbia Environmental Research Center gave us support and advice. We are grateful to Jim Randolph of the U.S. Army Corps of En-gineers, Tulsa District, for funding this research.The U.S. Fish and Wildlife Service and Kansas Department of Wildlife and Parks provided per-mits to collect and hold Neosho madtoms. Finally, we thank the manuscript reviewers for their con-structive comments, which include but are not lim-ited to David Edds, Dan Mulhern, and Vernon Ta-bor. This publication is a contribution from the Missouri Agricultural Experiment Station and from the Missouri Cooperative Fish and Wildlife Research Unit (U.S. Geological Survey, Missouri Department of Conservation, University of Mis-souri-Columbia School of Natural Resources, and the Wildlife Management Institute cooperating).
Reference to trade names does not imply endorse-ment by the U.S. Government.
References Beyer, W. H. 1981. CRC standard mathematical tables, 26th edition. CRC Press, Boca Raton, Florida.Blythe, B., L. A. Helfrich, W. E. Beal, B. Bosworth, and G. S. Libey. 1994. Determination of sex and mat-urational status of striped bass (Morone saxatilis) using ultrasonic imaging. Aquaculture 125:175-184.Bulger, A. G., and D. R. Edds. 2001. Population struc-ture and habitat use in Neosho madtom (Noturus placidus).
Southwestern Naturalist 46:8-15.Bulger, A. G., M. L. Wildhaber, and D. R. Edds. 1998.Population dynamics, reproductive biology, and be-havior of the Neosho madtom, Noturus placidus:
a state and federally listed threatened catfish. Final report submitted to the U.S. Army Corps of Engi-neers, Environmental Analysis and Support Branch, Tulsa District, Tulsa, Oklahoma.Bulger, A. G., M. L. Wildhaber, and D. R. Edds. 2002a.Effects of photoperiod on behavior and courtship of the Neosho madtom (Noturus placidus).
Journal of Freshwater Ecology 17:141-150.
Bulger, A. G., C. D. Wilkinson, D. R. Edds, and M. L.Wildhaber.
2002b. Breeding behavior and repro-ductive life history of the Neosho madtom, Noturus NEOSHO MADTOM REPRODUCTIVE CYCLE 229 placidus (Teleostei:
lctaluridae).
Transactions of the Kansas Academy of Science 105:106-124.
Burr, B. M., and J. N. Stoeckel.
1999. The natural history of madtoms (genus Noturus), North Amer-ica's diminutive catfishes.
Pages51-101 in E. R.Irwin, W. A. Hubert, C. F. Rabeni, H. L. Schramm, Jr., and T. Coon, editors. Catfish 2000: proceedings of the international ictalurid symposium.
Ameri-can Fisheries Society, Symposium 24, Bethesda, Maryland.Chan, M. D. 1995. Life history and bioenergetics of the brown madtom, Noturus phaeus. Master's thesis.University of Mississippi, Oxford.Clugston, J. P., and E. L. Cooper. 1960. Growth of the common eastern madtom, Noturus insignis in central Pennsylvania.
Copeia 1960:9-16.
Columbo, R. E., P. S. Wills, and J. E. Garvey. 2004.Use of ultrasound imaging to determine sex of shov-elnose sturgeon.
North American Journal of. Fish-eries Management 24:322-326.
Dinkins, G. R., and P. W. Shute. 1996. Life histories of Noturus baileyi and N. flavipinnis (Pisces; Ictaluri-dae), two rare madtom catfishes in Citico Creek, Monroe County, Tennessee.
Bulletin of the Alabama Museum of Natural History 18:43-69.Fuselier, L. C., and E. R. Edds. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleostei:
Ictaluridae:
Notrurs placidus).
Southwestern Natu-ralist 39:217-223.
Greeley, J. R. 1929. Fishes of the Erie-Niagara water-shed. Pages 150-179 in A biological survey of the Erie-Niagara system. Supplement to the Annual Re-port of the New York State Conservation Depart-ment 18, Albany.Hamman, R. L. 1982a. Induced spawning and culture of bonytail chub. Progressive Fish-Culturist 44: 201-203.Hamman, R. L. 1982b. Spawning and culture of hump-back chub. Progressive Fish-Culturist 44:213-216.
Hamman, R. L. 1985. Induced spawning of hatchery-reared razorback sucker. Progressive Fish-Culturist 47:187-189.
Hamman, R. L. 1986. Induced spawning of hatchery-reared Colorado squawfish.
Progressive Fish-Cul-turist 48:72-74.Jaffe, L. E 1985. Calcium explosions, waves, and pulses in activating eggs. Pages 127-165 in C. B. Metz and A; Monroy, editors. The fertilization of the egg, volume 3. Academic Press, New York.Karlsen, 0., and J. C. Holm. 1994. Ultrasonography, a noninvasive method for sex determination in cod (Gadus morhua). Journal of Fish Biology 44:965-971.Langlois, T. H. 1954. The western end of Lake Erie and its ecology. J. W. Edwards Brothers, Ann Arbor, Michigan.Martin, R. W., J. Meyers, S. A. Sower, D. J. Phillips, and C. McAuley. 1983. Ultrasonic imaging: a po-tential tool for sex determination in live fish. North American Journal of Fisheries Management 3:258-264.Martin-Robichaud, D. J., and M. Rommens. 2001. As-sessment of sex and evaluation of ovarian matura-tion of fish using ultrasonography.
Aquaculture Re-.search 32:113-120.
Mattson, N. S. 1991. A new method to determine sex and gonad size in live fishes by using ultrasonog-raphy. Journal of Fish Biology 39:673-677.
Mayden, R. L., and B. M. Burr. 1981. Life history of the slender madtom, Noturus exilis, in southern Il-linois (Pisces: Ictaluridae).
Occasional Papers of the Museum of Natural History, University of Kansas 93:1-64.Media Cybernetics.
1999. Optimas image analysis soft-.ware, version 6.51. Silver Spring, Maryland.Menzel, B. W., and E. C. Raney. 1973. Hybrid madtom catfish, Noturus gyrinus X Noturus miurus, from Ca-yuga Lake, New York. American Midland Naturalist 90:165-176.
Moghim, M., A. R. Vajhi, A. Veshkini, and M. Masou-difard. 2002. Determination of sex and maturity in Acipenser stellatus by using ultrasonography.
Jour-nal of Applied Ichthyology 18:325-328..
Moss, R. E. 1983. Microhabitat selection in Neosho River riffles. Doctoral dissertation.
University of Kansas, Lawrence.Pfingsten, D. G., and D. R. Edds. 1994. Reproductive traits of the Neosho madtom (Noturus placidus) (Pi-sces: Ictaluridae).
T1ransactions of the Kansas Acad-emy of Science 97:82-87.SAS. 1992. SAS/LAB software, release 6.08. SAS In-stitute, Cary, North Carolina.Saunders, J. W., Jr. 1982. Developmental biology: pat-terns, problems, and principles.
Macmillan, New York.Stoeckel, J. N. 1993. Controlled cultivation techniques for the recovery of threatened fishes in Virginia.Doctoral dissertation.
Virginia Polytechnic Institute and State University, Blacksburg.
Stoeckel, J. N., and B. M. Burr. 1999. A review of key reproductive traits and methods used to spawn ic-talurids.
Pages 141-159 in E. R. Irwin, W. A. Hub-ert, C. F. Rabeni, H. L. Schramm, Jr., and T. Coon, editors. Catfish 2000: proceedings of the interna-tional ictalurid symposium.
American Fisheries So-ciety, Symposium 24, Bethesda, Maryland.Stoeckel, J. N.., and R. J. Neves. 2001. Induced spawn-ing of the margined madtom. North American Jour-nal of Aquaculture 63:99-108.
Taylor, W. R. 1969. A revision of the catfish -genus No-turus Rafinesque, with an analysis of higher groups in the Ictaluridae.
U.S. National Museum Bulletin 282.USFWS (U.S. Fish and Wildlife Service).
1991. Neosho madtom recovery plan. USFWS, Denver.U.S. Naval Observatory; 1999-2001.
Sunrise-sunset data. U.S. Naval Observatory, Astronomical Ap-plications Department, Washington, D.C. Avail-able: http://aa.usno.navy.mil/.
Walsh, S. J., and B. M. Burr. 1985. Biology of the sto-necat, Noturus flavus (Siluriformes:
Ictaluridae), in central Illinois and Missouri streams, and compar-isons with Great Lakes populations and congeners.
Ohio Journal of Science 85:85-96.
230 BRYAN ET AL.Wildhaber, M. L., D. M. Papoulias, A. J. DeLonay, D. E. Tillitt, J. L. Bryan, M. L. Annis, and J. A.Allert. 2005. Gender identification of shovelnose sturgeon using ultrasonic and endoscopic imagery and the application of the method to the pallid stur-geon. Journal of Fish Biology 66:1-19.Zweibel, W. J., and R. Sohaey. 1998. Introduction to ultrasound.
Saunders Company, Philadelphia.
10 Am. Midl. Nat. 156:305-318 Influence of Water Flow on Neosho madtom (Noturus placidus)
Reproductive Behavior JANICE L. BRYAN 1 Department of rhemas and WdWlfr &Sencez, The School of N.atul Resour=4 Uufivesit of Missouri Columbia 65211 MARK L. WIDHABER U& Geological Sunw, Columbia Envuironmmental Resarch Cent, Columbia, Misouri 65201 AND DOUGLAS B. NOLTIE Department of Asheries and IUd*fe Sciences.
The School of Natural Rsourvet Uni t of Msou, Columbia 63211 Asmcr.-The Neosho madtom is a small, short-lived catfish species endemic to gravel ban of the Neosho River in Kansas, Oklahoma and Missouri, USA. It spawns during summer in nesting cavities excavated in gravel. Although the species has survived dam construction within the Neosho River basin, its declining numbers resulted in it being added to the federal threatened species list in 1991. To test how water flow affects the reproductive behavior of Neosho madtoms, we compared activities of male-female pairs in static versus flowing-water aquaria. Using a behavioral catalog, we recorded their behavior sequences during randomly selected 5-min nighttime periods. For males and females, Jostle and Embrace were the mast performed reproductive behaviors and the Jostle-Embrace-Carousel was the most performed reproductive behavior sequence.
Water flow decreased the mean frequency of occurrence, percentage of time spent and mean event.duration of male Nest Building.
Because Neosho madtom courtship, reproduction and parental care is a complex and extended process, disturbances such as heightened river flows during the species' spawning season may negatively affect nest quality and reproductive success.LntODuCnON Many environmental cues trigger spawning in temperate fish species, including fooc abundance, photoperiod, temperature, flooding, lunar cycles and social interaction (Bye 1984; Munro et aL, 1990). The nature of these cues and how they are used can differ froir family to family, and even between closely-related species, depending upon geographit location and prevailing environmental conditions (Bye, 1984). Within order Siluriformes the environmental factors that trigger reproduction include temperature, photoperiod ane water flow (Brauhn, 1971; Brauhn and McCraren, 1975; Vasal and Sundararaj, 1976; Davia et aL, 1986; Kelly and Kohler, 1996; Stoeckel and Burr, 1999). For the ictalurid genut Noturus flowing water has not been examined as a factor affecting reproduction, ever though most species occur in flowing water habitats (Taylor, 1969; Burr and Stoeckel, 1999)Altering a river's natural flow regime can affect fish spawning and reproductive success'`For example, discharges that mimic natural flow regimes facilitate reproduction in stripec bass (Morone saxatilis Zincone and Rulifson, 1991) and lake sturgeon (AcipenserfiLuescens
'Corresponding author present address: U. S. Geological Survey, Columbia Environmental Researcl Center, 4200 New Haven Road, Columbia, Missouri 65201. Telephone:
(573)-441-2953; FAX: (573)-876 5399; e-mail: jbryan~usgs.gov 306 THE AMEmcAN MIDLAND NkTu T 156 Auer, 1996). In contrast, alterations of the natural flow regime negatively affect fi spawning success by impeding upstream migration to the spawning grounds (Votinov a: Kas'yanov, 1978) and decreasing survival and growth of eggs and fry (Reiser and Whi.1990; Gomes and Agostinho, 1997).Numerous flood control and low-bead dams have been constructed in the Neosho Rn basin (U.S. Army Corps of Engineers, 1993), which is the only habitat of the Neos madtom, Noturus pladdus (Taylor, 1969; Cross and Collins, 1995). The effects of-dam a reservoir construction on this river's hydrology are typical. The low-head dams have chang the riverine habitat upstream and downstream, together with the associated fish semblages (Gillette, 2005), and Neosho madtoms are less abundant in gravel bars direc above and below these dams (Tiemann et aL, 2004). Following placement of the Jo Redmond flood control reservoir and dam on the Neosho River, the river exhibits I variable flow rates, increased winter flows, high flow events of greater and more varia]durations, and delayed maximum annual flows of more variable timing (Wildhaber a 2000). These changes are problematic because Neosho madtom density is correlated w the magnitude, duration, and timing of flow minima (Wildhaber ae aL, 2000). Despite scope, the aforementioned research has examined water flow modification impacts o at the population level. How water flow impacts Neosho madtom individuals dur reproduction has not been studied..Five species of madtoms are included on the U.S. Fish and Wildlife Service's threatened a.endangered species list (Burr and Stoeckel, 1999). The Neosho madtom was cla fled as threatened in 1991 (U. S. Fish and Wildlife, 1991; 55 FR 21148). A typical wild ak Neosho madtom is 35-70 mm in total length (Fuselier and Edds, 1994; Bulger and EU 2001). Most wild Neosho madtoms live 1 to 2 y (Bulger and Edds, 2001). Neosho madtc occur almost exclusively within mainstems of the Neosho River. Like most madtoms, t are typically found in association with gravel bars in areas of flowing water (Deacon, 19 Taylor, 1969; Moss, 1983; Fuselier and Edds, 1994; Cross and Collins, 1995; Bulger .Edds, 2001). During the breeding season, which spans May to August/September (Bulger;Edds, 2001), laboratory studies have shown that nest cavities are constructed under la objects in the gravel and that spawning occurs at temperatures ranging from 21 to 28 C (Bul et aL, 2002b; Bryan et aL, 2005); high turbidity in the Neosho River inhibits direct observati (Pfingsten and Edds, 1994). Male parental care is typical of madtoms (Burt and Stoec 1999), and Neosho madtom males provide 8-9 d of post-spawning parental care (Bulger ev 2002b). Compared to longer-lived fish species, Neosho madtoms approach semelpai because they appear to live only one or two years in nature (Bulger and Edds, 2001), d opportunities for reproduction seem to be limited.To better understand Neosho madtom reproduction and the effects of water flow on it conducted a laboratory study comparing Neosho madtom reproductive behavior under I and non-flow conditions.
We chose to study reproductive behaviors over other facet reproduction because: (1) examining spawning success in the wild is almost impossible
- to the high and continuously turbid in situ water conditions (Pfingsten and Edds, 1994), sacrificing individuals to obtain information regarding reproductive/gonadal status unwise due to the species' conservation status, necessitating work with live individuals, reproduction in this genus and species is both lengthy and complex (Fitzpatrick, 1!Bowen, 1980; Chan, 1995; Bulger et aL, 2002a), providing frequent opportunities for f related disruptions of spawning to occur and to be assessed and (4) madtoms in laboratory rarely complete the entire spawning process, from nesting to fry disix (Stoeckel and Burr, 1999; Bulger dt aL, 2002b; Bryan et aL, 2005), necessitating an empt on events culminating in spawning.
2006 BRYAN ET AL.: MADTOM REPRODUCTIVE BEHAVIOR METHODS GENERAL FISH MAINTENANCE The Neosho madtoms used in this study were collected from Kansas sections of Neosho (n = 24) and Cottonwood rivers (n = 34) in April-July 1998 (Bulger, 1999). At time of collection, these fish were assumed to be one-year-olds due to their small size (4(67 mm). In 1998 the fish were used in another reproductive behavior study conducted Bulger (1999). Our study of these fish began in the summer of 1999 and continued throt the summer of 2000, when the fish were assumed to be 2+ and 3+ years of age, respectiv The holding aquarium for the fish was a single living StreamR System (Frigid Units, IL Toledo, Ohio, 720 L, model LS.900) housed in an isolated photoperiod-and temperan controlled room. To minimize inter-gender aggression, we kept the genders separate in Living Stream using perforated partitions (four total compartments).
Water temperatm (9-27 C, winter vs. summer) and laboratory photoperiods (8-16 h of light). %manipulated throughout the year (Bryan et aL, 2005) to mimic the natural environm of the Neosho River at the latitude and longitude of Emporia, Kansas (38°26'N, 960192 Fish were fed to satiation thrice weekly [frozen brine shrimp (Arnemia salina), fro bloodworms (Chironomus sp.) or Sinking Carnivore Pellets (Hikari, Hayward, Californi excess food was removed routinely to prevent fouling of the water.EXPERIMENT Our work was conducted at the U.S. Geological Survey Columbia Environme:
Research Center in Columbia, Missouri, U.S.A. Behavioral monitoring of male/female p was conducted at the same time as the wild Neosho madtom spawning season (May thro July, Bulger and Edds, 2001). Each year's experiment continued until all signs of spawr activity ceased. Consequently, the laboratory spawning season lasted 63 d in 1999 (star May 22 at 19 C) and 44 d in 2000 (starting June 3 at 20 C). Photothermal manipulati successfully induced the gonad and secondary sexual characteristic development comr to breeding Neosho madtoms (Pfingsten and Edds, 1994; Bulger et a., 2002b; Bryan i 2005). Fish that exhibited the most pronounced secondary sexual characteristics v paired and randomly assigned to treatment groups (see below).Aquaria.--In 1999 we used 21.9-L aquaria (12 tanks: six flow and six non-flow).
In 200(increased the size of the aquaria to 43.7-L (six tanks: three flow and three non-flow) bea the fish had grown considerably.
While in the aquaria, our maintenance of the continued as described above.We provided each aquarium with a nesting cavity cover made of a length of PVC I halved lengthwise.
Larger nest cavity covers were used in 2000, again because the had grown (1999:13.3 cm long X 10.2 cm wide X 5.0 cm high; 2000:12.7 cm long X 14.1 wide X 6.4 cm high). For each aquarium, this nest cover was placed concave side down" one end positioned against the front wall of the aquarium to allow for direct viewing the nest. The substrate for each aquarium was natural chert-limestone gravel similar in to that of spawning sites in nature (1.37 +/- 0.52 cm; Bulger and Edds, 2001). Aquaria* illuminated during the day by eight overhead fluorescent bulbs (34 W, General Ele, Company, Cleveland, Ohio, 120 cm long X 3.8 cm diameter), and at night by six overt infrared illuminators (30 W, American Dynamics, Orangeburg, New York, model 1(3050), which allowed camera viewing of fish during darkness.Waterflows.-Each year, aquaria were randomly assigned to flow and non-flow treatme All aquaria were plumbed similarly and filled with well water. Current in each flow aquai was generated using an external recirculating pump plumbed to draw aquarium water I 308 THE A mitcA Mmt.Ln NATURALIST 156(l just below the water surface and to aim flows across the substrate directly into the nest cavit" Velocities of 30 cm/s at the nest cavity opening were established at the start of eaca experiment, a rate similar to that found at substrate level at adult Neosho madtor collection sites (Bulger and Edds, 2001). Corresponding outflow pipe velocities wer measured weekly. thereafter to avoid disrupting the fish in their nest cavities.
These fbo, measurements were also taken in the non-flow aquarium to ensure equal disturbanc between treatments.
Behavioral data. colleaion.-Fish behavior was recorded using a monochrome vide, multiplexer (American Dynamics, Orangeburg, New York, model 1480/16), time laps video recorder (Panasonic, Secaucus, NewJersey, model AG-6760P or Toshiba America C.I Inc., Buffalo Grove, Illinois, model KV-7168A) and six black and white cameras (Panasonic Secaucus, New Jersey, model WV-BP310).
The VCRs. alternately recorded successive 12.periods and were set to record 1/6 0'h of a second every 1/10'h s. Cameras were positione directly in front of each aquarium and provided a full aquarium width view that include a view underneath the nest cavity cover.Although we recorded around the dock, the Neosho madtoms proved to be nocturn: and, thus, largely inactive during the daytime (see also Bulger dt aL, 2002a). Consequently, w collected behavior data only from the nighttime.
Nights with disturbances (power/camera recording failure and/or feeding nights) were excluded from data collection, leavin 56 and 57% of the nights for behavior sampling in 1999 and 2000, respectively (35 out c 63 and 25 out of 44 nights). For each aquarium, we randomly selected a sub-sample of 1-7 the undisturbed nights for data collection (a 49% sub-sample for 1999 that we matched i 2000). This random selection yielded different subsets of nights for each aquarium.For each tank-night selected, a random 5--min period was further sub-sampled from eac.hour, with the night's hour count beginning the moment the lights were turned off. Becaw night length varied progressively through the experiment each night's duration was not a even multiple of 60 min. Consequently, we also excluded from consideration any hot within which the morning transition to light occurred.For every randomly selected 5-min period, each fish's behavior was observed and t: beginning and. ending time of every behavioral act was recorded by gender. For sing: individuals, we could not confidently distinguish single acts of long duration from sever continuous repetitions of the same act that lacked intervening pauses. Therefore, v considered all such cases as the former, which precluded the reporting of repeats of td same behavior.
Although the fish were not marked, we 'could always distinguish the ma from the female by their secondary sexual characteristics.
Gender was confirmed later usir.ultrasound examination (Bryan et aL, 2005).Behaviors.-Twenty-three Neosho madtom behaviors had previously been cataloged I Bulger et al. (2002a) which defined behaviors using position in the aquaria and weither or or both fish performed the behavior.
We modified this catalog by defining the behavio without aquaria position or performance by one or both fish (Table 1). Ten behaviors we: observed during videotape analyses; five additional behaviors were not seen during our su sampled time intervals, but were observed at other times. Four of the 10 observed behavio are considered Reproductive Behaviors in madtoms (Carousel, Embrace, Jostle, and Ne Building; Fitzpatrick, 1981; Stoeckel, 1993; Chan, 1995; Bulger et al, 2002a, b). Three of tl Reproductive Behaviors were further categorized here as Pair-Based, because both gende participated (Caiousel, Embrace and Jostle).For each of the 10 observed behaviors, we calculated an overall mean for each of thn parameters using SAS (1999-2001):
(1) frequency of occurrence
.(number of tim a behavior was performed during a 5-min period), (2) percentage of time (total time spe 2006 BRam rr AL.: MADTOM REPRODUCrvE BmmwOm 3 TA=LE 1.-Neosho madtom ethogram (modified from Bulger a aL, 2002a).Behavior Description Bite Fish approaches a conspecific head first, then closes its mouth on or against the point of contact, the latter being other than the mouth.Carousel'-
While oriented head to tail, both fish swim in a circular pattern, following one another, during which one fish rubs and nudges the caudal peduncle area of the conspecific.
Chase' Fish swims close behind a conspecific, rapidly following it around the tank.Embrace" With both fish oriented laterally head to tail, one or both curls caudal fin across and over other fish's snout.Feeding Fish is ingesting food accompanied by chewing-like mouth movements.
Jostle"2 While in contact with a conspecific, fish bumps and pushes a conspecific with its body during a series of twists and turns.Mouth Bite' With conspecifics oriented head-on, one approaches the other then clasps its mouth onto the other's, after which the two fish thrash their tails back and forth, while remaining joined..Nest Building' At a developing nest site, fish moves a substrate particle either by seizing the substrate particle with its open mouth then pushing or lifting it, or by levering its head under the particle then pushes it across the bottom.Nudge Fish swims toward a conspecific then bumps headfirst into it.Oral Flare Fish opens mouth and splays its' opercula, directing this display toward a conspecific.
Parallel Swim Display' With two fish positioned side by side, facinig in the same direction, one spreads its fins and laterally flexes its body resulting in a sinusoidal wave passing down the length of the body.Resting J, Fish is stationary on the bottom.Scratching' Fish swims forcefully, causing its body to graze against or/and glance off a stationary object.Spawningl" Gamete release by both the male and female.Swimming Fish propels itself around the tank using its caudal fin.'Indicates behavior that did not occur during data collection, consequently no analysis could done 2 Indicates Reproductive Behavior s Indicates Pair-Based Reproductive Behavior performing a behavior during a 5-min period divided by 5 min X 100), and (3) mean evs duration (total time spent performing a behavior during a 5-min period divided frequency of occurrence).
These parameter means were then used to test for an effect water flow on Neosho madtom behavior.DATA ANALYSS Individual behaviors.-This analysis was designed to determine for males and fema whether each separate Neosho madtom behavior parameter differed between the flow a non-flow treatments.
The three parameter values above were extracted from all 5-n samples for each behavior under consideration.
Then the three parameter means for e;aquarium/gender/behavior combination were generated for each hour, the hourly me;were then averaged within each night and these per-aquarium/per-night means were t.averaged over the sampled nights of the spawning season. Because some behaviors did i occur during many of the 5-min periods, the data was not normally distributed and '
310.THE AMERICAN MIDLAND NATURALIST 156(variance was not constant.
Consequently, we tested for treatment effects on each behavioi parameters using a multi-response permutation procedure (BLOSSOM, Version W2001.05 Cade and Richards, 2000), this being the non-parametric equivalent of a multivaria analysis of variance (Bonferroni-adjusted alpha level = 0.025, i.e., 0.05/2, accounting for td separate male and female tests of each parameter).
Behavior sequences.-This analysis was designed to test for each gender whether the foi Reproductive Behaviors were performed in any prevailing order and, if so, to, descril such ordering and' test whether it differed between flow and non-flow conditions.
Reproductive Behavior Sequence was defined as a series of successive acts performed 1 individual fish where both the first and last act were any one of the four Reproducti Behaviors.
Data analyzed were the sequences of all behaviors that each individual ft performed during the 5-min samples from 1999 and 2000. Data for each fish within d pairings were analyzed separately because male/female interactions were not the focus this analysis.Each Reproductive Behavior Sequence was designated using any two of the fo Reproductive Behaviors, the first being termed the "given" behavior, the last being term, the "target" behavior.
The sequences we evaluated included those where the target behavi occurred immediately after the given behavior (lag 1), two steps after (one interveni behavior, lag 2) or three steps after (two intervening behaviors, lag 3). Intervening behavic could be any of the behaviors we observed.
For example, the sequence Jostle-Swimmir.
Resting-Carousel would designate Jostle as the given behavior, Swimming and Resting intervening behaviors, and Carousel as the target behavior (lag 3). Any sequence with Nt Building as aý target or, given behavior was designated as a Nest Building Reproducti Behavior Sequence and any sequence with a Pair-Based Reproductive Behavior as both tf target and given behavior were designated as a Pair-Based Reproductive Sequence.
Sixte, different given/target combinations could be derived from the four Reproducti Behaviors; consequently, a total of 44 different combinations of sequences and lags we possible (16 sequences X 3 lags = 48 combinations minus 4 sequences that would ha required distinguishing repeats of the same behavior = 44).To examine behavior interdependence, we summed the Reproductive Behavior quence occurrence for each lag/gender/treatment/year combination and used lag quential analysis to test the degree to which some Reproductive Behaviors followed otho more frequently than expected by random using a Pearson chi-square test (i.e., tended occur in a sequence; GSEQ for Windows vesion 4.1.2 software; Sackett, 1979, 191 Bakeman and Quera, 1995; Bakeman and Gottman, 1997). A Bonferroni-adjusted alp level was used to assess significance (alpha = 0.000142 = 0.05/352 comparisons; 352 c square tests = 44 sequences X 2 y X 2 treatments X 2 genders).
We also tested the effect flow on the mean frequency of occurrence of each Reproductive Behavior Sequence/la gender/treatment combination using the same nonparametric -multi-response permutati procedure described above, and a Bonferroni-adjusted alpha level to assess significar (n = 18; alpha = 0.000568 = 0.05/88 comparisons; 88 permutation tests = 44 sequencet 2 genders).Finally, we also examined the 12 sequences that only contained the three Pair-Bas Reproductive Behaviors, designating, these as Pair-Based Only Reproductive Behav Sequence Trios (Carousel-Embrace-Jostle, Embrace-Carousel-Jostle, etc.). We tested I effect of flow on the mean frequency of occurrence of each again using the nonparamet multi-response permutation procedure and a Bonferroni-adjusted alpha level to assess nificance (alpha = 0.002083 = 0.05/24 comparisons; 24 permutation tests = 12 sequence;2 genders).
2006 BAN Er AL.: MADTOM REPRODUCTIVE BEHAVIOR RESULTS INDIVIDUAL BEHAVIORS Overall, our Neosho madtoms spent little time performing Reproductive Behavi (Table 2) and spent the vast percentage of time Swimming or Resting (unpubl. data). N Building was always performed substantially more often by males (Table 2). Because mos-the separate Reproductive Behaviors were usually performed by the fish together as a p the performance parameter values are similar for both genders (Table 2). For males 2 females, Jostle was generally the most frequently performed behavior (Table 2). BothJo;and Embrace were generally performed with the highest percentage of time and mean ev duration.Rm effects on inidvidual behaviors.--Because the BLOSSOM software precluded simultaneous testing of both year and flow effects, we tested first for the year effect. Femn were significantly more active (i.e., exhibited greater frequencies of occurrence percentages of time spent performing various behaviors) in 1999 than in 2000 (Table 3).negated the significant year effect using an alignment procedure (Mielke and Iyer, 19f which used each behavior/parameter/gender/year combination average and subtrac it from each respective behavior/parameter/gender observation within that year. 'I alignment procedure successfully removed the year influence for both males and ferm (Table 3), allowing the adjusted behavior/parameter observations to be compared betw, treatments with both years combined.
Post-alignment, male Nest Building was the c behavior significantly affected by water flow (Table 3), its' frequency of occurrer percentage of time spent and mean event duration all being significantly lower in f than in non-flow aquaria. Female Neosho madtoms ilso performed Nest Building be!iors, although the associated parameters showed no differences between the treatmnt (Tables 2, 3). 1, Using the same statistical procedure, we also examined only the 5-min periods wherei: least one Reproductive Behavior was performed, thereby eliminating most of the zero the original data set. The results (data not shown) were consistent with those above, exc that the mean event duration of male Nest Building behavior was not significantly differ between the flow and non-flow aquaria, and the frequency of occurrence of female Caroi behavior was significantly lower in the flow treatment.
BEHAVIOR SEQUENCES Of the 352 possible male and female Reproductive Behavior Sequences, 109 male 74 female sequences yielded expected occurrence values of 5 or more, necessary critei for valid chi-square testing (Bakeman and Quera, 1995). Fifty-one male and 39 fen Reproductive Behavior Sequences yielded significant chi-squared test results (7 out of the sequences that were different between males and females were Nest Building Reproduc Behavior Sequences due to the lack of Nest Building by females).
Because three of the I Reproductive Behaviors were Pair-Based, the results for each gender were very similar wl 46 male and 37 female Reproductive Behavior Sequences were performed significantly rr than expected.
Albers (2001) provides details regarding all the Reproductive Beha Sequences.
Seven out of the 51 significant male Reproductive Behavior Sequences were Nest Buik Reproductive Behavior Sequences.
For male Neosho madtoms under both flow conditi.the performance of Nest Building was followed most often by more Nest Buildin 1 significant sequences; lag 2:1999 Non-Flow Aý= 454.66, df= 1, P < 0.0001, 1999 Flow.265.16, df= 1, P < 0.0001; lag 3:1999 Flow XY = 16.75, df= 1, P = 0.0001). All four of TAALE 2.-Male and female Neosho madtom Reproductive Behavior parameters for 1999 and 2000 under flow and non-flow conditions for an average 5-min period. Entries are mean -standard deviation Frequency of occurrence
(#) Percentage of time (%) Event duration (s)Year/Treatment Behavior Male Female Male Female Male Female 1999 Non-Flow (n -6) Carousel 0.59 +/- 0.43 0.59 +/- 0.43 2.83 +/- 1.94 2.86 +/- 1.96 3.14 +/- 2.48 3.15 +/- 2.49 Embrace 0.35 +/- 0.28 0.34 +/- 0.28 3.01 +/- 3.76 3.02 +/- 3.78 4.65 +/- 5.99 4.69 +/- 6.10 Jostle 0.63 +/- 0.42 0.62 +/- 0.42 2.27 +/- 1.51 2.23 +/- 1.52 2.62 +/- 2.00 2.61 4- 2.01 Nest Building 0.73 +/- 0.45 0.08 +/- 0.11 3.91 +/- 2.44 0.49 +/- 0.69 4.96 +/- 2.61 0.96 +/- 1.34 1999 Flow (n = 6) Carousel 0.44 +/- 0.48 0.44 +/- 0.48 2.54 +/- 2.61 2.53 +/- 2.59 2.85 +/- 2.76 2.84 +/- 2.75 Embrace 0.28 +/- 0.29 0.28 +/- 0.29 3.37 +/- 3.90 3.36 +/- 3.88 5.41 +/- 6.16 5.38 +/- 6.12 Jostle 0.48 +/- 0.42 0.47 +/- 0.41 2.46 +/- 1.62 2.46 +/- 1.58 3.09 +/- 1.82 3.11 +/- 1.72 Nest Building 0.23 +/- 0.20 0.06 +/- 0.03 1.54 +/- 0.94 0.41 +/- 0.26 2.87 +/- 1.27 0.83 +/- 0.45 2000 Non-Flow (n 3 3) Carousel 0.28 +/- 0.33 0.28 +/- 0.33 0.79 +/- 1.04 0.83 +/- 1.12 1.33 +/- 1.86 1.41 +/- 2.00 Embrace 0.22 +/- 0.28 0.22 +/- 0.28 2.25 +/- 3.19 2.24 +/- 3.17 4.29 +/- 6.07 4.24 +/- 5.99 Jostle 0.67 +/- 0.78 0.67 +/- 0.77 3.84 +/- 4.32 3.79 +/- 4.24 3.92 +/- 4.28 3.90 +/- 4.24 Nest Building 0.21 +/- 0.25 0.06 +/- 0.06 1.24 +/- 1.42 0.38 +/- 0.35 1.86 +/- 2.31 0.73 +/- 0.64 2000 Flow (n -3) Carousel <0.01 +/- <0.01 <0.01 +/- <0.01 0.01 +/- 0.01 0.01 +/- 0.01 0.02 +/- 0.03 0.02 +/- 0.03 Embrace <0.01 +/- <0.01 <0.01 +/- <0.01 0.01 +/- 0.01 0.01 +/- 0.01 0.02 +/- 0.03 0.02 +/- 0.03 Jostle 0.05 +/- 0.02 0.05 +/- 0.02 0.20 -0.15 0.20 +/- 0.15 0.38 +/- 0.28 0.38 +/- 0.28 Nest Building 0.01 +/- 0.01 <0.01 +/- <0.01 0.02 +/- 0.03 <0.01 +/- <0.01 0.05 +/- 0.08 <0.01' +/- <0.01 H I I-I.,
2006 BRYAN ET AL.: MADTOM REPRODUCTIVE BEHAVIOR TABU 3.-Results of the permutation tests examining the effect of year and flow treatments on D and female Neosho madtom Reproductive Behaviors.
Entries are P-values (n = 18, alpha level = OX Year effect or Treatment Frequency of occurrence
(#) Percentage of time (%) Mean event duration effect by Behavior Sequence Male Female Male Female Male Femal Year, Pre-Alignment all behaviors except Resting 0.09 0.01b 0.09 0.02a 0.08 0.03 Year, Post-Alignment all behaviors except Resting 0.99 0.76 0.98 0.83 1 0.85 T1reatment, Post-Alignment Reproductive Behaviors Nest Building 0.01C 0.30 0.01C 0.51 0.01C 0.39 Pair-Based Carousel 0.38 0.59 0.53 0.52 0.47 0.41 Embrace 0.49 0.50 0.96 0.96 0.91 0.91 Jostle 0.21 0.20 0.44 0.45 0.68 0.6E Indicates summer 1999 value significantly exceeded summer 2000 value b Indicates summer 2000 value significantly exceeded summer 1999 value c Indicates non-flow treatment value significantly exceeded flow treatment value significant Nest Building Reproductive Behavior Sequences that involved Nest Building one Pair-Based Reproductive Behavior occurred less frequently than expected (all occw in the 1999 Non-Flow group; lag 1: Nest Building-Carousel )r= 21.44, df= 1, P < 0.0(Nest Building-Embrace X' = 31.26, df= 1, P < 0.0001, Embrace-Nest Building Y = 24 df= 1, P < 0.0001; lag 2: Nest Building-Jostle )e = 16.77, df= 1, P < 0.0001). None of 10 female Nest Building Reproductive Behavior Sequences expected at least 5 times V significantly different than expected (Bonferroni-adjusted alpha level of 0.000142; P-vM range 0.0049-0.7102).
Of the male Reproductive Behavior Sequences that were performed significantly ir than expected, 93.5% involved only the Pair-Based Reproductive Sequences, indica the Pair-Based Reproductive Behaviors occurred in conjunction with one another (43 ot 46 sequences; all P-values < 0.0001). Only 5 out of the 43 male Reproductive Beha Sequences that were perfomed significantly more than expected were not significant the female (lag 2: 1999 Non-Flow Embrace-Carousel )e = 12.37, df= 1, P = 0.0006; la 1999 Non-FlowJostle-Jostle e = 5.59, df= 1, P = 0.0172, Embrace-Jostle J---5.38, df- 1, 0.0193, Embrace-Embrace Ye = 9.54, df= 1, P = 0.0022, 2000 Non-Flow Embrace-Jostle.
12.16, df= 1, P = 0.0006). These five sequences are not involved in the two most perfon Pair-Based Only Reproductive Behavior Sequence Trio, consequently, our detailing of males' reproductive behavior patterns below is also applicable to the females.Irrespective of treatment, the most frequently performed Pair-Based Only Reproduc Behavior Sequence Trio was Jostle-Carousel-Embrace (211 out of 929 total performan When a male Neosho madtom performed a Jostle-Carousel lag I sequence I performances), 49.5% (or 211) of the times it was followed by Embrace. Additionally.
the Jostle-Embrace lag 2 sequence (212 performances), 99.5% (or 211) of the times intervening behavior was Carousel (212 performances of the Jostle-Embrace la sequence).
Irrespective of treatment, the second most frequent Pair-Based
(
314 THE AMICAN MmLA.ND NATuRAusr 136(.TALE 4.-Results of the permutation tests examining the effect of year and flow on the frequency male and female Neosho madtom Reproductive Behavior Sequences.
Entries are P-values (n = 18, alpi level = 0.000568).
NP Indicates a sequence of which was that combination was not possible, given hc we scored behaviors Year effect or Treatment Lag 1 Lag 2 Lag S effect by Behavior Sequence Male Female Male Female Male Ferm Year, Pre-Alignme,,t
'" all behavior sequences combined 0.12 0.06 0.05 0.08 0.07 0.0 Year, Post-Alignmentt all behavior sequences combined 0.14 0.15 0.19 0.20 0.24 0.1 Treatment, Post-Alignment Reproductive Behavior Sequences Nest Building Sequences Nest Building Carousel 0.22 0.05 0.07 0.46 0.21 0.2 Nest Building Embrace 0.64 0.64 0.28 0.64 0.06 0.3 Nest Building Jostle 0.01 0.40 0.17 0.32 0.04 0.5 Nest Building Nest Building NP NP 0.04 0.73 0.38 0.6 Carousel Nest Building 0.09 0.65 0.05 0.83 0.11 0.4 Embrace Nest Building 0.02 0.64 0.22 0.26 0.64 0.6 Jostle Nest Building 0.04 0.50 0.03 0.05 0.03 0.0 Pair-Based Sequences Carousel Carousel NP NP 0.41 0.39 0.33 0.2 Carousel Embrace 0.56 0.56 0.95 0.91 0.57 0.4 Carousel Jostle 0.26 0.26 0.21 0.30 0.10 0.(Embrace Carousel 0.67 0.74 0.45 0.51 0.54 01 Embrace Embrace NP NP 0.76 0.75 0.37 0..Embrace Jostle 0.30 0.09 0.26 0.44 0.41 0.Jostle Carousel 0.49 0.32 0.36 0.35 0.15 0.1 Jostle Embrace 0.81 0.80 0.64 0.68 0.81 0.1 Jostle Jostle NP NP 0.17 0.16 0.31 0.!Reproductive Behavior Sequence Trio was Carousel-Embrace-Jostle (170 out of 929 to performances).
When a male Neosho madtom performed a Carousel-Embrace lag sequence (451 performances), 37.7% (or 170) of the times it was followed by Jost Additionally, for the Carousel-Jostle lag 2 sequence (209 performances), 81.4% (or 170)the times the intervening behavior was Embrace.Rlow effects on sequences.--Due to the marginally significant year effects for male and fern sequences (Table 4), we used the same alignment procedure described above to test for fl effects on both the Reproductive Behavior Sequences and the Pair-Based Only Reproduct Behavior Sequence Trios. Post-alignment, occurrence of the male and female Reproduct Behavior Sequences did not differ between the non-flow and flow treatments usi a Bonferroni-adjusted alpha level (Table 4), as was the case for the male and female P: Based Only Reproductive Behavior Sequence Trios (Bonferroni-adjusted alpha level 0.002083; P-value range 0.27-0.91).
DISCUSSION Neosho madtom Reproductive Behaviors were generally uncommon in their occurren The only other study to quantify madtom behaviors, Bulger (2002a), also observed low lei 2006 BRYAN "rr AL.: MADTOM REPRODUCTIVE BEHAVIOR of reproductive behaviors.
The Pair-Based Reproductive Behaviors were performed n often (Carousel, Embrace, Jostle), followed by Nest Building.
Carousel and/or Embi have been observed in other madtom species (Bowen, 1980; Fitzpatrick, 1981; Stoec 1993; Chan, 1995), although, Jostle has only been observed in Neosho madtoms (Bid et aL, 2002a). Male Neohso madtoms were the primary nest builders with female Neo madtoms performing Nest Building at a much lower level. This observation paral previous findings for brown madtoms (Chan, 1995), brindled madtoms (Bowen, 191 freckled madtoms (Fitzpatrick, 1981) and Neosho madtoms (Bulger et aL., 2002a), wl, both genders have been observed performing nest building behaviors.
However thi contrary to the majority of previous madtom studies, where only male madtom nest builc was observed (Burr and Stoeckel, 1999).In the flow treatment, all of the male Nest Building behavior parameters were significa reduced (frequency of occurrence, percentage of time spent and mean event duratic Because Nest Building involves energetically costly movements like nudging or carr)large stones in the mouth, relative to their body size, heightened water flows that furl elevate energy expenditures during Nest Building are of concern from mult perspectives.
First, the nest site is occupied for an extended period: male madtoms sp at least 3 wk at a nest, from site selection to spawning to departure of young (Chan, 1.Bulger et aL, 2002b). Second, nests provide shelter from predators for the spawning pair eggs/young (Mayden et aL, 1980; Mayden and Burr, 1981) and, when disturbed, the egg fry are immediately consumed by predators or displaced by the current (Burr and Dimm 1981; Mayden and Burr, 1981; Walsh and Burr, 1985). In either case, the lower qualit a nest fashioned with less Nest Building may result in reduced reproductive success.No previous work has addressed the sequencing of madtom Reproductive Behaviors (I and Stoeckel, 1999). Our finding that the Jostle-Carousel-Embrace and Carousel-Embr Jostle Reproductive Behavior sequences were so prevalent in their occurrence broadens appreciation of madtom reproduction.
Even though Carousel and Embrace have previo been shown, to be prominent components of reproduction in other madtom spe (Bowen, 1980; Fitzpatrick, 1981; Stoeckel, 1993), our study demonstrated the linb between these behaviors and the Jostle behavior.
Biologically, these behaviors appear u important for madtom mate assessment (Jostle) and positioning the pair for the spawr act (Carousel and Embrace).The flow-related reduction in frequency of occurrence of Nest Building discussed at suggested that effects might be seen in our behavior sequence analyses.
However, majority of Nest Building Reproductive Behavior Sequences that involved one Pair-BI Reproductive Behavior occurred at frequencies below what were expected.
This lad a detectable flow effect on Nest Building.
Reproductive Behavior Sequences is li a consequence of the temporal organization of spawning behavior in Neosho madtoms explain, nest building typically occurs before egg deposition (Fitzpatrick, 1981; Chan, 19 and our analyses confirmed that it was unlikely to immediately precede or follow any of Pair-Based Reproductive Behaviors.
Consequently, detecting a reduction in the occurre of Nest Building Reproductive Behavior Sequences that involved Pair-Based Reprodu Behaviors was improbable, given that we only considered sequences that were a maximui four acts long (i.e., lag 3).Whereas dams typically hamper fish reproduction by impeding migration (Li et aL, 19 this is less relevant to madtoms since they are largely sedentary (Burr and Stoeckel, 19 Instead, madtom reproductive success is more apt to be negatively impacted by dam flood-related hydrologic changes that occur during the spawning season, similar to o stream-nesting fish species like salmonids, centrarchids, cyprinids and catostomids (See!
316 316 T'Itz AMERIcAN MmLAND NATuRM~un6(136(.and Gard, 1972; Noltie and Keenleyside, 1986; Pearsons et al., 1992; Lukas and Orth, 199.Jennings and Philipp, 1994). Since construction of the John Redmond Dam the chance .flooding during the Neosho spawning season has increased due to the delayed maximm annual discharge of the Neosho River (Wildhaber et al., 2000). Such high discharges durir the spawning season are apt to have deleterious effects on Neosho madtoms constructir and maintaining their nests. One solution to this problem might be to manipulate reservo discharges to duplicate the historic timing of peak flows (i.e., earlier in the year), in additic to maintaining the yearly minimal flows and annual variability suggested by Wildhaber et (2000).In conclusion, we found evidence that elevated water flows negatively affect Neost madtom reproductive behavior.
However, our study addressed only the relative impacts, moderate versus absent water flows on Neosho madtom reproductive behavior.
V recognize that Neosho madtoms in spawning condition have been found at substantial higher velocities than we employed [i.e., up to 71 cm/s at the substrate (Bulger and Edd 2001)]. Consequently, we suggest that future research extend the range of water floi examined and involve in situ studies at the population level. This would broaden oi understanding of how differing hydrologic regimes affect Neosho madtoms.Acmowedgmnits.-We thank David R. Edds of Emporia State University, Emporia, KS and Chadl F. Rabeni of the University of Missouri-Columbia for their advice and review comments.
Angela BuIg provided expertise and assistance, and the researchers and staff at the U.S. Geological Survey Columt Environmental Research Center gave us support and advice. We are grateful to Jim Randolph of t U.S. Army Corps of Engineers, Tulsa District, for funding this research.
The U.S. Fish and Wildb Service and Kansas Department of Wildlife and Parks provided permits to collect and hold Neosl madtoms. This publication is a contribution from the Missouri Agricultural Experiment Station, sa from the Missouri Cooperative Fish and Wildlife Research Unit (U.S. Geological Survey, Missoi Department of Conservation, The School of Natural Resources of the University of Missour-Columb and the Wrildlife Management Institute cooperating).
LUERATURE OCED ALraRsJ. L 2001. Influence of photoperiod and temperature on the Neosho madtom (Noturusplacidi reproductive cycle and the effects of flowing water on Neosho madtom reproductive behavi.M.S. Thesis, University of Missouri, Columbia, Missouri.
132 p.AumR, N. A. 1996. Response of spawning lake sturgeons to change in hydroelectric facility operatic Trans. Am. Fish. Sac., 125:66-77.
BAxEmAN, .L ANDJ. M. GoTrtn. 1997. Observing interaction:
an introduction to sequential analysis, 2 ed. Cambridge Univeristy Press, New York. 207 p.-AND V. QUERA. 1995. Analyzing interaction:
sequential analysis with SDIS and GSEQ. Cambrid University Press, New York. 155 p.BowrN, C. A.,.JL 1980. The life history of the brindled madtom Notunus miurus (Jordan) in Salt Cre Hocking and Vinton counties, Ohio. M.S. Thesis, Ohio State University, Columbus, Oh 195 p.BRxutN,J.
L 1971. Fall spawning of channel catfish. Prog. Fish-Cult., 33:150-152..
-MCRAprEN.
1975. Ovary maturation in channel catfish. Pmag. Fish-Cult., 37:209-212.
BReYAN,J.
L., M. L WnDHABEa AND D. B. NOLTI. 2005. Examining Neosho madtom reproductive biolc using ultrasound and artificial photothermal cycles. N. Am J. Aquac., 67:221-230.
BuLOG A. G. 1999. Population structure, habitat use, and breeding behavior of the Neosho madto Noturus placidus.
M.S. Thesis, Emporia State University, Emporia, Kansas. 83 p.-AND D. R. Eo)Ds. 2001. Population structure and habitat use in Neosho madtom (Notu placidus).
Sothurst.
Nat., 46:8-15.
2006 BRYAN ET AL.: MADTOM REPRODUCTIVE BEHAVIOR 3 , M. L. WniDHABER AND D. R. EDDS. 2002a. Effects of photoperiod on behavior and courtship oft Neosho madtom (Noturus placidus).
J. Freshwat.
Ecol., 17:141-150., C. D. WILKaNSON, D. R. Eons AND M. L WL.DABEL.
2002b. Breeding behavior and reproduct life history of the Neosho madtom, Noturus placidus (Teleostei:
Ictaluridae).
Trans. Kans. Ac S Sd., 105:106-124.
But.t, B. M. An W. W. D.MM4c. 1981. Nests, eggs, and larvae of the elegant madtom Noturus elegns fri Barren River drainage, Kentucky (Pisces: Ictaluridae).
Trans. Ky. Acad. Sc., 42:116-118.
AN J. N. STOECKEL.
1999. The natural history of madtoms (genus Noturus), North Amerc diminutive catfishes, p.51-101. In.: E. R. Irwin, W. A. Hubert, C. F. Rabeni, H. L Schram Jr. and T. Coon (eds.). Catfish 2000: proceedings of the international ictalurid symposin Symposium
- 24. American Fisheries Society, Bethesda, Maryland.BY, V.J. 1984. The role of environmental factors in the timing of reproductive cycles, p. 187-206.G. W. Potts and R. J. Wootton (eds.). Fish reproduction:
strategies and tactics. Acadex Press, London.CADE, B. S. Am J. D. Riciwns. 2000. User manual for BLOSSOM statistical software.
U.S. Ceologi Survey Midcontinent Ecological Science Center, Fort Collins, Colorado.
111 p. http www.mesc.usgs.gov/products/software/blossom/blossom.asp.
CHAN, M. D. 1995. Life history and bioenergetics of the brown madtom, Noturus phaeus. M.S. The University of Mississippi, Oxford, Mississippi.
177 p.CRoss, F. B. ANý J. T. Counss. 1995. Fishes in Kansas, 2nd ed., revised. University of Kansas Natt History Museum, Public Education Series No. 14, University Press of Kansas, Lawrence, Kan-315 p.DAvis, K. B., C. A. GoUmmB. A. Simco, R. MAcGRzooa, El Amn N. C. PARKFt. 10R.. Environrnei rulAic , influence u-d~c-cyc-4z ti-un- -iz-'2 .,2] , channel catfish (Idaturus punctaus).
PhysL Zoo!; 59,7-17-724..
DEACoN,J.
E. 1961. Fish populations; following a drought, in the Neosho and Marais des Cygnes river Kansas. Univ. Kan. Publ,, Mus Nat. Hut., 13:359-427.
FrrZPATIUCV, K. C. 1981. Comparative reproductive behavior of the slender (Noturus exilis), brindled miurus), and freckled (N. nocturnus) madtoms (Pisces: Ictaluxidae).
MA. Thesis, South Illinois University, Carbondale, Illinois.
66 p.Fusar.s., L. C. AND D. R. EDOs. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleot Ictaluridae:
Noturus placidus).
Southwest.
Nat., 39:217-223.
GIUzrrE, D. P., J. S. TIsmANN, D. R. EDns AND M. L WnIsmHA.a 2005. Spatiotemporal patterns of assemblage structure in a river impounded by low-head dams. Copeia, 2005: 539-549.Goms, L C. AND A. A. Axosnwruo.
1997. Influence of the flooding regime on the nutritional state.juvenile recruitment of the curimba, Prochilodus scrofa Steindachner, in upper Parana Ri Brazil. Fish. Manag. Eco, 4:263-274.
JENNINGs, M. J. AND D. P. PmLUpp. 1994. Biotic and abiotic factors affecting survival of early life hisi intervals of a stream-dwelling sunfish. Envisn. Biol. Fsh., 39:153-159.
KELLy, A. M. AND C. C. KoHLEt. 1996. Manipulation of spawning cycles of channel catfish in indoor w.recirculating systems. Prog. Fish-Cult., 58:221-228.
LiH. W., C. B. ScHRzcK, C. E. BornD ADm E. RxxsrAD. 1987. Factors influencing changes in assemblages of Pacific Northwest streams, p. 193-202. In:W.J. Matthews and D. C. Heins (ec Community and evolutionary ecology of. North American stream fishes. Univeristy Oklahoma Press, Norman, Oklahoma.Luxms,J. A. Amm D.J. Otmr. 1993. Reproductive ecology of redbreast sunfish Lomis auritus in a Virg stream. J. Freshwat.
EcoL, 8:235-244.
MAYDEN, R. L. AND B. M. BUas. 1981. Life history of the slender madtom, Noturus exilis, in soutk Illinois (Pisces: Ictaluridae).
Occas. Paper. Mus. Nat. Hist., Univ. Kans., 93:1-64.-AND S. L DEwKE 1980. Aspects of the life history of the Ozark madtom, Noturus albab southeastern Missouri (Pisces: Ictaluridae).
Am Midi. Nat., 104:335-340.
MiF- KE, P. W. %,tm H. K. Iaa. 1982. Permutation techniques for analyzing multi-response data f randomized block experiments.
Comm. Sat. Theor. Meth., 11:1427-1437.
S18 THE AmERucN MIDLAND NAxuxmuST 156(.Moss, R. E. 1983. Microhabitat selection in Neosho River riffles. Ph.D. Dissertation, University of Kanm Lawrence, Kansas. 294 p.MuNao, A. D., A. P. Scorr AmD T. J. LAM. 1990. Reproductive seasonality in teleosts:
environment influences.
CRC Press, Inc., Boca Raton, Florida. 254 p.NOLTIv, D. B. An M. H. A. KwEavsmE.
1986. Correlates of reproductive success in stream-dwelling ma rock bass, AmblopiUt rupestris (Centrarchidae).
Environ BioL Fish, 17:61-70.PIAtsoNS, T. N., W. L. HIRAM Ame G. A. LXmBaxmn.
1992. Influence of habitat complexity on resistance flooding and resilience of stream fish assemblages.
Trans. Am. Fish. Soc., 121:427-436.
Pnsmoim, D. G. Am D. R. EDDs. 1994. Reproductive traits of the Neosho madtom (Noturus piacidd (Pisces: Ictaluridae).
Trans. Kans. Acad. 54., 97:82-87.RaisER, D. W. AmD R. G. Wwm-r. 1990. Effects of streamflow reduction on chinook Salmon egg incubati(and fry quality. River, 1:110-118-SxcxarT, G. P. 1979. The lag sequential analysis of contingency and cyclicity in behavioral interacti(research, p. 623-649. InJ. D. Osofsky (ed.). Handbook of infant development.
Wiley, No York.1980. Lag sequential analysis as a data reduction technique in social interaction research, 300-340. In: D. B. Sawin, R. C. Hawkins IU, L 0. Walker andJ. H. Penticuff (eds.). Exceptior infant psychosocial risks in infant-environment transactions, Vol. 4. Brunner/Mazel, New Yoi SAS INsrnrrum 1999-2001.
SAS Software, Release 7.0. SAS Institute, Cary, North Carolina.SEEGisus D. W. Aim RK GARD. 1972. Effects of floods on trout in Sagehen Creek, California.
7Tans. A Fish. Soc., 101:478-482.
STOECiCEL,J.
N. 1993. Controlled cultivation techniques for the recovery of threatened fishes in V'igin Ph.D. Dissertation, Virginia Polytechnic Institute and State University, Blacksburg, Virgin 2 9 5 p.MD B. M. BURR. 1999. A review of key reproductive traits and methods used to spawn ictalurii p. 141-159. In:" E. R. Irwin, W. A. Hubert, C. F. Rabeni, H. L Schramm,Jr.
and T. Coon (eds Catfish 2000: proceedings of the international ictalurid symposium, Symposium
- 24. Americ Fisheries Society, Bethesda, Maryland.TAv-OYR, W. R. 1969 .A revision of the catfish genus Noturis Rafinesque, with an analysis of higher grot: in the Ictaluridae.
U. S. Mns. BulL, 282:1-315.
TIMmAm,J.
S., D. P. Gnirr, M. L WwnDAa- R AND D. K. EDDS. 2004. Effects of lowhead dams on rifl dwelling fishes and macroinvertebrates in a midwestern river. 7Tans. Am. FAsh. Sc., 133:705-7 U.S. CORPS OF ENrrIEEas.
1993. Civil works projects pertinent data. Tulsa District U.S. Army Coi of Engineers, Tulsa, Oklahoma.
146 p.U. S. Fish AND WiDtwus SERvimca 1991. Neosho madtom recovery plan. U. S. Fish and Wildlife Servi.Denver, Colorado.
42 p.VASAL, S. AND B. I. SutNAsjAR.
1976. Response of the ovary in the catfish, HeteropneustasfossiLis (Bloch), various combinations of photoperiod and temperature.
J. Exp. BioL, 197:247-264.
VoTiov, N. P. AND V. P. K.s'YANov.
1978. The ecology and reproductive efficiency of the Siber'sturgeon, Acipenser baer, in the Ob as affected by hydraulic engineering works. J. Jchd 18:20-29.WALSH, S.J. AND B. M. BURt. 1985. Biology of the stonecat, Noturusflavus (Siluriformes:
Ictaluridae), central Illinois and Missouri streams, and comparisons with Great Lakes populations a congeners.
OhioJ. Sci., 85:85-96.M. L., V. M. TABOR, J. E. WHITAKERm A. L. Au.lirr, D. W. MuuszmR, P.J. LAMBERSON AND K.POwELu. 2000. Ictalurid populations in relation to the presence of a main-stem reservoir a midwestern warmwater stream with emphasis on the threatened Neosho madtom. Trans., Fish. Soc., 129:1264-1280.L H., Ja. AND R. A. RULwSON. 1991. Instream flow and striped bass recruitment in the Lo&Roanoke River, North Carolina.
Rivers, 2:125-137.
SuiimrrmD 24 JAw'ARY 2005 Acm:~em. 9 MA~ci 21 Journal of Fish Biology (2006) 68, 1494-1512 doi: 10.1111 /j. 1095-8649.2006.01 037.x, available online at http://www.blackwell-synergy.com Habitat use by a Midwestern U.S.A. riverine fish assemblage:
effects of season, water temperature and river discharge D. P. GILLETTE*I, 1. S. TIEMANN*T, D. R. EDDS* AND M. L. WILDHABER§
- Department of Biological Sciences, Emporia State University, Emporia, Kansas 66801, U.S.A. and §U.S. Geological Survey, Columbia Environmental Research Center, Columbia, Missouri 65201, U.S.A.(Received I December 2004, Accepted 16 November 2005)The hypothesis that temperate stream fishes alter habitat use in response to Changing water temperature and stream discharge was evaluated over a I year period .in the Neosho River.Kansas, U.S.A. at two spatial scales. Winter patterns differed from those of all other seasons, with shallower water used less frequently.
and low-flow habitat more frequently, than at other times. Non-random habitat use was more frequent at the point scale (4.5 in 2) than at the larger reach scale (20-40 in), although patterns at both scales were similar. Relative to available habitats, assemblages used shallower, swifter-flowing water as temperature increased, and'shallower, slower-flowing water as river discharge increased.
River discharge had a stronger effect on assemblage habitat use than water temperature.
Proportion of juveniles in the assemblage did not have a significant effect. This study suggests that many riverine fishes shift habitats in response to changing environmental conditions, and supports, at the assem-blage level, the paradigm of lotic fishes switching from shallower, high-velocity habitats in summer to deeper, low-velocity habitats in winter, and of using shallower, low-velocity habitats during periods of high discharge.
Results also indicate that different species within temperate river fish assemblages show similar habitat use patterns at multiple scales in response to environmental gradients, but that non-random use of available habitats is more frequent at small scales. ,V 2006 The Fisheries Society of the British Isles (No claim to original US government works)Key words: dynamic landscape model; river discharge; river fishes; water depth; water flow;water temperature.
INTRODUCTION Recent conceptual models of temperate stream fish ecology, such as Schlosser's (1991, 1995) dynamic landscape model, emphasize habitat shifts in response to environmental variation.
In this model, stream fishes move among different tAuthor to whom correspondence should be addressed at present address: Department of Zoology, University of Oklahoma, 730 Van Vleet Oval, Room 314, Norman, Oklahoma 73019. U.S.A.Tel.: +1 405 325 7671; fax: +1 405 325 7560; email: dgillette@ou.edu JPresent address: Center for Biodiversity, Illinois Natural History Survey, Champaign, Illinois 61820.U.S.A.1494 c 2006 The Fisheries Society of the British Isles (No claim to original US government works)
RIVERINE ASSEMBLAGE HABITAT USE 1495 habitats in response to changing environmental gradients or life-history stages.In these temperate systems, strong environmental gradients are often generated by seasonal variation in water temperature and stream discharge.
In many cases, stream fishes respond to this variation by altering use of available water flow velocities and depths.Flow velocity has a strong influence on the energy expenditure required for a fish to maintain position in the water, and, for many fishes, on the supply of trophic resources.
In summer, during periods of potentially high growth, occu-pied habitats are typically those that maximize food intake and growth (Hughes &Dill, 1990; Hill & Grossman, 1993; Grossman et al., 2002), and are often characterized by high water flow velocities that Supply drifting invertebrate food resources (Hill & Grossman, 1993; Nakano, 1995). During winter, cold water temperatures characterized by low growth potential (Keast, 1985; Zapata& Granado Lorencio, 1993) predominate, making slower-flow velocity habitats most advantageous by allowing fishes to minimize energy expenditure (Cunjak, 1996). Likewise, as river discharge increases, fishes often seek refuge in slower-flowing waters (Harrell, 1978).Water depth determines which habitats serve ag refugia from biotic and abiotic threats. Shallow water provides protection from piscivorous fishes, and deeper water shelters fishes from avian and terrestrial predation (Power, 1987;Gorman, 1988). The importance of such refugia may vary with water tempera-lure; as metabolism of exothermic predators slows in colder weather, the impor-tance of refugia from predatory fishes may decrease.
Abiotic factors also become important in winter; when air temperatures become extremely cold, shallow water is likely to be colder than deep water, and in temperate systems subject to freezing, scouring from ice break-up can increase the risk of occupying shallow water (Brown et al., 2001). During periods of high stream discharge, however, fishes often inhabit shallow waters offering refuge from high-flow velocities (Ross & Baker, 1983; Matheney & Rabeni, 1995; Brown et al., 2001), and, in some cases, access to increased levels of trophic resources (O'Connell, 2003).The concepts above support a paradigm in which small-bodied fishes of temperate lotic systems are predicted to occupy shallower, higher-flow habitats in warm temperatures, and deeper, slower-flow habitats during cold-tempera-tures. Increased stream discharge also should lead to fishes using shallower, slower-flow habitat. Thus, the influence of water temperature and stream discharge represents a potential paradigm of predictable habitat shifts in response to environmental variation, as predicted by Schlosser's (1991, 1995)model. Although this paradigm is intuitive, its support has been drawn largely from autecological work on salmonids (Heggenes et al., 1993; Contour &Griffith, 1995; Young, 1999), or, less frequently, cyprinids (Matthews
& Hill, 1980; Lucas & Batley, 1996; Clough & Beaumont, 1998). Few studies, however, have compared characteristics of occupied and unoccupied habitats through-out the year for speciose temperate lotic fish assemblages (Matthews
& Hill, 1980; Bart, 1989). Such information is necessary before the paradigm above is applied at the assemblage level.Ontogeny can also have a strong effect on habitat use by fishes in lotic systems. Adults often occupy deeper habitats, with higher flow velocities, than t 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006, 68, 1494-1512 (No claim to original US government works) 1496 D. P. GILLETTE ET AL.juveniles (Schlosser, 1982; Gelwick, 1990). Thus, habitat use by riverine fish assemblages can be controlled both by extrinsic environmental factors and intrinsic characteristics of the assemblage itself.The present study examined year-round use of available water depths and. flow velocities at two spatial scales by fishes in the Neosho River, Kansas, U.S.A.This was done in two ways: (1) each species in each season was tested for patterns of non-random use of water depths and flow velocities; results for all species were then pooled by season to evaluate seasonal differences in frequency of these patterns.
Because patterns of habitat use by lotic fishes can vary with spatial scale (Baxter & Hauer, 2000; Mattingly
& Galat, 2002), the frequency of non-random habitat use at two different spatial scales was compared to evaluate the spatial extent of observed patterns and (2) for each collection, mean differ-.ence from average for water depths and flow velocities occupied by all fishes was calculated, then regressed against river discharge and water temperature.
To evaluate the effect of ontogeny, this value was also regressed against the propor-tion of juveniles in the assemblage.
The first approach assessed seasonal varia-tion in use of available water depths and flow velocities by individual species, and the second the extent to which assemblage-wide patterns of water depth and flow velocity use were associated with water temperature, stream discharge and assemblage ontogenetic composition.
MATERIALS AND METHODS STUDY SYSTEM The Neosho River, of the Arkansas River drainage, is fifth-order in the study reach. It flows south-east in Lyon County, KS, U.S.A., through mixed-grass prairie and cropland.Mature riparian vegetation, with associated canopy cover, occurs in some sections.Riffle-pool geomorphology occurs, and is especially pronounced during periods of low river discharge.
The substratum is rocky, composed primarily of gravel <64 mm in diameter, with'clay and silt, sand, boulder and some bedrock also present. Water is turbid all year-round, with mean turbidity of 37.7 nephelometric turbidity units (NTU)measured over the study period (D. Gillette & J. Tiemann, unpubl. data).Eight sites were sampled along a 34 km stretch of the Neosho River from Americus to Emporia, KS; one site was eliminated prior to analyses (a priori) because lack of landowner permission made it impossible to adequately assess available habitat. This resulted in seven sites retained for analysis (Fig. 1); the excluded site was located between sites I and 2. Study sites were chosen that were both representative of available habitat in this river section, and that could be sampled well with the methodology used. At each site, five permanent transects were fixed perpendicular to shore, spaced equally every 5 to 10 m, depending on length of the reach to be sampled. Width of cross-stream transects varied from 14 to 35 m.SAMPLING Each site was sampled monthly from November 2000 to October 2001, between the 9th and 22nd of each month, during daylight hours. Sampling order was randomized each month. Due to ice cover, sites 4 and 5 could not be sampled from December to February.In addition, flow velocities were too low at site 5 in September, October and November, site 6 in August and December, and site 4 in November, to be sampled effectively.
These collections were omitted from analyses, resulting in a total of 72 collections.
Sampling at each site proceeded from downstream to upstream transects, and from near shore to far shore points along each transect.
A maximum of five points were sampled along each 2006 The Fisheries Society of the British Isles, Journal of Fish Biology 2006, 68, 1494-1512 (No claim to original US government works)
RIVERINE ASSEMBLAGE HABITAT USE 1497 Site 2 Am~ericus N.T 0 3 km.Site 4 i6 Lyon County FIG. I. Map of study reach of the Neosho River in Lyon County, KS, U.S.A., showing seven study sites.transect, depending on river width and depth, and landowner permission.
Points along each transect were spaced at least 0-5 m apart to minimize disturbing adjacent points. At each point, fishes were sampled by kick-seining, using a 3 mm mesh seine. While one person fixed a 1-5 m seine at a sampling point, another person disturbed the substratum, moving downstream from a starting point 3 m upstream.
In this manner, fishes within a 4.5 m area were carried downstream and 'chased' into the seine. This methodology has been shown to effectively capture fishes from shallow, lotic habitats (Matthews, 1990;Wildhaber et al., 1999). Fishes were counted and identified as juvenile or adult, using a 30 mm total length (LT) maximum juvenile length for minnows (Campostoma, Pirnephales, Cyprinella and Notropis spp.) and darters (Etheostoma and Percina spp.), and a 50 mm LT maximum juvenile length for madtoms (Noturus spp.) and sunfishes (Lepomis spp.) (Gelwick, 1990).* Water depth and flow velocity at 60% depth were measured at each point. Water depth was measured with a metre-stick, reading depth from the downstream edge. Water flow velocity, was measured with a Global Flow Probe (Global Water Company, Inc., Gold River, CA, U.S.A.). For each site in each month (collection), surface water temperature was measured using a laboratory thermometer, and river discharge from the United States Geological Survey (U.S.G.S.)
gauging station in Americus, KS, U.S.A., was obtained via the U.S.G.S. web site (http://www.waterdata.usgs.gov/ks/nwis/rt).
DATA ANALYSIS Statistical analyses were performed using SPSS v.12 (SPSS, Inc.), SAS v.8 (SAS Institute, Inc.). and Resampling Stats v.5.0.2 (Resampling Stats, Inc.). Species occurring 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006. 68, 1494-1512 (No claim to original US government works) 1498 D. P. GILLETTE ET AL.in <10% of all collections were omitted from analyses because small sample sizes were unsuitable for statistical testing. Seasonal variation in available water depths and flow velocities was tested by one-way ANOVA, with collections as replicates and season as treatment.
Seasons were defined monthly as follows: autumn (September to November), winter (December to February), spring (March to May) and summer (June to August).Independence of water depth and flow velocity was evaluated using Pearson's r.Testing for non-random use of available water depths and flow velocities for each species in each season was conducted at two spatial scales. At the larger scale, mean depth and flow velocities of occupied stream reaches were compared to those of unoccu-pied stream reaches within each season, subsequently referred to as 'reach scale.' Each collection represented one stream reach, and mean reach depth and flow velocity were calculated by averaging all points sampled for a given collection.
At the smaller scale, depth and flow velocities of individual occupied points were compared to those of unoccupied points within each collection, referred to as 'point scale.' If habitat use by a species is random with respect to water depth or flow velocity, then the expected value for the difference between occupied and unoccupied habitats would be zero. Different methods were used to test this hypothesis at each spatial scale.At the reach scale, a randomization procedure (Manly, 1991) was used to test the null hypothesis that there was no difference in mean water depth and flow velocity between occupied and unoccupied reaches in a given season. For each species in each season, mean depth and flow velocity of occupied reaches was calculated, and the value for unoccupied reaches subtracted.
To generate the probability of this value occurring under the null hypothesis (random habitat use), all available stream reaches for that season were pooled, and 10 000 samples randomly selected with N equal to the number of reaches that the species being tested occurred in. For each sample, the difference between the means of selected and unselected depth and flow velocities was calculated.
The proportion of differences from these random samples > a positive observed difference, or < a negative observed difference, represented the probability of the observed differ-ence occurring under the null hypothesis, random habitat use (Manly, 1991). For exam-ple, ghost shiner Notropis buchanani Meek occurred in eight of 17 available stream reaches in autumn 2000. To test for non-random use of water depth, mean water depth was calculated for the eight occupied reaches, and mean water depth of the nine unoccupied reaches subtracted, producing a difference between means of occupied and unoccupied depths for N. buchanani in the autumn of 0.53 cm. To generate the prob-ability of this difference occurring under the null hypothesis (random habitat use), 10 000 random samples of eight water depths were generated from the 17 available reaches in autumn, and the difference between selected and unselected depths calculated for each.The proportion of random samples with differences
>0.53 was 0-473 (4730 out of 10 000 samples).
Thus Ho: (random use of available depths) was retained for N. buchanani in autumn, with a P value of 0-473.At the point scale, a one-sample t-test was used to test the same hypotheses.
For each species in each collection, mean depth and flow velocity of unoccupied points was subtracted from that of occupied points. The resulting values were pooled by season, and'the t-test used to generate the probability of the observed value occurring under Ho: (random use of available habitat).
A significant positive t indicated occupied > unoccupied, and a significant negative t indicated occupied < unoccupied.
The analyses above yielded a data set showing, for each species in each season, non-random patterns of habitat use at each scale, and the direction of such patterns (e.g.inhabited shallower or deeper water than expected at random). To test for differences in frequency of these patterns by season and spatial scale, further analyses were conducted on these results. Frequency was defined as the proportion of species showing a given habitat use pattern (e.g. the number of species using slower flow velocities), divided by the total number of species tested. For the seasonal analysis, this hypothesis was first tested for all non-random habitat use patterns combined [e.g. Ho: (combined frequency of all types of non-random habitat use does not vary by season)], then separately for each of the four specific patterns:
use of shallower habitats [e.g. Ho: (frequency of shallower habitat use does not vary by season)], use of deeper habitats, use of faster-flowing J 2006 The Fisheries Society of the British Isles, Journal of Fish Biology 2006, 68, 1494-1512 (No claim to original US government works)
RIVERINE ASSEMBLAGE HABITAT USE 1499 habitats and use of slower-flowing habitats.
For the purpose of this analysis, all rando-mization and t-tests-with P [HO: (occupied
= unoccupied)]
<0-05 were considered sig-nificant.
Although this rather liberal criterion increases probability of overall type I error, it does not affect results of testing for pattern frequency, as it was applied equally to all categories of comparison.
To test if frequency of non-random habitat use differed by season, the likelihood-ratio X test was used. Before proceeding, however, the appropriateness of combining results from point and reach scale analyses was first checked, using X2 to test whether or not* both scales showed the same frequency of non-random habitat use patterns by season. If both scales showed the same pattern, the results were pooled to test for seasonal differences.
The likelihood-ratio X 2 tests the hypothesis that the frequency of an event (non-random habitat use) is independent of the level of a categorical variable (season), and is the most appropriate test of this hypothesis for small to moderate sample sizes such as those in this study (Agresti & Finlay, 1997). For significant X2 comparisons, adjusted residual analysis*
was used to determine the cells responsible for the observed difference; this methodology is equivalent to using post hoc multiple comparison proce-dures to determine where differences lie for a significant ANOVA (Agresti & Finlay, 1997). To test if frequency of non-random habitat use was independent of scale, a z-test was used (Agresti & Finlay, 1997). To compare spatial scale of non-random habitat use to that of variation in habitat parameters, mean coefficient of variation (CV) for points within each collection and collections within each season was calculated for water depth and flow velocity.For the assemblage-level analysis, linear regression was employed to relate use of available water depths and flow velocities to river discharge and water temperature.
For each collection, the difference between occupied and unoccupied depths and flow velocities was calculated for all species, and these differences averaged to produce single values for water depth and flow velocity for each collection.
These values are referred to as *assemblage depth shift' and 'assemblage flow shift' and represent the extent to which species in a given collection collectively exhibited non-random use of available water depths and flow velocities.
These values for all collections were then regressed against water temperature, logio of mean daily river discharge and proportion of juveniles in the assemblage.
Simple linear regression was used to evaluate predictive ability of variables.
Independent predictive ability of collinear variables was assessed using regression on residuals.
Multiple linear regression with backwards elimination was used to compare predictive ability of all significant independent variables.
RESULTS Fourteen species occurred in at least 10% of collections, and were retained for analyses (Table I). Water temperature and river discharge varied seasonally (Fig. 2), leading to seasonal differences in available flow velocities and water depths (Fig. 3); available flow velocities were higher in winter and spring, and available depths greater in spring and summer. Sampled points ranged in depth from 2 to 121 cm, and in flow velocities from 0 to 1.7 m s-1. Water depth and flow velocity were not significantly correlated at the point (r = 0-045, P = 0-103) or reach (r = 0-031, P = 0!799) scale, allowing independent analysis of these two habitat measures.Non-random habitat use was frequent; 32% (61 of 193) of all comparisons were significant at cx = 0-05, and 9% (18 of 193) at sequential Bonferroni-corrected a (Rice 1989; Table 1). Non-random patterns of habitat use differed seasonally for three species. Neosho madtom Noturus placidus Taylor occupied deeper points with slower-flow velocities in winter, but shallower points with higher-flow velocities in summer. Channel darter Percina copelandi (Jordan)2006 The Fisheries Society of the British isles. Journal ofFish Biology 2006. 68, 1494-1512 (No claim to original US government works) 0;ý(0 NJ 0 0 0%-*1 2, (A 0 0 rb*0-C CC 0-,.0 o ~~. ~-ft;0 0 (A (TO CC.~-'-0-, LA'C.U, TABLE I. Results of seasonal test for non-random use of available water depths and flow velocities at two spatial scales for 14 fishes collected in the Neosho River, Lyon County, KS, U.S.A., from November 2000 to October 2001. Probabilities of the observed habitat use pattern occurring under Ho: (random habitat use) are shown. If significantly different from random, two-letter abbreviations describe occupied habitat relative to available habitat (Sh, shallower; De, deeper; SI, slower flow; Fa, faster flow). Capital letters denote conditions under which testing for non-random habitat use could not be conducted (A, occurred in all collections; AO. occurred in all but one collection; N, none collected; AB, absent at all but one site; F, degree of freedom I)Depth Flow Velocity Season Species Reach Point Reach Point Autumn 2000 Central stoneroller Campostonia anomahan (Rafinesque)
Red shiner Cyprinella lutrensis (Baird & Girard)Ghost shiner Notropis buchanani Meek Sand shiner Notropis stramineus (Cope)Bluntnose minnow Pimephales nolatus (Rafinesque)
Slim minnow Pinephales tenellus (Girard)Bullhead minnow Pimephales vigilax (Baird & Girard)Stonecat Noturusflavus Rafinesque Neosho madtom Noturus placidus Taylor Orangespotted sunfish Lepomis humilis (Girard)Orangethroat darter Etheostoma spectabile (Agassiz)Slenderhead darter Percina phoxocephala (Nelson)Logperch Percina caprodes (Rafinesque)
Channel darter Percina copelandi (Jordan)0-039 Sh*A 0-473 0-277 0-339 0-331 0-355 0-135 0-096 0.091 0-034 Sh*0-007 De*0.101 0-1.37 0.459 0.0014 Sh**0-318 0.877 0-608 0.281 0.808 0.021 Sh*0.144 0.230 0-085 0.080 0.317 0-391 0-048 Fa* 0-359 A 0-204 0-064 0-189 0.190 0-250 0-112 0-293 0.005 Sl*0-033 Fa*0-469 0-278 0-052 0-197 0.900 0-957 0-030 SI*0-362 0.065 0-054 0-214<0-0001 Sl**0-101 0.0004 Fa**0-0017 SI**F 0~0 C~)m-4-4-4 ttq eD 0~o.,'-, ,,o rm Io0 CD Winter 2000-2001 Spring 2001 C. anomalum C. lutrensis N. buchanani N. siramineus.
P. notatus P. tenellus P. vigilax N. flavus N. placidus L. humilis E. spectabile P. phoxocephala P. caprodes P. copelandi C. anomalum C. lutrensis N. buchanani N. stramineus P. notatus P. tenellus P. vigilax N. flavus N. placidus L. humilis E. spectabile P. phoxocephala P. caprodes P. copelandi 0.403 0.299 AB N 0.405 0.080 0.424 N 0.030 De*0.183 0.442 0.030 De*0.141 N 0.017 Sh*AO 0.208 0-0039 Sh**0.022 De*0.029 De*0.490 0.039 Sh*0.190 0.015 De*0.016 Sh*0-094 0.446 0.010 De*0-202 0-328 AB N 0.705 F 0-167 N 0-013 De*0.645 0-300 0-310 F N 0-396 0-031 Sh*0-041 De*0-234 0.014 Sh*0-258 0.956 F 0.135 0.012 Sh*0-0004 Sh**0-243 0-757 0.751 0-249 0.162 AB N 0-0042 Si**0.186 00001 Sl**N 0-481 0-0021 S!**0-074 0-250 0-047 SI*N 0-071 AO 0-219 0-073 0-354 0-331 0-428 0-115 0-253 0-331 0-167 0-290 0-113 0-0016 SI**0.040 Fa*0.0070 Sl**AB N 0.021 Sl*F 0.014 Sl*N 0-011 SI*0.0037 SI**0.835 0.450 F N 0.109 0.0024 SI**0.037 SI*0.595 0.0038 SI**0.074 0.026 Sl *F 0.449 0.0039 Sl**0-396 0.0018 Fa**0.438 0-146 m rrl rm rm-r-I I-'tJi 0 0~0,'-C, 0," E.I TABLE I. Continued Depth Flow Velocity Season Species Reach Point Reach Point Summer 2001 C. anomalum 0.007 Sh* 0.0005 Sh** 0.472 0.087 C. lutrensis A 0.818 A 0.996 N. buchanani 0.061 0.038 De* 0.015 Fa* 0.220 N. stramineus 0.083 0.172 0.255 0.525 P. notatus 0.331 0.108 0.091 0.031 SI*P. tenellus 0.209 0.858 0.407 0.212 P. vigilax 0.309 0.122 0.266 0.012 SI*N..flavus 0.008 ShO 0.112 0.259 0.038 Fa*N. placidus 0.025 ShO 0.744 0-390 0-023 Fa*L. humilis A 0.242 A 0.013 SI*E. speciabile 0.499 0.0010 Sh** 0.480 0-086 PA phoxocephala 0.441 0.167 0.013 Fa* 0.038 Fa*P. caprodes 0.330 0.013 Sh* 0.063 0.530 P. copelandi 0.200 0021 Sh* 0.411 0.867* Test significant at a = 0.05.** Test significant at sequential Bonferroni-corrected (x (Rice. 1989) applied seasonally.
mr m-I RIVERINE ASSEMBLAGE HABITAT USE 1503 35-30 25 20'~15 10 0'u 80'S60,.A 40-20 0 Nov Jan Mar May Jul Sep Month FiG. 2. (a) Mean + S.D. monthly surface water temperature from seven study sites and (b) daily river discharge measured at the United States Geological Survey gauging station on the Neosho River at Americus, KS, U.S.A., from November 2000 to October 2001.used shallower points in summer, but used deeper reaches, or used depths randomly, in all other seasons. Notropis buchanani used slower-flow points in spring, but higher-flow reaches in summer. In spring, two species showed scale-specific patterns of habitat use. Orangespotted sunfish Lepomis humilis (Girard)and bluntnose minnow Pihnephales notatus (Rafinesque) occupied reaches deeper than available at random, but points shallower than available at random.Reach and point scale analyses showed. the same seasonal patterns of non-random habitat use frequency (X 2 , 0.25 < P < 0-90 for all, shallower, deeper, slower and faster). That is, patterns of habitat use (e.g. shallower habitat use)occurring frequently in a given season at one scale, also occurred frequently in the same season at the other scale. Consequently, results from the point and reach scale were combined for analysis of seasonal frequency patterns.Combined frequency of non-random habitat use did not vary by season (Table II), nor did that of deeper or faster-flowing habitat use. Frequency of shallower habitat and slower-flowing habitat use did vary, however (Table 11).Both of these differences were driven by winter patterns; z-scores from adjusted residual analysis showed significant results for only two cells: proportion of species using shallower (z = 88, P = 0.030) and slower-flowing (z = 2.43, P = 0-008) habitat in winter (Table 11). Thus, during winter, fishes used* 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006, 68, 1494-1512 (No claim to original US government works) 1504 D. P. GILLETTE ET AL.0.8 , 0-6 b o0 b ab 0-4.aab 0-2.iC 0-0., 60'b b 40 0 a a20 E t i 00 FIo. 3. Mean + s.m. (a) flow velocity and (b) water depth by season for 72 collections from the Neosho River, Lyon Co., KS, U.S.A., from November 2000 to October 2001. ANOVA results were (a)F 3.6 9 = 3-828, P = 0.014 and (b) F3.6 9 = 5-325, P 0.002. Lowercase letters indicate results of post-hoc least-significant difference tests.shallower habitat significantly less frequently, and slower-flowing habitat significantly more frequently, than in other seasons.Non-random habitat use was significantly more frequent at the point scale (36-1%) than the reach scale (27.1%, z = 6.30, P < 0-001). Mean CV for reaches in each season (mean +/- S.D. water depth = 47-90 +/- 9-26 cm and flow velocity = 61.49 +/- 13.07 m s-) was similar to that for points within each reach (mean +/- S.D. water depth = 42.54 +/- 11.01, flow velocity = 85-32 +/- 35.94), indicating that similar levels of variation in available water depths and flow velocities existed at both scales.Assemblage-level analysis revealed a significant relationship between assemblage depth and flow shifts, and mean daily river discharge and water temperature (Fig. 4). As river discharge increased, assemblages used shallower, slower-flowing habitat than that available.
As water temperature increased, assemblages used shallower and swifter-flowing habitat than that available.
For both assemblage depth shift and flow shift, however, mean daily river discharge had the strongest effect, with standardized regression coefficients greater than those of water i. 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006. 68, 1494-1512 (No claim to original US government works)
RIVERINE ASSEMBLAGE HABITAT USE 1505 TABLE 11. Proportions of fish species collected in the Neosho River, Lyon County, KS, U.S.A., from November 2000 to October 2001 showing non-random habitat use patterns by season, with results of X 2 test for seasonal differences in frequency of each pattern Season Habitat use pattern Autumn Winter Spring Summer Likelihood-ratio X 2 results Combined 0.231 0.361 0.385 0.308 X 2= 3-29, d.f. = 3, P =-349 Shallower 0-148 0.0001* 0.308 0-269 X 2 = 10-92, d.f. = 3, P = 0-012*Deeper 0-037 0.167 0-192 0.038 X 2 = 5-67, d.f. = 3, P = 0-129 Slower 0-154 0-500** 0-231 0.115 X 2 9-38, d.f. = 3, P = 0-025*Faster 0-115 0-056 0-038 0.192 X 2 3.90, d.f. 3, P = 0-272* Comparisons significant at a = 0.05.** Cells differing significantly from expected values as determined by z-test from adjusted residual analysis.temperature for both multiple regressions (Table Ill).This difference was strongest for the depth shift regression, where water temperature did not have a significant effect when included in the multiple regression model with river discharge.
Water temperature and mean daily river discharge were not collinear (r = 0.114, P = 0-342). The proportion of juveniles in the assemblage had a significant effect 40 40* 0(b)3 0 (a)20. 20 10. .1.; .10 %..0 0w -10t.. ..".-20.
- 300 * -40 1.0 5 0.0 0:5 1:0 1-5 2-0 0 5 10 15 20 25 30 35 0-2 0-2(c)
- V 0.1. (d) **"~0-1*0-0 0-0. *~ * :2 ~ : u-0-1 .-0 b-0-2 2 b-03.3 ,-0-3.-0-4 .-04.4-0-5 5 --0-6 1-0-6--1-0 5 0-0 0-5 1-0 1-5 2-0 0 5 10 15 20 25 30 35 Logl 0 daily mean discharge (m 3 s s-1) Water temperature (C)FIo. 4. Linear regression plots of (a) assemblage depth shift and river discharge. (b) assemblage depth shift and water temperature, (c) assemblage flow shift and river discharge and (d) assemblage flow shift and water temperature for 72 collections from the Neosho River, Lyon Co.. KS. from November 2000 to October 2001. The curves were fitted by: (a) y = 1-08 52x (P < 0-001), (b) ' = 3-74 23x (P = 0.027). (c)y = 04 14x(P < 0-001) and (d)y = 14 + 0.005x (P = 0.004).ýr 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006. 68, 1494-1512 (No claim to original US government works) 1506 D. P. GILLETTE ET AL.TABLE 111. Standardized regression coefficients for multiple linear regressions of fish assemblage depth and flow velocity shifts against water temperature and.river discharge for 72 collections from seven sites in the Neosho River, Lyon County,. KS, U.S.A., from November 2000 to October 2001 Variable Dependent Independent Coefficient t P Depth shift Water temperature
-0,179 883 0.064 River discharge 529 -5.569 <0-001 Flow shift Water temperature 0-391 4.016 <0-001 River discharge 497 -5.105 <0.001 on depth shift (y 88 + 16-01x, r 2 0-14, FII, P= 0-001), but not on flow shift (F 1.7 1 , P = 0-47). The proportion ofjuveniles in the assemblage, however, was negatively correlated with daily river discharge (r = 49, P < 0-001). After removing the river discharge effect by taking, residuals of the discharge-depth shift regression, juvenile composition no longer had a significant effect on depth shift (F 1 , 7 1, P = 0-27). After removing the effect of juvenile composition, however, the effect of river discharge on depth shift was still strong (FI, 7 1 , P < 0-001). Thus, proportion of juveniles in the assemblage did not have an independent effect on either depth or flow shift.DISCUSSION Results of this study suggest that temperate riverine fishes use available water depths and flow velocities differently as environmental conditions change. This pattern fits conceptual models, such as Schlosser's (1991, 1995) dynamic land-scape model, that regard movement among habitats as a central feature of lotic fish ecology. As previous authors have noted (Fausch et al., 2002), application of this model in a conservation context necessitates recognition that species need multiple habitat types among which to move in response to changing environ-mental gradients.
Certain habitats may be used rarely at one time, but more frequently at another. For example, winter habitat use patterns in this study differed markedly from those of other seasons, with increased use of slower-flow habitat, and decreased use of shallower water habitat.The high frequency of non-random habitat use observed at two spatial scales in this study is similar to findings of previous authors at the microhabitat scale (Grossman
& Freeman, 1987; Grossman & Ratajczak, 1998). Collectively, such results indicate that habitat selection is a characteristic of many lotic fish assemblages.
Despite the large body of work demonstrating habitat selection by individual species (Fraser & Cerri, 1982; Holbrook & Schmitt, 1988;Schlosser, 1988; Fraser & Gilliam, 1992; Petty & Grossman, 1996; Utne et al., 1997; Thompson et al., 2001), studies documenting year-round patterns in entire fish assemblages are uncommon.At both spatial scales, results of the present study support the hypothesis that fishes use slower, deeper water during cold temperatures, and swifter-flowing, c 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006, 68, 1494-1512 (No claim to original US government works)
RIVERINE ASSEMBLAGE HABITAT USE 1507 shallower water during warm temperatures (Bjornn, 1971; Matthews & Hill, 1979; Cunjak, 1996; David & Closs, 2003). This conclusion is based on seasonal frequencies of individual species patterns, as well as regression of assemblage depth and flow shifts against water temperature.
The hypothesis above is based on the assumption that high-flow velocities are better foraging habitats (Gamer et al., 1998), but are more energetically expensive to occupy (Facey & Grossman, 1990); stream fishes often select habitats that maximize net energetic gain (Fausch & White, 1986).Although this model has been applied primarily to drift-feeding invertivorous fishes, it can logically be extended to species with different trophic ecologies.
Most fishes in this study feed on invertebrates, with the exception of the herbi-vorous central stoneroller Campostoina anomalum (Rafinesque) (Pflieger, 1997).Of these, darters, madtoms and some minnows are benthic pickers (Matthews, 1998), gleaning invertebrates from the benthos. Many of the organisms these fishes consume require flowing water with high levels of dissolved oxygen, and silt-free, rocky substrata on which to graze and find shelter (Merritt & Cummins, 1995). Consequently, flow may be a requirement of good foraging habitat for benthic pickers as well as for drift feeders. Shallower, swifter-flowing areas may also provide the best feeding habitats for grazers, as siltation can hinder algivory by fishes (Power, 1984; Gelwick el al., 1997), and light can limit benthic auto-troph production in some lotic systems (Murphy & Hall, 1982; Keithan & Lowe, 1985; Lowe et al., 1986). In turbid rivers such as the Neosho, deep water may limit light penetration for photosynthesis, potentially restricting C. anonialum foraging to shallow areas. Indeed, reaches occupied by C. anomalum were shallower than unoccupied reaches in all seasons except winter (Table I). Thus, shallower, swifter-flowing water may provide the best foraging habitat for most fishes in this study, predicting increased occupancy of such habitat during times of high energy expenditure in warmer months. In cold weather, conversely, shallower, higher velocity habitats become less advantageous as metabolism slows, reducing the amount of trophic resources required and making it more difficult to maintain position in swift river currents (Graham et al., 1996).Predation threat may also drive use of shallower water during warm months.Piscivorous fishes' high energetic requirements and associated feeding rates in summer could necessitate occupation of shallow water spatial refugia by small bodied fishes. In cold weather, predators' metabolisms slow, causing them to feed less (Little et al., 1998), and potentially decreasing predation risk in deeper water.In addition to temperature-driven seasonal variation, fish assemblage habitat use changed in response to river discharge.
This effect was stronger than that of water temperature, as indicated by comparison of standardized regression coef-ficients.
These results are consistent with findings of previous authors (Harrell, 1978; Ross & Baker, 1983; Matheney & Rabeni, 1995; Brown et al., 2001), showing that fishes of lotic systems respond to rising water levels by increasing use of shallower, slower-flowing habitat. Presumably, such behaviour decreases downstream displacement (Harvey, 1987; Gido et al., 1997; Brown el al., 2001), and may increase access to trophic resources along channel margins (O'Connell, 2003). This strong effect of discharge on habitat use highlights the potential impact of altered flow regimes on riverine fishes, suggesting that maintenance of 2006 The Fisheries Society of the British Isles, Journal of Fish Biologyv 2006., 68, 1494-1512 (No claim to original US government works) 1508 D. P. GILLETTE ET AL.natural flow regimes is an important part of conserving riverine fish faunas (Poff et al., 1997).Proportion of juveniles in the assemblage did not have an independent effect on assemblage-wide use of available depths and flow velocities.
This result differs from previous work (Schlosser, 1982; Gelwick, 1990) showing ontogenetic habi-tat use differences between conspecifics.
Although ontogeny had no effect in the present study, differences in habitat use between juveniles and adults may exist, but be overshadowed by assemblage-wide responses to changing discharge levels.Habitat-use patterns of individual species in this study generally did not differ by season or scale. Three species (N. placidus, P. copelandi and N. buchanan), however, showed non-random habitat use in one direction in one season, and in the opposite direction in another season. Two other species (P. notatus and L. humilis) showed non-random patterns of habitat use in one direction at one spatial scale, and in the opposite direction at the other scale. These four species illustrate the need for multi-scale, multi-season data to adequately describe habitat use patterns of certain taxa.In this study, non-random habitat use was more frequent at the smaller point scale than at the larger reach scale. Because different statistical approaches were used at each scale, this difference could potentially be a statistical artefact.Randomization tests, however, generally have power equal to that of appropri-ate parametric statistics (Manly, 1991). Coefficients of variation for. available depths and flow velocities were similar at both scales, indicating equivalent levels of habitat variation.
Thus, low frequency of non-random habitat use at the reach scale was not due to a lack of habitat variation.
In many aquatic systems, comparison of similar patterns at multiple scales is inappropriate, because causal mechanisms can differ (Fisher, 1992). The present study, however, suggests that similar mechanisms operated at both point and reach scales. X 2 analysis showed no difference in distribution of significant patterns by season; thus, frequency of use of each habitat type (shallower, deeper, slower- and faster-flowing) was distributed similarly among seasons at both scales. This indicates that the same pattern of habitat use occurred at both scales, but was stronger at the smaller point scale, leading to a higher frequency of significant differences.
In comparing occupied to unoccupied habitats, patterns can potentially be driven by changes in habitat availability, rather than by habitat selection.
If a species tracks a constant water depth throughout the year, occupied depths could be less than unoccupied depths during high river discharge, and greater than unoccupied depths during low discharge.
In the present study, however, this is probably not the case. The main seasonal pattern was increased frequency of slower-flow, deeper habitat use in winter. Winter flow velocities were fairly high (Fig. 3), so increased use of slower flows could potentially be a consequence of habitat availability.
Available flow velocities were even greater in spring, when slower-flow habitat was used less.than half as frequently.
If fishes were indeed occupying the same velocity regardless of available habitat, frequency of slower-flow habitat use should be greater in spring than in winter. Instead, results suggest that fishes were selecting habitat differently in the two seasons.Likewise, available water depths in autumn and winter were similar. No species, however, used shallower habitat in winter, whereas several species did in ,: 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006. 68, 1494-1512 (No claim to original US government works)
RIVERINE ASSEMBLAGE HABITAT USE 1509 autumn. Thus, seasonal patterns in this study were due to changes in fish habitat selection, rather than changes in habitat availability.
Habitat shifts related to changing water temperature and river discharge, such as those documented in the present study, are an important component of seasonal habitat use patterns by temperate fish assemblages, and should be explicitly incorporated into conceptual models and conservation plans for fishes of these systems. From a conservation perspective, it is important to note that these patterns may be applicable on an assemblage-wide, and not just a species-specific level. This suggests that assemblage-wide conservation and management plans may be applicable to the Neosho River and similar systems, a view important in light of recent calls for conservation plans encompassing whole communities, rather than focusing on individual species (Rohlf, 1991; Sergio el al., 2003).We thank B. Chance, J. Dean, L. Freeman, B. Harkins, J. Howard, S. Sherraden and 1. Singh for assistance in the field. W. and M. Leffler, P. and D. Matile, L. Schlessener, G. Guide, the City of Emporia, and Emporia State University (ESU) Natural Areas generously provided river access. D. Zelmer, L. Scott and.D. Moore provided valuable suggestions throughout the course of this study. S. Gillette helped with data entry and management.
We thank W. Matthews and three anonymous reviewers for reviewing an earlier version of the manuscript.
Funding for this study was provided by a Faculty Research and Creativity Grant and a Graduate Student Research Grant from ESU, and by the U.S. Geological Survey, Department of the Interior, under USGS Cooperative Agreement No. O0CRAGO025.
All fishes were collected under Kansas Department of Wildlife and Parks Scientific Collector's Permits SC-065-2000 (2000) and SC-033-2001 (2001), issued to D.R. Edds.References Agresti, A. & Finlay, B. (1997). Statistical Methods for the Social Sciences, Rev. edn.Upper Saddle River, NJ: Prentice Hall.Bart, H. L., Jr. (1989). Fish habitat association in an Ozark stream. Environmental Biology of Fishes 24, 173-186.Baxter, C. V. & Hauer, F. R. (2000). Geomorphology, hypoporheic exchange, and selection of spawning habitat. by bull trout (Salvelinus confluentus).
Canadian Journal of Fisheries and Aquatic Sciences 57, 1470-148 1.Bjornn, T. C. (1971). Trout and salmon movements in two Idaho streams as related to temperature, food, stream flow, cover, and population density. Transactions of the American Fisheries Society 100, 423-438.Brown, R. S., Power, G. & Beltaos, S. (2001). Winter movements and habitat use of riverine brown trout, white sucker and common carp in relation to flooding and ice break-up.
Journal of Fish Biology 59, 1126-1141.
doi: 10.1006/jfbi.2001.1725 Clough, S. & Beaumont, W. R. C. (1998). Use of miniature radio-transmitters to track the movements of dace, Leuciscus leuciscus (L.) in the River From, Dorset.Hydrobiologia 371/372, 89-97.Contour, R. C. & Griffith, 3. S. (1995). Nocturnal emergence of juvenile rainbow trout from winter concealment relative to light intensity.
Hydrobiologia 299, 179-183.Cunjak, R. A. (1996). Winter habitat of selected stream fishes and potential impacts from land-use activity.
Canadian Journal of Fisheries and Aquatic Sciences 53 (Suppl. 1), 267-282.David, B. 0. & Closs, G. P. (2003). Seasonal variation in diel activity and microhabitat use of an endemic New Zealand stream-dwelling galaxiid fish. Freshwater Biology 48, 1765-1781.
- 't 2006 The Fisheries Society of the British Isles, Journal of Fish Biology' 2006, 68, 1494-1512 (No claim to original US government works) 1510 D. P. GILLETTE ET AL.Facey, D. E. & Grossman, G. D. (1990). The metabolic cost of maintaining position for four North American stream fishes: effects of season and velocity.
Physiological Zoology 63, 757-776.Fausch, K. D. & White, R. J. (1986). Competition among juveniles of coho salmon, brook trout, and brown trout in a laboratory stream, and implications for Great ,Lakes tributaries.
Transactions of the American Fisheries Society 115, 363-381.Fausch, K. D., Torgerson, C. E., Baxter, C. V. & Li, H. W. (2002). Landscapes to riverscapes:
bridging the gap between research and conservation of stream fishes.Bioscience 52, 483-498.Fisher, S. G. (1992). Pattern, process and scale in freshwater systems: some unifying thoughts.
In Aquatic Ecology- Scale, Pattern. and Process: Proceedings.
of the 34"1* Symposium of the British Ecological Society with the American Society of Limnology and Oceanography (Giller, P. S., Hildrew, A. G. & Raffaelli, D. G., eds), pp. 575-591. British Ecological Society Symposium 34.Fraser, D. F. & Cerri, R. D. (1982). Experimental evaluation of predator-prey relation-ships in a patchy environment:
consequences for habitat use patterns in minnows.Ecology 63, 307-313.Fraser, D. F. & Gilliam, J. F. (1992). Nonlethal impacts of predator invasion:
facultative suppression of growth and reproduction.
Ecology 73, 959-970*Garner, P., Clough, S., Griffiths, S. W., Deans, D. & lbbotson, A. (1998). Use of shallow marginal habitat by Phoxinusphoxinus:
a trade-off between temperature and food?Journal of Fish Biology 52, 600-609.Gelwick, F. P. (1990). Longitudinal and temporal comparison of riffle. and pool fish assemblages in a northeastern Oklahoma Ozark stream. Copeia 1990, 1072-1082.
Gelwick, F. P., Stock, M. S. & Matthews, W. J. (1997). Effects of fish, water depth, and predation risk on patch dynamics in a north-temperate river ecosystem.
Oikos 80, 382-398.Gido, K. B., Propst, D. L. & Molles, M. C. (1997). Spatial and temporal variation of fish communities in secondary channels of the San Juan River, New Mexico and Utah.Environmental Biology of Fishes 49, 417-434.Gorman, 0. T. (1988). The dynamics of habitat use in a guild of Ozark minnows.Ecological Monographs 58, 1-18.Graham, W. D., Thorpe, J. E. & Metcalfe, N. B. (1996). Seasonal current holding performance of juvenile Atlantic salmon in relation to temperature and smolting.Canadian Journal of Fisheries and Aquatic Sciences 53, 80-86.Grossman, G. D. & Freeman, M. C. (1987). Microhabitat use in a stream fish assem-blage. Journal of Zoology, London 212, 151-176.Grossman, G. D. & Ratajczak, R. E. (1998). Long-term patterns of microhabitat use by fish in a southern Appalachian stream from 1983 to 1992: effects of hydrologic period, season and fish length. Ecology of Freshwater Fish 7, 108-131.Grossman, G. D., Rincon, P. A., Farr, M. D. & Ratajczak, R. E. (2002). A new optimal foraging model predicts habitat use by drift-feeding stream minnows. Ecology of Freshwater Fish 11, 2-10.Harrell, H. L. (1978). Responses of the Devils River (Texas) fish community to flooding.Copeia 1978, 60-68.Harvey, B. C. (1987). Susceptibility of young-of-the-year fishes to downstream displace-ment by flooding.
Transactions of the American Fisheries Society 116, 851-855.Heggenes, J., Krop, 0. M. W., Lindas, 0. R., Dokk, J. G. & Bremnes, T. (1993).Homeostatic behavioural responses in a changing environment:
brown trout (Salmo trutta) become nocturnal in winter. Journal of Animal Ecology 62, 295-308.Hill, J. & Grossman, G. D. (1993). An energetic model of microhabitat use for rainbow trout and rosyside dace. Ecology 74, 685-698.Holbrook, S. J. & Schmitt, R. J. (1988). The combined effects of predation risk and food reward on patch selection.
Ecology 69, 125-134.Hughes, N. F. & Dill, L. M. (1990). Position choice by drift-feeding salmonids:
models and test for Arctic grayling (Tlhvmallus arcticus) in subarctic mountain streams, interior Alaska. Canadian Journal of Fisheries and Aquatic Sciences 47, 2039-2048.
2006 The Fisheries Society of the British Isles, Journal of Fish Biology 2006, 68, 1494-1512 (No claim to original US government works)
RIVERINE ASSEMBLAGE HABITAT USE 1511 Keast, A. (1985). Growth response of the brown bullhead Ictalurus nebulosus to temperature.
Canadian Journal of Zoology 63, 1510-1515.
Keithan, E. D. & Lowe, R. L. (1985). Primary productivity and spatial structure of phytolithic growth in streams in the Great Smoky Mountains National Park, Tennessee.
Hydrobiologia 123, 59-67.Little, A. S., Tonn, W. M., Tallman, R. F. & Reist, J. D. (1998). Seasonal variation in diet and trophic relationships within the fish communities of the lower Slave River, Northwest Territories, Canada. Environmental Biology of Fishes 53, 429-445.Lowe, R. L., Golloday, S. W. & Webster, J. R. (1986). Periphyton response to nutrient manipulation in streams draining clearcut and forested watersheds.
Journal of the North American Benthological Society 5, 221-229.Lucas, M. C. & Batley, E. (1996). Seasonal movements and behaviour of adult barbel Barbus barbus, a riverine cyprinid fish: implications for river management.
Journal of Applied Ecology 33, 1345-1358.
Manly, B. F. J. (1991). Randomization and Monte Carlo Methods in Biology, Rev. edn.London: Chapman & Hall.Matheney, M. P. & Rabeni, C. F. (1995). Patterns of movement and habitat use by northern hog suckers in an Ozark stream. Transactions of the American Fisheries Society 124, 886-897.Matthews, W. J. (1990). Spatial and temporal variation in fishes of riffle habitats; a comparison of analytical approaches for the Roanoke River. American Midland Naturalist 124, 31-45.Matthews, W. J. (1998). Patterns of Freshwater Fish Ecology. New York: Chapman &Hall.Matthews, W. J. & Hill, L. G. (1979). Influence of physico-chemical factors on habitat selection by red shiners, Notropis lutrensis (Pisces: Cyprinidae).
Copeia 1979,* 70-81.Matthews, W. J. & Hill, L. G. (1980). Habitat partitioning in the fish community of a southwestern river. Southwestern Naturalist 25, 51-66.Mattingly, H. T. & Galat, D. L. (2002). Distributional patterns of the threatened Niangua darter, Etheostoma nianguae, at three spatial scales, with implications for species conservation.
Copeia 2002, 573-585.Merritt, R. W. & Cummins, K. W. (1995). An Introduction to the Aquatic Insects of North America, 3rd edn. Dubuque, IA: Kendall/Hunt.
Murphy, M. L. & Hall,J. D. (1982). Varied effects of clear-cut logging on predators and their habitat in small streams of the Cascade Mountains, Oregon. Canadian Journal of Fisheries and Aquatic Sciences 38, 137-145.Nakano, S. (1995). Competitive interactions for foraging microhabitats in a size-structured interspecific dominance hierarchy of two sympatric stream salmonids in a natural habitat. Canadian Journal of Zoology 73, 1845-1854.
O'Connell, M. T. (2003). Direct exploitation of prey on an inundated floodplain by cherryfin shiners (Lythrurus roseipinnis) in a low order, blackwater stream. Copeia 2003,635-645.
Petty, J. T. & Grossman, G. D. (1996). Patch selection by mottled sculpin (Pisces: Cottidae) in a southern Appalachian stream. Freshwater Biology 35, 261-276.Pflieger, W. L. (1997). The Fishes of Missouri, Rev. edn. Jefferson City, MO: Missouri Department of Conservation.
Poff, N. L., Allan, J. D., Bain, M. B., Karr, J. R., Prestegaard, K. L., Richter, B. D., Sparks, R. E. & Stromberg, J. C. (1997). The natural flow regime: a paradigm for river conservation and restoration.
Bioscience 47, 769-784.Power, M. E. (1984). The importance of sediment in the grazing ecology and size class interactions of an armored catfish, Ancistrus spinosus.
Environmental Biology of Fishes 10, 173-181.Power, M. E. (1987). Predator avoidance by grazing fishes in temperate and tropical streams: importance of stream depth and prey size. In Predation (Kerfoot, W. C. &Sih, A, eds), pp. 333-351. Hanover, NH: University Press of New England.c. 2006 The Fisheries Society of the British Isles. Journal of Fish Biology 2006, 68, 1494-1512 (No claim to original US government works) 1512 D. P. GILLETTE ET AL.Rice, W. R. (1989). Analyzing tables of statistical tests. Evolution 43, 223-225.Rohlf, D. J. (1991). 6 biological reasons why the Endangered Species Act doesn't work -and what to do about it. Conservation Biology 5, 273-282.Ross, S. T. & Baker, J. A. (1983). The response of fishes to periodic spring floods in a southeastern stream. American Midland Naturalist 109, 1-14.Schlosser, 1. J. (1982). Fish community structure and function along two habitat gradi-ents in a headwater stream. Ecological Monographs 52, 395-414.Schlosser, 1. J. (1988). Predation risk and habitat selection by two size classes of a stream cyprinid:
experimental test of a hypothesis.
Oikos 52, 36-40.Schlosser, 1. J. (1991). Stream fish ecology: a landscape perspective.
Bioscience 41, 704-712.Schlosser, 1. J. (1995). Critical landscape processes that influence fish population dynamics in headwater streams. Hydrobiologia 303, 71-81.Sergio, F., Pedrini, P. & Marchesi, L. (2003). Reconciling the dichotomy between single species and ecosystem conservation:
black kites (Milvus migrans) and eutrophica-tion in pre-Alpine lakes. Biological Conservation 110, 101-111.Thompson, A., Petty, J. T. & Grossman, G. D. (2001). Multi-scale effects of resource patchiness on foraging behaviour and habitat use by longnose dace, Rhinichthvs cataractae.
Freshwater Biology 46, 145-160..Utne, A. C. W., Brannas, E. & Magnhagen, C. (1997). Individual responses to predation risk and food density in perch (Percafluviatilis L.). Canadian Journal of Zoology 75, 2027-2035.
Wildhaber, M. L., Allert, A. L. & Schmitt, C. J. (1999). Potential effects of interspecific competition on Neosho madtom (Noturus placidus) populations.
Journal of Freshwater Ecology 14, 19-30.Young, M. K. (1999). Summer diel activity and movement of adult brown trout in high-elevation streams in Wyoming, U.S.A. Journal of Fish Biology 54, 181-189.doi: 10.1006/jfbi.
1998.0854 Zapata, S. C. & Granado Lorencio, C. (1993). Age, growth and feeding of the exotic species Lepomis gibbosus in a Spanish cooling reservoir.
Archivfur Hydrobiologie 4 (Suppl. 90), 561-573..c 2006 The Fisheries Society of the British Isles, Journal of Fish Biology 2006. 68, 1494-1512 (No claim to original US government works)
Copyrighted Material Protect accordingly C.p'a, 2005(3), pp. 539-549'V 4 Spatiotemporal Patterns of Fish Assemblage Structure in a River Impounded by Low-Head Dams DAVID P. GIuzETrE, JEREMY S. TIEMANN, DAVID R. EDDS, AND MARK L. WiLDHABER We studied spatiotemporal patterns of fish assemblage structure in the Neosho River, Kansas, a system impounded by low-head dams. Spatial variation in the fish assemblage was related to the location of dams that created alternating lotic and lentic stream reaches with differing fish assemblages.
At upstream sites close to dams, assemblages were characterized by species associated with deeper, slower-flowing habitat. Assemblages at sites immediately downstream from dams had higher abundance of species common to shallow, swift-flowing habitat. Temporal variation in assemblage structure was stronger than spatial variation, and was associated with fish life history events such as spawning and recruitment, as well as seasonal changes in environmental conditions.
Our results suggest that low-head dams can influence spatial patterns of fish assemblage structure in systems such as the Neosho River and that such assemblages also vary seasonally.
R rVERINE fish assemblage structure often varies along environmental gradients from headwaters to lower mainstem (Schlosser, 1982;Gelwick, 1990; Edds, 1993). When these gradi-ents are interrupted, however, alternate pat-terns can result. Such interruptions can be nat-ural, as is the case with waterfalls (Balon and Stewart, 1983; Maret et al., 1997) or rapids- (Bal-on and Stewart, 1983), or anthropogenic, as oc-curs with river fragmentation from damming (reviewed by Baxter, 1977; Dynesius and Nils-son, 1994; and Richter et al., 1997).Dams affect lotic systems in many ways, and their impacts are often reflected in the spatial patterns of fish assemblages.
Dam construction can fragment watersheds (Dynesius and Nilsson, 1994), affecting fish assemblages directly by eliminating or reducing movement of fishes, leading to reduced upstream species richness, especially.
for migratory species (ReyesGavilan et al., 1996; Holmquist et al., 1998; March et al.,.2003). Alteration of the natural flow regime also influences fish assemblage structure, reducing in abundance species reliant upon seasonal flow variation to complete their life cycle (Bonner and Wilde, 2000; Minckley et al., 2003). Hypo-limnial-release dams can also sharply decrease downstream water temperature, resulting in de-creased growth and increased time to maturity for native fishes (Clarkson and Childs, 2000)and often replacing native warmwater assem-blages with non-native coldwater assemblages over time (Quinn and Kwak, 2003). Impound-ment-induced changes in current velocity also cause habitat alteration, with decreased flow ve-locities and high siltation rates upstream of dams (Kondolf, 1997; Bennett et al., 2002) and increased flow velocities leading to substrate scouring downstream (Kondolf, 1997; Camargo and Voelz, 1998). Many of these habitat alter-ations create conditions favorable for non-na-tive species (Marchetti and Moyle, 2001), which can then further alter fish assemblage compo-sition via predation and competition (Richter et al., 1997; Godinho and Ferreira, 1998; Eby et al., 2003).Although the number of free-flowing riverine.ecosystems in the world continues to decline (Poff et al., 1997), more research is required to understand the spatial pattern of fish assem-blage structure in impounded rivers. Compli-cating the situation is the fact that not all types of dams affect lotic ecosystems similarly.
For ex-ample, deep hypolimnial release dams, with their associated cooling effect downstream, af-fect rivers in different ways than smaller flood-control. dams that do not alter river temperature (Petts, 1984).Low-head dams (<4 m in height) are a type of impoundment common to many North American rivers. For example, Kansas has the second-highest number of dams in the United States (5,699; Shuman, 1995), and the vast ma-jority of these are low-head dams. The Neosho River in eastern Kansas alone is impounded by 15 such structures.
Given the well-documented effects of other types of impoundments on lotic systems, low-head dams, in spite of their small size, appear likely to affect riverine fish assem-blages. Indeed, Porto et al. (1999) showed that low-head dams with heights less than 1.5 m can alter fish assemblage composition, with species richness declining immediately upstream.
How-ever, Dodd et al. (2003) and Raborn and Schramm (2003) documented habitat alteration in the presence of low-head dams, but little 0 2005 by the Amnetican Society of Ichthyologists and Herpetologists 540 COPEIA, 2005, NO. 3 Fig. 1. Map of study area in Lyon County, KS, showing eight study sites and two low-head dams-along the Neosho River from Americus to Emporia.'overall change to the fish assemblage.
Although they are common throughout North America, little is know about, the spatiotemporal patterns of fish assemblages within river systems im-pounded by low-head dams. Our objective was to quantify patterns of spatiotemporal variation in fish assemblages in the Neosho River, Kansas, and to evaluate the extent to which low-head dam impoundments affected these patterns.MATERIAtS AND METHODS Study area.-The Neosho River lies within the Prairie Parkland ecosystem province (Bailey, 1983), and is part of the Arkansas River drain-age. It is fifth-order in our study reach, draining mostly mixed-grass prairie and cropland, with.mature riparian vegetation along some sections.We sampled eight sites along a 34-km stretch of the river from Americus to Emporia in Lyon County, Kansas (Fig. 1). Sites were selected to be representative of the Neosho River based on location relative to two low-head dams (see be-low) and appropriate for our sampling meth-odology (i.e., relatively shallow, with consistent current velocity).
Overall stream gradient is low (0.44 m/kmn), although variation in gradient be-tween sites varied due to impoundments.
Gra-dient was lowest between sites 5 and 6 (negli-gible) and highest between sites 6 and 7 (1.31 m/km). There are no permanent tributaries along this stretch of the Neosho River. At each site, we fixed five permanent transects perpen-dicular to shore, spaced equally every 5 to 10 m, depending on site length.Sampling.-We sampled each site monthly from November 2000 to October 2001. Samples were taken between the 9 1h and 2 2"d of each month during daylight hours, and sampling order of sites was randomized each month. We were un-able to adequately sample Site 2 in January and February and Sites 5 and 6 from December through February dtie. to ice cover, and Site 7 in August and December because flow velocities were too low for our sampling methodology.
Sampling at each site proceeded from down-stream to upstream transects, and from near shore to far shore points along each transect.We sampled up to five points along each tran-sect, depending on river width and depth and landowner permission.
At Site 8, landowner per-mission was only obtained for one .side of the river. However, we were able to obtain represen-tative samples of the fish assemblage at this site in spite of this limitation.
All sampling points along each transect were spaced at least 0.5 m apart to minimize disturbing adjacent points. At each point we sampled fishes by kick-seining, using a 1.5-m length by 1.8-m height section of 3-mm mesh seine. Upon fixing the seine at a sampling point, we disturbed substrate begin-ning 3 m upstream.
In this manner, fishes with-in a 4.5-M 2 area were carried downstream into the seine. This methodology effectively captures both water column and benthic fish species (Matthews, 1990; Wildhaber et al., 1999). We counted and identified fishes as juvenile or adult, using a 30-mm total length (TL) maxi-mum juvenile length for minnows (Campostoma, Phenacobius, Pimephales, Cyprinelia, Notropis, and Lythrurus spp.) and darters (Etheostoma and Per-cina spp.), and a 50-mm TL maximum juvenile length for madtoms (Noturus spp.) and sunfish-es (Lepomis spp.), following Gelwick (1990).Other fishes were measured individually and.classified as juvenile or adult based on pub-lished accounts.
We used a 305-mm maximum juvenile cutoff length for Channel Catfish (Ic-talurus punctatus), 380 mm for Flathead Catfish (Pylodictis olivaris), and 280 mm for redhorses (Moxostoma spp.) based on work by Deacon (1961) in the Neosho River, 220 mm for Spot-ted Bass (Micropterus punctulatus) and White Bass (Morone chrysops), and 240 mm for carp-stickers (Carlander, 1969, 1977, 1997). We did not distinguish between juvenile and adult Gain-busia affinis. All fishes were held until sampling of the site was completed, then returned to the river. Juvenile redhorses and carpsuckers were difficult to identify.
in the field due to small size and were recorded as Moxostoma sp. and Carpi-odes sp., respectively.
It11 GILLETTE ET AL.-RIVERINE FISH ASSEMBLAGES 541 Habitat measurement.-We measured water depth, current velocity at 60% depth, substrate composition, and substrate compaction at all points along each cross-stream transect.
Velocity was measured using a Global Flow Probe (Glob-al Water Company, Gold River, CA). We visually estimated substrate at each.point as percentage composition of clay/silt, sand, gravel, pebble, cobble, boulder, and bedrock (Mullner et al., 2000). Definition of substrate categories and sampling methodology followed Bain (1999).Compaction, a surrogate of the amount of fine sediment surrounding larger substrate types, was quantified by tactile evaluation; each point was assigned a compaction index value from I to 4, with I representing loose substrate, 2 sub-strate lightly packed with clay/silt, 3 substrate tightly packed with clay/silt, and 4 bedrock (Fu-selier and Edds, 1996).After fish were collected, water quality was measured immediately upstream of transects.
We measured water temperature with a labora-tory thermometer, and dissolved*
oxygen and pH with a Hach kit model AL-36B. We then took a I-L sample of surface water for further analysis.
From this sample we measured alkalin-ity and hardness with a Hach kit model AL-36B;nitrate, ammonia, carbon dioxide, total acidity, and orthophosphate with a Hach Surface Wa-ters kit; chloride and sulfate with a Hach kit model DREL/iC; and turbidity with a Hach 2100P turbidimeter.
Two 100-ml portions of the 1-L sample were vacuum filtered using Pall-Gel-man Type A/E round 47-mm glass fiber filters, and the filtrate frozen at -10 C for future de-termination of chlorophyll a and particulate or-ganic carbon (POC). We measured chlorophyll a using a model 10-AU-005 Field Flourometer (Turner Designs, Sunnyvale, CA) and POC us-ing a Coulometrics Carbon Model 5014 Analyz-er (UIC, Inc.,Joliet, IL).Data analysis.--Analyses were performed using SASv.8 and SPSS v.7.5.1. Ordinations were con-ducted using PC-Ord v.4. We included only those taxa occurring in >5% of collections for ordination (Gauch, 1982).To examine spatiotemporal patterns of fish assemblage structure, we used correspondence analysis (CA) to ordinate collections from each site during each month. We included 84 collec-tions in the CA, composed of 17 juvenile and 16 adult taxa in addition to G. affinis (Table 1).We excluded samples taken in December at Site 8 andJanuary at Site 7 because these collections consisted of only one fish. We analyzed conspe-cific juveniles and adults as separate taxa be-cause spatial and trophic resource use varies on-togenetically for many stream fishes (Schlosser, 1982; Gelwick, 1990; Gido and Propst, 1999). To test for effects of low-head dam impoundments on fish assemblages, we grouped sites into one of three levels of an "impoundment treat-ment," based on proximity to these dams. Sites less than 2 km downstream from dams (Sites 3, 4, and 7) comprised a "downstream" level of treatment, sites less than 5 km upstream (Sites 2, 5, and 6) an "upstream" level, and sites great-er than 5 km from dams (Sites 1 and 8) a "dis-tant" level of treatment.
To test for impoundment and temporal ef-fects on CA 1 and 2 scores, we used the mixed linear model (SAS Proc Mixed). Scores from both axes were modeled separately as depen-dent variables, with month, impoundment, and the interaction between the two as fixed effects (Agresti and Finlay, 1997). Sites were modeled as repeated subjects nested within levels of im-poundment treatment.
The mixed linear model is a generalization of the general linear model that allows data to exhibit correlation and non-constant variability; fixed effect parameters are associated with known explanatory variables, as in the general linear model. Where appropri-ate, we used a Tukey-Kramer multiple compar-ison test on least-square means to distinguish significant differences among treatment levels.To examine relationships between environ-mental gradients and CA axes, we calculated Pearson's correlation coefficient between axis scores and environmental variables for each col-lection. To assess potential effects of hydrologi-cal variation on assemblage structure, we in-cluded in the correlation matrix river discharge for the day of each collection, measured at the U.S. Geological Survey gauging station on the Neosho River at Americus, KS.Because of the large number of environmen-tal variables measured, we used Principal Com-ponents Analysis (PCA) on a collections-by-en-vironmental variables matrix to eliminate re-dundant variables.
In the case of suites of vari-ables loading similarly on the first three PCA axes, we selected the one variable most biolog-ically meaningful to represent the group. In this analysis, water depth was selected from a group of variables including percent substrate com-position of clay/silt, gravel and pebble, POC and ammonia. Dissolved oxygen was selected from a suite of water chemistry variables includ-ing sulfate, alkalinity, dissolved carbon dioxide, hardness, pH, and nitrate. Current velocity was selected from a group including substrate em-beddedness and water turbidity, and percent cobble substrate composition from a group in-cluding percent boulder substrate composition.
54.2 COPEIA, 2005, NO. 3 TABLE 1. SPECIES COLLECTED FROM EIGHT SITES ON THE NEOSHO RIVER, LYON Co., KS, NOVEMBER 2000 TO OCTOBER 2001, SHOWING PERCENT COMPOSITION BY SPECIES, INCLUSION OF ADULT (A) AND JUVENILE (J) TAXA IN CA ORDINATION, AND CORRELATIONS (PEARSON'S R) OF TAXON ABUNDANCE WITH CA 1 AND 2. Criterion for inclusion was occurrence in at least 5% of collections.
(*Juvenile and adults were not distinguished for Cam-busia affinis).% Assemblage Taxa included Species composition in CA -CA I r CA 2 r Cyprinella lutrensis 45.04 A -0.15 -0.51* Notropis buchanani Pimephales notatus Lepomis humilis Percina phoxocephala imephales ifiiax Etheostoma spectabile Pimephales teneUus Phenacobius mirabilis Campostoma anomalum Notropis stramineus Ictalurus punctatus Percina caprodes Noturus placidus Gambusia affinis Percina copelandi Lepomis cyanetlus Lepomis macrochirus Noturus flavus Cyprinella camura Aplodinotus grunniens Moxostoma sp.Lepomis megalotis Fundulus notatus Pylodictis olivaris Lythrurus umbratilis Micropterus punctulatus Morone chrysops Dorosoma cepedianum Etheostoma flabeUare Carpiodes sp.11.13 10.88 7.67 7.56 4.94 3.04: 2.78 1.29 1.20 0.85 0.75 0.56 0.44 0.42 0.42 0.36 0.22 0.16 0.06 0.06 0.04 0.03 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.41-0.64-0.07-0.28.-0.01-0.22 0.01-0.26-0.30 0.04 0.25 0.40-0.17.-0.34-0.06-0.11-0.15-0.37-0.09 0.05-0.24-0.30-ý0.03 0.01 0.23-0.28-0.14 0.09 0.25 0.08-0.06.-0.19-0.14-0.10 0.19-0.02 0.05 0.36-0.17 0.13-0.45 0.06 0.02 0.10-0.15-0.12 0.03 0.04-0.53-0.24 0.06<0.01 0.03-0.34 0.02 0.03 0.13 0.09-0.02-0.15 0.17 0.24-0.16-0.10-0.05-0.12 4 This resulted in the following 10 variables re-tained for correlation analysis with CA axes: per-cent substrate composition of sand, cobble, and bedrock, water depth and current velocity, water temperature, dissolved oxygen, chloride, chlo-rophyll a and river discharge.
RESUI.TS We collected 15,215 fishes of 31 species, ac-counting for 44 taxa inclusive of juveniles and adults (Table 1). Ten families from five orders were represented; Cyprinidae had the greatest GILLETTE ET AL.-RIVERINE FISH ASSEMBLAGES 543 TABLE 2. MEAN AND STANDARD DEVIATION OF ENVIRONMENTAL VARIABLES FOR EIGHTY-FouR COLLECTIONS MADE OVER TWELVE MONTHS FROM EIGHT SITES. Eight collections were omitted from upstream sites (two from De-cember and three from both January and February), three from downstream sites (one in December, one in January, and one in August), and one from distant sites (December).
Upstream Downstream Distant Mean S. D. Mean S. D. Mean S. D.Substrate composition
(%)Bedrock 0.0 0.0 27.7 39.5 0.0 0.0 Boulder 1.6 3.4 0.2 0.5 0.0 0.0 Cobble 4.7 7.0 2.1 2.9 0.3 1.0 Pebble 33.4 8.7 26.1 17.1 36.4 7.1 Gravel 38.1 8.1 30.7 18.8 41.8 6.2 Sand 3.4 1.4 3.9 3.9 6.0 2.9 Clay/Silt 18.9 16.5 8.3 8.7 15.7 7.9 Other microhabitat variables Substrate compaction 2.2 0.3 2.4 1.0 1.8 0.3 Water depth (cm) 58.2 8.7 26.7 13.5 38.1 17.3 Flow velocity (m/s) 0.1 0.1 0.4 0.2 0.4 0.2 Water properties Dissolved oxygen (mg/L) 8.3 2.2 10.0 2.4 8.5 2.9 pH 8.0 0.1 8.0 0.2 8.0 0.2 Alkalinity (mg/L) 171.2 54.8 183.2 60.0 1.71.2 49.6 Hardness (mg/L) 232.8 42.8 239.7 51.4 237.9 47.9 Turbidity (NTU) 33.4 27.6 38.9 44.9 43.4 40.7 Dissolved carbon dioxide (mg/L) 10.2 3.5 10.4 4.0 10.7 4.2 Ammonia (mg/L) 0.0 0.0 0.0 0.0 0.0 0.0 Nitrate (mg/L) 0.0 0.0 0.0 0.0 0.0 0.0 Chloride (mg/L) 9.1 4.6 9.2 5.5 8.7 3.5 Sulfate (mg/L) 28.1 9.7 27.5 7.1 28.4 8.2 Particulate organic carbon 1555.8 931.7 1556.8 925.9 1793.7 1027.8 Chlorophyll a (tig/L) 704.0 604.7 510.7 620.8 506.5 487.1 number of species (10), followed by Percidae (5), Centrarchidae (5), and Ictaluridae (4).Habitat varied among upstream, downstream, and distant sites (Table 2). Upstream sites were deeper and slower-flowing than downstream and distant sites. Downstream sites were shallow-est, had the highest percentage substrate com-position of bedrock, and had the lowest per-centage composition of clay/silt.
Relative abundance of fishes varied with im-poundment treatment and season (Table 3).Cyprinelia lutrensis was the most abundant taxa at downstream and distant sites, and R notatus or N. buchanani most abundant at upstream sites, depending on season. Percina phoxocephala was more abundant at downstream and distant sites than at upstream sites, and Lepomis humilis more abundant at upstream sites than at down-stream or distant sites.Axis 1 of the CA showed a temporal pattern of fish assemblage structure, covering a gradient of 5.6 standard deviations with an eigenvalue of 0.399, and explaining 16.8% of the variance.Month significantly affected Axis 1 scores (Ta-ble 4), with winter collections scoring highest and early summer collections lowest (Figs. 2, 3).There were no significant impoundment or in-teraction effects (Table 4). Axis 1 was positively correlated with dissolved oxygen and negatively correlated with water temperature, chlorophyll a, water depth and current velocity, percent sub-strate composition of sand, and river discharge (Table 5). Axis. 1 thus represented a pattern of fish assemblage structure along a gradient from cold, shallow, and slow-flowing winter condi-tions to warm, deeper, and swifter-flowing sum-mer conditions.
Taxa characteristic of winter collections were positively correlated with Axis 1, and taxa char-acterizing summer collections were negatively correlated (Tables 1, 3). Strong positive corre-lates included G. affinis, which was only collect-ed in November and December, adult E. specta-bile, and juvenile C. lutrensis, R vigilax, and L.macrochirus (Table 1). Strong negative correlates of CA I included juvenile P caprodes, C. anom-544 COPEIA, 2005, NO. 3 TABLE 3. RELATIVE ABUNDANCE OF TAXA COLLECTED FROM UPSTREAM, DOWNSTREAM, AND DISTANT SITES DURING WINTER AND SPRING (A) AND SUMMER AND FALL (B). Abundance of some species differed by less than 0.1%;others were equal in abundance. (A = adult, J = juvenile).
Upstream Downstream Distant Rank Taxon % Rank Taxon % Rank Taxon %A. Winter-Spring 1 P notatusA 21.5 1 C. lutrensisA 43.9 1 C. lutrensisA 65.8 2 N. buchanani A 20.6 2 P. phoxocephala A 11.7 2 P notatua A 5.2 3 3 L. humitisJ 14.7 3 S N. buchananiA 10.7 3 P phoxocephalaA 4.8 4 P tene/us A 7.4 4 P notatus A 7.4 4 C lutrensisJ 4.7 5 C. lutrensis A 6.8 5 C. lutrensisJ 6.0 5 R teneus A 3.8 6 P vigilax A
- 5.7 6 L humilisJ 4.9 6 L. humilisJ 3.4 7 P notatusJ 5.6 7 P vigilaxA 4.0 7 P vWiaxA 3.0 8 P phoxocephalaA 4.9 8. E. spectabikeA 3.5 8 N. buchananiA 2.5 9 C. lutrensisJ 3.7 9 P mirabilisA 1.9 9 E. specabi/eA
.2.0 10 E. spectabileA 3.3 10. N. stramineusA 1.2 10 P copelandiA 0.9 B. Summer-Fall 1 N. buchananiA 18.6 1 C. lutrensisA 38.2 1 C. lutrensisA 20.5 2 C. lutrensisA 17.9 2 C. lutrenssisj 23.3 2 N. buchananiA 16.3 3 L. humilisJ 14.6 3 R phoxocephala A 7.3 3 C. lutrensisJ 15.4 4 C. lutrensisJ 13.6 4 P notatusA 5.4 4 P notatusA 8.9 5 P notatusA 8.9 5 E. spectabi eA 4.3 5 R phoxocephalaA 7.4 6 P notatusJ 4.5 6 N. buchananiA 4.1 6 P vigilaxA 4.5 7 P tenellus A 3.8 7 R vigilax A 3.0 7 P notatusJ 3.6 7 P phoxocephala A 3.8 8 L humilisJ 2.0 8 L. humilisJ 3.1 9 P vigilaxA 3.3 9 P mirabilisA 2.0 9 E. spedabi/eA 3.0 10 E. spectabileA 1.2 10 P vigilaxJ 1.9 10 PRmirabilisA 2.9 alum, and P phoxocephala these age-0 fishes were present only from June through early Fall. Low-scoring adult taxa included N. buchanani and P tenellur,, these two taxa were collected most fre-quently from April through August, with N.buchanani almost completely absent in other months.Axis 2 of the CA showed a spatial pattern of fish assemblage structure related to low-head dams, with a slight temporal component, cov-ering a gradient of 3.5 standard deviations with an eigenvalue of 0.315, and explaining 13.3%of the variance.
Impoundment, month, and im-poundment-by-month interaction significantly affected Axis 2 scores (Table 4). Upstream sites scored higher than downstream sites (Fig. 2).Relative position of distant site scores varied temporally, grouping with downstream sites in winter and spring, and upstream sites in sum-mer and fall (Fig. 2). Axis 2 was negatively cor-related with current velocity and dissolved oxy-gen and positively correlated with water depth and chloride (Table 5). Axis 2 thus represented a pattern of fish assemblage structure along a gradient from lotic habitat downstream from dams to lentic habitat upstream from dams.Correlations of fish taxa with Axis 2 differed between riffle species, predominant at down-stream sites, and pool species, prevalent at up-stream sites (Tables 1, 3). With the exception of TABLE 4. RESULTS OF Two-WAY REPEATED-MEASURES ANALYSIS OF CA I AND CA 2 SCORES FOR COLLECTIONS FROM THE NEOSHO RIVER, KS, 2000-2001, WITH MONTH AND IMPOUNDMENT AS TREATMENT.
Axis Effect Numerator d.f. Denominator d.f. F P CA I Impoundment 2 5 0.66 0.558 Month 11 45 31.34 <0.0001 Interaction 20 45 1.21 0.291 CA 2 Impoundment 2 5 24.51 0.003 Month 11 45 3.78 0.001 Interaction 20 45 2.59 0.004 GILLETTE ET AL.-RIVERINE FISH ASSEMBLAGES 545 1.0 0.5 S0.0-0.5-1.0 C4 2.0 1.5 1.0 0.5 0.0-0.5-1.0-1.5 a 1.0 -0.5 0.0 0.5 1.0 CA 1 (S.D.)1.5 1.0-0.5 o.o-0.0-0.5*/ 0 ov / o'0 ,Winter*/o 0 Upstream/ 0 Downstream Spring / v v Distant-2.0 .....-Nov Jan Mar May Jul Sep Fig. 3. Plots of monthly mean and standard devi-ation collection CA 1 scores. Months not differing sig-nificantly, as determined by Tukey-Kramer tests on CA 1 scores, share lowercase letters. Means for which sam-pie sizes were too small to calculate least square means used in Tukey-Kramer tests are denoted by an asterisk.N. buchanani adults, R vigilax juveniles, and G.affinis, loaded strongly on both Axis 1 and Axis 2, and thus were important in defining both temporal and impoundment gradients.
The fact that these taxa were strongly associated with len-tic habitat upstream from dams and also varied temporally in abundance likely contributed to the significant month effect on Axis 2 scores.DiscussioN This study indicated that low-head dams can influence structure of small-bodied fish assem--1.0-1.0 -0.5 0.0 0.5 1.0 1.5 CA 1 (S.D.)Fig. 2. Plot of CA I vs. CA 2 scores by season for collections from the Neosho River, KS, 2000-01, grouped by impoundment treatment.
Seasons were defined monthly as Winter (December-February), Spring (March-May), Summer (June-August), and Fall (September-November).
C. lutrensis, most strong negative correlates of Axis 2 were benthic riffle fishes, such as P phox-ocephaka, P mirabilis, C. anomalum, and P capro-des. Strong positive correlates included midwa-ter species most abundant in slow waters of up-stream sites, such as R notatusjuveniles, Lepomis spp.,. and N. buchanani adults (Table 1).Differences in seasonal abundance patterns among species led to the significant temporal and interaction effects on Axis 2 scores (Table 3). Several positively-correlated taxa (N. buch-anani adults, P notatusjuveniles, and Lepomis hu-milis juveniles) were more abundant, and neg-atively-correlated taxa (C. lutrensis adults, R phoxocephala adults, and P mirabilis adults) less abundant, at upstream sites than at downstream sites in all seasons (Table 3). Abundance at dis-tant sites, however, varied seasonally (Fig. 2). As a consequence of these temporal changes in fish species composition, Axis 2 scores for dis-tant sites were relatively higher in summer and fall than in winter and spring, leading to the significant interaction.
Several taxa, including TABLE 5. PEARSON'S CORRELATION COEFFICIENT OF SELECTED ENVIRONMENTAL VARIABLES WITH CA 1 AND CA 2 COLLECTION ScoREs. Correlations significant at a = 0.05 are denoted by an asterisk, and those sig-nificant at a = 0.01 by two asterisks.
CAI CA2 Percent substrate composition.
Sand -0.333** 0.006 Cobble -0.030 -0.041 Bedrock 0.197 -0.209 Other microhabitat variables Water depth -0.331"* 0.344**Water flow velocity -0.213* -0.558**Water chemistry variables Dissolved oxygen 0.540** -0.252*Chloride 0.021 0.249*Other variables Chlorophyll a -0.223* 0.182 Water temperature
-0.638** 0.086 River kilometer
-0.182 -0.214 River discharge
-0.471** -0.172 546 COPEIA, 2005, NO. 3 blages in shallow waters of rivers via habitat al-teration.
Sites upstream from dams were deep-est with slow-flow velocities and high siltation levels, with fish assemblages characterized by a high abundance of lentic habitat fishes. Sites downstream from low-head dams were shallow-est, with scoured substrata including bedrock, and low levels of silt accumulation.
Fish assem-blages at these sites showed a higher abundance of riffle species commonly found in shallow, high-current velocity habitats.
Fish assemblages intermediate to these two extremes occurred at sites distant from low-head dams. This pattern of upstream and downstream habitat alteration is similar to that shown for larger dams (e.g., Kondolf, 1997; Camargo and Voelz, 1998; Ben-nett et al., 2002), but ours is one of the first studies to document these patterns of habitat alteration and associated fish assemblage differ-ences in a river impounded by low-head dams.Our results differed slightly from those of previous investigators studying fish assemblages in systems with low-head dams. Raborn and Schramm (2003) and Dodd et al. (2003)showed differences in habitat, but not fish as-semblages, between dammed and free-flowing streams and stream segments.
The discrepancy between our results and theirs is likely due to spatial scale; we compared small sites within a river, as opposed to stream reaches or entire streams. Our sites were shorter than those of the above investigators and spaced adjacent to multiple impoundments, allowing detection of these smaller-scale alterations of the fish assem-blage. In rivers impounded by large dams, fish assemblages can be influenced by impound-ments for many kilometers downstream (Kin-solving and Bain, 1993). Effects of low-head dams, however, appear to be more localized, re-stricted to habitat alteration immediately up-stream and downstream.
In addition, low-head dams such as those in our study do not severely alter river temperature and discharge as hydro-electric dams do (Kinsolving and Bain, 1993;Clarkson and Childs, 2000). This may explain the high abundance of benthic fishes we ob-served downstream from dams, as compared to the low abundance of these fishes shown by Travnichek and Maceina (1994) downstream of a hydroelectric dam. Likewise, we did not find the pattern of decreased upstream species rich-ness shown by Porto et al. (1999); rather, fish assemblage structure followed repeated gradi-ents of lentic habitat upstream from dams to lotic habitat downstream.
This result is not sur-prising, given that the species we collected in the Neosho River are not migratory, thus elim-inating the need to cross these barriers for spe-cies to persist. In addition, the Neosho River is a much larger system than the Great Lakes trib-utaries studied by Porto et al. (1999), perhaps limiting downstream transport of fishes and providing sufficient habitat to maintain fish populations upstream.
Despite these differences with previous studies, our results do show that low-head dams can produce noticeable changes in the spatial pattern of lotic fish assemblages.
Study sites also exhibited a great deal of tem-poral variation in assemblage structure, as shown by CA Axis 1 scores. Because CA calcu-lates axes of decreasing ecological significance (Gauch, 1982), it may be inferred that assem-blage patterns associated with Axis 1 were stron-ger than those associated with Axis 2. Complete faunal turnover typically occurs across an axis length of 4 standard deviations (Gauch, 1982);thus, Axis l's length of 5.6 standard deviations represented a strong temporal pattern. The temporal nature of Axis 1 is confirmed by strong correlations of axis scores with environ-mental variables that vary seasonally, such as wa-ter temperature (Table 5). A high degree of overlap among multiple comparison groupings indicates that seasonal fish assemblages were not mutually exclusive, but rather components of a gradual assemblage shift over the study year. These results are consistent with Gelwick's (1990) conclusion that lotic fish assemblages in shallow water show a great deal of temporal var-iation. Separation of species into juvenile and adult taxa could inherently bias our study to-wards temporal variation because of natural processes such as recruitment.
However, a par-allel analysis on species only also showed assem-blage variation to be greater temporally than spatially (D. P. Gillette, 2002, unpubl. data).This supports the conclusion that, at least at the spatial scale of the present study, temporal pat-terns of shallow-water fish assemblages in the Neosho River, as measured by position in mul-tivariate space, are stronger than spatial pat-terns.Temporal assemblage variation came from two sources: fish life history processes and as-semblage responses to changing abiotic condi-tions. As an example of the former, N. buchan-ani was absent from our study sites until early summer, when it occurred in great numbers.This was likely a spawning migration from near-by pools; Pflieger (1997) stated that this pool species spawns over riffles from late April through August, dates corroborated in Kansas by Cross and Collins (1995). Reproduction of E.spectabile, P phoxocephala, P captnodes, P mirabilis, and C. anomalum also changed assemblage com-position through an influx of juveniles persist-I GILLETTE ET AL.-RIVERINE FISH ASSEMBLAGES 547 ing from June through September.
In addition to these patterns, many species also declined greatly in abundance or were absent during win-ter. This pattern appears to be unrelated to life history events, because all of these species spawn from late spring through summer in Kan-sas (Cross and Collins, 1995) and were present at our study sites during both early spring and late fall. Rather, this pattern is likely due to a sharp drop in water temperature from Novem-ber to December that, coupled with shallower river depths in winter, caused these species to vacate gravel bars and retreat to nearby pools.With the exception of N. flavus, all of these spe-cies have been shown to inhabit pools at various times. Noturusflavus spawns in pools with mod-erate current in Kansas (Cross and Collins, 1995), so it may also be able to use pool habitat when water temperatures on shallow gravel bars become too cold. After water temperatures rose sharply from March to April, these species re-turned to gravel bars. The fact that adults of many species survived the winter to spawn, but were not collected on gravel bars during winter, suggests that deeper water may play a major role in providing winter refugia for species that fre-quent gravel bars in warmer months. This con-clusion supports recent conceptual models of stream fish ecology emphasizing.
the spatial ar-rangement of habitat patches used by fishes un-der varying abiotic conditions and during dif-ferent life history stages (Schlosser, 1991, 1995;Fausch et al., 2002).Given the high degree of habitat variability, variable sampling efficiency among our study sites cannot be ruled out. Although there are no published accounts of sampling efficiency for the kick-set methodology we employed, Pe-terson et al. (2004) showed that estimation of salmonid abundance by multipass electrofishing varied with stream area and substrate composi-tion. As mentioned in Materials and Methods, we eliminated two collections because flow ve-locity was insufficient to allow effective sam-pling. However, water depth, flow velocity, and substrate composition varied among sites and months during our study, possibly resulting in variable sampling efficiency.
Few studies have examined spatial patterns of fish assemblage structure on scales large enough to assess assemblage response to mul-tiple impoundments.
Reyjol et al. (2001)showed that flow alteration by hydroelectric im-poundments along a salmoniform-cypriniform transitional gradient caused an oscillation in dominant taxa corresponding to alterations in current velocity.
With few free-flowing river sys-tems remaining in the world, other situations similar to that in the Neosho River likely exist where multiple impoundments affect the spatial pattern of riverine fish assemblages via localized habitat alteration.
Effective conservation of these lotic systems and their biota requires knowledge of the spatiotemporal structure of fish assemblages in response to~such alterations.
ACKNOWLEDGMENTS We thank B. Chance, J. Dean, L. Freeman, B.Harkins,J.
Howard, S. Sherraden, and I. Singh for assistance in the field. River access was gen-erously provided by Mr. and Mrs. W. Leffler, Mr.and Mrs. P. Matile, Mrs. L. Schlessener, Mr. G.Guide, the City of Emporia,and Emporia State University (ESU) Natural Areas. ArcView soft-ware assistance was provided by R. Sleezer.Throughout the course of this study, .D. Zelmer, L. Scott, and D. Moore provided valuable com-ments, and J. Mendoza assisted with statistical analysis.
Laboratory assistance was provided by S. Olson, B. Lakish, J. Albers, J. Fairchild, C.Witte, and A. Allert; S. Gillette helped with data entry and management.
We thank W. Matthews for critically reviewing an earlier version of the manuscript.
Funding for this study was provided by a Faculty Research and Creativity Grant and a Graduate Student Research Grant from ESU, and by the U.S. Geological Survey, Department of the Interior, under USGS Cooperative Agree-ment No. OOCRAGO025.
All fishes were collect-ed under Kansas Department of Wildlife and Parks Scientific Collector's Permits SC-065-2000 (2000) and SC-033-2001 (2001), issued to DRE.This study was conducted in accordance with ESU Animal Care and Use Committee guide-lines.LITERATURE CITED AGRESTI, A. AND B. FINLAY. 1997. Statistical Methods for the Social Sciences.
Prentice Hall, Upper Sad-dle River, New Jersey.BAILEY, R. G. 1983. Delineation of ecosystem regions.* Environ. Manage. 7:365-373.
BAIN, M. B. 1999. Substrate, p.95-100. In: Aquatic Habitat Assessment:
Common Methods. M. B. Bain and N.J. Stevenson (eds.). American Fisheries So-ciety, Bethesda, Maryland.BAL.ON, E. K., AND D. J. STEwART. 1983. Fish assem-blages in a river with unusual gradient (Luongo, Africa-Zaire system), reflections on river zonation, and description of another new species. Environ.Biol. Fish 9:225-252.
BAXTER, R. M. 1977. Environmental effects of dams and impoundments.
Annu. Rev. Ecol. Syst. 8:255-283.BENNEi-r, S.J., C. M. COOPER, J. C. RITC.HIE,J.
A. DUN-BAR, P. M. AI.I.EN, L. W. CAIDWELL, AND T. M.
548 COPEIA, 2005, NO. 3 MCGEE. 2002. Assessing sedimentation issues within aging flood control reservoirs in Oklahoma.
J. Am.Water Resour. As. 38:1307-1322.
BONNER, T. H., AND G. R. WILDE. 2000. Changes in the Canadian River fish assemblage associated with reservoir construction.
J. Freshwater Ecol. 15:189-198.CAMARGO, J. A., AND N.J. VOELZ. 1998. Biotic and abi-otic changes along the recovery gradient of two im-pounded rivers with different impoundment use.Environ. Monit. Assess. 50:143-158.
CARLANDER, K. D. 1969. Handbook of Freshwater Fish-ery Biology. Volume 1. Iowa State University Press, Ames, Iowa..1977. Handbook of Freshwater Fishery Biol-ogy. Volume 2. Iowa State University Press, Ames, Iowa..1997. Handbook of Freshwater Fishery Biol-ogy. Volume 3. Iowa State University Pressi Ames, Iowa.CLARtSON, R. W., AND M. R. CHILDS. 2000. Tempera-ture effects of hypolimnial-release dams on early life stages of Colorado River Basin big-river fishes.Copeia 2000:402-412.
CROSS, F. B., ANDJ. T. COLLINS. 1995. Fishes in Kansas.2 nd Edition, Revised. University Press of Kansas, Lawrence, Kansas.DEACON, J. E. 1961. Fish populations, following a drought, in the Neosho and Marais de Cygnes riv-ers of Kansas. Univ. KS. Pub., Mus. Nat. Hist. 13: 359-427.DODD, H. R., D. B. HAYES, J. R. BAYLISS, L. M. CARL, J. D. GOLDSTEIN, R. L. MCLAUGHLIN, D. L. G. NOAKES, L. M. PORTO, AND M. L. JONES. 2003. Low-head sea lamprey barrier effects on stream habitat and fish communities in the Great Lakes basin. J. Great Lakes Res. 29:386-402 Suppl. 1.DYNESIus, M., AND C. NILSSON. 1994. Fragmentation and flow regulation of river systems in the northern third of the world. Science. 266:753-762.
EBY, L. A., W. F. FAGAN, AND W. L. MINCKLEY.
2003.Variability and dynamics of a desert stream com-munity. Ecol. Appi. 13:1566-1579.
EDDS, D. R. 1993. Fish assemblage structure and en-vironmental correlates in Nepal's Gandaki River.Copeia 1993:48-60.
FAusCH, K. D., C. E. TORGERSON, C. V. BAXTER, AND H. W. Li. 2002. Landscapes to riverscapes:
bridging the gap between research and conservation of stream fishes. Bioscience 52:483-498.
FUSELIER, L., AND D. R. EDDS. 1996. Seasonal variation of riffle and pool fish assemblages in a short miti-gated stream reach. Southwest.
Nat. 41:299-306.
GAUCH, H. G., JR. 1982. Multivariate analysis in com-munity ecology. Cambridge University Press, Cam-bridge.GELW1CK, F. P. 1990. Longitudinal and temporal com-parison of riffle and pool fish assemblages in a northeastern Oklahoma Ozark stream. Copeia 1990:1072-1082.
GIDo, K. B., AND D. L. PROPST. 1999. Habitat use and association of native and nonnative fishes in the San Juan River, New Mexico and Utah. Ibid. 1999: 321-332.GODINHo, F. N., AND M. T. FERREIRA.
1998. The rela-tive influences of exotic species and environmental factors on an Iberian native fish community.
Envi-ron.. Biol. Fish. 51:41-51.J. G., J. M. SCHMIDT-GENGENBACH, AND B.B. YOSHIOKA.
1998. High dams and marine-fresh-water linkages:
effects on native and introduced fauna in the Caribbean.
Conserv. Biol. 12:621-630.
KINSOLVING, A. D., AND M. B. BAIN. 1993. Fish assem-blage recovery along a riverine disturbance gradi-ent. Ecol. AppI. 3:531-544.
KONDOLF, G. M. 1997. Hungry water: effects of dams and gravel mining on river channels.
Environ. Man-age. 21:533-551.
MARCH, J. G.,J. P. BENSTEAD, C. M. PRINGLE, AND F. N.SCATENA. 2003. Damming tropical island streams: problems, solutions, and alternatives.
Bioscience 53:1069-1078.
MARCHETTI, M. P., AND P. B. MOYLE. 2001. Effects of flow regime on fish assemblages in a regulated Cal-ifornia stream. Ecol. Appl. 11:530-539.
MARET, T. R., C. T. ROBINSON, AND G. W. MINSHALL.1997. Fish assemblages and environmental corre-lates in least-disturbed streams of the upper Snake River basin. T. Am Fish. Soc. 126:200-216.
MATrHEWS, W. J. 1990. Spatial and temporal variation in fishes of riffle habitats:
a comparison of analyti-cal approaches for the Roanoke River. Am. Midi.Nat. 124:31-45.
MINCKLEY, W. L, P. C. MARSH, J. E. DEACON, T. E.DOWUJNG, P. W. HENDRICK, W. J. MATrHEWS, AND G.MUELLER. 2003. A conservation plan for native fish-es of the lower Colorado River. Bioscience 53:219-234.MULLNER, S. A., W. A. HUBERT, AND T. A. WESCHE.2000. Visually estimating substrate composition at potential spawning sites for trout in mountain streams. J. Freshwater Ecol. 15:199-207.
PETERSON, J. T., R. F. THUROW, AND J. W. GUZEVITCH.
2004. An evaluation of multipass electrofishing for estimating the abundance of stream-dwelling sal-monids. T. Am. Fish. Soc. 133:462-475.
PErS, G. E. 1984. Impounded Rivers. John Wiley, Chichester, U.K.PFLIEGER, W. L. 1997. The Fishes of Missouri.
Revised Edition. Missouri Department of Conservation,Jef-ferson City, Missouri.PoFF, N. L., J. D. ALLAN, M. B. BAIN, J. R. KARR, K. L.PREsTEGAARD, B. D. RICHTER, R. E. SPARKS, ANDJ. C.STROMBERG.
1997. The natural flow regime: a par-adigm for river conservation and restoration.
Bio-.science 47:769-784.
PORTO, L. M., R. L. MCLAUGHLIN, AND D. L. G. NOAKES.1999. Low-head barrier dams restrict the move-ments of fishes in two Lake Ontario streams. N.Am.J. Fish. Manage. 19:1028-1036.
QUINN, J. W., AND T. J. KwAK. 2003. Fish assemblage changes in an Ozark river after impoundment:
a long-term perspective.
T. Am. Fish. Soc. 132:110-119.RABORN, S. W., AND H. L. SCHRAMM. 2003. Fish assem-blage response to recent mitigation of a channel-ized warmwater stream. River Res. Appl. 19:289-301.1 1 GILLElTE ET AL.-RIVERINE FISH ASSEMBLAGES 549.RyEWsGAviLAN, F. G., R. GARmDO, A. G. NlicENzA, M.M. TOLEDo, AND F. BRANA. 1996. Fish community variation along physical gradients in short streams of northern Spain and the disruptive effect of dams. Hydrobiologia 321:155-163.
REPOL, Y., P. LM, F. DAutA, P. BARAN, AND A. BELAUD.2001. Role of temperature and flow regulation on the Salmoniform-Cypriniform transition.
Arch. Hy-drobiol. 152:567-582.
RICrTEER, B. D., D. P. BRAUN, M. A. MENDELSON, AND L. L MASTER. 1997. Threats to imperiled freshwater fauna. Conserv. Biol. 11:1081-1093.
SCHLOSSER, I. J. 1982. Fish community structure and function along two habitat gradients in a headwater stream. Ecol. Monogr. 52:395-414.
- 1991. Stream fish ecology: a landscape per-spective.
Bioscience 41:704-712.
-1995. Critical landscape processes that influ-ence fish population dynamics in headwater streams. Hydrobiologia 303:71-81.
SHUMAN; J. R. 1995. Environmental considerations for assessing dam removal alternatives for river resto-ration. Regul. River. 11:249-261; TRAVNICHEK, V. H., AND M. J. MACEINA. 1994. Com-parison of flow regulation effects on fish assem-blages in shallow and deep-water habitats in the Tal-lapoosa River, Alabama. 1. Freshwater Ecol. 9:207-216.WILDHABER, M. L., A. L. ALLERT, AND C. J. SCHMrrr.1999. Potential effects of interspecific competition on Neosho madtom (Noturus placidus) populations.
Ibid. 14:19-30.(DPG, JST, DRE) DEPARTMENT OF BIOLOGICAL SCIENCES, EMPORIA STATE UNIVERSITY, EMPO-RiA, KANSAS, 66801; (MLW) COLUMBIA ENVt-RONMENTAL RESEARCH CENTER, COLUMBIA, MIssouRI 65201. PRESENT ADDRESSES: (DPG)DEPARTMENT OF ZOOLOGY, UNIVERSITY OF OKLAHOMA, NORMAN, OKLAHOMA 73019;(.ST) CENTER FOR BIODIVERSITY, ILLINOIS NATURAL HISTORY SURVEY, CHAMPAIGN, ILIJ-NOIS 61820. E-mail: (DPG) dgillette@ou.edu.
Send reprint requests to DPG. Submitted:
30 April 2004. Accepted:
6 April 2005. Section editor: C. M. Taylor.I$
Copyrighted Material Protect accordingly h(Transactions of the American Fisheries Sociery 133:705-717, 2004 0 Copyright by the American Fisheries Society 2004 Effects of Lowhead Dams on Riffle-Dwelling Fishes and Macroinvertebrates in a Midwestern River JEREMY S. TIEMANN*I AND DAVID P. GILLETTE 2 Emporia State University, Department of Biological Sciences.Emporia. Kansas 66801. USA MARK L. WILDHABER U.S. Geological Survey, Columbia Environmental Research Center, Columbia, Missouri 65201, USA DAVID R. EDDS Emporia State University, Department of Biological Sciences, Emporia, Kansas 66801, USA Abstract.-Many studies have assessed the effects of large dams on fishes and macroinvertebrates, but few have examined the effects of lowhead dams. We sampled fishes, macroinvertebrates,habitat, and physicochemistry monthly from November 2000 to October 2001 at eight gravel bar sites centered around two lowhead dams on the Neosho River. Kansas. Sites included a reference site and a treatment site both upstream and downstream from each dam. Multivariate analysis of variance indicated that habitat, but not physicochemistry, varied immediately upstream and down-stream from the dams, with resultant effects on macroinvertebrate and fish assemblages.
Compared with reference sites, upstream treatment sites were deeper and had lower velocities and downstream treatment sites were shallower and had higher velocities; both upstream and downstream treatment sites had greater substrate compaction than reference sites. Macroinvertebrate richness did not differ among site types, but abundance was lowest at downstream treatment sites and evenness was lowest at upstream treatment sites. Fish species richness did not differ among site types, but abundance was highest at downstream reference sites and evenness was highest at upstream sites.The abundance of some benthic fishes was influenced by the dams, including that of the Neosho madtom Noturus placidus, which was lowest immediately upstream and downstream from dams, and those of the suckermouth minnow Phenacobius mirabilis, orangethroat darter Etheostoma spectabile, and slenderhead darter Percinaphoxocephala, which were highest in downstream treat-ment sites. Although limited to one system during a 1-year period, this study suggests that the effects of lowhead dams on fishes, macroinvertebrates, and habitat are similar to those reported for larger dams, providing important considerations for riverine ecosystem conservation efforts.About 2 million dams exist in the United States, including 75,000 dams over 2 m in height; Kansas ranks second in dam number among all states, with 5,699 dams (Shuman 1995). Dams fragment rivers, reducing connectivity and resulting in negative ef-fects on stream biota upstream and downstream from the impoundment (Doeg and Koehn 1994;Rabeni 1996; Kanehl et al. 1997). Dams block movement of fishes and affect habitat and physi-cochemical conditions of streams by converting lotic habitats to lentic, changing streamflow, al-* Corresponding author: jtiemann@inhs.uiuc.edu I Present address: Illinois Natural History Survey, Center for Biodiversity, Champaign, Illinois 61820, USA.2 Present address: University of Oklahoma, Depart-ment of Zoology, Norman, Oklahoma 73019, USA.Received March 10, 2003; acccptcd November 19, 2003 tering water quality, and modifying channel mor-phology and bed structure by increasing siltation upstream and erosion downstream (Watters 1996;Helfrich et al. 1999; Porto et al. 1999). These al-terations cause changes in assemblage structure of fishes and macroinvertebrates via shifts in com-position, abundance, and diversity both upstream and downstream from the impoundment.
Although many studies have addressed effects of large dams on fishes (e.g., Martinez et al. 1994; Clarkson and Childs 2000; Wildhaber et al. 2000b), few have examined effects of lowhead dams (<4 m in height) (e.g., Benstead et al. 1999; Helfrich et al.1999; Beasley and Hightower 2000), and none has done so in a midwestern U.S. stream.Our objectives were to investigate possible ef-fects of two lowhead dams on the fish and mac-roinvertebrate assemblages, habitat, and physico-chemistry of the Neosho River, Kansas. We pre-705 706 TIEMANN ET AL.Site 1 Correll Dam W Site 2 QJ Am ericus 0 3km Ruggles Dam Site 7 Emporia FIGURE 1.-Study area along the Neosho River in Lyon County, Kansas, where fish and macroinvertebrate assemblages, habitat variables, and physicochemical variables were sampled in relation to two lowhead dams (Correll and Emporia).dicted that, because of differences in habitat and physicochemistry, fish and macroinvertebrate as-semblages would differ among treatment sites (ar-eas immediately upstream and downstream from lowhead dams) and reference sites (areas outside the direct zone of influence of dams). We had two a priori hypotheses.
First, we hypothesized that because of inundation, upstream treatment sites would be deeper and have lower velocities, greater siltation, and greater substrate compaction than reference sites, resulting in gravel bars with fewer lotic-type fishes (e.g., madtoms Noturus spp. and darters Etheostoma and Percina spp.) and macro-invertebrates (e.g., mayflies [Ephemeroptera], stoneflies
[Plecoptera], and caddisflies
[Trichop-tera]) but more lentic-type fishes (e.g., sunfishes Lepomis spp.) and macroinvertebrates (e.g., drag-onflies [Odonata]).
Upstream treatment sites would also have lower fish and macroinvertebrate abundance, richness, and evenness than reference sites due to habitat homogeneity.
Second, we hy-pothesized that because of scouring, downstream treatment sites would be shallower and have higher velocities and a higher proportion of large sub-strate than reference sites, resulting in gravel bars with lower fish and macroinvertebrate abundance, richness, and evenness due to habitat homogeneity.
We compared fish and macroinvertebrate abun-dance, richness, and evenness against 10 habitat variables and 7 physicochemical variables at up-stream and downstream treatment and reference sites to test for localized effects of lowhead dams.Methods Study area and sampling methodology.-Our study sites were eight gravel bars situated along a 34-km stretch of the Neosho River in Lyon County, Kansas (Figure 1), within the Prairie Parkland EFFECTS OF LOWHEAD DAMS 707 Province Ecoregion (Chapman et al. 2001). This portion of the Neosho River is a fifth-order stream impounded by three lowhead dams (Correll, Rug-gles, and Emporia) designed for water supply. The Neosho River basin is primarily agricultural, with principal crops of mixed grasses, corn, wheat, and soybeans, and small riparian zones lie adjacent to crop fields. We sampled a segment of river with a mean gradient of 0.54 m/km and mean widths ranging from 14 to 35 in. Council Grove Reservoir is located near the headwaters of the Neosho River, 39 km upstream from site I (Figure 1). Designed for flood control, the reservoir impounds 1,310 ha at conservation pool and has a 5-m-diameter, epi-limnetic outlet that regulates the flow of the Ne-osho River.We sampled the sites in random order during daylight hours monthly from November 2000 to October 2001 (Tiemann 2002). Our eight sites were comprised of four site types (upstream ref-erence, upstream treatment, downstream treat-ment, and downstream reference) positioned near two lowhead dams, Correll and Emporia (Figure 1). Given that the proportion of habitat made up of gravel bars is relatively constant along this length of the Neosho River, we selected sites based on presence of a gravel bar composed mainly of gravel smaller than 64 mm, proximity to the dams, and landowner permission.
We focused on gravel bars because fish assemblages in these habitats are most influenced by disturbances from impound-ments (Wildhaber et al. 2000b). The Correll Dam (38031 '19"N, 96-19'05"W) is situated in the upper part of the study area, is 2.3 in high and 45 m long, and impounds approximately 20 ha; this dam is no longer used for water supply. The Emporia Dam (38-26'1 l"N, 96°12'28"W), located downstream of the Correll Dam, is 3.7 m high and 22 m long, impounds approximately 25 ha, and is used as a water supply station. Because no pre-impoundment data on fish and macroinvertebrate assemblage structure were available, we chose treatment sites as the first samplable gravel bars directly upstream and downstream from each dam, and reference sites as the first samplable gravel bars outside the zone of direct dam influence on flow. Site I (up-stream reference) and site 2 (upstream treatment, or inundated) were located 7.0 and 1.9 km up-stream from the Correll Dam; site 3 (downstream treatment) and site 4 (downstream reference) were located 0.1 and 1.1 km downstream from the Cor-rell Dam. Site 5 (upstream reference) and site 6 (upstream treatment, or inundated) were located 4.1 and 2.7 km upstream from the Emporia Dam;site 7 (downstream treatment) and site 8 (down-stream reference) were located 0.1 and 7.0 km downstream from the Emporia Dam. We judged these sites to be appropriate andvalid standards for presently free-flowing portions of the Neosho River.We chose to examine dam effects on these rel-atively homogeneous mesohabitats because their inhabitants can be sampled more easily and effi-ciently and with more standardized methodology than bigger fishes of deep pools, for which several different types of gear would be necessary.
Also, we chose gravel bars over pools or runs, not only because gravel bars are more samplable, but also because we expected gravel bars to be more af-fected (e.g., change of lentic to lotic) than pools or runs, and because gravel bar fauna generally is more sensitive than those of pools or runs. We used the sampling methods of Wildhaber et al. (2000a), which are appropriate for gravel bars. At each site, we evenly spaced five transects at least 5 m apart perpendicular to the river channel along the length of the gravel bar, and sampled up to five points on each transect, maintaining a minimum of 0.5 m between points. To minimize disturbance, we sam-pled transects from downstream to upstream, sam-pled points from near shore to far shore, and sam-pled variables in the following order at each site: fishes, water depth and stream velocity, substrate compaction and composition, macroinvertebrates, and physicochemistry.
We could not sample sites.5 and 6 in December, January, or February, or site 2 in January or February, due to ice cover.Macroinvertebrates andfishes.--To collect fish-es, we kick-seined a 4.5-M 2 area at each point by disturbing the substrate 3 m upstream from a sta-tionary, 1.5-m, 3-mm-mesh seine and proceeding downstream to the seine. We identified, counted, and released all fishes upon completion of sam-pling at a site.We sampled macroinvertebrates at three random points per site in undisturbed substrate at the head of the gravel bar, in accordance with the strongly upstream-biased distribution of macroinverte-brates within gravel bars (Brown and Basinger-Brown 1984). We used a D-net to dredge a 0.09-m 2 area of substrate, and placed the sample into a bucket partially filled with water. We stirred the substrate for 2 min, strained the water through a I-mm-mesh net, and preserved the contents in 45%isopropyl alcohol. In the laboratory, we sorted samples to family, except nematodes, which were identified to order.Habitat quality and physicochemistry.-We as-708 TIEMANN ET AL.sessed water depth, stream velocity, substrate com-paction, and substrate composition at each point.We measured water depth with a meter stick and measured stream velocity with a Global Flow Probe FP1Ol current meter (Global Water, Gold River, California) positioned at 60% depth from the surface. We estimated substrate compaction by touch, and coded loose substrate as 1, medium as 2, firm as 3., and bedrock as 4 (Fuselier and Edds 1995). We sampled substrate with a shovel (Grost et al. 1991; Bain 1999) and estimated composition visually (Mullner et al. 2000) on a modified Went-worth scale to determine percentages of clay and silt, sand, gravel, pebble, cobble, boulder, and bed-rock (Cummins 1962). We verified field estimates in the laboratory based on reference samples taken at the beginning and end of the study.At the head of each gravel bar, upstream from the area sampled for fishes and macroinverte-brates, we measured temperature with a thermom-eter and dissolved oxygen with a Hach Model AL-36B kit (Hach Chemical Company, Loveland, Col-orado). We then collected a water sample for lab-oratory analyses of alkalinity (Hach Model AL-36B kit), ammonia (Hach surface waters kit), and turbidity (Hach 2100P turbidimeter).
By use of a vacuum pump and Pall type-A/C, glass-fiber filters, we filtered 100 mL of water through two filters and stored them at -10C for subsequent chlorophyll-a and particulate organic carbon (POC) analyses at the Columbia Environmental Research Center (CERC) in Columbia, Missouri.At the CERC, we used a model 10-AU-005 Field Fluorometer (Turner Designs, Sunnyvale, Califor-nia) to measure chlorophyll a and a Coulometrics Model 5014 carbon analyzer (UIC, Inc., Joliet, Il-linois) to measure POC in the filtered samples.Statistical analysis.-We averaged scores for all points to obtain a mean site value for each month and pooled these data for analysis at the treatment level (upstream reference, upstream treatment, downstream treatment, and downstream reference; Bain 1999; Wildhaber et al. 2000a). We used the Shapiro-Wilk test (Zar 1999) to evaluate distri-butions of means for normality, and we used Lev-ene's test (Milliken and Johnson 1984) to evaluate homogeneity of variance.
We log 1 0 transformed nonnormal variables and arcsine-square-root-transformed proportional variables (Zar 1999).Transformation normalized the data, and we ac-cepted the premise that F-statistics used to com-pare means of normally distributed variables are effective whether or not variances are equal, es-pecially when sample sizes are equal or nearly so (Milliken and Johnson 1984). We used Tukey's studentized range test for pairwise comparisons among treatments.
We eliminated fish species and macroinvertebrate taxa occurring in less than 5%of all samples (<5 of the 88 samples) from abun-dance analyses following Gauch (1982). We con-ducted all statistical tests in the Statistical Analysis System (SAS Institute, Inc., Cary, North Carolina)and considered them significant at P-values less than 0.05. Because of multiple tests, we applied a sequential Bonferroni correction (a = 0.05), where appropriate, to help control overall experimental type I error rate (Rice 1989).We performed separate three-way (site type, dam, and month) multivariate analyses of variance (MANOVA) to test for effects of lowhead dams on habitat and physicochemical variables and on fish and macroinvertebrate abundance (number per mi 2). We used Wilk's lambda (X; Zar 1999) to test for significance, the error term being the three-way interaction.
We followed significant MANOVAs with a step-down analysis of covariance (AN-COVA) (Tabachnick and Fidell 1983) to examine the contributions of individual variables.
As a measure of assemblage response in terms of species diversity, we calculated richness (num-ber of species) (Menhinick 1964) and evenness (equitability) (Williams 1964) of fishes and mac-roinvertebrates for each sample. Because richness values depend upon area sampled, we used rare-faction for fishes (unequal area sampled per site)but not for macroinvertebrates (equal area sampled per site) (Glowacki and Penczak 2000; Wildhaber et al. 2000a). We chose an evenness index that is independent of richness (Smith and Wilson 1996).We performed three-way MANOVAs on individ-ual habitat and physicochemical variables, in ad-dition to individual fish and macroinvertebrate taxa abundances, richness, and evenness, to further test for effects of lowhead dams. We also calcu-lated Pearson's correlation coefficient to examine potential relationships of statistically significant habitat and physicochemical variables with fish and macroinvertebrate abundance, richness, and evenness.Results Habitat Quality and Physicochemistry Habitat characteristics varied significantly among site types (MANOVA: X = 0.0003; n =88; P < 0.0001). Bedrock (step-down ANCOVA: F = 425.23; df = 60, 27; P < 0.0001) and substrate compaction (step-down ANCOVA: F = 16.29; df EFFECTS OF LOWHEAD DAMS 709 TABLE Il--Means (SDs in parentheses) and analysis of variance results (F-values, with P-values in parentheses) for habitat and physicochemical variables by site type in the Neosho River, Kansas, from November 2000 to October 2001;N is the number of samples per site type, lowercase letters within rows indicate significant Tukey's groupings, and asterisks indicate significant sequential Bonferroni-adjusted P-values.Habitat or physicochemical variable Water depth (cm)Stream velocity (nils)Substrate compaction Clay/silt
(<0.06 mm; %)Sand (0.06-1 mm; %)Gravel (2-15 mm; %)Pebble (16-63 mm; %)Cobble (64-256 mm; %)Boulder (>256 mm; %)Bedrock (solid bottom)Temperature
(°C)Dissolved oxygen (mg/L)Alkalinity (mg/L)Ammonia (mg/L)Chlorophyll a (4Lg/L)POC' (mg/L)Turbidityb Upsueam reference (N = 21)48.4 (13.1) z 0,24 (0.15) z 1L9 (0.2) z 20.0 (61) z 5.0 (2,4) z 41.7 (3.4) z 33.0 (6.7) z 0.3 (0.8) Z 0.0 (0.0) z 0.0 (0.0) Z 14.9 (9.9)8.9 (2.3)171.5 (49.2)0.01 (0.03)678.4 (708.8)170.1 (97.0)35.9 (35.8)Upstream treatment (N = 19)57.7 (4,4) y 0.05 (0.08) y 2.3 (0.3) y 11.5 (6.7) z 3.3 (1.4) z 38.4 (6.9) z 37.1 (5.7) z 7.2 (1.9) y 2.5 (2.6) y 0.0 (0.0) z 15.0 (9.5)8.3 (2.3)176.0 (58.8)0.02 (0.04)680.0 (521.6)164.6 (81.3)31.1 (21.8)Downstream treatment (N = 24)24.0 (12.4) x 0.42 (0.27) x 2.8 (0.4) y 4.1 (5.4) y 2.8 (2.3) y 24.5 (6.4) y 22.7 (7.6) y 4.3 (2.7) y 1.0 (2.0) z 40.7 (6.1) y 15.5 (10.8)10.0 (2.5)176.0 (54.4)0.02 (0.04)535.1 (574.4)179.7 (91.3)40.6 (47.6)Downstream reference (N = 24)35.5 (15.1) z 0.32 (0.15) z.1.7 t0.3) z 14.8, (7.2) z 5.4 (3.2) z 42.6 (6.9) z 37.1 (6.6) z 0.1 (0.2) z 0.0 (0.0) z 0.0 (0.0) z 15.3 (10.8)9.4 (2.2)179.6 (65.3)0.03 (0.05)421.0 (393.3)166.3 (74.7)40.5 (40.9)Site type F-value (df = 3, 27)57.16 (<0.0001)*
29.64 (<0.0001)*
99.77 (<0.0001)*
22.86 (<0.0001)*
5.83 (0.003)-85.85 (<0.000t)*
26.09 (<0.0001)*
54.08 (<0.0001)*
8.14 (0.0005)*425.25 (<0.0001)*
0.45 (0.72)1.70 (0.11)0.35 (0.79)0.42 (0.74)1.38 (0.27)0.68 (0.57)1.90 (0.15)a Particulate organic carbon.b Nephelometric turbidity units.= 3, 30; P < 0.0001) contributed significantly to the variation among site types. Multivariate AN-OVA indicated that all habitat variables differed significantly among site types (Table 1). Tukey's test indicated that upstream treatment sites were deeper and had slower velocities than reference sites, whereas downstream treatment sites were shallower and had faster velocities than reference sites (Figure 2); both treatment site types had high-er substrate compaction than reference sites (Fig-ure 2). Downstream treatment sites had a different particle size distribution compared to reference sites and upstream treatment sites. Tukey's test in-dicated that downstream treatment sites had lower percentages of clay/silt, sand, gravel, and pebble substrates, and a higher percentage of bedrock than reference sites and upstream treatment sites, whereas upstream treatment sites had a higher per-centage of boulder than reference sites; both treat-ment site types had higher percentages of cobble than reference sites had (Figure 3).Physicochemistry did not vary significantly among site types (MANOVA: X = 0.47; n = 82;P = 0.64). None of the seven variables differed significantly among site types (Table 1).Macro invertebrates and Fishes We collected 11,594 macroinvertebrates repre-senting 26 identified taxa (12 orders encompassing 25 families, plus the nematode order Rhabditida), of which 23 were sufficiently common to be re-tained for abundance analysis (Table 2). Aquatic insects comprised 94.9% of the macroinvertebrates sampled. Chironomidae (order Diptera) was the most abundant family collected (64.0%), followed by Hydropsychidae (order Trichoptera; 10.3%)and -eptageniidae (order Ephemeroptera; 6.5%).Macroinverlebrate abundance varied signifi-cantly among site types (MANOVA: X =0.000006; n = 88; P < 0.0001). Mean (+/--SD)mac-roinvertebrate abundance varied from 46.4 +/- 6.2 individuals/m 2 in upstream reference sites and 48.1+/- 12.4 individuals/m-2 in upstream treatment sites to 25.9 +/- 6.6 individuals/m 2 in downstream treat-ment sites and 55.4 +/- 1.8 individuals/m 2 in down-stream reference sites. Abundances of 12 of the 23 taxa were significantly different among site types (Table 2). Abundances of Culicidae (Diptera)(step-down ANCOVA: F = 40.84; df = 60, 27; P< 0.0001), Lestidae (order Odonata) (step-down ANCOVA: F = 7.48; df = 3, 33; P = 0.0002), Chironomidae (step-down ANCOVA" F = 9.62;df = 3, 32; P < 0.0001), and Heptageniidae (step-down ANCOVA: F = 9.86; df = 3, 1; P < 0.0001)contributed significantly to variation in abundance among site types. Tukey's test indicated that Cu-licidae, Chironomidae, and Lestidae had higher abundances at upstream treatment sites compared with other site types, Chironomidae had lower abundance at downstream treatment sites com-710 TIEMANN ET AL.75 50 25 0 a b r 0 E 0.0 C.100 80 60 40 20 0 1313-de, U Cobble WdPbble CIG-1c O~soo UR UT DT DR Site type 0.8 0.6 0.4 U 0.2 0.0 F a aa a UR UT DT DR Site type 4 E.2 01 0 b b a a UR UT DT DR Site type FIGURE 3.-Mean substrate composition percentages per site type (UR = upstream reference; UT = upstream treatment; DT = downstream treatment; DR = down-stream reference) in the Neosho River, Kansas, Novem-ber 2000-October 2001.correlated with percent bedrock substrate (Pear-son's correlation:
r = -0.32; P = 0.0003).Macroinvertebrate taxa richness did not signif-icantly differ among site types (ANOVA: F =0.74; df = 3, 27; P = 0.54), but evenness did (ANOVA: F = 8.37; df = 3, 27; P = 0.0004).Mean macroinvertebrate evenness varied from 0.48 -t 0.01 in upstream reference sites and 0.39+/- 0.18 in upstream treatment sites to 0.41 +/- 0.01 in downstream treatment sites and 0.47 +/- 0.03 in downstream reference sites. Tukey's test indicated that upstream treatment sites had lower evenness than reference sites and downstream treatment sites. Neither macroinvertebrate taxa richness nor evenness was significantly correlated with fish abundance or any habitat or physicochemical var-iable.In 88 samples, we caught 15,222 fish repre-senting 10 families, 19 genera, and 31 species, of which 21 species were sufficiently common to be retained for abundance analysis (Table 3). Eleven species occurred at all eight sites (central stone-roller, red shiner, ghost shiner, bluntnose minnow, bullhead minnow, channel catfish, orangespotted sunfish, bluegill, orangethroat darter, logperch, and slenderhead darter). Red shiner was the most abun-dant species collected (47.8%), followed by ghost shiner (10.6%) and bluntnose minnow (10.0%).Fish abundance varied significantly among site types (MANOVA: X = 0.0002; n = 88; P <0.0001). Mean fish abundance varied from 1.37 +/-0.29 fish/m 2 in upstream reference sites and 1.85* 0.52 fish/m-2 in upstream treatment sites to 2.68* 0.42 fish/m-2 in downstream treatment sites and 3.09 +/- 0.44 fish/m 2 in downstream reference sites.Abundances of 4 of the 21 species were signifi-UR UT DT DR Site type.FIGURE 2.-Mean water depth, stream velocity, and substrate compaction
(+/-SD) per site type (UR = up-stream reference; UT = upstream treatment; DT =downstream treatment; DR = downstream reference) in the Neosho River, Kansas, November 2000-October 2001. The lowercase letters in the lowest panel indicate significant groupings according to Tukey's test.pared to other site types, and Heptageniidae had higher abundances in reference sites than treatment sites. Macroinvertebrate abundance was positively correlated with percent pebble substrate (Pearson's correlation:
r = 0.46; P < 0.0001) and negatively EFFECTS OF LOWIEAD DAMS 711 TABLE 2.-Mean macroinvertebrate taxa abundance per square meter (SDs in parentheses) and analysis of variance results (F-values, with P-values in parentheses) by site type in the Neosho River, Kansas, from November 2000 to October 2001; N is the number of samples per site type, lowercase letters within rows indicate significant Tukey's groupings, and asterisks indicate significant sequential Bonferroni-adjusted P-values.Upstream Upstream Downstream Downstream Site type Benthic reference treatment treatment reference F-value invertebrates (N = 21) (N = 19) (N = 24) (N = 24) (df = 3, 27)Order Epbemeroptera Potamanthidae Baetidae Heptageniidae Order Plecoptera Perlidae Order Trichoptera Limnephilidae Hydropsychidae Order Odonata Gomphidae Lestidae Order Coleoptera Carabidae Dytiscidac Gyrinidae Order Heniiptera Corixidae Belostomatidae Order Diptera Chironomidae Chaoboridae Culicidae Simuliidae Order Oligochaeta Tubificidae Order Rhynchobdellida Glossiphoniidae Order Heterodonta Corbiculidae Order Gastropoda Lymnaeidae Order Decapoda Cambaridac Order Rhabditida 0.23 (0.03)1.38 (0.42) z 4.43 (0.85) z 0.64 (0.11)0.22 (0.00)0.40 (0.21) y 1.75 (1.14) y 1.39 (0.68)0.06 (0.02)0.24 (0.13) y 0.96 (0.19) y 0.14 (0.04)0.40 (0.03) 3.35 (0.03)1.94 (0.29) z 10.75 (<0.0001)*
4.18 (0.13) z 21.67 (<0.0001)*
0.72 (0.04)7.14 (0.001)0.01 (0.00) 0.04 (0.00) 0.03 (0.00) 0.10 (0.03) 0.79 (0.65)6.52 (0.50) z 1.66 (1.05) y 2.71 (0.91) y 6.75 (0.10) z 18.30 (<0.0001)*
0.00 (0.00) 0.12 (0.00) 0.00 (0.00) 0.00 (0.00) 7.17 (0.001)0.00 (0.00) z 0.95 (0.01) y 0.00 (0.00) z 0.00 (0.00) z 26.00 (<0.0001)*
0.75 (0.06) z 1.98 (0.56)0.73 (0.22) z 0.15 (0.08) y 0.26 (0.16)0.13 (0.07) y 0.31 (0.02) y 0.64 (0.31)0.15 (0.01) y 1.11 (o.oo) y 1.86 (0.12)0.74 (0.05) z 19.13 (<0.0001)*
5.36 (0.005)12.31 (<0.0001)*
0.00 (0.00) z 0.12 (0.03) y 0.00 (0.00) z 0.01 (0.00) 2 11.71 (<0.0001)*
0.00 (0.00) 0.18 (0.01) 0.01 (0.00) 0.04 (0.00) 5.56 (0.004)26.78 (2.91) z 0.73 (0.06)0.00 (0.00) z 0.87 (0.35)0.77 (0.11)34.57 (6.45) y 0.48 (0.05)1.15 (0.34) y 0.65 (0.27)1.20 (0.40)17.29 (3.96) x 0.31 (0.08)0.03 (0.00) z 1.38 (0.66)0.35 (0.11)33.71 (0.46) z 0.65 (0.07)0.00 (0.00) z 1.04 (0.05)0.76 (0.09)25.30 (<0.0001)*
1.04 (0.39)40.84 (<0.0001)*
0.31 (0.82)3.60 (0.03)0.00 (0.00) z 0.29 (0.10) y 0.03 (0.00) z 0.00 (0.00) z 19.70 (<0.0001)*
0.18 (0.00) z 0.00 (0.00) y 0.03 (0.00) y 0.90 (0.23) z 10.43 (<0.0001)*
0.00 (0.00)0.34 (0.09)0.32 (0.13)0.00 (0.00)6.30 (0.002)0.29 (0.01) z 1.69 (0.84) y 0.86 (0.00) y 0.22 (0.02) z 8.57 (0.0004)*0.07 (0.03) 0.24 (0.09) 0.08 (0.04) 0.21 (0.13) 0.52 (0.67)cantly different among site types (Table 3). Abun-dances of orangethroat darter (step-down AN-COVA: F = 14.86; df = 60, 27; P < 0.0001) and suckermouth minnow (step-down ANCOVA: F =5.96; df = 3, 72; P = 0.001) contributed signifi-cantly to variation in abundance among site types.Tukey's test indicated that the orangethroat darter, suckermouth minnow, and slenderhead darter were more abundant in downstream treatment sites com-pared to other sites, and that the Neosho madtom was less abundant in treatment sites compared to reference sites (Figure 4). Fish abundance was not significantly correlated with macroinvertebrate abundance or any habitat or physicochemical var-iable.Fish species richness did not significantly differ among site types (ANOVA: F = 2.83; df = 3, 27;P = 0.06), but evenness did (ANOVA: F = 4.83;df = 3, 27; P = 0.008). Mean evenness varied from 0.55 +/- 0.02 in upstream reference sites and 0.52 +/- 0.05 in upstream treatment sites to 0.44 +/-0.07 in downstream treatment sites and 0.45 +/-0.04 in downstream reference sites. Tukey's test indicated that both upstream site types had higher 712 TIEMANN ET AL.TABLE 3.-Mean fish species abundance per square meter (SDs in parentheses) and analysis of variance results (F-values, with P-values in parentheses) by site type in the Neosho River, Kansas, from November 2000 to October 2001;N is the number of samples per site type, lowercase letters within rows indicate significant Tukey's groupings, and asterisks indicate significant sequential Bonferroni-adjusted P-values.Upstream Upstream Downstream Downstream Site type reference treatment treatment reference F-value Fishes (N = 21) (N = 19) (N = 24) (N = 24) (df = 3, 27)Cyprinidae Central stoneroller Campostoma anomrahlm 0.013 (0.008) 0.037(0.023) 0.028 (0.011) 0.025 (0.000)Red shiner Cyprinella lutrensis 0.386 (0.002) 0.454 (0.069) 1.421 (0.031) 1.773 (0.095)Ghost shiner Notropis buchanani 0.310 (0.033) 0.333 (0.167) 0.193 (0.005) 0.156 (0.034)Sand shiner N. stramnineus 0.015 (0.009) 0.006 (0.000) 0.006 (0.002) 0.049 (0.031)Suckermouth minnow Phenacobius mirabilis 0.019 (0.01 1)a 0,001 (0.000)- 0.064 (0.00g)b 0.027 (0.008)2 Bluntnose minnow Pinephales notatus 0.200 (0.055) 0.321 (0.138) 0.215 (0.061) 0.341 (0.047)Slim minnow Pimephales tenellus 0.048 (0.032) 0.091 (0.000) 0.020 (0.001) 0.091 (0.035)Bullhead minnow Pimephales vigilax 0.091 (0.035) 0.057 (0.019) 0.041 (0.008) 0.233 (0.067)Catostomidae Golden redhorse Moxostoma erythrurum 0.001 (0.000) 0.003 (0.000) 0.001 (0.000) 0.001 (0.000)Ictaluridae Channel catfish lctaluruspunctatus 0.022 (0.003) 0.008 (0.005) 0.014 (0.000) 0.019 (0.008)Stonecat Noturusflavus 0.003 (0.000) 0.000 (0.000) 0.006 (0.001) 0.005 (0.000)Neosho madtom N. placidus 0.014 (0.007)a 0.003 (0.001)b 0.001 (0.000)b 0.021 (0.003)8 Poeciliidae Western mosquitofish Gamnbusio offinis 0.006 (0.000) 0.004 (0.001). 0.028 (0.019) 0.000 (0.000)Centrarchidae Green sunfish Lepomis cyanellus 0.000 (0.000) 0.009 (0.001) 0.010 (0.003) 0.012 (0.002)Orange spotted sunfish L. humilis 0.097 (0.045) 0.331 (0.041) 0.083 (0.027) 0.178 (0.039)Bluegill L. macrochirus 0.005 (0.002) 0.008 (0.001) 0.005 (0.003) 0.003 (0.001)Percidac Orangethroat darter Etheostoma specrabile 0.041 (0.028)a 0.048 (0.013)8 0.133 (0.010)b 0.037 (0.021)a Logperch Percina caprodes 0.009 (0.002) 0.014 (0.001) 0.023 (0.013) 0.006 (0.002)Channel darter P. copelandi 0.009 (0.000) 0,013 (0.000) 0.000 (0.000) 0.015 (0.008)Slenderhead darter P. pho.,ocephala 0.007 (0.017)a 0.102 (0.046)8 0.377 (0.2 1 6)b 0.095 (0.044)a Sciaenidae Freshwater drum Aplodinotus grunniens 0.000 (0.000) 0.002 (0.000) 0.002 (0.000) 0.002 (0.000)1.10 (0.37)4.53 (0.0i)0.86 (0.47)4.00 (0.02)14.38 (<0.0001)*
1.24 (0.31)6.42 (0.002)4.73 (0.009)0.56 (0.65)2.24 (0.74)1.12 (0.38)9.66 (0.0002)*0.93 (0.44)0.95 (0.42)3.73 (0.03)0.65 (0.59)14.86 (<0.0001)*
2.87 (0.05)2.65 (0.07)9.85 (0.0001).0.24 (0.87)
EFFECTS OF LOWHEAD DAMS 713 0.03 r.'4 0.02 0.01 0.00 mediately upstream and downstream, with resul-tant effects on macroinvertebrate and fish assem-blage structure.
Results were similar to those for lowbead dams in other parts of North America (e.g., Helfrich et al. 1999; Porto et al. 1999; Beas-ley and Hightower 2000), and to a lesser extent resembled those for large dams (e.g., Martinez et al. 1994; Camnargo and Voelz 1998; Wildhaber et al. 2000b).UR UT DT DR Site type 0.6 ,, 0.4 0.2 0.0 FIGURE 4.-tom (top pan orangethroat (triangles) (b type (UR =ment; DT =reference) in 2000-Octobe nificant grouj evenness th only upstrea stream refer ness nor eve macroinvert physicocher Habitat Quality and Physicochemistry b As a stream is deepened, water velocity is de-ab creased and its ability to carry sediment in the water column is reduced, generally resulting in increased sedimentation of the substrate (Kondolf a a a 1997; Wood and Armitage 1997). Our upstream treatment sites were deeper and had lower velocity and higher substrate compaction than the other site UR UT DT DR types, but percentages of fine substrates were not Site type significantly different from those of reference sites. Upstream treatment sites had a higher pro--Mean abundances
(+/--SD) of Neosho mad- portion of larger particles (cobble and boulder), el) and suckermouth minnow (asterisks), perhaps as a result of the parent material existing darters (circles), and slenderhead darters prior to inundation or the relationship between ottom panel) abundance per for each site prao tionlor the rion between upstream reference; UT = upstream treat- mean current velocity and the size of particles that downstream treatment; DR = downstream can be transported after inundation.
Our study the Neosho River, Kansas, November could not adequately address this issue. Down-r 2001. The lowercase letters indicate sig- stream treatment sites were shallower and had pings according to Tukey's test. higher velocities than other site types, and differed from other site types in 8 of 10 substrate charac-teristics.
Treatment sites immediately downstream han downstream treatment sites, but from the dams had greater substrate compaction am reference sites differed from down- and larger mean substrate size, which was reflected rence sites. Neither fish species rich- in more bedrock and lower percentages of clay/nness was significantly correlated with silt, sand, gravel, and pebble compared to refer-ebrate abundance or any habitat or ence sites. Over time, water flowing over these nical variable.
dams appears to have scoured finer substrates and taken the gravel bar down to bedrock, which ac-Discussion counted for the differences in bedrock among sites.correlation exists between habitat var- A coarsening of substrate can result from stream-sh and macroinvertebrate assemblages, bed erosion by "sediment-hungry" release waters es that fragmentation and modification with increased velocity, and this process typically ation, scouring, and channelization) of reduces habitat diversity (Kondolf 1997; Camargo itat can have profound effects on biotic and Voelz 1998). In rivers with large dams, effects cluding declines in abundance and di- on substrate size composition typically are greatest nacroinvertebrates and fishes (Neves immediately downstream from the dam, causing cier 1990; Dynesius and Nilsson 1994; scouring of organisms that sometimes leaves 1. 1999). Although our study was con- streambeds devoid of much of their fauna (Ca-during a period of 1 year and on a margo and Voelz 1998).our data demonstrate influences of Mean daily extraction from the Emporia water ms on habitat quality, macroinverte-supply station was about 30 million liters (-0.34 fishes in this midwestern stream. Cor- m 3/s) (City of Emporia 2001), whereas mean daily mporia dams affected water depth, discharge from Council Grove Reservoir during city, and substrate characteristics im- our study was approximately 1.73 m 3/s (USACE A strong iables and fi which impli (e.g., inunda riverine hab integrity, in.versity of n and Angerm Luttrell et a ducted only single river, lowhead da brates, and rell and Er stream velo 714 TIEMANN ET AL.2001). During August, site 7 (immediately down-stream from Emporia Dam) had periods of no flow due to water extraction levels that exceeded Coun-cil Grove Reservoir discharge.
We found no pre-vious reports of the effects of lowhead dams and water extraction.
Water extraction behind lowhead dams could indirectly degrade downstream sub-strate and affect benthic organisms.
Reduced dis-charge, whether natural or artificial, can expose portions of gravel bars and cause compaction by the drying of organic material in interstitial spaces.If the substrate remains compacted following re-turn to normal water levels, benthic organisms could be forced into less suitable areas, resulting in decreased survival (Wildhaber et al. 2000a; Bul-ger and Edds 2001).Physicochemistry values were within the range reported by Wildhaber et al. (2000a) and Bulger and Edds (2001) for undammed portions of the Neosho River, and there were no significant dif-ferences among site types for any of the seven physicochemical variables.
Unlike large dams (Wildhaber et al. 2000b), the lowbead dams we studied did not seem to affect physicochemistry of the Neosho River, perhaps because of lower water retention time. Although Hach kits do not provide sufficient accuracy or precision to be de-fensible, our results are comparable among our sites because we used the same kits throughout the study. In addition, our water quality data were comparable to those reported in other studies in the Neosho River.Effects of Adjacent Dams Ruggles Dam is another lowhead structure that impounds the Neosho River between the Correll and Emporia dams (Figure 1); we were unable to obtain landowner permission to sample around this dam. As with most North American rivers (Benke 1990), the Neosho River is highly regulated, hav-ing 2 reservoir dams and 15 lowhead dams in Kan-sas. We could not remove the potential effects of these other dams. Rather, in our analysis of the localized impacts of the Correll and Emporia dams, we chose reference sites outside the zone of direct dam influence on flow; these reference sites represented the normal condition for presently un-dammed portions of the Neosho River and there-fore acted as appropriate and valid standards.
Downstream effects of lowhead dams depend on dam size, hydrology, geology, faunal composition, and other factors (Baxter 1977).Council Grove Reservoir Dam, 39 km upstream from site 1. also affects the river. During our study, discharge of the Neosho River at Americus (Figure 1) (USGS 2001) mirrored releases from Council Grove Reservoir (USACE 2001). However, be-cause Council Grove is not a hydroelectric facility and because it has epilimnetic release, it produces no regular pulses of discharge and no alteration of the thermal regime downstream.
Given these fac-tors, the facility's relatively small size, and the considerable downstream distance of our study area from the reservoir, we felt confident in the assumption that our sites were outside the direct influence of this dam.Macroinvertebrates and Fishes Macroinvertebrate abundance was lowest at downstream treatment sites, perhaps as a result of substrate coarsening and reduced habitat diversity (Baxter 1977; Kondolf 1997; Camargo and Voelz 1998). Macroinvertebrate abundance depends upon presence of a mixture of heterogeneous grav-el, pebble, and cobble substrates, and moderate, consistent flow (Waters 1995), which were not characteristics of our downstream treatment sites.Macroinvertebrates inhabiting degraded stream-bed substrates are subjected to scouring, which could make them more susceptible to predation through dislodgment (Newcombe and MacDonald 1991). No macroinvertebrate taxon had higher abundance immediately downstream from the dams.Macroinvertebrate taxa richness was not signif-icantly different among site types, but evenness was lowest at upstream treatment sites, as the fau-na was dominated by some resilient lentic taxa (Merritt and Cummins 1996), including Culicidae, Chironomidae, and Lestidae, that reached their highest abundances or occurred only at these in-undated sites. More-sensitive, lotic taxa (Merritt and Cummins 1996), including Baetidae (Ephem-eroptera), Heptageniidae, and Hydropsychidae, were less abundant at upstream and downstream treatment sites than at reference sites. Given that these organisms are good environmental indicators (Brown and Basinger-Brown 1984; Brown and Brussock 1991; Merritt and Cummins 1996), our results suggest that the lowhead dams we studied have negatively impacted habitat quality of the Neosho River.Fish species richness did not differ significantly among site types. Evenness was lower at down-stream reference sites, mainly due to a February sample in which 606 of 607 fish were red shiners.Fish abundance was highest at downstream ref-erence sites and lowest at upstream reference sites.
EFFECTS OF LOWHEAD DAMS 715 Helfrich et al. (1999) suggested that a series of lowhead dams might present a serious cumulative challenge to fish passage, leading to gradual al-teration of fish assemblage structure in a river. Giv-en the presence of 17 dams on the Kansas portion of the Neosho River, such extensive modification could have a collective impact on fish populations.
Although our study did not specifically address this issue, we did not see a significant longitudinal effect on fish assemblages (Gillette 2002).Differences in fish assemblage structure were reflected mainly in abundances of benthic species immediately upstream and downstream from the dams. For example, abundance of the federally listed Neosho madtom was lower at upstream treat-ment sites, which had more cobble and boulder, deeper, slower water, and higher substrate com-paction than reference sites. The Neosho madtom and many other substrate-oriented fishes are hab-itat specialists whose abundances vary according to stream velocity and substrate composition (Cross and Collins 1995; Pflieger 1997). Down-stream treatment sites had shallower water depths, greater stream velocities, and lower percentages of clay/silt, sand, gravel, and pebble substrates, fa-voring the orangethroat darter, slenderhead darter, and suckermouth minnow. These species prefer ample stream velocities and sites free of silt (Pflie-ger 1997), and they dominated the assemblage in downstream treatment areas. However, as with macroinvertebrates, scoured downstream treat-ment sites lacked the loosely compacted substrate required by many substrate-oriented fishes. For ex-ample, the Neosho madtom prefers loose, clean gravel/pebble substrate in moderate water depths and stream velocities (Fuselier and Edds 1994;Bulger and Edds 2001). Compared with the situ-ation at reference sites, the abundance of this fish was significantly lower immediately downstream from the dams, where water was shallower and faster and where gravel and pebble substrates were less prevalent.
Compaction of substrate in down-stream treatment areas might force substrate-oriented fishes into less suitable areas, where they could experience lower survival rates (Bulger and Edds 2001). It is also possible that the larger in-terstitial spaces in the cobble, which was more abundant immediately upstream and downstream from the dams, might not offer as many macro-invertebrates to feed on or as much protection from predators for substrate-oriented fishes as do gravel and pebble (Wildhaber et al. 2000a). It should be noted that our upstream treatment sites were not located directly behind the dams because seining there was not possible.
By standardizing our sam-pling to kick-seining of gravel bars, our collections were efficient and comparable among site types.However, this sampling was most effective for small, lotic fishes, and probably underrepresented some Neosho River fishes, including the larger len-tic taxa like centrarchids and catostomids.
Conclusions Our findings suggest that lowhead dams cause differences in habitat immediately upstream and downstream, producing effects on fish and mac-roinvertebrate assemblages that are similar to, but less extensive than, the effects of large dams. The dams in our study were associated with significant differences in water depth, stream velocity, sub-strate compaction, and substrate composition that appear to affect macroinvertebrate and fish abun-dance and evenness, especially for habitat spe-cialists.
Our study contributes insights into the ef-fects of lowhead dams on riverine habitat and fish and macroinvertebrate assemblages in the Mid-west. Additional studies in other drainages and regions, with differing faunas and hydrologic re-gimes, should be conducted to gain a better un-derstanding of how lowhead dams affect the bi-ology and hydrology of stream ecosystems.
Knowledge of the effects of these barriers can be used in the conservation and protection ofriverine biotic integrity.
Acknowledgments This research was supported by the U.S. Geo-logical Survey (USGS) under USGS Cooperative Agreement Number OOCRAGO025, with addition-al funding from an Emporia State University (ESU) Faculty Research and Creativity Grant and an ESU Graduate Student Research Grant. We thank the landowners (G. Guide, W. Leffler and M. Leffler, P. Matile and D. Matile, L. Schlesener, Girl Scout Council of the Flint Hills, the City of Emporia, and ESU Natural Areas) for allowing access to the river. Thanks to D. Moore, L. Scott, and D. Zelmer of ESU for statistical advice and helpful discussions.
B. Johnson of Trout Unlimited and H. Dodd of the Illinois Natural History Survey (INHS) provided background information.
D. Mul-hem and V. Tabor of the U.S. Fish and Wildlife Service loaned a Hach kit. R. Ferguson, B. Flock, B. Harkins, G. Head, G. Sievert, R. Sleezer, L.Sneed, and L. Westerman of ESU and J. Anderson of the Kansas Biological Survey provided tech-nical assistance.
B. Chance, J. Dean, L. Freeman, J. Howard, S. Sherraden, and I. Singh of ESU as-716 TIEMANN ET AL.sisted with fieldwork, and J. Albers, A. Allert, J.Fairchild, B. Lakish, S. Olson, and C. Witte of the USGS assisted with water chemistry analysis.
G.Levin, T. Rice, C. Taylor, and D. Thomas of INHS reviewed the manuscript and provided construc-tive criticism.
References Bain, M. B. 1999. Substrate.
Pages95-103 in M. B.Bain and N. J. Stevenson, editors. Aquatic habitat assessment:
common methods. American Fisheries Society, Bethesda, Maryland.Baxter, R. M. 1977. Environmental effects of dams and impoundments.
Annual Review of Ecology and Sys-tematics 8:255-283.
Beasley, C. A., and J. E. Hightower.
2000. Effects of a low-head dam on the distribution and characteristics of spawning habitat used by striped bass and Amer-ican shad. Transactions of the American Fisheries Society 129:1316-1330.
Benke, A. C. 1990. A perspective on America's van-ishing streams. Journal of the North American Ben-thological Society 9:77-88.Benstead, J. P., J. G. March, C. M. Pringle, and F N.Scatena. 1999. Effects of a low-head dam and water abstraction on migratory tropical stream biota. Eco-logical Applications 9:656-668.
Brown, A. V., and K. Basinger-Brown.
1984. Distri-bution of insects within riffles of streams. Fresh-water Invertebrate Biology 3:2-11.Brown, A. V., and P. P. Brussock.
1991. Comparisons of benthic invertebrates between riffles and pools.Hydrobiologia 220:99-108.
Bulger, A. G., and D. R. Edds. 2001. Population struc-ture and habitat use in Neosho madtom (Noturus.placidus).
Southwestern Naturalist 46:8-15.Camargo, J. A., and N. J. Voelz. 1998. Biotic and abiotic changes along the recovery gradient of two im-pounded rivers with different impoundment use. En-vironmental Monitoring and Assessment 50:143-158.Chapman, S. S., J. M. Omernik, J. A. Freeouf, D. G.Huggins, J. R. McCauley, C. C. Freeman, G. Stei-nauer, R. T. Angelo, and R. L. Schlepp. 2001.Ecoregions of Nebraska and Kansas. [Color poster with map, descriptive text, summary tables, and photographs.]
U.S. Geological Survey, Reston, Vir-ginia.City of Emporia. 2001. Available:
ws/Public%20Works.nsf/waterplantmain.htm.(November 2001.)Clarkson, R. W., and M. R. Childs. 2000. Temperature effects of hypolimnial-release dams on early life stages of Colorado River basin big-river fishes.Copeia 2000:402-412.
Cross, F. B., and J. T. Collins. 1995. Fishes in Kansas, 2nd edition. University Press of Kansas, Lawrence.Cummins, K. W. 1962. An evaluation of some tech-niques for the collection and analysis of benthic samples, with special emphasis on lotic waters.American Midland Naturalist 67:477-504.
Doeg, T. J., and J. D. Koehn. 1994. Effects of draining and desilting a small weir on downstream fish and macroinvertebrates.
Regulated Rivers: Research and Management 9:263-277.
Dynesius, M., and C. Nilsson. 1994. Fragmentation and flow regulation of river systems in the northern third of the world. Science 266:753-762.
Fuselier, L., and D. Edds. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleostei:
lc-taluridae:
Noturus placidus).
Southwestern Natural-ist 39:217-223.
Fuselier, L., and D. Edds. 1995. An artificial riffle as restored habitat for the threatened Neosho madtom.North American Journal of Fisheries Management 15:499-503.
Gaueh, H. G., Jr. 1982. Multivariate analysis in com-munity ecology. Cambridge University Press, Cam-bridge, UK.Gillette, D. P. 2002. Spatiotemporal patterns of fish as-semblage structure and habitat use on gravel bars in the Neosho River, Lyon County, Kansas. Master's thesis. Emporia State University, Emporia, Kansas.Glowacki, L., and T. Penczak. 2000. Impoundment im-pact on fish in the Warta River: species richness and sample size in the rarefaction method. Journal of Fish Biology 57:99-108.
Grost, R. T., W. A. Hubert, and T. A. Wesche. 1991.Field comparisons of three devices used to sample substrate in small streams. North American Journal of Fisheries Management 11:347-351.
Helfrich, L. A., C. Liston, S. Hiebert, M. Albers, and K. Frazer. 1999. Influence of low-head diversion dams on fish passage, community composition, and abundance in the Yellowstone River, Montana. Reg-ulated Rivers: Research and Management 7:21-32.Kanehl, P. D., J. Lyons, and J. E. Nelson. 1997. Changes in the habitat and fish community of the Milwaukee River, Wisconsin, following removal of the Woolen Mills Dam. North American Journal of Fisheries Management 17:387-400.
Kondolf, G. M. 1997. Hungry water: effects of dams and gravel mining on river channels.
Environmental Management 21:533-551.
Luttrell, G. R., A. A. Echelle, W. L. Fisher, and D. J.Eisenhour.
1999. Declining status of two species of the ,Macrhybopsis aestivalis complex (Teleostei:
Cy-prinidae) in the Arkansas River basin and related effects of reservoirs as barriers to dispersal.
Copeia 1999:981-989.
Martinez, P. J., T. E. Chart, M. A. Trammell, J. G.Wullschleger, and R. P. Bergersen.
1994. Fish spe-cies composition before and after construction of a main-stem reservoir on the White River, Colorado.Environmental Biology of Fishes 40:227-239.
Menhinick, E. F 1964. A comparison of some species diversity indices applied to samples of field insects.Ecology 45:859-861.
Merritt, R. W., and K. W. Cummins. 1996. An intro-duction to the aquatic insects of North America, 3rd edition. Kendall/Hunt, Dubuque, Iowa.Milliken, G. G., and D. E. Johnson. 1984. Analysis of EFFECTS OF LOWHEAD DAMS 717 messy data, volume I. Designed experiments.
Wads-worth, Belmont, California.
Mullner, S. A., W. A. Hubert, and T. A. Wesche. 2000.Visually estimating substrate composition at poten-tial spawning sites for trout in mountain streams.Journal of Freshwater Ecology 15:199-207.
Neves, R. J., and P. L. Angermeier.
1990. Habitat al-teration and its effects on native fishes in the upper Tennessee River system, east-central USA. Journal of Fish Biology 37:45-52.Newcombe, C. P., and D. D. MacDonald.
1991. Effects of suspended sediments on aquatic ecosystems.
North American Journal of Fisheries Management 11:72-82.Pflieger, W. L. 1997. The fishes of Missouri, 2nd edition.Missouri Department of Conservation, Jefferson City.Porto, L. M., R. L. McLaughlin, and D. L. G. Noakes.1999. Low-head barrier dams restrict the move-mcnts of fishes in two Lake Ontario streams. North American Journal of Fisheries Management 19: 1028-1036.
Rabeni, C. F 1996. Prairie legacies:
fish and aquatic resources.
Pages 111-124 in F. B. Samson and F. L.Knoph, editors. Prairie conservation:
preserving North America's most endangered ecosystem.
Is-land Press, Washington, D.C.Rice, W. R. 1989. Analyzing tables of statistical tests.Evolution 43:223-225.
Shuman, J. R. 1995. Environmental considerations for assessing dam removal alternatives for river res-toration.
Regulated Rivers: Research and Manage-ment 11:249-261.
Smith, B., and J. B. Wilson. 1996. A consumer's guide to evenness indices. Oikos 76:70-82.Tabachnick, B. G., and L. S. Fidell. 1983. Using mul-tivariate statistics.
Harper and Row, New York.Tiemann, J. S. 2002. Effects of lowhead dams on fish and benthic invertebrate assemblage structure in the Neosho River, with comments on the threatened Ne-osho madtom, Noturus placidus.
Master's thesis.Emporia State University, Emporia, Kansas.USACE (U.S. Army Corps of Engineers).
2001. Available:
http://www.swt.usace.mil!COUNcharts.html. (Novem-ber 2001.)USGS (U.S. Geological Survey). 2001. Gauging station 07179730:
Neosho River near Americus.
Kansas.Available:
http://ks.water.usgs.gov/nwis/uv/?site-no
=07179730&PARAmeterTcd=00065,00060. (Novem-ber 2001.)Waters, T. F. 1995. Sediment in streams: sources, bio-logical effects, and control. American Fisheries So-ciety, Monograph
- 7. Bethesda, Maryland.Watters, G. T. 1996. Small dams as barriers to freshwater mussels (Bivalvia:
Unionoida) and their hosts. Bi-ological Conservation 75:79-85.Wildhaber, M. L., A. L. Allert, C. J. Schmitt, V. M.Tabor, D. Mulhern, K. L. Powell, and S. P. Sowa.2000a. Natural and anthropogenic influences on the distribution of the threatened Neosho madtom in a Midwestern warmwater stream. Transactions of the American Fisheries Society 129:243-261.
Wildhaber, M. L., V. M. Tabor, J. E. Whitaker, A. L.Allert, D. W. Mulhern, P. J. Lamberson, and K. L.Powell. 2000b. Ictalurid populations in relation to the presence of a main-stem reservoir in a Mid-western warmwater stream with emphasis on the threatened Neosho madtom. Transactions of the American Fisheries Society 129:1264-1280.
Williams, C. B. 1964. Patterns in the balance of nature.Academic Press, London.Wood, P. J., and P. D. Armitage.
1997. Biological effects of fine sediment in the lotic environment.
Environ-mental Management 21:203-217.
Zar, J. H. 1999. Biostatistical analysis, 4th edition. Pren-tice-Hall, Upper Saddle River, New Jersey.
TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE Vol. 107, no. 1/2 p. 17-24 (2004)Correlations among densities of stream fishes in the upper Neosho River, with focus on the federally threatened Neosho madtom Noturus placidus JEREMY S. TmMANN,l8 DAVID P. GILLETrE,Ib MARK L. WRDH"ER 2 AND DAVD R.*EDDS 1 1. Emporia State University, Department of Biological Sciences, Emporia, KS 66801 2. U.S. Geological Survey, Columbia Environmental Research Center, Columbia, MO 65201 a) Present* address: Illinois Natural History Survey, Center for Biodiversity, Champaign, IL 61820 o'tiemann@inhs.uiuc.edu) b) Present address: University of Oklahoma, Department of Zoology, Norman, OK 73019 We sampled fishes monthly from November 2000 to October 2001 at four gravel bar sites along a 34-km stretch of the upper 14eosho River in Lyon County, Kansas. We assessed the potential for interspecific competition among stream fishes, with focus on the federally threatened Neosho madtom, Noturus placidus, by using Pearson's correlation analysis with sequential Bonferroni correction of alpha to examine relationships among fish densities.
Of the 19 fish species analyzed, there were six significant positive and no significant negative correlations.
Abundance of N. placidus did not vary significantly with total abundance of fishes or with abundance of any of these potential competitors.
The lack of significant negative correlations at these sites*at this time might reflect an assemblage in equilibrium or one controlled abiotically rather than by ongoing active competition.
Keywords:
interspecific competition, stream fishes, gravel bar, Neosho madtom, Noturus placidus, upper Neosho River, Kansas.INTRODUCTION Several studies have addressed competition in stream fish assemblages (e.g. Matthews 1982;Roell and Orth 1994; Grossman et al. 1998), but only one (Wildhaber, Allert and Schmnitt 1999) has focused on potential interspecific competition with the federally threatened Neosho madtom Noturus placidus.
This small (generally
<75 mm in total length)ictalurid presently is distributed discontinuously in the Neosho (Grand) -Spring River system, which is located within the Prairie Parkland Province and Ozark Upland Province ecoregions in Kansas, Missouri and Oklahoma (Wildhaber, Allert and Schmitt 1999). Individuals typically are found in riffles and sloping gravel bars in moderate current, and prefer deposits of loosely compacted gravel where they nocturnally feed on insects (USFWS 1991;Cross and Collins 1995). The U.S. Fish and Wildlife Service (USFWS) listed N. placidus as threatened in 1991 (55 FR 21148), and suggested that populations might be limited by competition for resources (e.g. food and habitat) with other fishes (USFWS 1991). To address this question, Wildhaber, Allert and Schmitt (1999) sampled 12 gravel bars in the Neosho (from near the confluence with the Cottonwood to the Grand Lakes of the Cherokees) and Cottonwood (near Emporia, Kansas) rivers, and 20 gravel bars in the Spring River (downstream from its 18 1iemann, Gillette, Wildhaber and Edds confluence with its North Fork to the Grand Lakes of the Cherokees).
They concluded that interspecific competition was not limiting N. placidus populations based on positive correlations between densities .of N.placidus and other stream fishes with habitat preferences similar to those of N. placidus (e.g. suckermouth minnow Phenacobius mirabilis, juvenile channel catfish Ictalurus punctatus, and slenderhead darter Percina phoxocephala).
Many fishes inhabiting the Neosho River system potentially are interspecific competitors for resources (Wildhaber, Allert and Schmitt 1999), and their presence suggests that competition could limit N.placidus populations.
Noturus placidus densities are generally higher in the Cottonwood and Neosho rivers than in the Spring River (Wilkinson et al. 1996;Wildhaber, Allert and Schmitt 1999); within the Neosho River, densities are higher in the upper Neosho (upstream from its confluence with the Cottonwood) than in the lower Neosho (Wildhaber, Tabor et al. 2000). We asked whether the patterns observed by Wildhaber, Allert and Schmitt (1999) would hold for the upper portion of the Neosho River based on the premise that higher densities result in greater potential for interspecific competition (Strange, Moyle and Foin 1992). We examined correlations among densities of stream fishes, with focus on N. placidus, in the upper Neosho River.Little is known of potential interspecific
- competition with N. placidus in the upper Neosho; therefore, this study could aid the species' recovery.MATERIALS AND METHODS We sampled fishes monthly from November 2000 to October 2001 on four gravel bars along a 34-km stretch of the upper Neosho River in Lyon County, Kansas (Fig. 1). This segment of the Neosho River lies within the Prairie Parkland Province Ecoregion Site 1 coxoeUDm&l Americus IN 0 'I ]3 IIo.md Site 3 "q_ Emporia Dam WS Figure 1. Sampling sites along the Neosho River in Lyon County, Kansas.(Chapman et al. 2001), and is a 5th order stream with a mean gradient of 0.54 m/kin.To maximize the probability that N. placidus would be collected, collection sites were selected based on the presence of a gravel bar composed mainly of gravel < 64 mm in size (Fuselier and Edds 1994; Wildhaber, Allert and Schmitt 1999).Depending on depth (depths >1.25 m were not sampled) and landowner permission, three to five cross-channel transects perpendicular to the river channel were spaced equally along each gravel bar, and up to five sampling points were spaced equally along each transect.
At each point, we collected fishes from a 4.5 m 2 area by disturbing the gravel substrate 3 m upstream from a. stationary 1.5 rn long, 3 mm mesh seine. To minimize disturbance, we sampled transects from downstream to upstream, and points from near shore to far shore. All fishes were identified and counted upon completion of a point, and were released upon completion of a site.We pooled point data for each month at each site, and calculated mean total abundance of Transactions of the Kansas.Academy of Science 107(1/2), 2004 19 fishes per 100 M 2 , in addition to mean abundance of individual fish species per 100 in 2.We assumed that potential competitors were equally vulnerable to capture by our sampling method (Wildhaber, Allert and Schmitt 1999). We eliminated fishes occurring in < 5% of the samples (< two samples) from analyses (Gauch 1982). We compared densities at the site level to assess potential competition among fish species..Because of multiple tests, sequential*
Bonferroni correction of a standard a = 0.05 was applied to help control overall experimental Type I error (Rice 1989). All statistical tests were conducted using SYSTAT,'for Macintosh, Version 5.2 (SYSTAT, Inc., Evanston, IL). Distribution of means was evaluated for normality using the Shapiro-Wilk test (Zar 1999), and for homogeneity of variance using Levene's test (Milliken and Johnson 1984); non-normal variables were loglO transformed (Zar 1999). We used one-way analysis of variance (ANOVA) to test for differences in mean fish abundance among sites, and Pearson's correlation analysis to assess correlations among non-zero fish densities (Wildhaber, Allert and Schmitt 1999).RESULTS We collected 26 fish species representing 15 genera and nine families from 45 samples (Site 3 was frozen in December, January, and February).
Seven species occurred in < two samples, leaving 19 species for analysis.(Table 1). Mean total abundance of fishes was 315.1 (SE = 464.0) fish per 100 m 2.Mean individual species' abundance ranged from 1.2 (SE = 0.2) per 100 M 2 (longear sunfish Lepomis megalotis) to 120.1 (SE = 223.7)per 100 M 2 (red shiner Cyprinella Iutrensis);
mean N. placidus abundance was 3.6 (SE =1.8) per 100 m 2 (Table 1).ANOVAs did not indicate significance by site for total abundance of fishes or for any of the 19 species, thus allowing, meaningful comparisons by Pearson's correlation analysis.
With sequential Bonferroni adjustment of alpha, there were six significant positive and no significant negative correlations (Table 2). At cc = 0.05, there were an additional 12 (nine positive and three negative) correlations (Table 2). There were no significant correlations between N.placidus abundance and total abundance of fishes or with abundance of any of the other 18 fish species (range: P = 0.14 to P = 0.96).DiscussioN Even though some degree of competition among fishes undoubtedly occurs on a gravel bar, and changes with differing physical conditions (Fausch and White 1986), coexisting species can segregate into distinct microhabitats and partition resources (Ross 1986; Matthews 1998). Strange, Moyle and Foin (1992) suggested that deterministic (density-dependent) factors, including interspecific competition, occur when stochastic (density-independent) factors, including natural and anthropogenic disturbances, are not occurring.
In our study, the only significant correlations were positive, suggesting limited interspecific competition, among stream fishes at these sites at this time. All 11 species having significant correlations (Table 2) inhabit streams with permanent flow, moderate gradient, and gravel substrate, but can utilize different resources (Cross and Collins 1995;Pflieger 1997). Through coexistence, the fish assemblage might have evolved to where each species now demonstrates slight differences in food. (e.g. size or timing of.food eaten) or habitat preferences (e.g.velocity or substrate composition), thus reducing the level of competition (Matthews 1998). The lack of significant negative correlations likely reflects abiotic control (Grossman, Moyle and Whitaker 1982) or an evolved equilibrium (e.g. non-linear competitive hierarchies) in resource partitioning among members of the 20 Tiemann, Gillette, Wildhaber and Edds Table 1. Fish species and their mean abundances per 100 m 2 (standard error) collected in the upper Neosho River, November 2000 to October 2001. Asterisks
(*) indicate species that occurred in <5% of the 45 samples (< two samples) and were excluded from analysis.Mean Family Scientific name Common name anan abundance Campostoma anomalum Central stoneroller 5.1 (4.9)Cyprinella camura* Bluntface shiner 0.0 (0.0)Cyprinella lutrensis Red shiner 120.1 (223.7)Notropis buchanani Ghost shiner 38.9 (50.7)Family Cyprinidae Notropis stramineus Sand shiner 9.6 (9.6)Phenacobius mirabilis Suckermouth minnow 4.8 (4.5)Pimephales notatus Bluntnose minnow 33.4 (39.8)Pimephales tenellus Slim minnow 12.4 (8.5)Pimephales vigilax Bullhead minnow 25.0 (46.9)Family Catostomidae Moxostoma erythrurum*
Golden redhorse 0.0 (0.0)Ictalurus punctatus Channel catfish 6.5 '(9.3)Family Ictaluridae Noturus placidus Neosho madtom 3.6 (1.8)Noturus flavus Stonecat 2.4 (1.7)Family Fundulidae Fundulus notatus* Blackstripe topminnow 0.0 (0.0)Family Poeciliidae Gambusia affinis* Western mosquitofish 0.0 (0.0)Family Moronidae Morone chrysops*
White bass 0.0 (0.0)Lepomis cyanellus Green sunfish 6.4 (5.3)Lepomis humilis Orangespotted sunfish 16.9 (30.2)Family Centrarchidae Lepomis macrochirus Bluegill 2.2 (1.0)Lepomis megalotis Longear sunfish 1.2 (0.2)Etheostoma spectabile Orangethroat darter 8.5 (8.8)Etheostoma flabellare*
Fantail darter 0.0 (0.0)Family Percidae Percina caprodes Logperch 2.8 (1.8)Percina copelandi Channel darter 3.6 (2.2)Percina phoxocephala Slenderhead darter 12.0 (13.1)Family Sciaenidae Aplodinotus grunniens*
Freshwater drum 0.0 (0.0)V Transactions of the Kansas Academy of Science 107(1/2), 2004 21 Fish species r (P-value)combination Pimephales vigilax 0.98 (<0.0001)Lepomis humilis Campostoma anomalum 0 Phenacobius mirabilis
.89 (0.0001) *Pimephales notatus Percina phoxocephala 0.82 (0.001) *Pimephales tenellus Lepomis megalotis Percina caprodes 0,8 (0.003)*Notropis buchanani* P. mirabilis P mirbilis0.77 (0.004)*Noturus flavus C. anomalum N. flavus 0.74 (0.006)P notatus P tenellus 0.72 (0.009)P. tenellus P tenllus0.70 (0.01)Percina copelandi P. notatus P noatus0.70 (0.01)P copelandi C. anomalum 0.68 (0.02)P caprodes P mirabilis N. buchanani C. anomalum 0 (0.03)N. buchanani P. notatus L. ntatuds 0.63 (0.03)L. me ga/otis Ictalurus punctatus 0.59 (0.04)Lepomis macrochirus Etheostoma spectabil
-0.58 (0.04)P caprodes E. spectabile
-0.58 (0.04)P copelandi E. spectabile
-0.58 (0.04)P notatusTable 2. Pearson's correlation analysis [r (P-value)]
between mean site densities of significant fish species combinations collected in the upper Neosho River, November 2000 to October 2001. Asterisks (*) indicate correlations significant at sequential-Bonferroni adjusted alpha value.assemblage (Connell 1980) rather than ongoing active competition.
Differentiation between these two premises was beyond the scope of the present study.Wildhaber, Tabor et al. (2000) reported greater mean overall densities of N. placidus (number per 100 m 2) in the upper Neosho (19.8) than in the lower Neosho (5.6). Their collections were made between August and October, a time when N. placidus densities are typically highest due to young-of-year recruitment (Moss 1983; Wilkinson et al.1996). Our mean density of 3.6 was calculated from monthly collections throughout the year. Fuselier and Edds (1994), sampling throughout the year, reported a density of 3.3 in the Cottonwood River. Bulger and Edds (2001), sampling from April to October, recorded a density of 4.5 in the upper Neosho and 1.9 in the Cottonwood.
Other reports of N. placidus densities in the Neosho River mainstem did not differentiate upper and lower portions of the river, including Moss (1983) with 11.7 from July to October, Wenke et al. (1992)with 6.8 in December and March, Eberle and Stark (1995) with 22.3 in October, and Wildhaber, Allert and Schmitt (1999) with 12.0 from August. through October. Densities in Spring River are generally lower (Edds and Dorlac 1995 with 0.9; Wilkinson et al. 1996 with 2.4; Wildhaber, Allert and Schmitt 1999 with 3.3.). Differences in densities reported in these studies could be attributed to seasonal or annual variation, or to differences in collectors, sampling efficiency, quadrat size, or habitat quality.
22 Given the lack of significant negative correlation, interspecific competition does not appear to be limiting N. placidus populations at these sites at this time in the upper Neosho River, contrary to the hypothesis of the USFWS (1991). This finding is similar to that of Wildhaber, Allert and Schmitt (1999), who found significant positive correlations between N. placidus and three fishes with habitat preferences similar to those of N. placidus (P mirabilis, L punctatus, and P phoxocephala) in the Neosho, Cottonwood, and Spring rivers combined.
However, they found significant negative correlations between N. placidus and three fishes (bluntnose minnow Pimephales notatus, slim minnow P tenellus, and bullhead minnow P vigilax) with habitat preferences dissimilar to those of N.placidus, whereas we found no significant negative correlations.
Ross (1986) suggested that more distantly related species (similar to.Pimephales sp. and N. placidus in Wildhaber, Allert and Schmitt 1999) segregated more on resources (e.g. space or time) than did closely related species.For two species to coexist, they need to segregate along one or more resources (e.g.separation of feeding activity), which would reduce competition to a level at which both species could persist (Gause 1934). For example, Noturus species are dominant food consumers during the night, whereas other fishes (e.g. minnow and darters) are dominant food consumers during the day, thus avoiding direct competition for food resources (Burr and Stoeckel 1999). Noturus placidus abundance was positively correlated with macroinvertebrate abundance along the same stretch of river, but was limited by habitat, as was macroinvertebrate abundance (Tiemann 2002). Previous research on N. placidus has suggested that anthropogenic factors, including impoundments (Wildhaber, Tabor et al. 2000; Tiemann 2002), and environmental contaminants (Wildhaber, Allert et al. 2000), are limiting N. placidus populations, and Tiemann, Gillette, Wildhaber and Edds might reduce opportunities for deterministic biotic interactions, including competition.
Given the influence of these stochastic factors, it is difficult to assess effects of deterministic factors on N. placidus populations in the field; however, additional research could be conducted in the laboratory.
and field to better understand effects of deterministic factors on N. placidus.ACKNOWLEDGEMENTS This research was supported by the U.S.Geological Survey (USGS), Department of Interior, under USGS Cooperative Agreement No. OOCRA0025, with addition funding from an Emporia State University (ESU) Faculty Research and Creativity Grant and the ESU Department of Biological Sciences.
K.Cummings, G Levin, T. Rice, C. Taylor, and D.Thomas, of the Illinois Natural History Survey provided comments.and suggestions, and the Kansas Department of Wildlife and Parks and the U.S. Fish and Wildlife Service issued collecting permits. Landowners (G Gulde, P.and D. Matile, L. Schlesener, and ESU Natural Areas) allowed access to the river on their property.
D. Moore and D. Zelmer of ESU gave statistics and study design advice, R.Sleezer and L. Sneed of ESU helped create the map of the study site, and ESU students B.Chance, J. Dean, B. Harkins, and S. Sherraden assisted with fieldwork.
LrrEiATuRE cnmr Bulger, A.G and Edds, D.R. 2001. Population structure and habitat use in Neosho madtom (Noturus placidus).
Southwestern Naturalist 46(1), p. 8-15.Burr, B.M. and Stoeckel, J.N. 1999. The natural history of madtoms (Genus Noturus), North America's diminutive catfishes.
In Irwin, E.R., Hubert, W.A., Rabeni, C.F., Schramm, H.L. Jr. and Coon, T. (eds.), Catfish 2000: Proceedings of the international ictalurid symposium, p. 51-0 Transactions of the Kansas Academy of Science 107(1/2), 2004 23 101. American Fisheries Society, Symposium 24, Bethesda, Maryland.* Chapman, S.S., Omernik, J.M., Freeouf, J.A Huggins, D.G., McCauley, J.R., Freeman C.C., Steinauer, G., Angelo, R.T and-* Schlepp, R.L. 2001. Ecoregions of Nebraska and Kansas (color poster with map, descriptive text, summary tables, ai photographs).
U.S. Geological Survey, Reston, Virginia.* Connell, J.H. 1980. Diversity and the coevolution of competitors, or the ghosi competition past. Oikos 35(2), p. 131-1 Cross, F.B. and Collins, J.T. 1995. Fishes i Kansas, 2nd ed. University Kansas Museum of Natural History, Lawrence, p.Eberle, M.E. and Stark, W.J. 1995.Distribution and abundance of the Neosl madtom (Noturus placidus) in Kansas, an assessment of the amount of suitable habitat. Unpub. Report to the Kansas Department of Wildlife and Parks, Pratt, P.Edds, D.R. and Dorlac, J.H. 1995. Survey the fishes of the Spring River Basin in Missouri, Kansas, and Oklahoma, with emphasis on the Neosho madtom. Unpu Report to the Kansas Department Wildli and Parks, Pratt, 43 p.Fausch, K.D. and White, R.J. 1986.Competition among juveniles of coho salmon, brook trout, and brown trout in laboratory stream, and implications for Great Lakes tributaries.
Transactions American Fisheries Society 155(3), p.381.Fuselier, L. and Edds, D.R. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleostei:
Ictaluridae:
Noturuw placidus).
Southwestern Naturalist 39(: p. 217-223.Gauch, H.G Jr. 1982. Multivariate analysh community ecology. Cambridge Univer Press, Cambridge, United Kingdom, 29E Gause, G.F. 1934. The struggle for existen Macmillan (Hafner.Press), New York, 11 p.Grossman, GD., Moyle, P.B. and Whitaker, J.O. Jr. 1982. Stochasticity in structural and functional characteristics of an Indiana stream fish assemblage:
A test of community theory. American Naturalist 120(4), p. 423-454.Grossman, GD., Ratajczak, R.E. Jr., Crawford,.id ' M. and Freeman, M.C. 1998. Assemblage organization in stream fishes: effects of environmental variation and interspecific interactions.
Ecological Monographs t of 68(3), p. 395-420.138. Matthews, W.J. 1982. Small fish community n structure in Ozark (Arkansas and Missouri,*' USA) streams: Structured assembly 315 patterns or random abundance of species?American Midland Naturalist 107(1), p. 42-54.1o Matthews, W.J. 1998. Patterns in freshwater wVith fish ecology. Kluwer Academic Publishers, Norwell, Massachusetts, 756 p.Milliken, G.G. and Johnson, D.E. 1984.10 Analysis of messy data, Volume I: Designed experiments.
Wadsworth, Inc., Belmont, of California, 473 p.Moss, R.E. 1983. Microhabitat selection in Neosho River riffles. Unpub. Doctoral b. dissertation, University of Kansas, ife Lawrence, 287 p.Pflieger, W.L. 1997. The fishes of Missouri, 2nd ed. Missouri Department of Conservation, Jefferson City, 372 p.I Rice, W.R. 1989. Analyzing tables of statistical tests. Evolution 43(1), p. 223-225.363- Roell, M.J. and Orth, D.J. 1994. The roles of predation, competition, and exploitation in the trophic dynamics of a warmwater stream: a model synthesis, analysis, and application.
Hydrobiologia 291(3), p. 157-3), 178.Ross, S.T. 1986. Resource partitioning in fish 3 in assemblages:
a review of field studies.sity Copeia 1986(2), p. 352-388.1 p. Strange, E.M., Moyle, P.B. and Foin, T.C.ce. 1992. Interactions between stochastic and 63 deterministic processes in stream fish 24 community assembly.
Environmental Biology Fishes 36(1), p. 1-15.Tiemann, J.S. 2002. Effects of lowhead dams on fish and benthic invertebrate assemblage structure in the Neosho River, with comments on the threatened Neosho madtom, Noturus placidus.
Unpub.Master's thesis, Emporia State University, Emporia, Kansas, 71 p.United States Fish and Wildlife Service. 1991.Neosho madtom recovery plan. U.S. Fish and Wildlife Service, Denver, Colorado, 42 p.Wenke, T.L., Eberle, M.E., Ermsting, G.W. and Stark, W.J. 1992. Winter collections of the Neosho madtom (Noturus placidus).
Southwestern Naturalist 37, p. 330-333.Wildhaber, M.L., Allert, A.L. and Schmitt, C.J.1999. Potential effects of interspecific competition on Neosho madtom (Noturus placidus) populations.
Journal Freshwater Ecology 14(1), p. 19-30.Wildhaber, M.L., Allert, A.L., Schmitt, C.J., Tabor, V.M., Mulhern, D.W., Powell, K.L.Tiemann, Gillette, Wildhaber and Edds and Sowa, S.P. 2000. Natural and anthropogenic influences on the distribution of the threatened Neosho madtom in a midwestern warrnwater stream.Transactions American Fisheries Society 129(1), p. 243-261.Wildhaber, M.L., Tabor, V.M., Whitaker, J.E., Allert, A.L., Mulhern, D.W., Lamberson, P.J. and Powell, K.L. 2000. Ictalurid populations in relation to the presence of a main-stream reservoir in a midwestern warmwater stream with emphasis on the threatened Neosho madtom. Transactions American Fisheries Society 129(6), p.1264-1280.
Wilkinson, C., Edds, D.R., Dorlac, J., Wildhaber, M.L., Schmitt, C.J. and Allert, A. 1996. Neosho madtom distribution and abundance in the Spring River.Southwestern Naturalist 41(1), p. 78-81.Zar, J.H. 1999. Biostatistical analysis, 4th ed.Prentice-Hall, Upper Saddle River, New Jersey, 663 p./
TANSACToNs OF THE KANSAS ACADEMY OF SCIENCE 105(3-4), 2002. pp. 106-124 Breeding Behavior and Reproductive Life History of the Neosho Madtom, Noturus placidus (Teleostei:
Ictaluridae)
ANGELA G. BULGER,' CRIusToPHER D. WILKINSON, 2 AND DAVID R. EDDS Department of Biological Sciences, Emporia State University Emporia, Kansas 66801, e-mail: AGBulger@pbsj.com MARK L. WILDHABER U.S. Geological Survey, Columbia Environmental Research Center Columbia, Missouri 65201 The Neosho madtom, Noturus placidus, is a small catfish listed by the U. S. Fish and Wildlife Service as threatened.
Little is known of its breeding biology and behavior because high turbidity and flow during its spawning season prevent direct observation in the field, and captive propagation has met with limited success. We held Neosho madtoms in laboratory aquaria in 1996 and 1998 to study sexual dimorphism during breeding season, court-ship and nesting behavior, egg and clutch size,a*and embryological and larval development.
We also attempted to induce. spawning.
Courtship behaviors were recorded on videotape, including "carousel" and "tail curl" displays in which the fish spun in circles, head to tail, then quivered, with the male's tail wrapped around the female's head. Three clutches were observed, all in nest cavities that had been excavated by the fish under a structure; one clutch (1996) consisted of approximately 60 eggs, with a mean chorion diameter of 3.1 mm, and two (1998) consisted of approximately 30 eggs, with mean diameter of 3.7 mm. In all situations, eggs hatched after eight or nine days, and yolk-sacs.
were fully depleted seven days later. One spawn (1998) oc-curred after two days of injection with synthetic hormone. Male parental care of eggs and larvae was observed in 1996. Larvae remained in the nest until yolk-sacs were absorbed, after which they dispersed throughout the tank. Dissection of two females that laid clutches in this study revealed previtellogenic eggs in the lumen of ovaries, with a mean chorion diameter of 0.9 mm. Swollen lips of males, distended abdomen of females, and dif-ferences in head shape, premaxillary tooth patch coloring, and genital pa-pillae of breeding males and females were documented during spawning periods.2 Present address: PBS&J, 206 Wild Basin Road, Suite 300, Austin, Texas 78746-3343.
2 Present address: Department of Water Resources, Fish Passage Improvement Program, P.O.Box 942836, Sacramento, California 94236-0001.
IOLUME 105, NUMBERS 3-4 107 INTRODUCTION Madtoms (Noturus) are a group of small North American catfishes (Ic-aluridae).
Information about madtom life histories was scarce until concern"or these catfishes, several of which are protected, prompted numerous eco-ogical studies in the 1980s (see reviews in Dinkins and Shute, 1996 and 3urr and Stoeckel, 1999). No direct observations of spawning behavior have)een made in the wild, thus descriptions of breeding behavior within the group are limited. However, clutch size, mean chorion diameter, embryonic levelopment, time to hatching, and larval development and growth have een investigated by transferring clutches from the field to the lab for study)r via captive propagation (Clark, 1978; Bowen, 1980; Mayden, Burr, and Dewey, 1980; Burr and Dimmick, 1981; Mayden and Burr, 1981; Burr and Vayden, 1982, 1984; Mayden and Walsh, 1984; Starnes and Starnes, 1985;Vives, 1987; Baker and Heins, 1994; Pfingsten and Edds, 1994; Chan, 1995;Dinkins and Shute, 1996). Although captive propagation has had limited;uccess (Shute, Shute, and Rakes, 1993), it has allowed breeding behavior:o be described for the brindled madtom, N. miurus (Bowen, 1980), and the 3rown madtom, N. phaeus (Chan, 1995), during laboratory spawns.Little is known about breeding biology of the Neosho madtom, N. pla-7idus. The species occurs only in the Neosho, Cottonwood, and Spring rivers:f Kansas, Oklahoma, and Missouri, and it was listed by the U. S. Fish and Wildlife Service (USFWS) as threatened on 22 May 1990 (55 ER. 21148).Understanding breeding biology and behavior of the Neosho madtom is zritical to its recovery (USFWS, 1991); however, high turbidity and flow during its spawning season prevent field observations.
Moss (1981) exam-ined museum specimens and characterized the spawning season of this spe-cies as beginning in March with egg development and continuing through July, when young-of-year first appear in samples. Sexual dimorphism is present during spawning season; characteristics include reddening of the premaxillary tooth patch and swelling of the genital papilla of males and females, swelling of lips and cephalic epaxial muscles of males, and disten-tion of the abdomen of females (Moss, 1981; Pfingsten and Edds, 1994;Edds and Wilkinson, 1996; Wilkinson and Edds, 1997). However, determin-ing Neosho madtom sex using external characteristics is difficult, even when secondary sex characteristics are well developed (Bulger and Edds, 2001), as in many other madtom species (Burr and Mayden, 1984; Simonson and Neves, 1992). Improved ability to sex Neosho madtoms is necessary to determine sex ratios, to evaluate differences in habitat use between the sexes during spawning season, and to pair individuals for captive propagation.
Previous attempts to induce spawning in captive Neosho madtoms led to discovery of one clutch of 63 eggs deposited under a cinder block in a flowing aquarium; however, these eggs did not develop and may not have 108 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE been fertilized (Pfingsten and Edds, 1994). We held Neosho madtoms in aquaria at the Columbia Environmental Research Center (CERC), Columbia, Missouri, and Emporia State University (ESU), Emporia, Kansas, to observe spawning behavior, nesting, parental care, clutch size, egg size, and embry-onic and larval development and growth, to define characteristics of sexual dimorphism during spawning season, and to investigate use of synthetic hormone to induce spawning.MATERIALS AND METHODS 1996 ESU On 16 May 1996, we collected 10 adult Neosho madtoms from the Ne-osho River, Lyon County, Kansas, and brought them into the lab at ESU.One male and one female were placed into each of three 38-L static aquaria, and two males and two females were placed in a 700-L static aquarium.
The bottom of each aquarium was covered with river gravel, and two 13-cm lengths of 10.5-cm diameter PVC pipe, cut in half lengthwise, were added to each aquarium to provide cover, while allowing visual observation.
We used aquarium heaters to slowly raise water temperature from 21.5 to 25"C between 23 May and 13 June, and photoperiods were held at a 13.5-h light: 10.5-h dark cycle using timed fluorescent lights. Red lights (25 W) were illuminated prior to the dark cycle to allow for nighttime behavioral obser-vations; Boujard, Morean, and Lugnet (1992) demonstrated that other cat-fishes displayed normal nocturnal activity under exposure to red light if it was the lowest intensity light throughout the photoperiod.
We periodically recombined, individuals in an attempt to match males and females that, based on development of secondary sex characteristics, seemed to have the greatest potential to breed. Fish were fed frozen brine shrimp and wild-caught chi-ronomid larvae.1998 CERC We collected adult Neosho madtoms from the Cottonwood River, Lyon and Chase counties, Kansas, on 23 June and 7 July 1998 and transported them to the CERC. Individuals were sexed according to development of secondary sex characteristics, and seven male/female pairs were placed in separate 29.5-L aquaria housed in an isolation chamber (Bulger, Wildhaber, and Edds, 2002). The chamber prevented entry of external light and damp-ened sound disturbances.
Each aquarium had an airstone, a standing drain-pipe to maintain approximately 20-cm water depth, and inflow from a well, creating a turnover rate of ca. 8.6 L h-1. Gravel was placed on the bottom of each aquarium and a 12.5-cm PVC pipe (10-cm diameter, cut in half lengthwise) was provided for shelter. We held water temperatures between 24 and 28°C using aquarium heaters, and photoperiod was a 16-h light: 8-h OLUME 105, NUMBERS 3-4 109 ark cycle using timed fluorescent lights mounted in the chamber. Fish were-d frozen brine shrimp six days a week and live blackworms once a week.ifra-red lights and time-lapse video equipment were used to continuously lonitor fish behavior (Bulger, Wildhaber, and Edds, 2002). Cavity enhance-ient and courtship behaviors were defined using the ethogram established y Bulger, Wildhaber, and Edds (2002).998 ESU We collected adult Neosho madtoms from the Neosho and Cottonwood ivers, Lyon and Chase counties, Kansas, from 26 April to 7 July 1998 and"ansported them to the lab at ESU. Individuals were sexed, paired, and laced in static aquaria ranging from 38-L to 192-L. The bottom of each quarium was covered with gravel. Each aquarium had an airstone, a flow-arough charcoal filter, and structure in the form of large, flat cobble (10 to.5 cm), freshwater mussel halves, half PVC pipes, aluminum cans or com--inations of these structures.
Fish were held in water temperatures ranging rom 24 to 31'C, controlled with aquarium heaters, and in a 16-h light: 8-h ark photoperiod, controlled by overhead lights on a timer. Fish were fed rozen brine shrimp and wild-caught chironomids every 1 to 3 days and, ince each week, aquatic insect larvae collected from nearby rivers. Red ights (25 W) were illuminated prior to the dark cycle to allow for nighttime iehavioral observations.
On July 12 (after no breeding had occurred) individuals were given a.25-pd preliminary injection of Ovaprim, a synthetic hormone used to nduce breeding in many fishes (Syndel Laboratories, Vancouver, B.C., ca.0.5 ml kg-') (J. Stoeckel, Arkansas Tech Univ., pers. comm.); 4 h later, each vas given a full dose of 2.5 jil. We administered injections of 2.5 pLl daily intil 20 July, when dosage was increased to 3 pl; dosage was increased on!2 and 23 July to 5 pil each and on 24 July to 10 ipl.)ex determination In addition to recording sexual dimorphism in the Neosho madtoms used luring this study, we examined preserved individuals from the ESU study:ollection for development of secondary sex characteristics, especially the ,enital opening, then dissected them to determine sex internally.
We de-;cribed and compared differences between males and females with well-leveloped secondary sex characteristics, as well as individuals without.RESULTS 1996 ESU On 8 July 1996, a clutch of approximately 60 viable eggs and eight yolk-ess membranes was discovered in a 38-L static aquarium with a 61-mm 110 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE Table 1. Characteristics of three N. placidus clutches from this study compared with data from a previous study (Pfingsten and Edds, 1994).X chorion Time to yolk Water Clutch diameter Time to hatching, absorption, 9 TL temp.Clutch size (SD, n) 9 TL (SD, n) (SD, n) 0 C ESU 1996 -60 3.1mm ---8 to 9 d 7 d 25.0 (0.15, 3) 6.8 mm (0.27, 4) 13.3 mm (0.94, 3)ESU 1998 -32 3.7 mm 9 d 9 d 25.0 (0.10, 7) 8.8 mm (0.20, 2) 13.0 mm (0, 2)CERC -.28.0 Pfingsten and Edds 63 3.1 mm 26.5 (1994) (0.20, 10)total length (TL) female and a 78-mm TL male. Water temperature in the tank was 25TC at the time the eggs were discovered.
The eggs had been laid in a gravel depression under the half PVC pipe. Gravel had been manipulated by one or both adults to form a depression to the glass tank bottom and was piled against the open ends of the PVC pipe, almost entirely blocking the entrances to the nest. After the male was observed chasing the female from the nest, she was removed from the tank. The male remained in the nest with the clutch and was observed fanning the eggs with his tail and rubbing them vigorously with his mouth and body.Three eggs were removed from the tank and placed in a 1.9-L glass aquarium with an airstone.
At the time of discovery, the eggs had chorion diameters of 3.3 mam, 3.0 mm, and 3.1 mm (Table 1), and the embryos already were undergoing organogenesis, with their tails separated from their yolk sacs and moving regularly from side to side. Rudimentary optic and otic vesicles were observed in the head of embryos and myomeres were visible in the tail. Based on embryonic development in Noturus exilis (May-den and Burr, 1981) and N. hildebrandi (Mayden and Walsh, 1984), the embryos appeared to be approximately 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> old at the time of discovery.
During the next four days, we video recorded daily observations of the three isolated eggs under a dissecting microscope.
Development was rapid;on 10 July (Day 6 embryo) eyes had developed further and vitelline vessels were present, but the eggs had become infected with fungus. We treated the eggs by immersing them in a solution of 1.9 ml formalin/3.79 L water for five minutes. After treatment, eggs were suspended in a small plastic basket over the airstone to receive constant aeration.
On the morning of 11 July (Day 7 embryo) distinct pupils in the eyes, pectoral fin buds, rudimentary barbels, and rhythmic opercular movements were observed in all embryos.In addition, tail movements were less regular and much slower. On 12 July (Day 8 embryo), the embryos appeared crowded in their eggs and exhibited "OLUME 105, NUMBERS 3-4 III ttle movement except for periodic shuddering, during which they flexed ieir tails back and forth vigorously.
Opercular movements and blood flow 1 the vitelline veins were continuous.
Two of these three eggs hatched on 2 July. The eggs that had developed in the nest with the male began hatch-.ig on 13 July, and all had hatched by the morning of 14 July. We suspect 3at handling of the three eggs used for observation resulted in premature atching of those eggs (Burr and Dimminck, 1981).We recorded the hatching events on video and made measurements of the olk-sac larvae at time of hatching, using National Institutes of Health im-ging software.
Total lengths of four larvae at hatching were 6.41 mm, 6.92 am, 6.96 mm, and 6.99 mm (9 = 6.82 mm, SD = 0.27), and few mela-Lophores were visible on the heads. Growth and development of larvae were apid. On 15 July (Day 3 larva), two larvae had total lengths of 11.4 mm nd 12.8 mm. They had melanophores over the front half of their body,'elvic fins were beginning to form, and their yolk sacs were noticeably educed in size. On 20 July (Day 7 larva), larvae had a mean TL of 13.3 am (SD = 0.94, n = 3), their dorsal side was covered with melanophores,.nd the yolk sac was entirely depleted in most individuals.
Approximately half of the larvae were kept in a separate 38-L static aquar-um ("nursery tank"). The larvae in this tank were observed first eating ecently hatched brine shrimp nine days after hatching.
They were subse-luently fed brine shrimp and crushed frozen chironomid larvae twice daily'or about 30 days. Larvae left in the breeding tank were fed brine shrimp md chironomid larvae sporadically, but we made periodic introductions of iver water and insects into the tank to provide a supplemental food source.kfter hatching, the yolk-sac larvae left in the breeding tank remained in the lest, forming a loose aggregation.
Although we did not observe the male'anning the larvae as he had the eggs, he often was observed hovering over hem. Approximately 8 to 10 days after hatching (21 July to 23 July), yolk-,acs were absorbed, and larvae dispersed throughout the tank and burrowed nto the gravel.Growth rates we observed in the lab may not be typical, because different rowth rates occurred in the two tanks (Wilkinson and Edds, 1997). Larvae cept in the nursery tank were fed brine shrimp and chironomid larvae more)ften and grew more rapidly than those left in the breeding tank. On 31 ruly, mean TL of larvae in the nursery tank was greater than that in the lesting tank (F,.2 = 18.7, P = 0.0001).1998 CERC At CERC, from late June to mid July, male and female pairs were ob-;erved via time-lapse video "carouselling," or swimming in circles head to:ail over the gravel substrate under the PVC structure (Fig. 1A). This be-iavior was followed typically by the "tail curl," in which individuals lay 112 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE Figure 1. A. "Carousel," courtship behavior of Neosho madtom. Male and female swim in circles head to tail near substrate.
B. "Tail curl," courtship behavior of Neosho madtom.Male and female lie above substrate with tail of male wrapped around head of female while both quiver.with the male's tail wrapped around the female's head (Fig. 1B). In this position the two would quiver slightly then separate.
After separation, the process was repeated frequently, beginning with the "carousel," although occasionally the female was chased from under the structure.
Cavity en-hancement behaviors (Bulger, Wildhaber, and Edds, 2002) also were ob-served: males and females nudged rocks with their heads, removed rocks in their mouths, and spun in circles alone over the gravel substrate under the PVC structure.
Specific details of cavity enhancement, including depth of cavity and mean rock size within cavities, were described by Bulger, Wild-haber, and Edds (2002).On 18 July, a clutch of approximately 30 eggs was discovered in a tank containing a male (61 mm TL) and female (59 mm TL) that had been captured on 23 June; water temperature was 28*C (Table 1). Eggs were spherical, with a yellow yolk in the center surrounded by a clear chorion, and they were adhered to one another in a cluster. Both male and female VOLUME 105, NUMBERS 3-4 113 Table 2. Ovary length and number of eggs in three reproductive female N. placidus.
CERC and ESU individuals had bred and eggs in ovaries were previtelline.
Specimen from ESU teaching collection was a gravid female, and eggs contained yolk.Fish TL Ovary length Number I chorion diameter Specimen (mm) (mm) of eggs (SD, n)CERC 59 9.0, 9.0 13, 17 0.8 (0.13, 12)ESU 67 12.0, -21, 23 0.9 (0.11, 10)Teaching collection 68 16.0, 16.0 39, 40 2.5 (0.19, 10)were present with the egg mass under the PVC structure at the time of discovery, but the female was removed from the tank when the male was observed biting and chasing her away after she ate approximately eight eggs.Following removal of the female, the male rested near the eggs, hovered over them, and fanned them with his tail. The eggs disappeared between the second and third day, at which time the male had a distended abdomen; it is presumed he ate the eggs.Gravel pushed against the front of the tank by the fish and the position of the fish in a depression (nest) under the structure prevented detailed ob-servation of spawning.
Unlike other N. placidus observed in captivity (Bul-ger, Wildhaber, and Edds, 2002), video record showed that the pair spent the majority of that day in the nest together and were active, performing the"'carousel" followed by the "tail curl," which lasted from 1 to 7 min, several times throughout the day. When resting, the two usually lay side by side, often touching.
After dark (2122 h), one individual, presumably the male, remained in the nest and the other, presumably the female, left and returned to the nest several times. As night progressed, that individual left the nest less frequently, and circles made during the "carousel" became smaller and faster. Beginning at approximately 0300 hours0.00347 days <br />0.0833 hours <br />4.960317e-4 weeks <br />1.1415e-4 months <br />, while both fish were under the structure "carouselling" and resting, the tail of one was seen flipping above the substrate of the nest. This was observed until the female was chased from the nest shortly after the lights came on in the chamber at 0524 h. Several times prior to discovery of the clutch, the female returned to the nest and was chased away by the male. Dissection of the female revealed two, 9.0-mm long, slightly pink ovaries containing 13 and 17 white, prev-itelline eggs, with a mean chorion diameter of 0.8 mm (Table 2).1998 ESU At ESU, on 14 July, after one preliminary and two full dosage injections of Ovaprim, a clutch of eggs was discovered in a 38-L aquarium; water 114 14TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE temperature was 27°C. The clutch consisted of two clusters of eggs, one of 14 and one of 18. Mean chorion diameter was 3.7 mm (SD = 0.10, n = 7)(Table 1); perivitelline space was approximately 0.4 mm. Eggs were spher-ical, with a yellow yolk in the center surrounded by a clear chorion and, except for division of the two clusters, adhered to one another. Both the male (73 mm TL) and female (67 mm TL) had been captured on 7 July.They were resting near the eggs in a slight cavity or depression (nest) in the gravel under the only cover in the tank, a flat rock (8 X 14 cm). After discovery of the eggs, the female was removed and examined.
Her urogenital pore was brown with coagulated blood, rather than white, as before, and her abdomen was not as distended as it had been; dissection revealed ovaries that contained 21 and 23 white, previtelline eggs, with a mean chorion di-ameter of 0.9 mm (SD = 0.11, n = 10) (Table 2).The 14-egg cluster was removed and placed in a hatching apparatus; the 18-egg cluster was left in the nest with the male. By Day 3, however, it was apparent the eggs in the nest were not being cared for; eight eggs had cloud-ed, indicating embryonic development had ceased. By Day 3, nine eggs in the hatching apparatus had broken or clouded, so all remaining eggs were removed and placed in a watch glass with an airstone; temperature varied from 21 to 27TC. By Day 4, only two eggs contained a developing embryo.Beginning on Day 5, these eggs were treated for fungus daily by immersing them in 0.5% formalin for 2 to 7 min.Based on development of N. exilis (Mayden and Burr, 1981) and previous observations of N. placidus from 1996, the eggs were estimated to be less than 10 h old at the time of discovery.
Cleavage was in progress, the animal pole consisted of approximately 18 cells (blastomeres) in a small cluster (Fig. 2A), and the chorion was clear, with a rough surface. By approximately 24 h post-fertilization, approximately 35 blastomeres at the animal pole had extended into the perivitelline space (Fig. 2B). At approximately 92 h post-fertilization, organogenesis had begun. A head, with pronounced eye cups, was present and partially separated from the yolk; the tail also was separated from the yolk, and it whipped back and forth in the perivitelline space.Somites were differentiated from just posterior of the head to the tip of the tail, and vitelline veins extended on the yolk from either side of the fish (Fig. 2C). Approximately 120 h post-fertilization, the head was larger, eye lenses were developed, and the spinal cord was visible from the head to the tip of the tail (Fig. 2D). At approximately 140 h, opercles were beginning to form at the base of the head, as were barbels near the mouth (Fig. 2E).At approximately 165 h, somite differentiation throughout the length of the body was no longer as distinct, opercles were better defined, pectoral spines were developing at the base of the head, and fin-forming caudal ray pri-mordia were present at the tip of the tail (Fig. 2F). Approximately 190 h post-fertilization (estimated 10 h prior to hatching), barbels were well de-VOLUME 105, NUMBERS 3-4 115 Figure 2. Development of Neosho madtom eggs laid 14 July 1998. A and B are lateral views, C through G are dorsal views. A = Day 1 (10 h), B = Day 2 (24 h), C = Day 4 (92 h), D = Day 5 (120 h), E = Day 6 (140 h), F = Day 7 (165 h), G = Day 8 (190 h).veloped, eyes were more dorsally situated on the head, and pectoral fins were better defined (Fig. 2G); however, eggs were covered with fungus by this time, making observation difficult.
On the morning of 22 July (approx-imately 200 h post-fertilization, Day 9), in water 25.5°C, both eggs hatched.One individual was 9.0 mm TL and the other was 8.6 mm TL. Both were light yellow with black eyes. and had a well-developed yolk sac (Fig. 3).Following hatching, larvae sought cover in gravel added to the dish. Yolk sacs were depleted by 30 July, when individuals were 13.0 mm TL. Stellate melanophores were spread over the body, and pelvic fins were completely developed by the time of yolk sac depletion, 9 days after hatching.
116 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE Figure 3. Larval Neosho madtom on day of hatching, 22 July 1998; TL = 9.0 mm. Two pairs of barbels are present, pectoral spines are beginning to develop; and vitelline veins are visible on yolk sac.Twice on 14 July and once on 15 July, it was observed that the female of a breeding pair was spent, her belly no longer distended, as it had been the day before, her pore was red or brown, rather than white, and that the male had a large, full belly. In such situations, it was presumed the male had eaten all of the eggs laid by the female, although the female also might have eaten some eggs.Sex determination Neosho madtoms at CERC and ESU developed secondary sex character-istics. Distinct differences were observed in the heads of breeding individ-uals; males developed swollen cephalic epaxial muscles on broad, flat heads with swollen lips, whereas the heads of females remained conical in shape and lips did not swell, similar to those of nonbreeding individuals.
Red or pink premaxillary tooth patches were present in both sexes, but usually were brighter red in males. Gravid females had distended abdomens.
The genital papilla in males became elongated and swollen. Tissues adjacent to the gen-ital papilla in females swelled, and the vent became swollen and rounded (Fig. 4).Neosho madtom genital papillae (Fig. 4) were drawn from representative preserved individuals in the ESU teaching collection; differences depicted generally were difficult to determine on live specimens.
Dissection of five preserved specimens (56, 58, 59, 65, and 68 rnm TL) that appeared gravid revealed enlarged ovaries (X ovary length = 14.9 mm, SD = 1.20, n = 10), containing amber-colored eggs. The ovaries of two of these individuals were dissected and inspected for eggs. The left and right ovaries, both 16 mm in length, of the 68 mm TL individual contained 39 and 40 yellow eggs, re-spectively, with a mean chorion diameter of 2.5 mm (SD = 0.19, n = 10)(Table 2). One other individual (65 mmn TL), had a 16-mm ovary -containing eggs with a mean chorion diameter of 2.6 mm (SD = 0.12, n = 3). The number of these larger eggs could not be determined because they fell apart when disturbed within the ovary. Approximately 31 small, whiteeggs, with VOLUME 105, NUMBERS 3-4 117 C Figure 4. Genital papillae of preserved Neosho madtoms: A, breeding female; B, breeding male; C, nonbreeding individual.
mean diameter of 0.7 mm (SD = 0.13, n = 10), were observed in the lining of the ovary. The other ovary (12 mm) contained only small, white eggs in the lining; mean chorion diameter was 0.8 mm (SD = 0.08, n = 13).DISCUSSION There is a paucity of information about courtship behaviors among No-turus species. They, however, may be similar as a result of evolutionary constraints within the lineage (Mayden and Walsh, 1984; Burr and Stoeckel, 118 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE 1999). The successful spawning of a single pair of Neosho madtoms in the ESU lab during the summer of 1996 allowed the first direct observation of reproduction in the species. In 1998, at CERC, courtship behaviors were recorded and, at ESU, two more clutches were observed.
In general, the results of our studies indicate that N. placidus, similar to other madtom species, is a cavity spawner, with male parental care of eggs and fry and rapid development of embryos and larvae. After hatching, we observed lar-vae in the nest, and they formed a loose aggregation until yolk sacs were depleted (13 mm TL). Burr and Stoeckel (1999) reported that, -within the genus Noturus, most young are guarded in the nest until they are 12-15 mm TL. We also observed both male and female N. placidus performing nest building or cavity enhancement activities, behaviors that have been observed in other madtom species as well.The "carousel" and "tail. curl" observed in this study were identical to behaviors described for the brindled madtom (Bowen, 1980) and: the brown madtom (Chan, 1995) and have been referred to by other researchers as"mutual caudal embrace" and "male-only caudal embrace," respectively (Burr and Stoeckel, 1999). Other ictalurids, such as the channel catfish (Ic-talurus punctatus), brown bullhead (Ameiurus nebulosus), and flathead cat-fish (Pylodictis olivaris), have breeding behaviors similar to those observed in N. placidus (Breder and Rosen, 1966). Breeding behavior described by Breder (1935) for the brown bullhead was similar to. that observed in N.placidus; in each, the "carousel" and "tail curl"- behaviors were repeated several times prior to spawning.The three clutches laid in captivity in 1996 and 199.8 were similar to each other in appearance.
At ESU in 1996, the clutch was laid in a depression that had been made in gravel under a PVC structure at a water temperature of 25°C. The clutch at ESU in 1998 was laid in a depression that had been made in the gravel under a large flat stone at a water temperature of 27°C.At the CERC the clutch was laid under a PVC structure in a slight depression on a gravel bottom at 28°C. In all examples, eggs adhered to each other but not to the substrate, similar to clutches described by Pfingsten and Edds (1994) for N. placidus and by Mayden and Burr (1981) for N. exilis, the slender madtom. Clutch size (30 to 32 eggs) differed only slightly between the two spawns in 1998, but were smaller than the ca. 60 observed in 1996 and the 63 reported by Pfingsten and Edds (1994). Other Noturus species have been reported to have clutch sizes ranging from 14 (N. leptacanthus, the speckled madtom; Clark, 1978) to 124 (N. exilis; Burr and Mayden, 1984; Burr and Stoeckel, 1999). Mean chorion diameter of eggs in 1998 (3.7 mm) was slightly larger than that of 3.1 mm observed in 1996 and that reported by Pfingsten and Edds (1994). Because clutch size and egg size in fish are related inversely (Jobling, 1995), larger egg size in 1998 could be associated with smaller clutches.
However, it is possible that the male or DLUME 105, NUMBERS 3-4 119-male ate some eggs before they were discovered.
Another possibility is tat the clutches were not the first to be laid by these females that year, as iese fish were collected later in the breeding season (23 June and 7 July)ian those collected in May 1996 and by Pfingsten and Edds (1994), who.so collected fish in May. Mean chorion diameters observed in our study l1 and 3.7 mm) are. consistent with that reported by Burr and Stoeckel[999) for Noturus species (3.6 mm).Several studies have suggested polyandry in Noturus species (N. exilis, layden and Burr, 1981; N. nocturnus, the freckled madtom, Burr and May-en, 1982; N. hildebrandi, the least madtom, Mayden and Walsh, 1984; N.haeus, Chan, 1995; N. baileyi, the smoky madtom, and N. flavipinnis, the ellowfin madtom, Dinkins and Shute, 1996), but this mating strategy has ot been confirmed.
In our study, eggs remaining in the ovaries of both-males were smaller than fertilized eggs or those observed in gravid mu-,um specimens, and had not yet undergone vitellogenesis.
The presence of:ss developed eggs in the ovaries of females is consistent with the hypoth-sis that multiple clutches may be laid in one season (Mayden and Burr, 981) and consistent with a hypothesis of a polygamous mating strategy, ihich could include polyandry, as suggested by Burr and Stoeckel (1999).alternatively, these eggs could be laid the following year (Baker and Heins, 994); however, field data suggest most Neosho madtoms do not live to Age (Edds and Wilkinson, 1996; Bulger and Edds, 2001).In 1996, the male Neosho madtom was observed caring for the clutch of ggs, including rubbing them with his head and belly and fanning them with is tail. These behaviors were seen until hatching, at which time the larvae emained in the nest with the male hovering over them until yolk sacs were.epleted (13 mm TL). Rubbing, fanning, and hovering behaviors also have een noted in male brown madtoms (Chan, 1995), slender madtoms (May-[en and Burr, 1981), additional madtom species (Burr and Stoeckel, 1999), nd other ictalurid species (Breder and Rosen, 1966). The clean and healthy ppearance of the eggs that were cared for by the male, relative to those hat were kept isolated, suggests that the attention provided by the male kept hem from becoming fouled or infected with fugus. It is not clear whether he male consumed any of the eggs in 1996. The male was not observed aking eggs into his mouth, but we counted 15 to 20 eggs fewer at the time)f hatching, which" suggests that he might have eaten some. During the time he male was guarding the nest, we saw him eat on only one occasion, when ome frozen brine shrimp came to rest at the edge of the nest. Although we:ontinued to introduce food into the nesting tank, the male did not regularly eave the nest to feed while he was guarding the eggs or larvae until about me week after the larvae left the nest.The reason for lack of male parental care in 1998 and consumption of:ggs by the male at CERC is unknown; however, males of other madtom 120 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE species have eaten egg masses in captivity (e.g., N. leptacanthus, Clark, 1978; N. miurus, Bowen, 1980; N. insignis, the margined madtom, J. Stoeck-el, pers. comm.), as four males are believed to have done in this study.Perhaps males ate the eggs because of stress caused by captivity or, in 1998, by the hormone injection.
It also is possible that the male at the CERC ate the clutch of eggs because they were not fertilized, although they appeared to have been. It is possible that the fish were simply hungry; however, food was readily available in all aquaria. Regardless of the reason or reasons, the behaviors seen in this study suggest that, in efforts at captive propagation, at least a portion of the egg mass should be removed from the nesting cavity and hatched separately, as was done in our study. Success rate using this method was low in 1998, most likely because of excessive agitation in the hatching apparatus, temperature fluctuation, and fungal infection.
Others have reported low success rates in rearing madtom eggs (N. baileyi, Shute, Shute, and Rakes, 1993; N. phaeus, Chan, 1995; J. Stoeckel, pers. comm.).It has been suggested that loss of eggs from fungus is controlled in the wild by parental care (Breder, 1935; Fontaine, 1944). Our efforts in the lab, how-ever, indicated that parental care in captivity is generally lacking and con-sumption rates are high.In 1996, the breeding female N. placidus was removed from the breeding tank after the male was observed chasing her from the nest. In 1998, the female at the CERC was removed from the tank after she was observed eating eggs and being chased from the nest by the male. Removal of the female after spawning is recommended for survival of the clutch in other ictalurids (Breder and Rosen, 1966); however, it is not known whether the female plays a role in survival of the eggs, early in their development.
May-den and Burr (1981) noted that female slender madtoms remain in the nests from between 12 to 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> after spawning, and in 1996 we did not remove the female from the spawning aquarium until approximately 72 h post-spawning and had higher success hatching eggs, compared to attempts in 1998.Egg development, hatching, and yolk resorption proceeded much as de-scribed previously for other madtoms (N. exilis, Mayden and Burr, 1981; N.hildebrandi, Mayden and Walsh, 1984; N. eleutherus, the mountain madtom, Starnes and Starnes, 1985; N. baileyi and N. flavipinnis, Dinkins and Shute, 1996; Burr and Stoeckel, 1999). One difference was the earlier appearance of the vitelline veins in 1998. During the 1996 breeding study, development of vitelline veins was observed on Day 6 (ca. 144 h post-fertilization), but, in 1998, vitelline veins were visible approximately 92 h post-fertilization.
Mayden and Burr (1981) reported the appearance of vitelline veins 102 to 104 h after fertilization in N. exilis, and Mayden and Walsh (1984) observed their formation by 130 h post-fertilization in N. hildebrandi.
The slight dif-ferences in development rates observed in our study could be attributed to
)LUME 105, NUMBERS 3-4 121 gher temperatures in 1998 (28°C) than in 1996 (25°C). In addition, larval eosho madtoms were observed breaking through the chorion head first in)ntrast to observations of other madtoms hatching tail first (Burr and toeckel, 1999). The behavior of hatchling Neosho madtoms was consistent ith that of other madtoms in that they schooled in the nest until their yolk tcs were absorbed, and they exhibited negative phototaxis and positive-otropism (Burr and Stoeckel, 1999). On Day 10 post-hatching, when their Ak sacs were absorbed completely, juveniles dispersed throughout the ravel in the aquarium.
This dispersal behavior seems to be an important fe history trait, also documented by field observations that young-of-year eosho madtoms tend to favor loosely compacted gravel substrate (Bulger.id Edds, 2001), within which they presumably find refuge. However, be-avior and distribution of young after leaving the nest is a topic that requires irther study for N. placidus and other madtoms (Burr and Stoeckel, 1999).Accurate sexing of N. placidus is easier when secondary sex character-tics are well developed, usually in late May through mid July (Bulger and dds, 2001). During breeding season, secondary sex characteristics were milar to those previously described for the Neosho madtom (Moss, 1981;fingsten and Edds, 1994; Edds and Wilkinson, 1996). Differences between-xes were seen in head shape, lip size, redness of tooth patches, abdomen istention, and genital papillae.
Other authors have described similar sec-ndary sex characteristics in madtom species (N. exilis, Mayden and Burr, 981; N. nocturnus, Burr arid Mayden, 1982; N. flavater, the checkered iadtom, Burr and Mayden, 1984; N. hildebrandi, Mayden and Walsh, 984), and these are. consistent with those reported by Burr and Stoeckel 1999). Burr and Mayden (1982) and Dinkins and Shute (1996) reported olor change in males during the breeding season in N. exilis and N. baileyi,-spectively.
In our study, no obvious color changes occurred in N. placidus uring breeding season..Sneed and Clemens (1959) indicated that one to seven injections of human horionic gonadotrophin are required to stimulate spawning in channel cat-sh. At ESU, in 1998, one successful spawn and three suspected spawning vents occurred after three or four days of injection with Ovaprim.
Success ite using hormone injection was low (4 out of 15), even when dosage was icreased and injections were administered over a 12-d period. It should be oted, however, that injections were administered late (12 July), probably fter the peak of spawning season. It is important to inject before ovulation, ecause later hormone injections will not be as effective (J. Stoeckel, pers.omm.).Our observations of N. placidus reproductive behavior are consistent with ie conclusion of Burr and Stoeckel (1999) that madtoms exhibit a suite of ncestral behaviors common among ictalurids and exhibit other behaviors hat have evolved independently in the madtom and flathead catfish clade 122 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE Jescribed by Lundberg (1992). Our research has added to the knowledge of V. placidus reproductive biology and behavior, but many questions remain.The role of the female in parental care is unknown, as is mating strategy;if polygamous, is it polyandrous or polygynous?
Breeding habitat in nature remains to be discovered, and environmental variables that trigger breeding are unknown, as is the number of clutches laid in a lifetime.
Although ob-servation of N. placidus spawning behavior in our study occurred in captiv-ity, such courtship behavior is likely representative of behavior in natural zonditions; Porterfield (1998) reported similar spawning behavior in lab and natural settings for eight species of darters (Etheostoma).
Lab investigations of small secretive fishes in turbid water can be important to understanding their breeding biology and behavior.ACKNOWLEDGMENTS.
Thanks to B. Bishop, R. Brammell, M. Nelson, and C. Pendergraft for allowing access to their properties for collection of Neosho madtoms, and to those who helped with fish collection:
A. Babbit, S. Bulger, M. Combes, T. Hirata-Edds, P. Lamberson, B. Smith, and J. Whitaker.
At CERC, thanks to A. Allert, E. Greer, P. Lamberson, D. Papoulias, J. Whitaker, and D.Zumwalt. At ESU, thanks to J. Bartley, P. Fillmore, J. Halstead, K. How-deshell, D. Saunders, L. Sievert, K. Smalley, E. Finck, and J. Witters. In 1996, T Mosher and J. Stephen, Kansas Department of Wildlife and Parks, provided technical expertise in raising embryonic and larval madtoms.Thanks to J. Stoeckel, Arkansas Technical University, for advice on breed-ing, injecting, and raising madtoms. Thanks to S. Bulger for assistance in drawing Neosho madtoms. Thanks to M. Annett for assistance with elec-tronics, and to C. Annett and V. Annett for support. Thanks to D. Mulhern, USFWS, for help with permission to transfer fish to CERC, and to the USFWS and Kansas Department of Wildlife and Parks for permits to collect and hold Neosho madtoms. Funding was provided by the U.S. Army Corps of Engineers through a National Biological Service cooperative agreement (No. 14-45-CA03-97-908) and by the ESU Faculty Research and Creativity Committee.
We dedicate this paper to the memory of Frank B. Cross, a member of the Neosho madtom recovery team, who provided great insight for development of strategies for the recovery of the species. He and his life's work encouraged our study of the "little fishes." LrERATuRE CrrED Baker, J. A., and D. C. Heins. 1994. Reproductive life history of the North American madtom catfish, Noturus hildebrandi (Bailey and Taylor 1950), with a review of data for the genus. Ecol. Freshwater Fish 3:167-175.
Boujard, T., Y. Moreau, and P. Luquet. 1992. Diel cycles .in Hoplosternum littorale (Teleostei):
entrainment of feeding activity by low intensity colored light. Environ. Biol. Fishes 35(3):301-309.
)LUME 105, NUMBERS 3-4 123)wen, C. A., Jr. 1980. The life history of the brindled madtom Noturus miurus (Jordan) in Salt Creek, Hocking and Vinton counties, Ohio. unpubl. masters thesis, Ohio State Univ., 195 pp.eder, C. M., Jr. 1935. The reproductive habits of the common catfish, Ameiurus nebulosus (LeSueur), with a discussion of their significance in ontogeny and phylogeny.
Zoologica 19(3): 143-185."eder, C. M., Jr., and D. E. Rosen. 1966. Modes of reproduction in fishes. The Natural History Press, Garden City, New York, 941 pp.ilger, A. G., and D. R. Edds. 2001. Population structure, and habitat use in Neosho madtom (Noturus placidus).
Southwest Nat. 46(1):8-15.
ilger, A. G., M. Wildhaber, and D. R. Edds. 2002. Effects of photoperiod on behavior and courtship of the Neosho madtom (Noturus placidus).
Freshwater Ecol. 17(1):141-150.
arr, B. M., and W. W. Dimmick. 1981. Nests, eggs and larvae of the elegant madtom Noturus elegans from Barren River Drainage, Kentucky (Pisces: Ictaluridae).
Kentucky Acad.Sci. Trans. 42(3-4):116-118.
urr, B. M., and R. L. Mayden. 1982. Life history of the freckled madtom, Noturus nocturnus, in Mill Creek, Illinois (Pisces: Ictaluridae).
Univ. Kansas Mus. Nat. Hist. Occasional Pap. 98:1-15.urr, B. M., and R. L. Mayden. 1984. Reproductive biology of the checkered madtom (Noturus flavater) with observations on nesting in the Ozark (N. albater) and slender (N. exilis)madtoms (Siluriformes:
Ictaluridae).
Am. Midl. Nat. 112(2):408-414.
urr, B. M., and J. N. Stoeckel.
1999. The natural history of madtoms (genus Noturus), North America's diminutive catfishes.
Pages51-101 in Irwin, W. A., C. E Rabeni, H. L.Schramm, Jr., and T. Coon, eds., Catfish 2000: Proc. Intern. Ictalurid Symp., Am. Fish-eries Soc. Symp. 24 (Bethesda, Maryland).
han, M. D. 1995. Life history and bioenergetics of the brown madtom, Noturus phaeus.unpubl. masters thesis, Univ. Mississippi, 177 pp.lark, K. E. 1978. Ecology and life history of the speckled madtom, Noturus leptacanthus (Ictaluridae).
unpubl. masters thesis, Univ. Southern Mississippi, 134 pp.,inkins, G. R., and P W, Shute. 1996. Life histories of Noturus baileyi and N. flavipinnis (Pisces: Ictaluridae), two rare madtom catfishes in Citico Creek, Monroe County, Ten-nessee. Alabama Mus. Nat. Hist. Bull. 18:43-69.dds, D., and C. Wilkinson.
1996. Population status, breeding biology, age structure and habitat use of the Neosho madtom within the Cottonwood and Neosho rivers of Lyon and Chase counties, Kansas. Rept. Kansas Department of Wildlife and Parks (Pratt), 37 pp.ontaine, P. A. 1944. Notes on the spawning of the shovelhead catfish, Pilodictis olivaris (Rafinesque).
Copeia 1944(1):50-51.)bling, M. 1995. Environmental biology of fishes. Chapman & Hall, New York, 455 pp..undberg, J. G. 1992. The phylogeny of ictalurid catfishes:
a synthesis of recent work. Pages 392-420 in Mayden, R. L., ed., Systematics, historical ecology, and North American freshwater fishes. Stanford Univ. Press, Stanford, California.
layden, R. L., and B. M. Burr. 1981. Life history of the slender madtom, Noturus exilis, in southern Illinois (Pisces: Ictaluridae).
Univ. Kansas Mus. Nat. Hist. Occasional Pap. 93: 1-64.layden, R. L., and S. J. Walsh. 1984. Life history of the least madtom Noturus hildebrandi (Siluriformes:
Ictaluridae) with comparisons to related species. Am. Midl. Nat. 112(2): 349-367.4ayden, R. L., B. M. Burr, and S. L. Dewey. 1980. Aspects of the life history of the Ozark madtom, Noturus albater, in southeastern Missouri (Pisces: Ictaluridae).
Am. Midl. Nat.104(2):335-340.
24 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCE 4oss, R. 1981. Life history information for the Neosho madtom (Noturus placidus).
Report to Kansas Department of Wildlife and Parks (Pratt), 33 pp.Ifingsten, D. G., and D. R. Edds. 1994. Reproductive traits of the Neosho madtom, Noturus placidus (Pisces: Ictaluridae).
Kansas Acad. Sci. Trans. 97(3-4):82-87.
'orterfield, J. C. 1998. Spawning behavior of snubnose darters (Percidae) in natural and labo-ratory environments.
Environ. Biol. Fishes 53(4):413-419.
- hute, J. R., P. W. Shute, and P. L. Rakes. 1993. Captive propagation and population monitoring of rare southeastern fishes by Conservation Fisheries, Inc. Rept. Tennessee Wildlife Re-sources Agency, 31 pp.:imonson, T. D., and R. J. Neves. 1992. Habitat suitability and reproductive traits of the or-angefin madtom Noturus gilberti: (Pisces: Ictaluridae).
Am. Midl. Nat. 127(l):115-124.
- need, K. E., and H. P. Clemens. 1959. The use of human chorionic gonadotrophin to spawn warm-water fishes. Prog. Fish-Cult.
21(3):117-120.
- tarnes, L. B., and W. C. Starnes. 1985. Ecology and life history of the mountain madtom, Noturus eleutherus (Pisces
- Ictaluridae).
Am. Midl. Nat. 114(2):331-341.
J. S. Fish and Wildlife Service. 1991. Neosho madtom recovery plan. U.S. Fish and Wildlife Service (Denver, Colorado), 42 pp.lives, S. P. 1987. Aspects of the life history of the slender madtom Noturus exilis in north-eastern Oklahoma (Pisces: lctaluridae).
Am. Midl. Nat. 117(l):167-176.
Wrilkinson, C., and D. Edds. 1997. Breeding biology of the Neosho madtom, with assessment of population status and habitat use in the Cottonwood and Neosho rivers of Lyon and Chase counties, Kansas. Report to Kansas Department of Wildlife and Parks (Pratt), 63 pp.I Effects of Photoperiod on Behavior and Courtship of the Neosho madtom (Noturus placidus)Angela G. Bulgera Department of Biological Sciences Emporia State University Emporia, KS 66801 USA Mark Wildhaber Columbia Environmental and Contaminants Research Center USGS, BRD Columbia, MO 65201 USA and David Edds Department of Biological Sciences Emporia State University Emporia, KS 66801 USA ABSTRACT To test effects of long and short day-length on behavior of the Neosho madtom (Noturus placidus), we held six pairs of fish in separate tanks under 16 hr (L): 8 hr (D) (long-day) and six pairs under 12 hr (L): 12 hr (D) (short-day) photoperiods.
An ethogram was created and behavior was electronically and continuously recorded.
Two-minute intervals for each hour over four 4-day periods were examined, and proportion of time active and performing specific behaviors in each tank was analyzed to compare differences between treatments.
Individuals held under 16 L, 8 D were more active during the light cycle than those in 12 L, 12 D. Specific behaviors examined included resting, swimming, feeding, aggression, cavity enhancement, and courtship; A higher proportion of time was spent performing cavity enhancement, cavities were deeper, and gravel size in cavities was smaller for those fish in the long-day treatment.
Throughout the experiment various courtship behaviors were observed in male-female pairs held in 16 L, 8 D, but no such behaviors were observed in 12 L, 12 D. The relationships between a long photoperiod and activity, cavity enhancement, and courtship behaviors illustrate the influence of photoperiod on the Neosho madtom reproductive cycle.INTRODUCTION Little is known about the behavior of diminutive stream fishes (Matthews 1998), especially the madtoms, a group of small, nocturnal North American catfishes of the genus Noturus. Information regarding the-effects of photoperiod on activity and behavior is lacking in this group, especially with regard to spawning.
The Neosho madtom, Noturus placidus, is listed as threatened by the U.S. Fish and Wildlife Service (USFWS). The USFWS (1991) Neosho madtom recovery plan regarded understanding the reproductive biology and behavior as critical for recovery of this species.Due to high river turbidity and flow, behavioral observations in the field are nearly impossible during the presumed spawning season (late May through early July), thus no Neosho madtom spawning or nests have been observed in the wild (Pfingsten and Edds 1994). Attempts at captive propagation have had limited a Present address: 14801 W. 149th Ct., Olathe, KS 66062 USA 141 Journal of Freshwater Ecology, Volume 17, Number I -March 2002 success. Of four clutches laid in captivity, one did not develop and was likely never fertilized (Pfingsten and Edds 1994), one resulted in 43 surviving fish (Wilkinson and Edds 1997), one was presumably consumed by the spawning male (Bulger 1999), and one resulted in two surviving Neosho madtoms (Bulger 1999).Understanding environmental cues that trigger spawning could increase success of captive propagation, which is essential for studying N. placidus breeding biology and behavior, and could be necessary for successful reintroduction efforts. Photoperiod is one important factor in stimulating sexual maturation and ovulation in many fishes (Wootton 1990), including madtoms (Dinkins and Shute 1996), and may play an important role in triggering captive spawning.
In addition, there is need to understand how manipulating photoperiod affects other behaviors important to the fish in captivity.
For example, effects of increased photoperiod on feeding and aggression need investigation, as well as influences on overall activity levels.Our research focused on the effects of a long and short photoperiod on behavior of the Neosho madtom. Objectives were to investigate the influence of photoperiod on captive propagation by examining activity and the following specific behaviors:
resting, swimming, feeding, aggression, cavity enhancement, and courtship.
METHODS AND MATERIALS Fish were collected from the Cottonwood River, Chase and Lyon counties, Kansas, and transported to the Columbia Environmental Research Center (CERC)in Columbia, Missouri.
Eight individuals were captured on 13 August 1996 and 21 individuals were collected on 17 and 18 May 1997. Fish were kept in four 59-1 holding tanks at 18 to 21 0 C under a 13.5 hr (L): 10.5 hr (D) photoperiod.
We placed twelve 29.5-1 aquaria in an isolation chamber which was divided with black plastic to create two treatment conditions, with regard to photoperiod.
Well water (pH = 7.5) in each tank was maintained at 20 cm depth and 25- 27,C.Water was delivered continuously with a turnover rate of approximately 8.61 h .The bottom of each tank was covered approximately 4 cm deep with 2 to 24 mm.diameter gravel. Structure was provided by cutting 12.5-cm PVC pipes in half lengthwise, which resulted in a U-shaped shelter, the PVC provided cover while allowing observation from the front of the tank. Light was provided by fluorescent bulbs mounted in the chamber. Fish were fed a diet of live amphipods (Hyalella azteca) and blackworms (Lumbriculus sp.) every two to three days in both treatment aquaria and holding tanks.On 28 May 1997, we placed two fish into each of six experimental tanks at 18 to 21"C and 13.5 L: 10.5 D photoperiod.
Attempts were made to determine sex of each individual based on secondary sex characteristics (Pfingsten and Edds 1994) so that each tank would contain one male and one female. Individuals not placed in study tanks remained in holding tanks. Over the next three weeks we acclimated all fish (treatment and holding tanks alike) to the experimental temperature of 25°C. From 12 to 14 June, while water temperature was 21"C, we collected baseline behavior and activity data on all fish in the treatment chamber.On 16 June, we adjusted day-length to treatment settings of 16 hr (L): 8 hr (D), the long-day treatment, or 12 hr(L): 12 hr (D), the short-day treatment On 23 June, we removed five individuals due to health problems or lack of development of secondary sexual characteristics, and replaced them with 142 individuals from the holding tanks. After a five-day acclimation period, the first experimental period began 28 June. On 9 July, we replaced nine fish with individuals from the holding tanks. After a five-day acclimation period, a second experimental period began 14 July.Three Panasonic closed circuit black and white TV cameras were mounted on each side of the isolation chamber so that each camera recorded activity in two tanks. Infrared illuminators were mounted above the tanks to allow recording of nighttime behaviors.
A real-time and time-lapse VCR was used to record Neosho madtom behavior 24 hr per day.In addition to recording the control activity (12 to 14 June), we filmed behavior continuously during four 4-day periods, which comprised two experimental periods (first experimental period: 28 June -1 July; 4 July -7 July;second experimental period: 14 July -17 July; 20 July -23 July), simultaneously for both long- and short-day treatments.
We then analyzed2 amin, selected randomly, of each hour for each tank during the control and experimental periods;observations were not made during feeding or tank maintenance.
Each time the behavior of an individual changed during the 2-min interval, we recorded the time and new behavior.
Attempts were made to record the behaviors of each individual separately, but due to poor film quality and the small size of the fish this was not always possible; however, this collection method allowed us to determine the time spent performing each behavior in each tank. Behaviors recorded (Table 1) w=a taken from an ethogram created from a combination of observations of Neosho madtoms by A. Bulger, observations made during previous attempts at Neosho madtom captive breeding (Pfingsten and Edds 1994, Wilkinsonand Edds 1997), and from descriptions of spawning behavior of the brown madtom (Noturus phaeus; Chan 1995). Following completion of the study, we euthanized all fish and examined them internally to verify sex.We classified each behavior as either active or inactive (resting).
The proportion of time spent active in each tank in control -groups was compared by using a Wilcoxon-Mann-Whitney two-tailed test of ranks with the null hypothesis that there was no difference in activity between the two treatments (c--0.05 for all analyses).
The proportion of time spent active in light versus dark hours was also tested in each treatment group by using a Wilcoxon-Mann-Whitney one-tailed test under the alternative hypothesis that activity of this nocturnal fish was higher during the dark than during the light cycle of the photoperiod.
A Wilcoxon-Mann-Whitney two-tailed test was also performed on the proportion of time spent active in each tank with the null hypothesis that activity did not differ between the two experimental periods.To assess effects of photoperiod on behavior, we assigned each behavior from the ethogram one of six specific behavior types: resting, swimming, feeding, aggression, cavity enhancement, and courtship (Table 1); resting behaviors were not included in analysis, as they are the complement of active behavior.
We compared the proportion of time spent performing each behavior type in light and dark hours between treatment groups for both experimental periods by using a Wilcoxon-Mann-Whitney two-tailed test.To examine effects of photoperiod on cavity enhancement for nesting, we measured depth of the gravel substrate under and outside the structure at the end of the study. We also measured the diameters of three randomly chosen pieces of gravel from under the structure, and three from the rest of the tank. These 143 Table 1. Ethogram of Neosho madtom Behaviors performed by either or both fish during experimental periods. Each behavior is labeled as I = inactive or A = active (resting), and categorized as a behavior type: resting, swimming, feeding, cavity enhancement (cav enhan), aggression, and courtship.
Behavior Description Activity Type Performed by either fish:.Upside down resting upside down under structure Quiet in resting quietly under structure Quiet out resting quietly out of structure Restless in moving slightly about under structure Restless out moving slightly about outside of structure Circle alone swimming in circles against glass at front, back, or side of tank Swim swimming in no particular pattern Feeding feeding Headstand vertical in water nudging rocks with head Rock move moving a rock in its mouth (picks up rock and drops it in another place)Spin swimming in circular pattern under structure Fanning fanning tail while resting under structure Performed by both fish.Quiet in both both fish resting quietly under structure Quiet out both both fish resting quietly out of structure Restless in both both fish slightly moving about under structure Restless out both both fish slightly moving about out of structure Circle chase one fish chases other in circular pattern in front, back, or side of tank. Individuals periodically meet and have some sort of phvsical contact (rub. bite. or nudge)Bite one fish bites at body of other fish Chase one fish chases other in no particular pattern I I I I I A A A A A A A resting resting resting resting resting swimming swimming feeding cay enhan cav enhan cay enhan courtship I I I resting resting resting resting A swimming A A aggression aggression aggression Nudge Jostle Carousel Tail Curl one fish nudges resting individual and swims A away or rests next to it. Nudged individual may swim or remain resting fish switch positions back and forth under A structure between short periods of rest fish swim together head to tail in small A circular pattern under structure fish lay side by side, head to tail; male tail A wrapped around head of female and both fish quiver. This behavior was only seen following carousel courtship courtship courtship 144 a measurements allowed us to compare the depth of the cavity and gravel size within the cavity in each tank, compared to the rest of the tank bottom, and to compare cavity depth and gravel size between treatments by performing Wilcoxon-Mann-Whitney two-tailed tests. A Pearson correlation coefficient was calculated to examine strength of the relationship between cavity depth and -gravel size within the cavity.RESULTS Throughout the control and treatment periods, 36 pairs of Neosho madtoms was observed in the isolation chamber. Subsequent internal examination allowed sexing of 23 pairs and revealed 12 as male/female, 10 as female/female, and one as male/male; sex of one or more individuals in other pairs was equivocal (Table 2). Each treatment group had two male/female pairs during each data collection period. In addition, during the first experimental period, camera failure caused uneven sample sizes; two short-day tanks were not continuously monitored.
Table 2. Sex of Neosho madtoms in study tanks during study periods (M=male, F=female, U=undetermined), with depth of cavity (mm) under structure (depth of gravel outside structure
-depth of gravel under structure) and difference between mean diameter (mm) of gravel outside structure and under structure at end of experiment.
Tank long-day 1 2 3 4 5 6 short-day 7 8 9 10 11 12 Control Period M/F F/F F/F M/F U/U U/U M/F M/U U/U M/F U/F F/F Experimental Period 1 M/F F/F F/F MI/F U/U U/F M/F M/U U U/F U/F M/F Experimental Period 2 M/F F/F F/F M/F U/F F/F M/F M/U M/M M/F F/F: F/F Cavity Depth 15 37 25 33 11 22 5 4 1 16 3 11 Gravel Size Difference 4.3 7.3 2.3 4.4 4.0 5.0 2.3-3.7-2.0 1.0 7.7 1.5 Results from the control period, 13.5 hr (L): 10.5 hr (D), showed no significant difference between groups in the proportion of time spent active in the dark or light (Mann-Whitney U: Z=-0.88, P=0.38 and Z=-0.61, P=0.54, respectively).
We performed analyses on each experimental period separately for three reasons. First, the Wilcoxon-Mann-Whitney test showed a significant difference in the proportion of time spent active by individuals in the short-day treatment; more activity was seen during the second experimental period than during the first in the dark cycle of the photoperiod (Z=-2.24, P--0.03).
Second, the sample sizes were uneven because of a camera failure during the first experimental period. Third, individuals in both short- and long-day treatment 145 groups had been moved or replaced, thus the test subjects were different.
Additionally, based on what is known of the life history of the Neosho madtom, the fish has a short spawning period, mainly during June and July (Bulger and Edds, 2001). The time lag between 28 June and 14 July might impact spawning behavior.Neosho madtoms spent a significantly higher proportion of time active during dark hours versus light hours (long-day:
first experimental period Z=2.80, P=0.005; second experimental period Z=2.80, P--0.005; short-day:
first experimental period Z=2.17, P=0.03; second experimental period Z=2.80, P=0.005, Fig. 1). Comparison of the proportion of time spent active during the second experimental period showed individuals in long-day treatment were more First Experimental Period 4)C.4)5C 4C 1C W.C Varic Light Short-Day Trot Long-Day Tint Second Experimental Period 4)4)2 0 4)U S.-43 0..J Dark Light Figure 1. Percent of time active in dark and light of each treatment (short-day and long-day) during the first and second experimental periods.146 active during the light cycle than those in short-day treatment (Z=2.00, P=0.05, Fig. 1). We observed individuals in long-day treatment swimming about or foraging for food I to 2 h before the light cycle ended. This difference in activity was not statistically significant during the first experimental period (Z=-1.81, P-0.07). During dark hours, there was no significant difference in activity levels (first experimental period Z=-0.96, P=0.34; second experimental period Z=0.40, P=0.58, Fig. 1).We compared the proportion of time spent performing the following behaviors between the two treatment groups in dark and light hours for each collection period: swimming, feeding, aggression, and cavity enhancement.
The proportion of time spent swimming, feeding, and performing aggressive behaviors was not different between treatments.
During the second experimental period, individuals held under the long-day photoperiod spent a higher proportion of time performing cavity enhancement in dark hours than those in the short-day photoperiod (Z-=2.00, P=0.05). Although not significant in the first experimental period or in light, the proportion was consistently higher in long-day treatment.
Specific cavity enhancement behaviors included the "spin," in which one individual would spin in circles just above the gravel under the structure; the"headstand," in which one individual would hover at approximately 45 and nudge rocks from under the structure by using its head; and the "rock move," in which gravel was carried in the mouth from under the structure and dropped outside.Both males and females were observed performing the spin and the headstand, but only males were observed doing the rock move.No spawning was observed during this study; however, based on observations of other madtom species (Bowen 1980, Fitzpatrick 1981, Chan 1995), and observations during previous Neosho madtom breeding studies (Pfingsten and Edds 1994, Wilkinson and Edds 1997), behaviors were seen in this study that indicated courtship (Bulger 1999). No statistical analyses were performed on courtship behaviors because of small sample size. However, throughout the course of the study male/female pairs held in long-day photoperiod were observed performing courtship behaviors (Table 1), including the "carousel" (88 times), the "tail curl" (36 times), the "jostle" (four times), and the "fan" (once);these behaviors were never observed in male/female pairs held under short-day conditions.
Depressions under structures indicated cavity enhancement.
Cavity depth ranged from 1 to 37 mm (Table 2). Cavities deeper than 20 mm were made in four of 12 tanks; all four were in long-day treatment.
In addition, the deepest-cavity was constructed by females in a long-day treatment.
In long-day treatment, cavities were deeper (Z=-2.32, P--0.02) and gravel size within cavities were smaller (Z=1 .79, P--0.02) than in short-day treatment Mean gravel diameters within cavities in long-day treatment ranged from 12.0 mm to 16.7 mm (5=14.5, SD=3.42);
in short-day treatment means ranged from 14.0 mm to 23.7 nun (7=18.1, SD=4.84).
Pearson's correlation .coefficient between gravel size and cavity depth suggested a negative relationship (r=-0.55), but was not significant (P=0.07).147 their mouths, and Fitzpatrick (1981) observed a male brindled madtom remove gravel from a can by taking rocks into his mouth and dropping them outside the can opening. No other accounts of behaviors used to enhance cavities have been reported for Worurus species; however, brown bullheads (Amieurus nebulosus) and flathead catfish (Pylodictis olivaris) have been observed moving rocks in their mouths for nest construction (Breder and Rosen 1966).The male/female pairs held in long-day treatment performed courtship behaviors, including the carousel, tail curl, jostle, and fan, but those held in short-day conditions did not perform such behaviors.
Ostlund and Ahnesj6 (1998)reported that male courtship displays, such as fanning and body shakes, influenced female mate choice and hatching success in fifteen-spined sticklebacks (Spinachia spinachia).
Thus, the increase of such courtship displays by individuals held in long-day treatment could indicate the importance of photoperiod in stimulating such behavior.Sundararaj and Sehgal (1970) found a long photoperiod to be important in stimulating the ovarian cycle of a seasonally breeding catfish, Heteropneustes fossilis, which is native to areas where seasonal day length varies by only 4 h.Likewise, de Vlaming (1972) reported photoperiod to be a major environmental cue triggering reproductive cycles of salmonids and gasterosteids.
The importance of photoperiod to stimulating the breeding cycle is most likely related to the benefits of timing the spawn to coincide with juvenile food availability, which may optimize survival of offspring (Jobling 1995).Our study provided the first ethogramn and quantitative observations of the effects of photoperiod on Neosho madtom behavior including resting, swimming, feeding, aggression, cavity enhancement, and courtship.
The proportion of time spent performing cavity enhancement behaviors was higher in fish held under the long photoperiod, and more courtship behaviors were observed in those individuals.
Results of this study were consistent with the hypothesis that photoperiod plays a role in the breeding cycle of this fish.ACKNOWLEDGMENTS We sincerely thank the many people who assisted with this study, including:
A. Allert, C. Annett, M. Annett, V. Annett, J. Bartley, S. Bulger, L.Burress, M. Combes, D. Conwell, C. Edds, K. Edds, M. Fillmore, E. Finck, P.Fillmore, E. Greer, J. Halstead, T. Hirata-Edds, W. Jensen, P. Lamberson, D.Mulhern, M. Nelson, D. Papoulias, C. Pendergraft, L. Scott, C. Wilkinson, and J.Witters, J. Whitaker, D. Zumwalt, and the City of Emporia. Funding was provided by the Emporia State University Department of Biological Sciences, and by the U.S. Army Corps of Engineers through a National Biological Service cooperative agreement (No. 14-45-CA03-97-908).
LITERATURE CITED Bowen, C. A., Jr. 1980. The life history of the brindled madtom Noturus miurus (Jordan) in Salt Creek, Hocking and Vinton counties, Ohio. M.S. thesis, Ohio State University, Columbus.
195 pp.Breder, C. M., Jr. and D. E. Rosen. 1966. Modes of reproduction in fishes. The Natural History Press, Garden City, New York. 941 pp.Bulger, A. G. 1999. Population structure, habitat use, and breeding behavior of the Neosho madtom, Noturusplacidus.
M.S. thesis, Emporia State 149 University, Emporia, KS. 83 pp.Bulger, Angela and David Edds. 2001. Population structure and habitat use in Neosho madtom (Noturus placidus).
Southwestern Naturalist Vol. 46:1.Burr, B. M. and R. L. Mayden. 1982. Life history of the freckled madtom, Noturus nocturnus, in Mill Creek, Illinois (Pisces: Ictaluridae).
Occas.Pap. Mus. Nat. Hist. Univ. Kans. 98:1-15.Chan, M. D. 1995. Life history and bioenergetics of the brown madtom, Noturus phaeus. M.S. thesis, University of Mississippi, Vicksburg.
177 pp.Clark, K. E. 1978. Ecology and life history of the speckled madtom, Noturus leptacanthus (Ictaluridae).
M.S. thesis, University of Southern Mississippi, Hattiesburg.
134 pp.Cochran, P. A. 1996. Cavity enhancement by madtoms (genus Notzrus).
J.Freshwater Ecol. 11:521-522.
de Vlaming, V. L. 1972. Environmental control of teleost reproductive cycles: a brief review. J. Fish Biol. 4:131-140.
Dinkins, G. R. and P. W. Shute. 1996. Life histories of Noturus baileyi and N.flavipinnis (Pisces: Ictaluridae), two rare madtom catfishes in Citico Creek, Monroe County, Tennessee.
Bul. Ala. Mus. Nat. Hist.18:43-69.
Fitzpatrick, K. C. 1981. Comparative reproductive behavior of the slender (Noturus exilis), brindled (N. miurus) and freckled (N. nocturnus) madtoms (Pisces: Ictaluridae).
M.S. thesis, Southern Illinois University at Carbondale, Carbondale.
66 pp.Jobling, M. 1995. Environmental biology of fishes. Chapman and Hall, New York. 455 pp.Matthews, W. J. 1998. Patterns in freshwater fish ecology. Chapman and Hall, New York. 756 pp.Mayden, R. L. and B. M. Burr. 1981. Life history of the slender madtom, Noturus exilis, in southern Illinois (Pisces: Ictaluridae).
Occas. Pap. Mus.Nat. Hist. Univ. Kans. 93:1-64.Ostlund S. and I. Ahnesj6. 1998. Female fifteen-spined sticklebacks prefer better fathers. Animal Beh. 56:1177-1183.
Pfingsten, D. G. and D. R. Edds. 1994. Reproductive traits of the Neosho madtom, Noturus placidus (Pisces: Ictaluridae).
Trans. Kans. Acad. Sci.97:82-87.Stames, L. B. and W. C. Starnes. 1985. Ecology and life history of the mountain madtom, Noturus eleutherus (Pisces: Ictaluridae).
Amer. Midi. Nat.114:331-341.
Sundararaj, B. I. and A. Sehgal. 1970. Effects of a long or an increasing photoperiod on the initiation of ovarian recrudescence during the preparatory period in the catfish, Heteropneustesfossilis (Block). Biol. of Reprod. 2:413-424.
U. S. Fish and Wildlife Service. 1991. Neosho madtom recovery plan. U.S. Fish and Wildlife Service, Denver, Colorado.
42 pp.Wilkinson, C. and D. Edds. 1997. Breeding biology of the Neosho madtom, with assessment of population status and habitat use in the Cottonwood and Neosho rivers of Lyon and Chase counties, Kansas. Report to Kansas Department of Wildlife and Parks, Pratt. 63 pp.Wootton, R. J. 1990. Ecology of teleost fishes. Chapman and Hall, London.404 pp.150 Received:
25 October 2000 Accepted:
20 January 2001 Copyrighted Material Protect accordingly 11 STrarnactions of the American Fisheries Society 129:243-261, 2000 C Copyright by the American Fisheries Society 2000 Natural and Anthropogenic Influences on the Distribution of the Threatened Neosho Madtom in a Midwestern Warmwater Stream MARK L. WILDHABER,*
ANN L. ALLERT, AND CHRISTOPHER J. SCHMITT U.S. Geological Survey, Columbia Environmental Research Center, 4200 New Haven Road, Columbia, Missouri 65201, USA VERNON M. TABOR AND DANIEL MULHERN U.S. Fish and Wildlife Service, 315 Houston Street, Suite E, Manhattan, Kansas 66502, USA KENNETH L. POWELL Westwood Professional Services, Inc., 7599 Anagram Drive, Eden Prairie, Minnesota 55344, USA SCOTT P. SOWA Missouri Cooperative Fish and Wildlife Research Unit, The School of Natural Resources, University of Missouri, 302 ABNR Building.
Columbia.
Missouri 65211, USA Abstract.-We attempted to discern the contributions of physical habitat, water chemistry, nu-trients, and contaminants from historic lead-zinc mining activities on the riffle-dwelling benthic fish community of the Spring River, a midwestern warmwater stream that originates in Missouri and flows into Kansas and Oklahoma.
The Spring River has a fish community that includes the Neosho madtom Noturusplacidus, a species federally listed as threatened.
Although anthropogenic factors such as contaminants limited populations and densities of fishes, an integrated assessment of natural and anthropogenic factors was necessary to effectively estimate the influence of the latter. Fish populations in the Spring River, especially Neosho madtoms, seem to be limited by the presence of cadmium, lead, and zinc in water and in benthic invertebrate food sources and by physical habitat. The population density and community structure of fish in the Spring River also seem to be related to water chemistry and nutrients.
Concurrently, diminished food availability may be limiting fish populations at some sites where Neosho madtoms are not found. Many of the natural factors that may be limiting Neosho madtom and other riffle-dwelling fish populations in the Spring River probably are characteristic of the physiographic region drained by the upper reach and many of the tributaries of the Spring River. Our results indicate that competition between the Neosho madtom and other species within the riffle-dwelling fish community is an unlikely cause of Neosho madtom population limitation in the Spring River.Relationships between stream fish communities and their habitats have been well documented (An-germeier and Karr 1984; Matthews and Heins 1987; Kessler and Thorp 1993). Physical habitat complexity has been correlated with fish species diversity (Gorman and Karr 1978). Habitat factors such as water depth, velocity, and substrate com-position are important to stream fishes (Aadland 1993). Moreover, habitat utilization by stream fish-es varies with community composition (Fauscb and White 1981; Finger 1982), and water chem-istry and nutrients affect the distribution and abun-* Corresponding author: mark..wildhaber@usgs.gov Received March 19, 1998; accepted May 14, 1999 dance of stream fishes (Layher and Maughan 1985;Layher et al. 1987; Maret et al. 1997). Habitat has been the primary focus of studies that target factors limiting the distribution and density of stream fish-es, especially threatened and endangered fishes (Kessler and Thorp .1993; Freeman and Freeman 1994).The Neosho madtom Noturusplacidus is a small (<75 mm total length) ictalurid first described as a species in 1969 (Taylor 1969). Neosho madtoms have been found in the highest numbers in riffles during daylight in late summer and early fall, after young of the year are estimated to have recruited to the population (Moss 1983; Luttrell et al. 1992;Fuselier and Edds 1994). Neosho madtoms prefer the interstitial spaces of unconsolidated pebbles 243 244 WILDHABER ET AL.FIGURE 1.--Sampling sites on the Neosho, Cottonwood, and Spring rivers in 1994. Triangles represent sites where Neosho madtoms were collected; squares represent sites where they were not collected.
and gravel, moderate to slow flows, and depths averaging 0.23 m (Moss 1983). Neosho madtoms feed on larval insects among stones at night (Cross and Collins 1995). The Neosho madtom was listed as threatened by the U.S. Fish and Wildlife Service (USFWS) in May 1990, and a recovery plan was approved in September 1991 (USFWS 1991). The USFWS (1991) hypothesized that habitat and po-tential fish competitors of the Neosho madtom, such as other ictalurids, darters (Percidae), and other riffle-dwelling benthic fishes, may limit Ne-osho madtom populations.
Currently, Neosho madtoms are found in main stems of the Neosho, Cottonwood, and Spring rivers in Kansas, Mis-souri, and Oklahoma (Luttrell et al. 1992; Cross and Collins 1995; Wilkinson et al. 1996) (Figure 1). The density of Neosho madtoms is much great-er in the Neosho system (i.e., the Neosho and Cot-tonwood rivers combined) than in the Spring River (Moss 1983; Wilkinson et al. 1996). Cross and Collins (1995) described the Spring River drainage as supporting 20 fishes not found anywhere else in Kansas. Except for one small population just upstream of Baxter Springs, Kansas (Pflieger 1975; Barks 1977; Wilkinson et al. 1996), Neosho madtoms have only been collected from the Spring River upstream of the primary sources of pollution from lead (Pb) and zinc (Zn) mining (Figure 2).Studies of the effects of contaminants on fish populations have generally focused on the contam-inants (McCormick et al. 1994) and given little attention to other concurrent factors (Neves and Angermeier 1990; Hall et al. 1996; Scott and Hall 1997). Hall et al. (1996) assessed habitat factors along with contaminants; however, they empha-sized overall ecological health and biological in-tegrity of the fish community, not specific popu-lations of fish. Contaminants and physicochemical characteristics differ between the Neosho and Cot-tonwood rivers (Neosho system) and the Spring River (Moss 1983; Spruill 1987; Allen and Black-ford 1995). All are affected by similar anthropo-INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 245 FIGURE 2.-Enlargement of the Spring River study area with sampling sites in 1994. Triangles represent sites where Neosho madtoms were collected; squares represent sites where they were not collected.
Lead-zinc mining and processing occurred within shaded areas.genic factors (agricultural runoff and municipal waste inputs) (Allen and Blackford 1995; Kiner et al. 1997). The Spring River is also impacted by runoff from historic Pb-Zn mining and related ac-tivities that have resulted in elevated levels of Pb, Zn, and cadmium (Cd) (Barks 1977; Czarneski 1985; Spruill 1987; Smith 1988; Schmitt et al.1993) and by industrial inputs from chemical man-ufacturing and industrial facilities (Kiner et al.1997). Lead, Zn, Cd, arsenic (As), iron (Fe), mer-cury (Hg), and manganese (Mn) are also a concern in the Neosho system. However, concentrations of Pb and Zn in fish and sediments of the Neosho system are much lower than those historically found in Center and Turkey creeks, tributaries of the Spring River. Further, As, Fe, Hg, and Mn have relatively low concentrations in the Neosho system (Spruill 1987; Smith 1988; Schmitt et al. 1993;Allen and Blackford 1995). Most of the metals of concern in the Neosho and Spring River systems can be toxic to fish, and water quality standards for protection of aquatic life have been established for them (USEPA 1986); therefore, they must be considered in any comprehensive evaluation of these river systems. Previous studies (Moss 1983)indicate the Spring River tends to be less turbid and has lower un-ionized ammonia (NH 3), chloride (Cl), and sulfate (SO 4) concentrations than the Ne-osho system. Turbidity may provide protection to the Neosho madtom from predators; NH 3 , Cl, and SO 4 may have both natural and anthropogenic sources (Wetzel 1983).Detrimental effects of Pb, Zn, and Cd on fish have been well documented, and all three can be acutely toxic (USEPA 1986; Eisler 1988). Effects have been documented for waterborne (Eisler and Hennekey 1977; Weber 1993; Bryan et al. 1995)and dietary exposures (Thomas and Juedes 1992;Woodward et al. 1994). Lead affects heme syn-thesis (Johansson-Sjobeck and Larsson 1979). res-piration (Somero et al. 1977), and reproductive behavior (Weber 1993) of fishes. High concentra-tions of Zn cause hyperglycemia (Wagner and McKeown 1982), behavioral avoidance (Wood-ward et al. 1995, 1997), increased heterozygosity of specific allozymes (Roark and Brown 1996),
246 WILDHABER ET AL.and reduced survival (Eisler and Hennekey 197.7).Cadmium can affect the immune system (Lemaire-Gony et al. 1995), the kidney (Gill et al. 1989), and behavior (Bryan et al. 1995).The primary objective of this paper is to eval-uate natural and anthropogenic factors that may be limiting the Neosho madtom and other riffle-dwelling benthic fishes in the Spring River. We wanted to determine if lower densities of Neosho madtoms in the Spring River than in the Neosho system were a result of metals contamination, low-er-quality physicochemical habitat, biotic inter-actions, or some combination of these factors.Study Area The study area included the main stems of the Neosho (Grand) and Cottonwood rivers in Kansas and Oklahoma and the Spring River in Kansas, Missouri, and Oklahoma (Figures 1, 2). All are part of the Arkansas River system. Part or all of the main stems of these rivers are in the Prairie Parkland Province (Bailey 1995) and the Central Irregular Plains (Omernik 1987). The Neosho sys-tem and the lower Spring River drain mainly mixed-grass prairie with mature riparian vegeta-tion along some sections, whereas the upper Spring River and many of its tributaries primarily drain deciduous forests of the Ozark Uplands Province ecoregion (Moss 1983). The Spring River and its tributaries drain parts of the Tri-State Mining Dis-trict in Missouri, Kansas, and Oklahoma (Spruill 1987), which was mined for Pb and Zn from 1850 to the 1960s (Barks 1977). The Spring River drains approximately half the land area, has 70% of the mean annual discharge and 1.7 times the gradient of the Neosho system; however, all three rivers in this study possess similar riffle-pool habitat (Moss 1983; Kiner et al. 1997). The Cottonwood and Ne-osho rivers join near Emporia, Kansas; the Neosho and Spring rivers join near Miami, Oklahoma, in what is now Grand Lake of the Cherokees (Figure 1). The Cottonwood River, Neosho River upstream of its confluence with the Cottonwood River, and Spring River are fifth-order streams. Downstream of its confluence with the Cottonwood, the Neosho River is a sixth-order stream. The Neosho and Cot-tonwood rivers are regulated by reservoirs.
The Spring River is essentially unregulated until its confluence with Shoal Creek in Cherokee County, Kansas, in a power plant cooling reservoir.
Methods We quantified Neosho madtom distribution, Ne-osho madtom habitat, and the benthic communities associated with Neosho madtoms in the Neosho system to compare them with those in the mining-affected Spring River. We collected data on the aquatic community (fish and invertebrate species richness and density of potential competitors), physical habitat (depth, velocity, and substrate size), water chemistry (temperature.
turbidity, pH, dissolved oxygen, hardness, alkalinity, conductiv-ity, SO 4 , and Cl) nutrients (un-ionized NH 3 , nitrite plus nitrate [NO 2 + NO], and phosphate
[PO4]), and metals and metalloids (As, Cd, Fe, Hg, Mn, Pb, and Zn) in water, invertebrates, or both. This list of measurements was compiled from what oth-er researchers had previously identified as factors of concern in the Neosho and Spring River sys-tems, as discussed in the introduction (e.g., Barks 1977; Spruill 1987; Smith 1988; Allen and Black-ford 1995; Kiner et al. 1997). We used an empirical model based on physical habitat, water chemistry, and nutrients measured in the Neosho system dur-ing 1991 to predict the Neosho madtom distribu-tion for that system and Spring River in 1994 with-out information on metals or metalloids.
We then compared predicted and observed values from both river systems and different years to assess the ex-tent to which basic environmental quality and met-als contamination limited Neosho madtom distri-bution in the Spring River. We also used the 1994 data to compare the Neosho system to the Spring River and to compare sites on the Spring River with Neosho madtoms (madtom sites) to sites on the Spring River without Neosho madtoms (no-madtom sites). We compared differences in habitat and benthic communities between the Neosho sys-tem and the Spring River relative to differences in madtom versus no-madtom sites within the Spring River in an attempt to separate system differences from within-Spring River differences.
The methods we used to model the Neosho sys-tem are supported by the work of others (Layher and Maughan 1985; Leftwich et al. 1997). Based on previous research (Moss 1983; USFWS 1991;Luttrell et al. 1992; Fuselier and Edds 1994), we assumed that the abundance of Neosho madtoms on gravel bars during daylight in late summer-early autumn is an index of their overall abundance at a site. The discrete nature of the summer-fall distribution of the Neosho madtom and its com-paratively specialized habitat requirements facil-itated investigation and habitat modeling.
Layher and Maughan (1985) stated that habitat models are generally more successful for species with narrow niche requirements than for generalists, and that they are better applied within than across ecore-INFLUENCES ON NEOSHO gions. Because the lower Spring River represents an ecotone, the model we developed should .be effective.
We selected sites on the Neosho, Cottonwood, and Spring rivers that maximized the probability of collecting Neosho madtoms. In the Neosho sys-tem, 12 shoreline gravel bars comprising stones generally less than 38-mm diameter and known to harbor Neosho madtoms were selected by the USFWS for monitoring Neosho madtom popula-tions (USFWS 1991). All 12 sites on the Neosho and Cottonwood rivers were sampled in 1991.Eleven sites, many the same sites sampled in 1991, were again sampled in 1994. In the Spring River, 20 gravel bars between the North Fork confluence and Grand Lake of the Cherokees (most of the bars in the river) were selected.
In 1991 and 1994, sam-pling at all sites occurred during daylight between August and October.At each 1991 site, three to five transects per-pendicular to the river channel were spaced equal-ly from downstream to upstream along the length of the gravel bar. In most instances, five stations were spaced equally but at least 2 m apart along each transect.
Fewer than five stations were estab-lished when the river channel was less than 10 m wide or when a station would be too deep to seine (>1.25 m). Transects on each gravel bar were sam-pled in order from downstream to upstream.
On each transect, stations were sampled in order of their distance from the gravel bar. To minimize impacts of samples on each other, sampling pro-ceeded in the following order at each station: fish-es, substrate, water depth, water velocity, and sur-face water. Fishes were collected from a 4.5-M 2 area by disturbing the gravel substrate.
We started 3 in upstream of a stationary seine (3.0-mm 2 mesh)and proceeded downstream to the seine. All ictalu-rids, including Neosho madtoms, were identified (Pflieger 1975) and released back into the river.Substrate was collected from an undisturbed area adjacent to the fish sampling location with a 13-cm-deep X 10-cm-diameter cylindrical grab sam-pler. The substrate sample was sieved and cate-gorized into five size-classes
(<2 mm, 2 to <9 mm, 9 to <19 mrm, 19 to < 38 mm, and ->38 mm), which were then weighed. Water depth and water velocity at 60% of water depth were measured with a Marsh-McBirney model 201 current meter. After all station samples were collected at a site, a single surface water grab sample was collected and an-alyzed with a Hach model DREL/IC portable col-orimeter for pH. alkalinity, hardness, conductivity, turbidity, NH 3 , NO, + NO 3 , SO 4 , PO 4 , and Cl.MADTOM DISTRIBUTION
- 247 In 1994, the 1991 sampling procedures were re-peated except that pore waters and benthic inver-tebrates also were collected, and these samples along with surface water samples were analyzed for metals. Access to certain sites was limited and we obtained complete data for only 6 of the II Neosho system sites. At each station, sampling proceeded in the following order to minimize im-pacts of samples on each other: fishes, benthic in-vertebrates, substrate, pore water, water depth, and water velocity.
As in 1991, all ictalurids were iden-tified in the field and released.
Voucher specimens of other taxa and unidentifiable fishes were pre-served in ethanol for later identification.
Benthic invertebrates were collected in undisturbed sub-strate adjacent to the fished area with a modified Hess sampler (0.1- or 0.037-M 2 bottom area; the smaller one was used for water depths generally shallower than 0.19 m) with a 0.3-mm-mesh col-lection bag. Substrate within the Hess sampler was disturbed for 2 min. Benthic invertebrates were preserved in 80% ethanol for later identification to the lowest taxonomic level possible (Merritt and Cummins 1984) except chironomids and oligo-chaetes were not identified below the family level.Pore water was extracted directly from undisturbed substrate with a vacuum pump system upstream of the Hess sample collection site and adjacent to the fish collection site. A Hydrolab Surveyer II was used to measure temperature, dissolved oxygen, pH, and conductivity during pore-water extraction.
Pore-water samples were composited by transect for subsequent analyses.
Each composite sample was distributed between two acid-cleaned, high-density polyethylene bottles. One subsample was analyzed by inductively coupled argon plasma transmission spectroscopy (ICAP) for As, Cd, Fe, Hg, Mn, Pb, and Zn. The second subsample was analyzed for alkalinity, hardness, and Cl by titra-tion; for turbidity with a Hach 2100A turbidimeter; for NH 3 with an Orion EA940 meter; and for NO 2+ NO 3 , SO 4 , and P0 4 with a Hach DR 2000 spec-trophotometer (APHA et al. 1992). All pore-water sampling equipment was acid-cleaned between sites. Water velocity at 60% of water depth was measured with a Swoffer Instruments model 2100 current meter.In 1994, after all station samples were collected at a site, we collected surface water and benthic invertebrate samples for metals analyses and mea-sured geospatial coordinates.
A surface water grab sample was collected from the midpoint of the center transect for analysis of metals and water chemistry.
Because pore water was extracted on 248 WILDHABER ET AL.TAIALE I.-Riffle-dwelling fish taxa collected in the Ne-osho, Cottonwood, and Spring rivers that were assumed to be benthic competitors of the Neosho madtom based on habitat use and feeding descriptions as given by Pflieger (1975).Family and scientific name Common name Catostomidae c),cleptus elongatus Hypentelium nigricans Moxostoma duquesnei Aloxostoina ervrhrurum Mfoxosiorno macrolepidotum Moxosiorma spp.Sciaenidae Aplodinotus grunniens Cyprinidae Erimystar x-punctatus Nolropis spp. or Pimephales spp.Phenacobius mirabilis Pimephales notawns Pinephales tenellus Pimephales vigilax Ictaluridae Icialurus punclatus oton'rus exilis Noturs foavus Noturus iniurus Notn-um nocturnus Pylodictis olivaris Conidae Coitus carolinae Percidae Etheostoma blennioider Etheosromaflabellare Etheostorna nigrum Etheosloma stignaeum Etheostoma speciabile Etheostoma whipplei Etheostoma zonale Percina caprodes Percina copelandi Pei'cina phoxocephala Percina shumardi Blue sucker Northern hog sucker Black redhorse Golden redhorse Shorthead redhorse Freshwater drum Gravel chub Suckennouth minnow Bluntnose minnow Slim minnow Bullhead minnow Channel catfish Slender madtom Stonecat Brindled madtom Freckled madtom Flathead catfish Banded sculpin Grcenside darter Fantail darter Johnny darter Speckled darter Orangethroat darter Redtin darter Banded darter Logperch Channel darter Slenderhead darter River darter"Megaloptera" (dobsonflies), and "others" (gen-erally molluscs).
Although Neosho madtoms would not eat adults of these large taxa, these taxa were selected to represent concentrations of toxic metals in detritivorous and predatory invertebrates upon which they do feed. Benthic invertebrate samples of less than 5 g were analyzed for metals without partitioning.
Geospatial coordinates of the gravel bar were determined with a Trimble Path-finder Plus geographical positioning system.Statistical Analyses We analyzed the data at the site level to assess differences between the Neosho system and the Spring River and between madtom and no-madtom sites in the Spring River. Arithmetic site means were calculated for depth, velocity, and pore-water chemistry and metals. For each metal, we included only samples with concentrations above the de-tection limit in the mean because we considered these samples a measure of the maximum possible exposure at a site. For benthic invertebrates, we calculated species richness and Ephemeroptera, Plecoptera, and Trichoptera richness (EPT) at each site. Previous studies have demonstrated the ef-fectiveness of these metrics for documenting en-vironmental impacts (Kerans and Karr 1994). We calculated site densities of Neosho madtoms and, as a group, potential competitors (Table 1). We calculated fish densities by dividing the total num-ber of Neosho madtoms or potential competitors collected at a site by the total area sampled with the kick seine. We determined the list of potential competitors based on habitat preferences and food habits of each species, as described by Pflieger (1975). For each site, we calculated species rich-ness as a general measure of the natural and an-thropogenic impacts on the fish community.
Be-cause species richness values depend highly on the level of effort (sampling time, area, or both), we also calculated species rarefaction, which adjusts species richness estimates to a constant level of effort (Hurlbert 1971; James and Rathbun 1981), as suggested by Ludwig and Reynolds (1988): E(S)= 1;E(S) = expected number of species;? = total number of fish collected; n, = total number of fish collected in species i;N = sample size;gravel bars from the same interstitial spaces where Neosho madtoms are found and because surface water and pore-water measurements were similar (see Wildhaber et al. 1996), only pore-water con-centrations are presented here. However, surface water measurements were incorporated into esti-mates of Neosho madtom densities because no pore-water measurements were collected in 1991.Benthic invertebrates for metals analyses were col-lected from seines used for fish sampling, aug-mented with kick-net collections when necessary.
Invertebrates were placed in acid-washed plastic bags with acid-cleaned, Teflon-coated forceps.They were analyzed by ICAP for the same metals as pore waters except As and Hg were not ana-lyzed. For metals, benthic invertebrate samples were partitioned into "Decapoda" (crayfish), S = total number of species collected.
INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 249 We used rarefaction to calculate expected number of species at a site [E(S)] when a given number of fish (N) are collected.
The number of stations per site sampled for fish ranged from 10 to 25, so we used species rarefaction to make species richness comparable among sites. Comparable species rich-ness values among sites were produced by using the same sample size (N) for each site in all rar-efaction calculations.
The sample size (N) used in all rarefaction calculations was the lowest number of fish collected at any one site. We did not cal-culate any similarity indices; these were reported by Schmitt et al. (1997).For substrate, we calculated size category means at each site by dividing total weight of a size cat-egory by total weight of all size categories.
We also calculated the substrate geometric mean and fredle index (geometric mean adjusted for distri-bution of particle sizes) at each site, as suggested by McMahon et al. (1996), to characterize sub-strate suitability for Neosho madtoms. The fredle index relates potential permeability of sediment to water and hence is an indirect index of dissolved oxygen transport within sediment, and it has been correlated with the emergence success of salmonid alevins (Platts et al. 1983, citing other sources).Composite site means for metal concentrations in benthic invertebrates were calculated by summing the product of metal concentration (pLg/g) and bio-mass of a taxonomic category (g) over all taxo-nomic categories and dividing the sum by the total biomass of all taxonomic categories combined (g).We first checked site means for normality and then tested homogeneity of variance for river sys-tem differences using Levene's test, as recom-mended by Milliken and Johnson (1984). In 1994, the number of madtom and no-madtom sites in the Spring River were almost equal (9 and I I sites, respectively).
Therefore, for tests between madtom and no-madtom sites, we assumed that F-statistics and I tests for comparisons of normally distributed variables would be effective whether or not vari-ances were equal, as suggested by Milliken and Johnson (1984). Any variable with nonnormal site means was log 1 0-transformed.
The absence of Ne-osho madtoms (density = 0) at II of 20 sites in the Spring River in 1994 made it impossible to normalize densities through transformation even with the addition of a constant before transfor-mation. Thus, for 1994 data, we restricted corre-lation and regression analyses to madtom sites, which precluded development of multiple-regres-sion models.Stepwise multiple linear regression and Krus-kal-Wallis tests were used to compare observed and predicted densities of Neosho madtoms from the Spring River in 1994 based on observed den-sities of Neosho madtoms from the Neosho system in 1991 (SAS Institute 1990). Stepwise multiple linear regression with forward selection was used to develop a model based on physical habitat, wa-ter chemistry, and nutrient measures from 1991 Neosho system data. The variable list used in-cluded depth, water velocity, substrate size cate-gories, geometric mean of substrate size, fredle index, and surface water chemistry.
Inclusion of individual variables in the model was based on an a = 0.15 criterion and a final model in which all variables were significant at a = 0.05. The model based on the 1991 data were used to estimate Ne-osho madtoms densities at sites sampled in 1994.Kruskal-Wallis tests were used to validate the USFWS 1991 model from 1994 Neosho system data and to assess distributional differences be-tween observed and predicted 1994 Neosho mad-tom densities in the Spring River upstream and downstream of Center Creek (i.e., most upstream source of mining-derived contaminants).
The statistical methods used to make primary comparisons within 1994 data included analysis of variance (ANOVA), correlation analysis, multi-variate ANOVA (MANOVA), principal compo-nents analysis (PCA), and discriminant analysis (SAS Institute 1990). Separate one-way ANOVAs were performed on site means for each variable between river systems and between madtom and no-madtom sites. Along with testing the composite metal concentration for benthic invertebrates, we tested if either of the major groups, Decapoda and Megaloptera, biased our composite results. We tested for differences in metal concentration be-tween Decapoda and Megaloptera at sites where both groups were represented.
We also tested for significant differences between river systems and between madtom and no-madtom sites for Decap-oda and Megaloptera concentrations separately.
Because log, 0-transformation of NO 2 + NO 3 pore-water concentrations did not produce equal vari-ances between river systems, NO 2 + NO 3 concen-trations were analyzed with a Welch (1951) vari-ance-weighted ANOVA. Correlation analyses were used to assess relationships between nonzero Neosho madtom densities and other variables.
We used the multivariate tests to verify the results of the individual ANOVA tests and to determine if the significant differences identified by ANOVA effectively characterized river system and mad-tom-no-madtom differences.
In our discriminant 250 WILDHABER ET AL.c-o E wE z*0 0 X 0 10 100 1,000 Predicted Neosho madtorn density (fish/100 in 2)X Neosho River system 0 Spring River below Center Creek 0 Spring River above Center Creek -Predicted equals observed FIGURE 3.-Predicted versus observed Neosho madtoms densities in Neosho and Spring river systems. Predicted values are based on a regression model developed from 1991 USFWS data. Observed values were obtained during 1994 collections.
Observed densities of the Neosho and Cottonwood rivers combined (Neosho system) were sig-nificantly higher than densities found the Spring River below Center Creek (P = 0.0007; N = 21, all comparisons here are from Kruskal-Wallis tests), whereas predicted values did not differ between river systems (P = 0.28; N= 16). Observed and predicted Neosho madtom densities in the Neosho system were not significantly different for observed and predicted densities, respectively, at Spring River sites above Center Creek (P = 0.057, N = 21;P = 0.19, N = 16). Observed Neosho madtom densities at Spring River sites above Center Creek were significantly higher than densities at sites below Center Creek (P = 0.008; N = 20), whereas predicted values were not (P =0.94; N = 20).analyses, we used stepwise discriminant analyses with forward selection followed by removal to pro-duce a discriminant function.
We then tested how well the resulting discriminant function described the observed data.Because we were required to have more obser-vations than variables before we did any multi-variate analyses, we shortened the list of variables used for MANOVA, PCA, and discriminant anal-ysis in three ways. First, we excluded from mul-tivariate analyses any metal that was detected at fewer than 75% of our sites. Second, we used the fredle index to represent all substrate categories.
Third, we used only variables with P-values less than 0.05 in one-way ANOVAs.Results Predicted Neosho Madtom Densities Stepwise regression with forward selection of 1991 USFWS data from the Neosho system pro-duced the following equation for predicting Ne-osho madtom densities from physical habitat, wa-ter chemistry, and nutrient measurements:
D = 10-1.447-0.892Ioglo(G38)-0.0897CI; D = density of Neosho madtoms (number/100 M 2);G38 = weight proportion of substraten
- 38 mm;Cl = chloride ion concentration (mg/L).For the equation, r2 = 0.72; N = 11; P < 0.017 for G3; and P < 0.0026 for Cl. Based on a Bon-ferroni-adjusted ct = 0.0025 (0.05/20 compari-sons), Cl was highly correlated with 50 4 (r = 0.89;P = 0.0003; N = 11), conductivity (r = 0.83; P= 0.0015; N = I1), and hardness (r = 0.82; P =0.0022; N = 11). As a result of these strong cor-relations, only Cl significantly added to the vari-ance in the 1991 data that was accounted for by the overall regression model.At the six 1994 sites where water quality and substrate composition were measured in the Ne-osho system, predicted Neosho madtom densities ranged from 12.1/100.m 2 less to 42.6/100 m 2 more than observed densities (Figure 3). Despite the wide range, observed and predicted 1994 densities for the Neosho system were not significantly dif-ferent (Kruskal-Wallis test for distributional dif-ferences:
P = 0.92; df = 1). Likewise, predicted INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 251 and observed densities of Neosho madtoms at Spring River sites above Center Creek did not dif-fer significantly.
Below Center Creek. however, observed densities were markedly lower than pre-dicted. Of the combined 26 sites on the Neosho, Cottonwood, and Spring rivers, the Spring River sites at the mouth of Center Creek and downstream.
at Willow Creek were predicted to have the two highest densities of Neosho madtoms. The ob-served average density of Neosho madtoms was 100% of the predicted average density above Cen-ter Creek but only 1% of prediction below Center Creek. Above Center Creek, the predicted average density of Neosho madtoms was 13% of that pre-dicted for the Neosho system, whereas below Cen-ter Creek the average predicted density was 364%of the density predicted for the Neosho system.Neosho System versus Spring River Fishes and invertebrates.-The aquatic com-munities of the Neosho system and Spring River differed, as illustrated by fish densities and by fish and invertebrate community composition (Table 2). Nonzero Neosho madtom densities were higher in the Neosho system tha.n in the Spring River (Table 2). Furthermore, Neosho madtoms were collected at only 9 of 20 sites in the Spring River as opposed to 10 of II sites in the Neosho system.Density of potential competitors was also greater in the Neosho system than in the Spring River. In contrast, fish species rarefaction was greater in the Spring River than in the Neosho system. Neither species richness of fish and benthic invertebrates nor EPT differed between river systems (Table 2).Physical habitat, water chemnistry, and nutri-ents.-The Neosho system and Spring River differ in their physical habitat, water chemistry, and nu-trient concentrations.
Most of the substrate mea-surements and indices indicate that Spring River substrate consists of coarser gravel than that of the Neosho system (Table 2). Pore waters of the Ne-osho system were warmer, harder, had higher NH 3 and SO 4 concentrations, and were more conduc-tive, alkaline, and turbid than those of the Spring River (Table 2). Pore waters from all sites except those on the Cottonwood River were typically al-kaline (pH 7.5-8.5, alkalinity 100-160 mg/L) and hard (150-220 mg!L); the Cottonwood River was particularly high in alkalinity (about 200 mg/L)and very hard (>330 mg/L) (Schmitt et al. 1997).No doubt reflecting the dissolution of naturally occurring gypsum in central Kansas (Spruill 1987), pore-water concentrations of sulfate were more than twofold greater in the Cottonwood River (132-145 nag/L) than in the Neosho River (49-58 mg/L) and more than threefold greater than in reaches of the Spring River and its tributaries not affected by mining (Schmitt et al. 1997). In con-trast, NO, + NO 3 concentrations were greater in the Spring River than the Neosho system (Table 2). Depth., velocity, substrate 9 to <19 mm, and pore-water pH, dissolved oxygen, P0 4 , and CI did not differ between river systems (Table 2).Metals.-Concentrations of various metals in pore waters and benthic invertebrates differed sig-nificantly between the Neosho system and Spring River. Concentrations of Fe and Mn in pore water were higher in the Neosho system, Cd was only detected in Spring River pore waters at the mouth of Turkey Creek, and Pb was not detected in any pore-water sample (Table 2). Concentrations of Cd and Pb were higher in composite samples of ben-thic invertebrates from the Spring River than in those from the Neosho system (Table 2). Detect-able concentrations of As, Hg, and Zn in pore wa-ters and of Fe and Zn in composite invertebrate samples did not differ significantly between river systems (Table 2).Except for a few inconsistencies among com-posite, Decapoda, and Megaloptera metal concen-trations, taxonomic group analyses generally sup-ported the results of river system comparisons based on composite samples (Table 2). Concen-trations of Fe and Mn were significantly higher in Megaloptera than in Decapoda (respectively:
F =62.57 and 9.86; P = 0.0001 and 0.0032; N = 42).Concentrations of Cd in Decapoda were not quite significantly different between river systems. Lead was detected in the Neosho system at only one site for Decapoda and at no sites for Megaloptera.
Con-centrations of Zn in Megaloptera were greater in the Spring River than in the Neosho system Physical habitat, water chemistry, nutrients, and metals combined.-The shortened list of variables used in multivariate analyses included pore-water temperature, conductivity, turbidity, alkalinity, and hardness; NH 3 , NO, + NO 3 , SO 4.and Mn pore-water concentrations; the fredle index; and Cd, Mn, and Zn concentrations in composite in-vertebrate samples. Results of MANOVA dem-onstrated a significant difference between river systems (Wilks' lambda: P < 0.002; N =22). Prin-cipal components analysis of the same variables accounted for more than 63% of the variability in the data with only the first two principal compo-nents (Figure 4); the first component effectively separated Neosho system sites from Spring River sites. Based on the same 22 sites used in MAN-252 WILDHABER ET AL.TABLE 2.-Means and one-way analysis of variance test results (P-values) for comparisons between the Neosho River system and Spring River and between Spring River sites with and without Neosho madtoms. Abbreviations used are: ND = none detected; NA = not applicable; DL = detection limit or detection limit range.Within-Spring River comparisons Between-river comparisons Neosho Neosho Neosho Spring madtoms madioms system: River: P-value present: absent: P-value Measurement mean (N) mean (N) (F) mean (N) mean (N) (F)Community 12.00(10) 3.26(9) 0.042 (4.83)Neosho madtom density (per 100 m 2)Density of potential benthic fish competitors (per 100 m 2)Fish species raretaction Fish species richness Invertebrate taxa richness Emphemcroptera, Plecoptera, Tricoptera (EPT) richness Water depth (m)Velocity at 60% of depth (mis)Pore-water temperature (0 C)Pore-water turbidity (nephelometric units)Substrate
->38 mm (weight %)Substrate
<38 to 2-19 mm (weight %)Substrate
<19 to ->9 mm (weight %)Substrate
<9 to -2 mm (weight %)Substrate
<2 mm (weight %)Substrate geometric mean Fredle index pH Dissolved oxygen (mg,/L)Conductivity (1smhos/cm)
Alkalinity (mg/L)Hardness (rng/L)Un-ionized NH 3 (mg/L)NO, + NO 3 (mg/L)SO4 (mg/L)P0 4 (mg/L)CI (mg/L)As (DL = 12.3)Cd (DL = 0.59)Fe (DL = 6.52)Hg (DL = 0.10)Mn (DL 0.06)Pb (DL = 4.12)Zn (DL = 10.9)Cd (DL = 0.038-0.27)
Composite Decapoda Megaloptera Fe (DL = 3.81-27.02)
Composite Decapoda Megaloptera Mn (DL = 0.15-1.08)
Composite 301.94(11) 6.18(11)16.64(11)28.2 (5)120.89 (20) 0.0045 (9.46)8.10 (20) 0.0053 (9.08)16.35 (20) 0.88 (0.02)33.15(20) 0.17 (2.05)18.40 (5) 17.65 (20) 0.78 (0.08)Habitat 0.33 (11) 0.35 (20) 0.56 (0.35)0.44(11) 0.47(20) 0.56 (0.35)25.84 (6) 22.05 (20) 0.012 (7.40)254.90(6)12.64 (6)26.67 (6)23.71 (6)23.08 (6)13.90(6)I 1.74 (6)5.96 (6)7.85 (6)8.36(6)0.56(6)165.61 (6)230.04 (6)0.12 (61 0.22 (6)72.80 (6)0.23 (6)16.20 (6)20.60 (2)ND 75.97 14)0.22 (3)83.21 (6)ND 44.54 (6)59.27 (20) 0.0007 (15.21)25.72 (20) 0.026 (5.64)35.24 (20) 0.030 (5.35)17.66 (20) 0.052 (4.27)14.52 (20) 0.0085 (8.21)6.86 (20) 0.0009 (14.49)20.83 (20) 0.0034 (10.58)10.56(20) 0.0079 (8.39)Pore-water chemistry 7.84 (20) 0.97 (0.00)7.36 (20) 0.064 (3.77)0.40 (20) 0.0019 (12.15)135.90 (20) 0.0008 (14.78)169.36 (20) 0.0005 (15.89)0.04(20) 0.0001 (29.89)1.47 (20) 0.0006 (15.44)31.80(20) 0.0007(15.11) 0.34 (20) 0.074 (3.49)16.39 (20) 0.94 (0.01)Pore-water metals (gug/L)16.38 (4) 0.24 (1.94)0.73(1) NA 16.78 (7) 0.0045 (14.15)0.11(2) 0.35 (1.23)32.55 (20) 0.041 (4.65)ND NA 55.34 (20) 0.27 (1.30)3.26 (9)195.49 (9)8.11 (9)19.22 (9)37.00 (9)21.33 (9)0.37 (9)0.44(9)20.56 (9)86.83 (9)22.05 (9)36.85 (9)19.10 (9)14.56 (9)7.44 (9)19.34(9)10.09 (9)7.77 (9)7.86 (9)0.39 (9)142.93 (9)170.94 (9)0.06 (9)1.33 (9)26.71 (9)0.34 (9)17.63 (9)13.40 (2)ND 13.93 (4)0.11 (t)40.11 (9)ND 47.72 (9)0 (Il)81.59(11)8.09(11)14.00(11)30.00(11)14.64(11)0.39 (11)0.50o(1)23.26(11)43.37(11)28.72(1I)33.93(11)16.480(1)14.49(11)6.39(1I)22.05(11)10.96(11)7.90(t])6.95 (11)0.41 (11)130.15 (11)168.07(11) 0.04(11)1.590 I1)36.68(11)0.35(I1)15.45(I1)19.35 (2)0.73 (1)21.50 (3)0.10(1)27.43(11)ND 62.470(1)NA 0.0094 (8.46)0.98 (0.00)0.05 (4.42)0.021 (6.46)0.0038 (11.02)0.54 (0.40)0.28 (1.23)0.055 (4.23)0.044 (4.79)0.25 (1.42)0.46 (0.56)0.30 (1.13)0.98 (0.00)0.57 (0.34)0.38 (0.81)0.71 (0.15)0.35 (0.93)0.093 (3.15)0.53 (0.40)0.0092 (8.51)0.70 (0.16)0.0053 (10.03)0.46 (0.57)0.11 (2.90)0.91 (0.10)0.32 (1.04)0.11 (7.64)NA 0.32 (1.22)NA 0.34 (0.96)NA 0.26 (1.33)Benthic invertebrate metals (p.g/g)0.10(6) 0.23 (16) 0.02 (6.34)0.10 (4) 0.23 (12) 0.079 (3.60)0.07 (4) 0.24 (13) 0.019 (6.92)153.91 (6) 147.37 (20) 0.87 (0.03)99.59 (5) 102.69 (18) 0.87 (0.03)198.97(4) 272.30(15) 0.16 (2.14)47.73 (6) 115.39(20) 0.0021 (I 1.83)0.14 (6) 0.32(10) 0.021 (6.80)0.12(5) 0.43(7) 0.016 (8.48)0.11 (4) 0.33(9) 0.042 (5.30)147.17(9) 147.54(11) 0.99 (0.00)119.71 (9) 88.10(91 0.092 (3.211 296.99 (6) 256.98 (9) 0.53 (0.42)130.31 (9) 104.47(11) 0.40 (0.74)
INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 253 TABLE 2.-Continued.
Within-Spring River comparisons Between-river comparisons Neosho Neosho Neosho Spring madtoms madtoms system: River: P-value present: absent: P-value Measurement mean (N) mean (N) (F) mean tN) mean (N) (F)Decapoda 40.19 (5) 94.10 (18) 0.0085 (8.43) 110.89(9) 79.85 (9) 0.29 (1.22)Megaloptera 54.51 (4) 207.01 (15) 0.0003 (20.25) 241.55 (6) 186.81 (9) 0.42 (0.70)Pb (DL = 0.38-2.70)
Composite 0.73 (2) 2.01 (15) 0.021 (6.61) 1.58 (6) 2.36 (9) 0.17 (2.15)Decapoda 0.68(l) 1.90(8) 0.15 (2.56) 1.46(2) 2.07 (6) 0.53 (0.45)Megaloptera ND 2.71 (10) NA 1.01 (2) 3.48 (8) 0.11 (3.26)Zn (DL = 0.76-5.40)
Composite 27.64 (6) 40.14 (20) 0.18 (1.92) 26.64 (9) 56.13(11) 0.005 (10.24)Decapoda 27.43 (5) 36.02 (18) 0.32 (1.05) 27.35 (9) 47.47 (9) 0.032 (5.54)Megaloptera 24.2414) 55.34 (15) 0.023 (6.23) 32.64 (6) 78.65 (9) 0.0043 (11.87)OVA, stepwise discriminant analysis produced a list of four significant variables:
pore-water al-kalinity, NH 3 , SO 4 , and temperature.
The resulting discriminant function successfully categorized by river system the 26 sites at which all variables were measured (0% error rate).Neosho Madtoin versus No-Neosho Madlom Sites in the Spring River Fish and invertebrates.-The following metrics were significantly greater at Spring River madtom sites than at no-madtom sites: potential competi-tors, fish species richness, benthic invertebrate taxa richness, and EPT (Table 2). Fish rarefaction did not differ between madtom and no-madtom sites (Table 2).Physical habitat, water chemistrv, and nutri-ents.-Water chemistry and nutrient measure-ments revealed a few differences between madtom and no-madtom sites, and no differences in phys-ical habitat were evident. Madtom sites had higher NH 3 , alkalinity, and turbidity than no-madtom sites (Table 2), but depth, water velocity, all sub-strate size categories, temperature, pH, dissolved oxygen, conductivity, hardness, NO 2 + NO 3 , SO 4 , P0 4 , and CI in pore water did not differ signifi-cantly between madtom and no-madtom sites.Metals.-Only concentrations of metals in ben-thic invertebrates differed between madtom and no-madtom sites. Cadmium and Zn concentrations in benthic invertebrates were higher at no-madtom sites than at madtom sites, whereas detectable con-centrations of Fe, Mn, and Pb in benthic inverte-brates did not differ significantly between madtom and no-madtom sites (Table 2). As with the be-tween-river systems analyses, concentrations of Fe and Mn were significantly higher in Megaloptera than in Decapoda (respectively, F = 57.60 and 11.98; P = 0.0001 and 0.0019; N = 28). Separate analyses of invertebrate taxonomic groups pro-duced the same results as the composite analysis and thus supported use of the composite analyses.As noted earlier, Cd in pore water was only de-tected at the mouth of Turkey Creek, where no madtoms were collected.
Detectable concentra-tions of pore-water As, Fe, Hg, Mn, and Zn were not significantly different between madtom and no-madtom sites. Although its concentrations did not differ significantly between madtom and no-madtom sites, Zn in pore water was elevated at the mouths of Center Creek (116 t+/-g/L) and Turkey Creek (369 pýg/L) (Schmitt et al. 1997).Physical habitat, water chemistry, nutrients, and metals combined.-The shortened list of variables used in multivariate analyses included pore-water turbidity, alkalinity, NH 3 , EPT (which paralleled invertebrate taxa richness), and Cd and Zn con-centrations in composite invertebrate samples. Re-sults of MANOVA demonstrated a significant dif-ference between madtom and no madtom sites (Wilks' lambda: P < 0.051; N = 16). Principal components analyses of the variables used in MANOVA accounted for more than 84% of the variability in the data with only the first two prin-cipal components; the first component effectively separated most Neosho madtom sites from no-madtom sites (Figure 5). Based on the same 16 sites used in MANOVA, invertebrate Zn concen-tration was the only significant variable in stepwise discriminant analysis.
The resulting discriminant 254 WILDHABER ET AL.CY 01 CL E 0.4, 3 2 1I 0'-1'smaller substrate; higher turbidity.
alkalinity, hardness, conductivity, un-ionized ammonia, sulfate, and manganese and lower nitritel, nitrate in water; lower cadmium, manganese, and zinc In Invertebrales S 3-2-N greater Emphemeroplera.
Plecoplera, and Tricoptera richness;higher alkalinity and turbidity In water; lower cadmium and zinc In invertebrates N S 5 S S NN S N S S S CM r 0 N CIL E 0 0 Q.2 C*C 0.N Y N N NN N y Y Y N N Y-2-N-21-3t-3 S 6 7, 2 -1 0 1 2 3 4 5 Principal Component 1-3 1 0 1 Principal Component 1 2 3 N Neosho/Cotlonwood Rivers S Spring River FIGURE 4.-Principal components (PC) analysis of combined Neosho, Cottonwood, and Spring river data based on the shortened list of variables used for MAN-OVA. For the 22 sites at which all PC variables were measured, detected, or both, PCI accounted for 43% of the variability in the data and PC2 accounted for over 19%. Correlations (or loadings) of the variables used for PCI were pore-water temperature (r = 0.17, P 0.45), turbidity (r = 0.72, P = 0.0002), alkalinity (r 0.85, P = 0.0001), hardness (r = 0.82, P = 0.0001), con-ductivity (r = 0.71, P = 0.0002), NHW (r = 0.78, P =0.0001), NO 2 + NO 3 (r = -0.69. P = 0.0004). SO 4 (r= 0.61, P -0.0024), Mn (r = 0.60, P = 0.003); fredle index (r = -0.50, P = 0.017); and invertebrate Cd (r= -0.67, P = 0.0007), Mn (r = -0.57, P = 0.006), and Zn (r --0.62. P = 0.002).function based on invertebrate Zn successfully cat-egorized as madtom or no-madtom sites 16 of the 20 Spring River sites (20% error rate).Discussion Through this study, we have shown that an in-tegrated approach is necessary to differentiate the effects of natural and anthropogenic factors on fish populations and communities.
Fishes of the Spring River., especially the Neosho madtom, may be di-rectly limited by the presence of Pb, Zn, and Cd in water and indirectly limited by the concentra-tions of these metals in benthic invertebrate food sources as a result of historic Pb-Zn mining. In N No Neosho madtoms collected I Y Neosho madtoms collected FIGURE 5.-Principal components (PC) analysis of Spring River sites with and without Neosho madtoms based on the shortened list of variables used for MAN-OVA. For the 16 sites at which all PC variables were measured, PCI accounted for 6 3% of the variability in the data and PC2 accounted for 19%. Correlations (or loadings) of the variables used for PC1 were pore-water turbidity (r = 0.60, P = 0.013), alkalinity (r = 0.84, P= 0.0001), and NH 3 (r = 0.49, P = 0.056); EPT (r =0.92, P = 0.0001); and invertebrate Cd (r = -0.88, P= 0.0001) and Zn (r = -0.95, P = 0.0001).the Spring River. Neosho madtom populations may also be directly limited by lower benthic inverte-brate abundance (i.e., food) at sites where Neosho madtoms were not collected, possibly as an indi-rect result of contaminants.
The Neosho madtom population numbers also appear limited by avail-able physical habitat, they may be affected by ba-sic water chemistry and nutrients in the Spring River. In contrast, our results suggest that com-petition between Neosho madtoms and other fishes is not limiting Spring River Neosho madtom pop-ulations.According to estimates from the model gener-ated for the Neosho system and based on habitat and water quality of the Spring River, observed Neosho madtom densities in the Spring River above Center Creek were as expected (i.e., low), whereas below Center Creek observed densities INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 255 were much lower than expected (Figure 3). The Spring River below Center Creek appears to con-tain habitat that could support Neosho madtom densities higher than was found, on average, in the Neosho system. Thus, Neosho madtom densities in the upper portion of the Spring River appear limited only by habitat, whereas densities below Center Creek appear limited not by physical hab-itat but by the presence of contaminants.
Further-more, variation in the accuracy of predicted (rel-ative to observed) densities of Neosho madtoms in the Neosho system and Spring River above Cen-ter Creek (Figure 3) suggest that other environ-mental factors not accounted for in the model also affect Neosho madtom densities.
Highest concentrations of Pb, Zn, and Cd in the Spring River (all media) occurred below the con-fluence with Center Creek at sites where Neosho madtoms were not found (Wildhaber et al. 1996, 1997; Schmitt et al. 1997). In pore water, Pb was never detected and the only detectable Cd was at the mouth of Turkey Creek. The two highest Zn levels in pore water occurred at the mouth of Tur-key Creek (highest) and at the mouth of Center Creek. Average concentration of Zn in pore water at the mouth of Center Creek was 1.42 times great-er than that of the next highest site. Concentrations of metals in benthic invertebrates paralleled those in water, invertebrates having their highest Pb, Zn, and Cd levels at the mouths of Turkey and Center creeks. During 1993, dissolved Zn concentrations in the Spring River just below the confluence with its North Fork were 2.5-80 1+/-g/L during low river flow and 50-80 ILg/L during high flow (Dames and Moore 1993). Dissolved Pb was never greater than I pg/L, and Cd never historically exceeded 0.2 pLg/L, but higher concentrations occurred in Tur-key, Center, and Short creeks (Dames and Moore, Inc., Denver, Colorado, unpublished data). Al-though we did not detect either Pb or Cd by ICAP, others have documented elevated concentrations of these elements in the Spring River and its trib-utaries by more sensitive analytical methods. In the Spring River below Baxter Springs, dissolved Pb averaged 70 ltg/L from 1974 to 1978, and dis-solved Cd averaged about 2 pRg/L. From 1979 to1991, dissolved Pb averaged 24 l.g/L and dis-solved Cd averaged 3 ltg/L (Dames and Moore 1993). The dissolved Zn concentration in Center Creek was 264 l.g/L in the summer of 1989 (Schmitt et al. 1993). Zinc concentrations as great as 200,000 R.g/L have been reported in Short Creek (Spruill 1987).Of the mining-derived metals, Zn concentrations in pore water were sufficiently high to be toxic to Spring River fishes. Based on the USEPA (1987)chronic water quality criteria for Zn, which is hard-ness-dependent, pore-water concentrations of Zn exceeded the chronic criterion by 78% at the mouths of Center and Turkey creeks. Furthermore, because toxicities of heavy metals may be cu-mulative (Sprague and Ramsay 1965; Wildhaber and Schmitt 1996), concentrations of Pb, Zn, and Cd that may not be individually toxic may be cu-mulatively toxic in the Spring River.Higher metal concentrations in benthic inver-tebrates and lower densities of potential compet-itors at sites where Neosho madtoms were not found suggest that Spring River fishes are exposed to metals indirectly via their food as well as di-rectly via the water. As with waterborne metal con-centrations, Pb, Zn, and Cd concentrations in ben-thic invertebrates were greatest at the mouths of Center and Turkey creeks (Wildhaber et al. 1997).Working with laboratory rainbow trout Oncorhvn-chus inykiss fed a diet containing a mixture of Pb, Zn, Cd, and copper (Cu), Farag et al. (1994) re-ported scale loss and accumulation of metals in pyloric caeca, and Woodward et al. (1994) dem-onstrated reduced growth and tissue accumulation of metals. The concentrations of Pb, Zn, and Cd in invertebrates found by Farag et al. (1994) to be detrimental to fish were less than those we ob-served in benthic invertebrates at the mouth of Center and Turkey creeks; the concentration of Cu was slightly higher (Table 3). The concentrations of Pb and Cd in food found to be detrimental to fish by Woodward et al. (1994) were less than those we observed in benthic invertebrates at the mouth of Turkey and Center creeks; the experimental con-centrations of Zn and Cu were 121% and 150% of the concentrations we observed at the mouth of Turkey Creek (Table 3). Other studies have dem-onstrated detrimental effects of foodborne Pb (Thomas and Juedes 1992) and Cd (Rhodes et al.1985). We were not able to measure concentrations of metals in fish and our invertebrate metal anal-yses were done on taxa that may or may not be food of Neosho madtoms. However, Czarneski (1985) and Schmitt et al. (1993) reported elevated concentrations ofPb, Zn, and Cd in black redhorse from Center Creek, and it is therefore likely that other benthic fishes are similarly contaminated.
Consequently, our results and those of the studies cited here suggest that dietary metals play a role in constraining the Spring River Neosho madtom population.
In the Spring River, depauperate invertebrate 256 WILDIHABER ET AL.TABLE 3.-Concentrations of selected metals in invertebrate food sources from this study, Farag et a]. (1994), and Woodward el al. (1994). The concentrations presented for Farag et al. (1994) and Woodward et al. (1994) are the minimum levels at which a detrimental effect on fish was observed.
Concentrations are given as ýig/g wet weight (wet)or l.g/g dry weight (dry).Location or study: measurement type Cadmium Copper Lead Zinc Turkey Creek: wet 0.72 21.20 4.22 104.29 Turkey Creek: dry 3.04 90.14 17.93 443.47 Center Creek: wet 1.15 18.77 6.06 126.85 Center Creek: dry 4.26 69.29 22.36 468.35 Farag et al. (1994): wet 0.24 26.13 1.77 68.99 Woodward et al. (1994): dry 1.20 109 9.69 655 abundance may directly limit riffle-dwelling ben-thic fishes, including the Neosho madtom. Most of the riffle-dwelling benthic fishes in Spring River feed on benthic invertebrates, including the young, small instars of those used in the EPT index (Pflie-ger 1975; Mayden et al. 1980; Burr and Mayden 1982; Starnes and Starnes 1985). Our data illus-trate the greater numbers of EPT invertebrates at madtom sites than at no-madtom sites (Wildhaber et al. 1996). The similarity in habitat and the dif-ferences in contaminant concentrations between madtom and no-madtom sites suggest that ob-served benthic invertebrate patterns resulted from contaminants.
Phipps et al. (1995) demonstrated sensitivity of aquatic invertebrates to waterborne Pb, Zn, and Cd with Zn concentrations only 21 and 35% of the surface water concentration we observed at the mouths of Turkey and Center creeks, respectively (see Schmitt et al. 1997 for actual values).Depth, velocity, and substrate are important to Neosho madtoms (Moss 1983; Fuselier and Edds 1994). Our study showed no significant differences in either depth or velocity between river systems or between Spring River madtom and no-madtom sites. However, our study differs from previous investigations in that our analyses are based on overall site means and not microhabitat values, which may vary greatly within a site. The specific substrate composition needs of fishes have been demonstrated by many researchers (Moyle and Vondracek 1985; Wood and Bain 1995) and some studies have focused on threatened and endangered species (Kessler and Thorp 1993; Freeman and Freeman 1994), including other madtoms (Simon-son and Neves 1992). Substrate particle size in the Neosho system tended to be smaller than in the Spring River, but there was no difference in par-ticle size distribution between madtom and no-madtom sites within the Spring River (Table 2).Our observation of smaller substrate sizes in the Neosho system than in the Spring River parallels previous observations of a preference for moder-ate- to fine-grained substrate by the Neosho mad-tom (Moss 1983; Fuselier and Edds 1994). The larger average particle size in the Spring River and the significant negative regression coefficient for particle sizes larger than 38 mm in the model used to predict Neosho madtom densities suggest sub-strate limitations for the Neosho madtom and other riffle-dwelling benthic fishes in the Spring River, especially above Center Creek. Perhaps the larger interstitial spaces in Spring River gravel do not afford as much protection from predators or as much food for Neosho madtoms as the Neosho system provides, but offers habitat and food for other species such as stonecats.
Basic water chemistry and nutrients differed be-tween river systems; most important are those that differed between madtom and no-madtom sites (alkalinity, NH 3 , and turbidity).
The water chem-istry and nutrient patterns we observed in the two river systems parallel those observed by Moss (1983). We know little about the importance of alkalinity and NH 3 to Neosho madioms. The high correlation found among the various water chem-istry and nutrient measurements makes any dis-cussion of the importance of alkalinity and NH 3 by themselves highly speculative.
However, the potential importance of basic water quality to Ne-osho madtoms populations is suggested by the in-clusion of Cl, which was highly correlated with conductivity, hardness, and SO 4 , in the predictive model for Neosho madtom densities and by the significant differences between madtom and no-madtom sites in alkalinity, NH 3 , and turbidity.
Like other prairie stream fishes (Layher et al.1987), Neosho madtoms seem to prefer higher tur-bidities.
Higher turbidities may afford Neosho madtoms more protection from predators and more opportunity to capture prey with good visual acu-ity. There was significantly higher turbidity at INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 257 Spring River madtom sites than at no-madtom sites but the difference was minimal compared to the fourfold greater turbidity in the Neosho system than in the Spring River.Many of the physical habitat, water chemistry, nutrient, and community differences observed be-tween the Neosho system and the Spring River likely are due to the physiographic regions drained. Although the main-stem reaches we sam-pled in these three rivers are all found in the Prairie Parkland Province ecoregion, the upper reach and many of the tributaries of Spring River drain the very different Ozark Uplands Province (Bailey 1995). This ecoregional effect, which has been documented by others (e.g., Layher and Maughan 1985; Rabeni and Sowa 1996; Leftwich et al.1997), is an important consideration in understand-ing how Neosho madtom populations are being affected in the Spring River. The reach of the Spring River supporting Neosho madtoms is the most prairie-like because it is influenced by the North Fork of the Spring River and Cow Creek, which are prairie streams (Figure 1). The Ozark Uplands Province, part of which is drained by some Spring River tributaries and the upper reach-es of the main stem, has many spring-fed streams and is composed of limestone that contains large quantities of coarse chert and flint, unconsolidated chert acting as a water filter (Pflieger 1975). More than one-third of Missouri fishes have their dis-tribution centered in the Ozark Uplands (Pflieger 1975). The spring-fed nature, coarse substrate, clear water, and high species diversity of the Ozark Uplands are the likely reasons why Spring River has lower temperature, larger substrate, lower tur-bidity, and higher fish species rarefaction, respec-tively, than the Neosho system.The Neosho and Spring River systems differ substantially in soil types and land use (Moss 1983), differences that are reflected in variables such as conductivity, hardness, alkalinity, SO 4 ,and metals such as Mn. The comparatively high con-centrations of dissolved constituents in the Cot-tonwood River reflect the rocks and soil in its wa-tershed (Hem 1985) and the contribution of the Chase-Council Grove aquifer, the waters of which are characteristically high in SO 4 and other ions (Baker and Hansen 1988). Consequently, naturally high SO 4 concentrations are typical of the Neosho River system (Kenny and Snethen 1993). In the carbonate-dominated Spring River, some elevation of SO 4 occurs from the weathering of pyrite (iron sulfide) in the Pennsylvanian-age shales that over-lie the western part of its watershed (Spruill 1987).In the Tri-State District, ground and surface waters in the Spring River drainage are affected by mining to varying degrees, and SO 4 is an indicator of min-ing-derived water pollution (Barks 1977; Spruill 1987). In these areas, SO 4 results from the oxi-dation of pyrite as well as from the weathering of sulfide ore minerals (sphalerite and galena).Contrary to what the USFWS (1991) suggested, the observed fish community pattern suggests that interspecific competition is not limiting Neosho madtoms. Fish species richness of Spring River madtom sites was higher than that of no-madtom sites, which is likely due to the lower fish densities at no-madtom sites. After species richness was ad-justed for density by rarefaction, there was no dif-ference between Spring River madtom and no-madtom sites. The Spring River did have greater rarefaction than the Neosho system, as expected from descriptions by Cross and Collins (1995).The significant positive correlation between Ne-osho madtom density and potential competitors as a group indicated that Neosho madtom densities increase along with the density of other fishes. If interspecific competition was a primary factor lim-iting Neosho madtom populations, Neosho mad-tom densities should have decreased as densities of potential competitors increased.
Previous re-search has supported (Gilliam et al. 1993; Winston 1995) and refuted (Angermeier 1982; Grossman and Freeman 1987) interspecific competition as a determinant of fish community structure.
A likely scenario is one of alternating interspecific com-petition (density-dependent factors) and environ-mental impacts such as flooding or pollution (den-sity-independent factors) as determinants.
Inter-specific competition becomes important when den-sity-independent factors are not limiting (Strange et al. 1992); currently, Neosho madtoms seem to be limited by density independent factors such as contaminants and habitat quality. More detailed studies and analyses of interspecific relationships between the Neosho madtom and other species are necessary to further define the role of competition in regulating Neosho madtom populations.
Other factors not measured could affect Neosho madtom populations.
We focused on the benthic aquatic communities of gravel bars where Neosho madtom are found and did not attempt to assess communities in pools or other habitats.
This de-cision was based on our primary focus of collect-ing fishes with similar environmental preferences and the scarcity of Neosho madtoms in any other habitat (Fuselier and Edds 1994). Our focus on riffle-dwelling benthic fish species precluded col-258 WILDHABER ET AL.lecting any data on fish predators.
Predators (i.e., black and temporate basses) in all three rivers are similar (Pflieger 1975; Cross and Collins 1995), but we do not know if predator density differs between rivers. If it does, it could have influenced some of the patterns we have documented.
Conclusions When one evaluates limiting factors for rare fishes such as the Neosho madtom, it is important to consider anthropogenic factors as well as phys-ical habitat, basic water chemistry, and nutrients.
Knowledge of either low habitat quality or envi-ronmental contamination alone does not necessar-ily lead to effective management decisions that will stop suspected declines of fish populations.
Habitat improvement may not improve population status or community composition in a stream if the stream is also heavily contaminated.
Conversely, removal of contaminants may also not affect spe-cies of concern because physical habitat or basic water quality may be marginal for those popula-tions or communities.
Our results suggest that anthropogenic and nat-ural factors limit Neosho madtom populations in the Spring River. Where metals contamination is minimal, Neosho madtom densities seem to be lim-ited primarily by physical and chemical habitat quality and availability.
Where contamination has occurred, Neosho madtoms seem to be limited pri-marily by the presence of contaminants acting di-rectly (via mortality or avoidance) or indirectly (by suppressing, contaminating, or both the benthic invertebrate food base).Future research into understanding the popula-tion dynamics of the Neosho madtom should in-clude a more detailed look at regional and local factors. A regional factor that may be important to Neosho madtom populations is the regulation of water levels through impoundments on the Ne-osho and Cottonwood rivers. Local factors include more comprehensive investigations of the effects of microhabitat-scale environmental quality on Neosho madtom distribution across a gravel bar and an evaluation of fish communities, including predators, found in all habitats associated with gravel bars where Neosho madtoms are found.Acknowledgments This study was jointly funded and undertaken by the U.S. Environmental Protection Agency (USEPA), Region 7; the U.S. Geological Survey, through its Columbia Environmental Research Center (CERC); and the USFWS, through its Eco-logical Services (ES) Field Office in Manhattan, Kansas. Assistance with field collections and data processing was provided by A. Bissing, D. Har-desty, P. Heine, P. Lovely, B. Mueller, S. Olson, B. Poulton, B. Scharge, S. Russler, T. Thorn, R.Walton, and D. Whites of CERC; by M. Legg and D. Munie, contracted through USEPA; by C. Char-bonneau from USFWS-ES Field Office in Colum-bia. Missouri; by B. Wilkerson of Oklahoma De-partment of Wildlife Conservation; and by D.Wright of Missouri Southern State College (MSSC). P J. Lamberson assisted compiling ref-erences and reviewed the initial draft of this manu-script. We thank J. Messick of MSSC for providing laboratory space during the study. We gratefully acknowledge the cooperation of the many private landowners in Kansas, Missouri, and Oklahoma who granted us permission to sample on their prop-erties. This manuscript was greatly improved by comments from D. F Woodward of the CERC and three anonymous reviewers.
References Aadland, L. P. 1993. Stream habitat types: their fish assemblages and relationship to flow. North Amer-ican Journal of Fisheries Management 13:790-806.
Allen, G. T., and S. H. Blackford.
1995. Contaminants evaluation of Neosho madtom habitats in the Ne-osho River drainage in Kansas. U.S. Fish and Wild-life Service Contaminants, Region 6 Contaminants Program, Contaminant Report R6i513M/95, Man-hattan, Kansas.Angermeier, P. L. 1982. Resource seasonality and fish diets in an Illinois stream. Environmental Biology of Fishes 7:251-264.
Angermeier, P. L., and J. R. Karr. 1984. Relationships between woody debris and fish habitat in a small warmwater stream. Transactions of the American Fisheries Society 113:716-726.
APHA (American Public Health Association), American Water Works Association, and Water Pollution En-vironment Federation.
1992. Standard methods for the examination of water and wastewater, 18th edi-tion. APHA, Washington, D.C.Bailey, R. G. 1995. Description of the ecoregions of the United States. U.S. Department of Agriculture Mis-cellaneous Publication 1391.Baker, C. H., Jr., and C. V. Hansen. 1988. Kansas ground-water quality. U.S. Geological Survey Wa-ter-Supply Paper 2325:259-264.
Barks, J. H. 1977. Effects of abandoned lead and zinc mines and tailings piles on water quality in the Jop-lin area, Missouri.
U.S. Geological Survey Water Resources, Investigations 77-75:1-49, Rolla, Mis-souri.Bryan, M. D., G. J. Atchison, and M. B. Sandheinrich.
1995. Effects of cadmium on the foraging behavior INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 259 and growth of juvenile bluegill, Lepomis macro-chirus. Canadian Journal of Fisheries and Aquatic Sciences 52:1630-1638.
Burr, B. M., and R. L. Mayden. 1982. Life history of the brindled madtom Noturus miurus in Mill Creek, Illinois (Pisces: Ictaluridae).
American Midland Naturalist 107:25-41.
Cross, F. B., and J. T. Collins. 1995. Fishes in Kansas, 2nd edition. University Press of Kansas, Lawrence.Czarneski, J. M. 1985. Accumulation oflead in fish from Missouri streams impacted by lead mining. Bulletin of Environmental Contamination and Toxicology 34:736-745.
Dames and Moore. 1993. Final remedial investigation for Cherokee County. Kansas CERCLA site, Baxter Springs/Treece subsites.
Dames and Moore, Inc., Denver.Eisler, R. 1988. Lead hazards to fish, wildlife, and in-vertebrates:
a synoptic review. U.S. Fish and Wild-life Service Biological Report 85(1.14).Eisler, R., and R. J. Hennekey.
1977. Acute toxicities of Cd2", Cr 6", Hg 2* and Zn2÷ to estuarine macro-fauna. Archives of Environmental Contamination and Toxicology 6:315-323.
Farag, A. M., C. J. Boese, D. F. Woodward, and H. L.Bergman. 1994. Physiological changes and tissue metal accumulation in rainbow trout exposed to foodborne and waterborne metals. Environmental Toxicology and Chemistry 13:2021-2029.
Fausch, K. D., and R. J. White. 1981. Competition be-tween brook trout (Salvelinusfontinalis) and brown trout (Salmo trutta) for positions in a Michigan stream. Canadian Journal of Fisheries and Aquatic Sciences 38:1220-1227.
Finger, T. R. 1982. Interactive segregation among three species of sculpins (Cotius).
Copeia 1982:680-694.
Freeman, B. J., and M. C. Freeman. 1994. Habitat use by an endangered riverine fish and implications for species protection.
Ecology of Freshwater Fish 3: 49-58.Fuselier, L., and D. R. Edds. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleostei:
Ictaluridae:
Nolurus placidus).
Southwestern Natu-ralist 39:217-223.
Gill, T. S., J. C. Pant, and H. Tewari. 1989. Cadmium nephropathy in a freshwater fish, Puntius conchon-ius Hamilton.
Ecotoxicology and Environmental Safety 18:165-172.
Gilliam, J. F, D. F Fraser, and M. Alkins-Koo.
1993.Structure of a tropical stream fish community:
a role for biotic interactions.
Ecology 74:1856-1870.
Gorman, 0. T., and J. R. Karr. 1978. Habitat structure and stream fish communities.
Ecology 59:507-515.
Grossman, G. D., and M. C. Freeman. 1987. Micro-habitat use in a stream fish assemblage.
Journal of the Zoological Society of London 212:151-176.
Hall, L. W., Jr., M. C. Scott, W. D. Killen, Jr., and R.D. Anderson.
1996. The effects of land-use char-acteristics and acid sensitivity on the ecological sta-tus of Maryland coastal plain streams. Environ-mental Toxicology and Chemistry 15:384-394.
Hem, J. D. 1985. Study and interpretation of the chem-ical characteristics of natural water, 3rd edition.U.S. Geological Survey Water-Supply Paper 2254.Hurlbert, S. H. 1971. The non-concept of species di-versity: a critique and alternative parameters.
Ecol-ogy 52:577-586.
James, F C., and S. Rathbun. 198 1. Rarefaction, relative abundance, and diversity of avian communities.
Auk 98:785-800.
Johansson-Sjobeck, M. L., and A. Larsson. 1979. Ef-fects of inorganic lead on delta-aminolevulinic acid dehydratase activity and hematological variables in the rainbow trout, Solmo gairdnerii.
Archives of En-vironmental Contamination and Toxicology 8:419-431.Kenny, J. F., and D. H. Snethen. 1993. Kansas stream water quality. U.S. Geological Survey Water-Sup-ply Paper 2400:277-284.
Kerans, B. L., and J. R. Karr. 1994. A benthic index of biotic integrity (B-IBI) for rivers of the Tennessee valley. Ecological Applications 4:768-785.
Kessler, R. K., and J. H. Thorp. 1993. Microhabitat segregation of the threatened spotted darter (Eth-eosloma maculatum) and closely related orangefin darter (E. belium). Canadian Journal of Fisheries and Aquatic Sciences 50:1084-1091.
Kiner, L. K., C. Vitello, and K. Hash. 1997. Spring River basin inventory and management plan. Missouri Department of Conservation, Jefferson City.Layher, W. G., and E. Maughan. 1985. Relations be-tween habitat variables and channel catfish popu-lations in prairie streams. Transactions of the Amer-ican Fisheries Society 114:771-781.
Layher, W. G., E. Maughan, and W. D. Warde. 1987.Spotted bass habitat suitability related to fish oc-currence and biomass and measurements of physi-ochemical variables.
North American Journal of Fisheries Management 7:238-251.
Leftwich, K.N., P. L. Angermeier, and C. A. Dolloff.1997. Factors influencing behavior and transfera-bility of habitat models for a benthic stream fish.Transactions of the American Fisheries Society 126: 725-734.Lemaire-Gony, S., P. Lemaire, and A. L. Pulsford.
1995.Effect of cadmium and hcnzo(a)pyrene on the im-mune system, gill ATPase and EROD activity of European sea bass Dicentrarchus lahrax. Aquatic Toxicology 31:297-313.
Ludwig, J. A., and 1. F. Reynolds.
1988. Statistical ecol-ogy: a primer on methods and computing.
Wiley, New York.Luttrell, G. R., R. D. Larson, W. J. Stark, N. A. Ash-baugh, A. A. Echelle, and A. V. Zale. 1992. Status and distribution of the Neosho madtom (Noturus placidus) in Oklahoma.
Procedures of the Oklahoma Academy of Science 72:5-6.Maret, T. R., C. T. Robinson, and G. W. Minshall.
1997.Fish assemblages and environmental correlates in least disturbed streams of the upper Snake River basin. Transactions of the American Fisheries So-ciety 126:200-216.
Matthews, W. J., and D. C. Heins. 1987. Community 260 WILDHABER ET AL.and evolutionary ecology of North American stream fishes. University of Oklahoma, Norman.Mayden, R. L., B. M. Burr, and S. L. Dewey. 1980.Aspects of the life history of the Ozark madtom, Noturus albater, in southeastern Missouri (Pisces lctaluridae).
American Midland Naturalist 104: 335-340.McCormick, F. H., B. H. Hill, L. P. Parrish, and W. T.Willingham.
1994. Mining impacts on fish assem-blages in the Eagle and Arkansas rivers, Colorado.Journal of Freshwater Ecology 9:175-179.
McMahon, T. E., A. V. Zale, and D. J. Orth. 1996.Aquatic habitat measurements.
Pages83-120 in B.R. Murphey and D. W. Willis, editors. Fisheries techniques, 2nd edition. American Fisheries Soci-ety, Bethesda, Maryland.Merritt, R. W., and K. W. Cummins. 1984. Introduction to the aquatic insects of North America, 2nd edition.Kendall/Hunt, Dubuque, Iowa.Milliken, G. G., and D. E. Johnson. 1984. Analysis of messy data, volume 1: designed experiments.
Wads-worth, Belmont, California.
Moss, R. E. 1983. Microhabitat selection in Neosho River riffles. Doctoral dissertation.
University of Kansas, Lawrence.Moyle, P. B., and B. Vondracek.
1985. Persistence and structure of the fish assemblage in a small California stream. Ecology 66:1-13.Neves, R. J., and P. L. Angernieier.
1990. Habitat al-teration and its effects on native fishes in the upper Tennessee River system, east-central U.S.A. Journal of Fish Biology 37:45-52.Omernik, J. M. 1987. Ecoregions of the conterminous United States. Annals of the Association of Amer-ican Geographers 77:118-125.
Pfiieger, W. L. 1975. The fishes of Missouri, 2nd edition.Missouri Department of Conservation, Jefferson City.Phipps, G. L., V. R. Mattson, and G. T. Ankeley. 1995.Relative sensitivity of three freshwater benthic mac-roinvertebrates to ten contaminants.
Archives of En-vironmental Contamination and Toxicology 28: 281-286.Platts, W. S., W. F. Megahan, and G. W. Minshall.
1983.Methods for evaluating stream, riparian, and biotic conditions.
U.S. Forest Service General Technical Report INT-138.Rabeni, C. F., and S. P. Sowa. 1996. Integrating bio-logical realism into habitat restoration and conser-vation strategies for small streams. Canadian Jour-nal of Fisheries and Aquatic Sciences 53:252-259.
Rhodes, L., and six coauthors.
1985. Interactive effects of cadmium polychlorinated biphenyls, and fuel oil on experimentally exposed English sole (Parophrvs vetuhis).
Canadian Journal of Fisheries and Aquatic Sciences 42:1870-1880.
Roark, S., and K. Brown. 1996. Effects of metal con-tamination from mine tailings on allozyme distri-butions of populations of great plains fishes. En-vironmental Toxicology and Chemistry 15:921-927.SAS Institute.
1990. SAS procedures guide, volume 6, 3rd edition. SAS Institute, Cary, North Carolina.Schmitt, C. J., M. L. Wildhaber, J. B. Hunn, T. Nash, M. N. Tieger, and B. L. Steadman.
1993. Biomon-itoring of lead-contaminated Missouri streams with an assay for erythrocyte d-aminolevulinic acid de-hydratase (ALA-D) in fish blood. Archives of En-vironmental Contamination and Toxicology 25: 464-475.Schmitt. C. J., M. L. Wildhaber, A. Allert, and B. C.Poulton. 1997. The effects of historic zinc-lead mining and related activities in the tri-states mining district on aquatic ecosystems supporting the Ne-osho madtom, Notur-us placidus, in Jasper County, Missouri; Ottawa County, Oklahoma; and Cherokee County, Kansas. Final Report to U.S. Environmen-tal Protection Agency, Region 7, Kansas City, Kan-sas.Scott, M. C., and L. W. Hall, Jr. 1997. Fish assemblages as indicators of environmental degradation in Mary-land coastal plains streams. Transactions of the American Fisheries Society 126:349-360.
Simonson, T. D., and R. J. Neves. 1992. Habitat suit-ability and reproductive traits on the orangefin mad-tom Noturus gilberti (Pisces: Ictalauridae).
Ameri-can Midland Naturalist 127:115-124.
Smith, B. J. 1988. Assessment of water quality in non-coal mining areas of Missouri.
U.S. Geological Sur-vey, Water Resources Investigations 87-4286, Rol-la, Missouri.Somero, G. N., P. H. Yancey, T. J. Chow, and C. B.Snyder. 1977. Lead effects on tissue and whole or-ganism respiration of the estuarine teleost fish, Gil-lichthys mirabilis.
Archives of Environmental Con-tamination and Toxicology 6:349-354.
Sprague, J. B., and B. A. Ramsay. 1965. Lethal levels of mixed copper-zinc solutions for juvenile salmon.Journal of the Fisheries Research Board of Canada 22:425-432.
Spruill, T. B. 1987. Assessment of water resources in lead-zinc mined areas in Cherokee County, Kansas and adjacent areas. U.S. Geological Survey Water-Supply Paper 2268.Starnes, L. B., and W. C. Starnes. 1985. Ecology and life history of the mountain madtom, Notur-us eleutherus (Pisces Ictaluridae).
American Midland Naturalist 114:331-341.
Strange, E. M., P. B. Moyle, and T. C. Foin. 1992. In-teractions between stochastic and deterministic pro-cesses in stream fish community assembly.
Envi-ronmental Biology of Fishes 36:1-15.Taylor, W. R. 1969. A revision of the catfish genus No-turus Rafinesque, with an analysis of higher ver-tebrate groups in the lctaluridac.
U.S. National Mu-seum Bulletin 282.Thomas, P., and M. J. Juedes. 1992. Influence of lead on the glutathione status of Atlantic croaker tissues.Aquatic Toxicology 23:11-30.USEPA (U.S. Environmental Protection Agency). 1986.Ambient water quality criteria.
USEPA, 440/5-86-001, Washington, D.C.USEPA (U.S. Environmental Protection Agency). 1987.
INFLUENCES ON NEOSHO MADTOM DISTRIBUTION 261 Ambient water quality criteria for: zinc. USEPA, 4405-87-003, Washington, D.C.USFWS (U.S. Fish and Wildlife Service).
1991. Neosho madtom recovery plan. USFWS, Denver.Wagner, G. F, and B. A. McKeown. 1982. Changes in plasma insulin and carbohydrate metabolism of zinc-stressed rainbow trout, Salmo gairdneri.
Ca-nadian Journal of Zoology 60:2079-2084.
Weber. D. N. 1993. Exposure to sublethal levels of wa-terborne lead alters reproductive behavior patterns in fathead minnows (Pirnephales promelas).
NeuroToxicology 14:347-358.
Welsh, B. L. 1951. On comparison of several mean val-ues: an alternate approach.
Biometrica 38:330-336.
Wetzel, R. B. 1983. Limnology.
Saunders.
Philadelphia.
Wildhaber, M. L., and C. J. Schmitt. 1996. Estimating aquatic toxicity as determined through laboratory tests of Great Lakes sediments containing complex mixtures of environmental contaminants.
Environ-mental Monitoring and Assessment 41:255-289.
Wildhaber, M. L., C. J. Schmitt, and A. L. Allert. 1996.Historic zinc-lead mining and related activities and their effects on aquatic ecosystems supporting the Neosho madtom, Noturus placidus, in Jasper Coun-ty, Missouri; Ottawa County, Oklahoma:
and Cher-okee County,.Kansas.
Appendix A: tables A I-A35.Draft Final Report to U.S. Environmental Protection Agency, Region 7, Kansas City, Missouri.Wildhaber, M. L., C. J. Schmitt, and A. L. Allert. 1997.Elemental concentrations in benthic invertebrates collected from the Neosho, Cottonwood, and Spring rivers. Final Report to U.S. Fish and Wildlife Ser-vice, Region 6, Manhattan, Kansas.Wilkinson, C., D. Edds, J. Dorlac, M. L. Wildhaber, C.J. Schmitt, and A. Allert. 1996. Neosho madtom distribution and abundance in the Spring River.Southwestern Naturalist 41:78-81.Winston, M. R. 1995. Co-occurrence of morphologi-cally similar species of stream fishes. American Nat-uralist 145:527-545.
Wood, B. M., and M. B. Bain. 1995. Morphology and microhabitat use in stream fish. Canadian Journal of Fisheries and Aquatic Sciences 52:1487-1489.
Woodward, D. E, W. G. Brumbaugh, A. J. DeLonay, E.E. Little, and C. E. Smith. 1994. Effects on rainbow trout fry of a metals-contaminated diet of benthic invertebrates from the Clark Fork River, Montana.Transactions of the American Fisheries Society 123: 51-62.Woodward, D. F., J. N. Goldstein, A. M. Farag, and W.G. Brumbaugh.
1997. Cutthroat trout avoidance of metals conditions characteristic of a mining water site: Coeur d'Alene River, Idaho. Transactions of the American Fisheries Society 126:699-706.
Woodward, D. F., J. A. Hansen, H. L. Bergman, E. E.Little, and A. J. Delonay. 1995. Brown trout avoid-ance of metals water characteristic of the Clark Fork River, Montana. Canadian Journal of Fisheries and Aquatic Sciences 52:2031-2037.
Animal Behavior Sodety..)týABS Conservation Committee The Conservation Behaviorist, an electronic news-update, informs ABS members about the Conservation Committee's activities, research trends in behavior and conservation, and relevant scientific news in conservation research where behavior plays an bnportant role.www.aninalbehavior.org/Committees/Conservation Vol.4 No. 1 -May2006 Building a Case for Conservation Behavior In this special issue of The Conservation Behaviorist (TCB)we include articles published since 2003 and new essays. The area of conservation behavior has grown significantly during the past decade and the Animal Behavior Society Conservation Committee has played an important role in this process,.
Besides TCB, the Committee has created the E. 0. Wilson Conservation Award, three data bases available online (funding opportunities for behavioral research, publications in conservation behavior, mentors in conservation behavior), and has also sponsored scientific events at the Society's annual meetings.
We are confident that our efforts benefit ABS, the conservation community, and the public at large.The ABS Conservation Committee Created in.1997, the Conservation Committee aims to encourage ABS members to participate in research programs addressing the interface between animal behavior and conservation science. By identifying and evaluating the areas in which behavioral research has contributed to conservation, as well as the fields that need development, the Committee seeks to generate discussion and promote studies in behavior and conservation.
Interact with the Conservation Committee Send letters, announcements, comments and contributions to The Conservation Behaviorist gpazymino@worcester.edu Deadlines for articles are the 15t of the month preceding the next news update. The next deadline is October 15th. Contributions submitted by members of the Animal Behavior Society and judged by the Conservation Committee to be appropriate will be published in The Conservation Behaviorist The publication of such material does not imply ABS or Conservation Committee endorsement of the opinions expressed by contributors.
Editor Guillermo Paz-y-Miflo C.Associate Editor Allison C. Alberts ABS Conservation Committee Members Guillermo Paz-y-Mlfio C., Chair Worcester State College Allison C. Alberts Zoological Society of San Diego Daniel T. Blumstein University of California Los Angeles Richard Buchholz University of Missispp Colleen Cassady St. Clair University of Alberta, Canada Elizabeth V. Lonsdorf Lincoln Park Zoo J. Cully Nordby University of Califomia Berkeley Debra M. Shier University of Davis Ronald R. Swaisgood Zoological Society of San Diego Ilonka von Lippke University of California Los Angeles Mark L Wildhaber Columbia Environmental Research Center The Conservation Behaviorist 2006 Vol. 4 (1): 1
- 3. Stoinski T.S., Rosenbaum S., Fawcett K.A. (submitted).
Reproductive behavior of male mountain gorillas:
Shared opportunities or limited control?4. van Lawick Goodall J. 1968. Behavior of free-living chimpanzees of the Gombe Stream area. Animal Behavior Monographs 1:163-311 5. Nishida T. 1968. The social group of wild chimpanzees in the Mahale Mountains.
Primates 9:167-224 6. Goodall, J 1986. The Chimpanzees of Gombe: Patterns of Behavior.
Cambridge:
Harvard University Press 7. Wrangham R.W. 1979. Sex differences in chimpanzee dispersion.
In: The Great Apes (Ed by DA Hamburgh & ER McCown), pp. 480-489. Menlo Park, California:
Benjamin/Cummings
- 8. Pusey AE. 1990. Behavioral changes at adolescence in chimpanzees.
Behavior 115: 203-246 9.Nishida T., Corp N., Hamai M., Hasegawa T., Hiraiwa-Hasegawa M., Hosaka K., Hunt K.D., Itoh N., Kawanaka K., Matsumoto-Oda A, Mitani J.C., Nakamura M., Norikoshi K., Sakamald T., Turner L., Uehara S., Zamma K. 2003. Demography, female rife history and reproductive profiles among the chimpanzees of Mahale. American Journal of Primatology 59:99-121 10. Czekala N., Robbins M. M. 2001. Assessment of reproduction and stress through hormone analysis in gorillas.
In: Mountain gorillas:
three decades of research at Karisoke (Ed by MM Robbins, P Sicotte, KJ Stewart) pp. 317-339. Cambridge:
Cambridge University Press 11. Ammarm K. 2001. Bushmeat hunting and the great apes. In: Great apes & humans: the ethics of coexistence. (Ed by BB Beck, TS Stoinski, M Hutchins, TL Maple, 8 Norton, A Rowan, EF Stevens, A Artuke) pp. 71-85. Washington DC: Smithsonian Institution Press 12. Wilkie D., Shaw E., Rotberg F., Morelli G., Auzel P. 2000. Roads, development, and conservation in the Congo Basin. Conservation Biology 14:1614-1622
- 13. Watts D.P. 1989. Infanticide in mountain gorillas:
new cases and a reconsideration of the evidence.
Ethology 81:1-18 14. Stokes E.J., Parnell R.J., Olejniczak C. 2003. Female dispersal and reproductive success in wild western lowland gorillas (Gon/la gonrla gorilla).
Behavioral Ecology and Sociobiology 54:329-339
- 15. Tulin C., Stokes E., Boesch C., Morgan D., Sanz C., Reed T., Biona A., Walsh P., Blake S., Kormos R. 2005. Regional Action Plan for the Conservation of Chimpanzees and Gonillas in Western Equatorial Africa. Washington, DC: Conservation International
- 16. Pusey A.E, Pintea L., Wilson M.W., Kamenya S., and Goodall J. In press. 'The contribution of long-term research at Gombe National Park to chimpanzee conservation.
Conservation Biology 17.wilson M.W., Wrangham R.W. 2003. Intergroup relations in chimpanzees.
Annual Review of Anthropology 32: 363-392 18. Wilson M.L., Hauser M.D, Wrangham R.W. 2001. Does participation in intergroup conflict depend on numerical assessment, range location or rank for wild chimpanzees.
Animal Behavior 61: 1203-1216 19. Goossens, B., Setchell, J.M., Tchidongo E., Dflambaka E., Vidal C., Ancrenaz M., Jamart A. 2005. Survival, interactions with conspecifics and reproduction in 37 chimpanzees released into the wild. Biological Conservation 123: 461-475 20. Goodall J. 1983. Population dynamics during a 15 year period in one community of free-living chimpanzees in the Gombe National Park, Tanzania.
Zeitschrlt fur Tierpsychologie 61: 1-60 21. Walsh P.D., Abemethy KA., Bermejo M., Beyers R., De Wachter P., Ella Akou M., Huijbregts B., Idiata Mambounga D., Kamdem Toharn A., Kilboum A.M. et al. 2003.Catastrophic ape decline in western equatorial Africa. Nature 422:611-414
- 22. Homsy J. 1999. Ape Tourism and Human Diseases:
How Close Should We Get? A critical review of the rules and regulations governing park management and tourism for the wild mountain gorilla, Gorilla gorilla beringei.
Report to the Intemational Gorilla Conservation Programme 23. Guerrera W., Sleeman J.M., Jasper S.B., Pace L.B., Ichinose T.Y., Reif J.S. 2003.Medical survey of the local human population to determine possible health risks to the mountain gorillas of Bwindi Impenetrable Forest National Park, Uganda. International Journal of Primatology 24:197-207
- 24. Lonsdorf E.V., Travis D., Pusey A.E., Goodall, J. In press. Using retrospective health data from the Gombe chimpanzee study to inform future monitoring efforts. American Journal of Primatology:
Special Topics Issue on Disease Risk Analysis An imal Behavi r Society 43rd Annhual Meeting 12-1 &August, 2006 www.animalbehavior.orgIABS/MeetingslSnogwbirdO6/
The Role of Reproductive Behavior in the Conservation of Fishes: Examples from the Great Plains Riverine Fishes By Mark L. Wildhaber*
Recovery efforts for threatened and endangered fish species are hampered by lack of knowledge on their reproductive ecology. Habitat requirements and environmental stimuli necessary for reproduction are unknown and vary widely among species. For Great Plains riverine fishes, this is often complicated by the high turbidity of the system where the species occur, which precludes direct visual observation of behavior.
Innovative methods for collecting behavioral data are required to better understand the conditions necessary for successful reproduction.
To this goal, I will discuss four fish species on which I have worked in collaboration with university and agency researchers, graduate students, state and federal resource managers, and private landowners.
The species are: Topeka shiner (Notropis topeka -Gilbert 1884), a headwater and low-order stream species, Neosho madtom (Noturus piacidus Taylor 1969), a middle-size river species, and pallid (Scaphirhynchus albus Forbes and Richardson 1905) and shovelnose sturgeon (S. platotynchus Rafinesque 1828), large river species. These species demonstrate the variety of physical requirements necessary for successful reproduction in Great Plains riverine fishes. The recovery plans for these fishes indicate that information on behavior and habitat requirements for spawning is lacking'".
Topeka Shiner The Topeka shiner was listed as an endangered species in 19993. It is a small, stout minnow (<75 mm total length -TL) characteristic of small, low order (headwater) prairie streams. Topeka shiners occur in pool and run areas of streams, seldom being found in riffles. They are pelagic, occurring in mid-water and surface areas, and are primarily considered schooling fish 4.Clean gravel, cobble and sand are the predominant substrates within Topeka shiner streams. Kems 5 found that this species primarily feeds on insects while Hatch 6 found it to be omnivorous (flowering-plant seeds are common in the diet). Topeka shiners are broadcast spawners (i.e. eggs are released over open substrate) in pool habitats, over green sunfish (Lepomis cyanellus) and orangespotted sunfish (L humilis) nests, with males establishing small territories on the edges of these nests'A.Topeka shiner (Notropls topeka). Konrad Schmidt@ photo The Topeka shiner is affected by habitat destruction, degradation, modification, and fragmentation resulting from siltation, reduced water quality, tributary impoundment, stream channelization, in-stream gravel mining, changes in stream hydrology, and introduced predaceous fishes 3.The historic distribution of Topeka shiners included low order tributary streams throughout the central prairie regions of the United States. Topeka shiner occurrences have declined by 80 percent (50 percent within the last 40 years); isolated and fragmented populations now exist in less than 10 percent of its original range.The Conservation Behaviorist 2006 Vol. 4 (1): 15 Limited reproductive success is considered a potential cause for the decline of the species 3.My research focuses on the effects of temperature and photoperiod on reproductive development and behavior, as well as substrate particle size preference.
Approach The small size of adult Topeka shiners makes laboratory studies a relatively easy task. Under controlled conditions, adults are exposed to various combinations of photoperiod, temperature, and substrate to determine which combination is most effective at stimulating reproduction.
For these studies, adult fish came from hatchery ponds run by state and federal resource managers.The experiments included individually controlled and monitored experimental chambers and simulated winter conditions to assess stimulation of reproductive development Six females and one male were placed in a tank under specific temperature and photoperiod combinations.
Each tank was monitored with video cameras to minimize experimenter's disturbance and to record counting and spawning behaviors, defined as presence and successful hatching of eggs.Neosho madtom fNoturus placidus).
Janice L Bryan 0 photo Topeka shiner experiment tank. Christopher S. Witte 0 photo Information gained Preliminary results suggest that the combination of photoperiod and temperature are important factors influencing reproduction.
Longer photoperiods and temperatures between 22-280 C enhance reproductive development, while 310 C hinder the process. The next step in this research will be to determine substrate preferences under photoperiod and temperature combinations in which spawning behavior and success are highest This research should provide the U.S. Fish and Wildlife Service (USFWS)with information on the spawning requirements of the Topeka shiner, it will also help to identify suitable habitats for reintroductions and plan large-scale production for reintroductions, which ultimately will contribute to recover the species 3., Neosho Madtom The Neosho madtom was listed as threatened in 19911. It is a small (<75 mm TL) ictaludd fish endemic to the mainstems of the Neosho and Cottonwood rivers in Kansas and Oklahoma and the Spring River in Kansas and Missouri 7". This species occupies portions of riffles with mean flows of 79 cm/sec, mean depths of 0.23 m, and unconsolidated pebble and gravel (2-64 mm in diameter)'
0.Neosho madtoms feed at night on larval insects found among the gravel 8.High abundance of this species has been documented in riffles in late summer and early fall, after young-of-year (YOY) are estimated to have recruited to the population 7', 0.1 1.Previous research suggests that the Neosho madtoms have an annual lifecycle with recruitment of YOY into adult collection gear about the time the adults begin to disappear from collections' 1.Once distributed throughout the Spring-Neosho (Grand) River system, this species is now restricted to portions of the Neosho and Cottonwood Rivers in Kansas and Oklahoma, with one remnant population in the Spring River in Kansas. Much of Neosho madtom's historic habitat has been inundated by impoundments'.
Additional habitats have been degraded by in-stream gravel mining, feedlot operations, and lead-zinc mining1 2.Reservoir operations have affected reproduction and survival1 3.Similar methods to those described for Topeka shiner (above)'4 ,'1 7 , 9 have been used to examine the effects of photoperiod, temperature, and water flow on the reproductive behavior of Neosho madtom. In this specific study, the main goal was to determine the temperature' range, light period within which spawning occurs, and if excessive water flow limits spawning.Approach The small size of the Neosho madtom allowed for laboratory work under controlled conditions.
Adults were exposed to various combinations of photoperiod, temperature, and water flow to determine the most effective at stimulating reproduction.
Since production of offspring in the laboratory has been limited, for these studies individuals had to be obtained from the wild.The collection of data employed time-lapsed videography for monitoring behavior, individual controlled and monitored experimental chambers, and simulated winter conditions to stimulate reproductive development.
One female and one male were placed in a tank under a specific combination of temperature, photoperiod, and flow, and supplied with a gravel substrate and a PVC nesting objects. Each tank was monitored with video cameras to minimize human disturbance and to document courting, spawning, and rearing behaviors'.
4 1 9.The nest building habits of Neosho madtoms facilitated the collection of up-close spawning behaviors using an additional camera placed inside each nest2°. In initial studies, sex was determined through secondary sexual characteristics and internal examination upon completion of the study.In later studies designed to document changes in reproductive state under varying temperature and photoperiod, a medical ultrasound unit was used to confirm sex and to estimate fecundity of the same individuals over several annual cycles. Presence and successful hatching of eggs indicated successful spawning.Information gained The studies demonstrated that Neosho madtoms' proportion of time spent performing cavity enhancement was higher, cavities were deeper, and gravel size in cavities was smaller for fish given a longer photoperiod' 4.Courtship behaviors were observed in male-female pairs held in longer photoperiods, but not in shorter photoperiods.
Under flowing water conditions, there was a decreased average frequency, proportion of time, and event duration of male nest building behavior1 9.Water flow decreased the overall frequency of occurrence of reproductive behavior sequences.
Spawning was observed between 21 to 28D C, with most occurring at 250 C. Temperature and The Conservation Behaviorist 2006 Vol. 4 (1): 16 photoperiod influenced the reproductive cycle and increased river flows during spawning could have affected reproductive success negatively.
Knowledge of how photoperiod, temperature, and water flow affect Neosho madtom reproductive success will provide information to the USFWS and the U.S. Army Corps of Engineers on how flow regulation in concert with natural photothermal changes can be used to improve species recovery plans.Pallid and Shovelnose Sturgeon The pallid sturgeon was listed as endangered by USFWS in 19902.Although the shovelnose sturgeon is not listed by the USFWS, as either threatened or endangered, it has been listed as vulnerable by the World Conservation Commission 2 r. The pallid is a mid-sized sturgeon reaching up to 30 kg in weight, the shovelnose is smaller (<3 kg); both are native to the Missouri and Mississippi Rivers=22.
The shovelnose sturgeon feeds primarily on invertebrates, while the larger pallid sturgeon starts out feeding on invertebrates but shifts later to a fish diet 2 4"-. Pallid sturgeons are adapted to large, turbid, riverine environments and do not frequent tributaries or Clear-water riverine habitats, used by shovelnose sturgeon 2 7.Spawning habitat preferences of pallid and shovelnose sturgeon are not known; both species are assumed to spawn in current over coarse substrate 2 7 2 8.Like most sturgeon species, pallid and shovelnose sturgeon are suspected to be broadcast spawners where the eggs become adhesive soon after release and attach to the substrate until hatch 2 g. Biologists speculate that spawning runs are dependent on river flow 2 8 ,3O, 3 1.Spawning behavior, habitat, and environmental cues necessary to elicit spawning have not been documented.
Morphological, physiological and genetic similarities indicate that pallid and shovelnose sturgeon are closely related,22.-.
Therefore, research on the shovelnose sturgeon may be also applicable to the conservation of the pallid sturgeon.Neosho madtom experiment tank. Janice L Bryan @ photo Pallid sturgeon (Scaphlrhynchus albus). Steven Krentz @ photo Ultrasound use on Neosho madtom. Mark L Wildhaber 0 photo Shovelnose sturgeon (Scaphirhynchus platorynchus).
Aaron J. DeLonay @photo As with many sturgeon species, habitat alteration and destruction are limiting factors for pallid and shovelnose sturgeon 3 5 r. The shovelnose sturgeon may also be threatened by commercial over-harvest for the caviar industry, which has eliminated it from part of its range 2.The USFWS recovery plan for the pallid sturgeon lists rehabilitation of habitat as necessary for reproduction and recruitment 2.The shovelnose sturgeon is more common and widespread than the pallid sturgeon 2 8 I Past distribution of the species includes the Mississippi, Missouri, Ohio, and Rio Grande Rivers and their tributaries.
There has been a 30% reduction in the shovelnose sturgeon range, with an additional 30% reduction in population predicted for the next 10 years (three generations) 2 1. If the shovelnose and pallid sturgeon are to be conserved and recovered, their limited reproduction will be the primary obstacle to overcome 2.The goal of this research is to determine the ecological requirements for successful reproduction of pallid and shovelnose sturgeon in the Missouri River. The specific objectives are to: (1) determine the direction, magnitude, and habitat used during spawning migrations, (2) understand the reproductive physiology prior to and after successful and unsuccessful spawning, and (3)evaluate the effect that a semi-natural increase in flow has on the reproductive status, movements, and habitat use.Neosho madtom spawning event, Janice L Bryan @ photo The Conservation Behaviorist 2006 Vol. 4 (1): 17 Approach The approach of this study is interdisciplinary and integrates physiology, behavior, habitat use, and physical habitat assessment to document sturgeon spawning and assess the effects of environmental variables on spawning success. In the field, as many as 100 sturgeon were collected and assessed for reproductive state, fecundity of females, and gonadosomatic index using ultrasonic and endoscopic methods 3 7.Blood samples were taken for hormone analyses.
Female sturgeon that were ready to spawn were tagged both with ultrasonic telemetry tags (for relocating fish) and data storage tags (DSTs) that continuously monitor depth and temperature from within the fish's body cavity.This study took place in two different (ca. 640 km each) segments of the 1280 km Lower Missouri River. One of the river segments is highly influenced by controlled flows while the other has more natural flows, which allowed a comparison of the effects of natural and artificial flows on reproductive behavior.The tagged fish were located repeatedly throughout the spawning season.Using mapping equipment, a 3 km stretch of the river centered on a fish location was mapped for depth, velocity, and substrate to provide not only fish habitat use but also local habitat availability.
Continuous temperature loggers were placed in the Missouri River and tributaries where fish were collected.
Gravel and rock deposits were located within the thalweg of the Missouri River, from the mouth at St Louis to Sioux City, Iowa (during low water conditions).
After spawning season, the fish were recaptured to assess spawning success and retrieve the DST tags.Fish movement and habitat use data, along with the physical habitat data, were analyzed using a combination of discrete-choice and utilization distribution model3. Multivariate statistical analyses were conducted to determine predictor and explanatory variables (both environmental and physiological) indicative of spawning success.Information gained The majority of shovelnose sturgeon recaptured did spawn successfully, suggesting that the methodology did not compromise spawning behavior.Furthermore, data indicate'that shovelnose sturgeon may travel over 640 km from point of tagging during their spawning migration.
The measurements of water conditions and habitat characteristics will be important in qualitative and quantitative description of habitat used during pre-spawn and spawning periods. Fish internal temperature (from DSTs), compared with the temperature measured by the continuous temperature loggers, will indicate whether fish are selecting seasonal habitats based on thermal preferences and the role of temperature as a spawning cue. This comparison will also indicate whether fish ascended river tributaries.
The discrete-choice and utilization distribution modeling will contribute to determine if fish are selecting one habitat over another among those available on a local level, particularly during spawning.Blood chemistry data will be used to assess spawning or failure to spawn.A combined analysis of the hormone data with environmental data may point to potential spawning cues. Tracking reproductively mature fish will provide data on the timing and magnitude of spawning movements, and the potential spawning habitats.
Environmental and physical habitat data, obtained together with tracking gravid and post-spawn females, will be critical to understand where, and under what conditions sturgeon spawn. Results will be used to quantify existing spawning habitat and develop management strategies to create suitable and sufficient spawning habitat. This information will be critical to design adequate habitat alterations and experimental flow manipulations intended to promote reproduction.
Telemetry locations of implanted fish and the associated habitat and water quality measurements will be incorporated into a GIS format and made available tothe U.S. Army Corps of Engineers (USACE), the USFWS and others for use in the redirection of sturgeon assessment and monitoring efforts.The USACE, USFWS, numerous Tribes, state agencies, and stakeholders are involved in efforts to define operational changes that will minimize jeopardy and contribute to survival of the pallid sturgeon.
Management actions to alter the flow regime or morphology of the Missouri River and provide benefits to the pallid sturgeon need to be designed with a comprehensive and detailed understanding of how sturgeon might respond.Final Comment It is important to realize the crucial role that behavior can play in the conservation of Great Plains fishes. I hope this article provides an overview of the exciting approaches that are being used in the conservation of native fishes. This research could inspire similar conservation projects on other fish species where analogous questions and logistical problems arise.*U.S. Geological Survey, Columbia Environmental Research Center, USA mwidhaber@usgs.gov References
- 1. U.S. Fish and Wildlife Service (USFWS). 1991. Neosho madtom recovery plan. U.S.Fish and Wildlife Service, Denver, Colorado.
42 pp 2. USFWS. 1993. Recovery plan for the pallid sturgeon (Scaphirthynchus albus): U.S.Fish and Wildlife Service, Bismarck, ND, 55 pp 3. USFWS. 1998. Final rule to Eist the Topeka shiner as endangered.
Federal Register 63:69008-69021
- 4. Pflieger, W.L. 1997. The fishes of Missoun: Missouri Department of Conservation, Jefferson City, MO, 372 p 5. Kems, H. A. 1983. Aspects of the life history of the Topeka shiner, Notbpis topeka (Gilbert), in Kansas. unpublished M. S. Thesis, University of Kansas, Lawrence 6. Hatch, J. T., and S. Besaw. 2001. Food use in Minnesota populations of the Topeka shiner (Notropis topeka). Journal of Freshwater Ecology 16: 229-233 7. Luttrell, G. R., Larson, R.D., Stark, W.J., Ashbaugh, NA, Echelle, A.A. and A.V. Zale.1992. Status and distribution of the Neosho madtom (Noturus placidus) in Oklahoma.Proceedings of the Oklahoma Academy of Science 725-726 8. Cross, F. B., and J. T. Collins. 1995. Fishes in Kansas. Second Edition. Lawrence, Kansas: University Press of Kansas; Public Education Series 9. Wilkinson C., Edds, D.R., Dorlac, J., Wildhaber, M.L., Schmitt, C.J., and A. Ailert.1996. Neosho madtom distribution and abundance in the Spring River. The Southwestern Naturalist 41: 78-81 10. Moss, R. E. 1983. Microhabitat selection in Neosho River riffles. Doctoral dissertation.
University of Kansas, Lawrence, Kansas 11. Fuselier, L.., and D. Edds. 1994. Seasonal variation in habitat use by the Neosho madtom (Teleostel:
Ictaluridae:
Notunus placidus).
The Southwestern Naturalist 39: 217-223 12. Wildhaber, M.L, Allet, A.L., Schmitt, CJ., Tabor, V.M., Mulhem, D., Powell, K.L, and S.P. Sowa. 2000a. Natural and anthropogenic influences on the distribution of the threatened Neosho madtorn in a midwestern warmwater stream. Transactions of the American Fisheries Society 129: 243-261 13. Wildhaber, M.L., Tabor, V.M., Whitaker, J.E., AileR, A.L, Mulhem, D., Lamberson, P.J., and K.L Powell. 2000b. Ictalurid populations in relation to the presence of a main-stem reservoir in a midwestern warmwater stream with emphasis on the threatened Neosho madtom. Transactions of the American Fisheries Society 129:1264-1280
- 14. Bulger, A. G., Wildhaber, M.L., and D. R. Edds. 2002a. Effects of photoperiod on behavior and courtship of the Neosho madtorn (Noturus pladdus).
Journal of Freshwater Ecology 17:141-150
- 15. Pfingsten, D. G., and 0. R. Edds. 1994. Reproductive traits of the Neosho madtom, Noturus placidus (Pisces: lctaluridae).
Transactions of the Kansas Academy of Science.97:82-87 16. Bulger, A. G., Wilkinson, C.D., Edds, D.R., and M. L Wildhaber.
2002b. Breeding behavior and reproductive life history of the Neosho Madtom, Noturus placidus (Teleostei:
Ictaluridae).
Transactions of the Kansas Academy of Science 105:108-124
- 17. Bulger, A. G. and D. R. Edds. 2001. Population structure.
and habitat use In Neosho madtom (Noturus placidus).
The Southwestern Naturalist 46:8-15 18. Bryan, J.L., Wildhaber, M.L., and D.B. Noltie. 2005. Examining madtom reproductive biology using ultrasound and artificial plhotothermal cycles. North American Journal of Aquaculture 67:211-230
- 19. Bryan, J. L., Wildhaber, M.L., and D. B. Notie. In press. Influence of water flow on Neosho madtom (Noturus placidus) reproductive behavior.
The American Midland Naturalist
- 20. Albers, J.L. and M.L. Wildhaber.
2002. Neosho madtom spawning.
U.S. Geological Survey, Biological Science Report 2002-0002.
Columbia Environmental Research Center, Columbia, MO TheConservation Behaviorist 2006 Vol. 4 (1): 18
- 21. Surprenaut, C. 2004. Scaphilrynchus platorynchus, in, IUCN 2004, 2004 IUCN Red List of Threatened Species: < http://www.iucnredlist.org>, Downloaded on 17 October 2005 22. Bailey, R.M., and F.B. Cross. 1954. River sturgeon of the American genus Scaphirhynchus:
characters, distribution and synonymy:
Papers of the Michigan Academy of Science, Arts and Letters 39:169-208
- 23. Bemis, W.E., Findeis, EK. and L. Graide. 1997. An overview of Acipenseriformes.
Environmental Biology of Fishes 48: 25-71 24. Modde, T., and J.C. Schmulbach.
1977. Food and feeding behavior of the shovelnose sturgeon, Scaphirhynchus platorynchus, in the unchannelized Missouri River, South Dakota: Transactions of the American Fisheries Society 106: 602-608 25. Keenlyne, K.D. 1997. Life history and status of the shovelnose sturgeon, Scaphirhynchus platorynchus:
Environmental Biology of Fishes 48: 291-298 26. Carlson, D.M., Ptlieger, W.L., Trial, .L., and P.S. Havertand.
1985, Distribution, biology, and hybridization of Scpahirhyrnchus albus and Scaphirhynchus platorynchus In the Missouri and Mississippi River: Enviionmental Biology of Fishes 14: 51-59 27. Mayden R.L, and B.R. Kuhajda. 1997. Threatened fishes of the world: Scaphirhynchus albus (Forbes & Richardson, 1905) (Acipenseridae):
Environmental Biology of Fishes 48: 420-421 28. Becker, G.C. 1983. Fishes of Wisconsin:
University of Wisconsin Press, Madison, 1053 pp 29. Breder, C. M., Jr. and D. E. Rosen. 1966. Modes of reproduction in fishes. The Natural History Press, Garden City, New York. 941 pp 30. Keenlyne K.D., and LG. Jenkins. 1993. Age at sexual maturity of the pallid sturgeon:
Transactions of the American Fisheries Society 122: 393-396 31. USFWS. 2000. Biological Opinion on the Operation of the Missouri River Main Stem Reservoir System, Operation and Maintenance of the Missouri River Bank Stabilization and Navigation Project, and Operation of the Kansas River Reservoir System: U.S. Fish and Wildlife Service, Bismarck, ND 32. Campton, D. E., Bass, A.L., Chapman, FA, and BW. Bowen. 2000. Genetic distinction of pallid, shovelnose, and Alabama sturgeon:
emerging species and the US Endangered Species Act: Conservation Genetics 1: 17-32 33. Simons, A.M., Wood, R.M., Heath, L.S., Kuhajda, B.R., and R.L. Mayden. 2001.Phylogenetics of Scaphirhynchus based on mitochondrial DNA sequences:
Transactions of the American Fisheries Society 130: 359-366 34. Snyder, D.E. 2002. Pallid and shovelnose sturgeon larvae -morphological description and identification:
Journal of Applied Ichthyology 18: 240-265 35. Birstein, V.J. 1993. Sturgeons and paddlefishes:
threatened fish in need of conservation:
Conservation Biology 7:773-787 36. Birstein, V.J., Bemis, W.E. and J.R.Waldman.
1997. The threatened status of acipenseriform species: summary. Environmental Biology of Fishes 48:427-435
- 37. Wildhaber M.L., Papoulias D.M., DeLonay A.J., Tillitt D.E., Bryan J.L., Annis M.L, and J.A. Allert. 2005. Gender identification of shovelnose sturgeon using ultrasonic and-endoscopic-imagery-and the application of the method to the pallid sturgeon:
Journal of Fish Biology 67:114-132
- 38. Millspaugh, J.J., and J.M. Marzlugg.
2001. Radio Tracking of Animal Populations:
Academic Press, San Diego, CA. 474 pp Kidnapping the Don Juans of Guant namo By Allison C. Alberts*Temporary removal of dominant males and careful manipulation of a population's social structure could help conservation behaviorists reduce the effects of inbreeding.
The technique may be most effective for small genetically-compromised endangered species that show strong polygyny, with a few dominant males monopolizing territories and females. After the "Don Juans" are removed from their home ranges, new males take over their roles and females have access to a more diverse set of matest.West Indian rock iguanas (genus Cyclura) are among the most endangered lizards in the world, with five of the eight species considered critically endangered by IUCN. Introduced mongooses, feral cats and dogs, and free-ranging hoofstock have decimated once teeming populations of iguanas by predating young and degrading native vegetation.
Rock iguanas, as herbivores, play a crucial role in Caribbean dry forest ecology: they promote foliage growth through cropping, provide nutrients to developing seedlings, and disperse seeds into new habitats.In the mid-1 990s, we spent a year documenting hormones and behavior in a group of iguanas inhabiting the U.S. Naval Base at Guantbnamo Bay. Our behavioral observations revealed that 80% of adult males engaged in aggressive interactions with other males. We classified males winning more than 50% of encounters as high-ranking, and those winning less than 50% of encounters as low-ranking.
The remaining 20% of males never participated in agonistic interactions (non-ranking).
High-ranking males exhibited higher testosterone levels and were significantly larger in body length, weight, head size, and scent gland diameter than low-ranking males. High-ranking males vigorously defended small but well-defined home ranges that overlapped the ranges of various females. Non-ranking males occupied peripheral home ranges with very limited access to females and tended to avoid movement to escape the notice of more aggressive individuals.
Low-ranking males did not defend territories, instead they moved extensively throughout the study area while suffering constant chases by. high-ranking males. Analysis of mean distances between pairs of individuals indicated that each of the resident females on the site was closer to a high-ranking male than to a low- or non-ranking male. Headbob displays, chases, and mouth gaping, behaviors usually performed in the context of territorial defense, were exhibited by high-ranking males significantly more often than by low-ranking males. There was also a trend for courtship to be performed more often by high-ranking males than by other males. Although it is impossible to be certain in the absence of genetic studies, our results suggested that high-ranking males, through their more robust body morphology and behavioral dominance, had better access to mates than low and non-ranking males.We conducted an experiment to determine whether temporary alteration of local social structure could increase the probability that sexually mature but genetically under-represented male iguanas could improve chances to mate.During the 1994 breeding season, we temporarily removed the five highest-ranked males from the study site. Removal of these "Don Juans' produced immediate and dramatic changes in male social structure.
Within a few days, the five largest previously low-ranking males began to win more than half of their encounters and could be classified as high-ranking.
All of the previously non-ranking males began to move throughout, the study site and fight extensively with other males, behaving like low-ranking individuals.
The newly dominant males showed increased rates .of headbob display and chases associated with territorial defense, as well as testosterone levels typical of high-ranking males during the breeding season. Active courtship of females was seen in both the newly dominant males as well as the low ranking males.Once the previously dominant males were removed from the site, the five males that achieved high-ranking status in their absence defended territories that were strikingly spatially similar, to those vacated by the removed individuals.
At the dose of the breeding season, we returned the Don Juans to the study site. Our behavioral observations and home range mapping for five weeks following the release of the dominant males indicated no long-term disruption of behavior or social relationships.
/West Indian rock iguana The Conservation Behaviorist 2006 Vol. 4 (1): 19