ML20079N063
ML20079N063 | |
Person / Time | |
---|---|
Site: | McGuire, Mcguire |
Issue date: | 09/30/1982 |
From: | Adair W, Hogan J DUKE POWER CO. |
To: | |
References | |
RTR-NUREG-1437 AR, DUKEPWR-82-02, DUKEPWR-82-2, NUDOCS 9111110066 | |
Download: ML20079N063 (473) | |
Text
{{#Wiki_filter:- _ _ - _ _ _ _ _ _ DUKE PWR/82 - 02 LAKE NORMAN ' SU fARY VOLUMEI CHA L T , NORTH I SEPTEMBER 1982
- -_________ M
o lad NORMAN
SUMMARY
Volume 1 e-Edited by J. E. Hogan and W. D. Adair 1982 O Duke Power Company Technical Report DUKE PWR/82-02 Duke Power Conipany Production Support Department Production Environmental Services Route 4, Cox 531 Huntersville, ilt 28078 O
m... . _ -. _ . _ - _ - . . . _ _ _ . . _ _ . - _ _ _ _ _ _ , . _ . - _ _ . . _ . _ _ . _ . . . _ _ . _ . . _ . . . _ _ _ _ . , d 'h 4 -
\
+u 101 . i k $c i W DEDICATION This report was prepared and distributed with the belief that others might ,
- p. benefit from our work. In that belief, we dedicate its use to the memory.
.g of three of our fellow Duke Power Company environmentalists who lost their l-lives while sampling on Lake Norman in January 1980.
- i. :
i Robert Lynn Green i Elaine Faulk Jones David Wayne Revill . r - . - s h O :
...,_,_.._._,._,_..,,,,_._....w..r... ..,_.,,~._,..~.,_.J.,., w,,,,,.y_,..
~ .
i (} PREFACE Duke Power Company and other investigators have studied the aquatic environment of Lake Norman and vicinity since prior to its impoundment in the early 1960's. The purpose of the Duke Power studies was to develop a database of physical, chemical, and biological chat acteristics that would be useful in assessing the aquatic environmental impact of McGuire Nuclear Station, scheduled for operation in 1981, and of Marshall Steam Station, operating on the lake since 1965. This document summarizes thost environmental data collected by Duke Power from 1974 through 1980 at over fif ty locations on Lake Norman and Mountain Island Lake Volumes 1 and 2 contain these summarien, including statistical analyses, for representative portions of the databate. The entire database is reproduced on microfiche in Volume 3. O O i
TABLE OF CONTENTS Page Volume Number Number PREFACE . . . . . . . . . . . . . . . . . . . . . . . , . . i 1 TABLL OF CONTENTS . . . . . . . . . . . . . . . . . . . . 11 1 EXtCUTIVE
SUMMARY
. . . . . . . . . . . . . . . . . . . . . iii 1 INTRODUCTION ............ . .' I . ...... I 1 THERMAL REGIMES . . . . . . . . . . . . . . . . . . . . . . 34 1 WATER CHEMISTRY . . . . . . . . . . . . . . . . . . , . . 107 1 PHYTOPLANKTON , . . . . . . . . . . . 154 ......... 1 PHYTOPLANKTON - BI0 ASSAY ,, .... ............ 261 2 PERIPHYTON .... - ................... 278 2 MACR 0PHYTES , 315 . . . . . , . . . . . . . . . . . . . . 2 ZOOPLANKTON . . ......,,.............. 323 2 BENTHIC MACR 0 INVERTEBRATES . ............. .. 354 2 FISH ..... ......,........... ... 411 2 APPENDIX ,...-.................... 3 pa ii
by increased concentrations of ammonia, iron, manganese, and alkalinity. Nutrient concentrations also exhibited typical seasonal cycles with nitrate plus nitrite being the dominant nitrogen species in winter and spring, and ammonia the dominant g species during fall. Total phosphorus concentrations exhibited seasonal trends similar to turbidity. Trace element concentrations were within the North Carolina water quality standards for the protection of aquatic life, with occasional exceptions of zinc, mercury, and cadmium. The phytoplankton community of Lake Norman was dominated by diatoms from late fall to mid-spring, cryptophytes in late spring, and green algae and dino-flagellates from summer to mid-fall. Blue-green algae were present in the summer at relatively low densities. The abundance of phytoplankton, especially diatoms, was fairly consistent among locations south of the Davidson Creek-Catawba River confluence, but was slightly greater at uplake locations. Spatial variation, however, was generally outweighed by temporal changes. Phytoplankton abundance was usually greatest in the spring and occasionally high again in the fall. Phytoplankton were patchily to uniformly distributed throughout the water column during the cool months. During the thermally stratified period, however, phytoplankton were heavily concentrated in the near-surface waters. Annual mean chlorophyll a concentrations have declined since 1974, perhaps due to a similar decline of total phosphorus concentrations. The productiun rates of the phytoplankton community were apparently regulated primarily by light and temperature during the isothermal period, and by the ava1 'ity of phosphorus during the stratified period. The latter observation was supported by the results of phytoplankton nutrient bioassays. Furthermore, when McGuire discharge conditions were simulated, bioassays showed that algal growth response in the laboratory was more strongly related to temperature increase than to increase in nutrient concentrations. The periphyton community of Lake Nonnan was dominated by diatoms; green and blue-green algae were generally second and third, respectively, in abundance and biomass. Algal carbon and chlorophyll a were the best estimators of periphyton biomass. Biomass was generally h'ighest in late spring and lowest in winter; periphyton growth appeared to be primarily a function of light and temperature, except during late summer when low nutrient availability and/or increased predation may have caused lower bicmass. Spatial variation was small compared to seasonal changes. Macrophytes were not abundant in Lake Norman. In the Ramsey Creek area they contributed less than one percent to the total annual production of plants and had a negligible contribution to lake metabolism, even with regard to oxygen depletion during their decomposition. The zooplankton community of Lake Norman was dominated numerically by rotifers most of the year, although cladocerans and copepods were of ten abundant in the spring. Copepods usually dominated the community biomass. Zooplankton abundance
- and biomass were generally high in spring and fall; the spring pulse occurred l for all major zooplankton groups, whereas only rotifers contributed substantially to the fall pulse.
Differences in zooplankton community structure among sampling locations were small, although cladocerans and rotifers were generally more abundant uplake iv
bO' l i 3 EXECUTIVE
SUMMARY
(G This report summarizes aquatic ecosystem data collected by Duke Power Company from over fif ty sampling locations on Mountain Island Lake and Lake Norman, especially the latter, from 1974 through 1980. During this period Marshall Steam Station on the upper end of Lake Norman was operating, and McGuire When Nuclear Station on the lower end of the lake was under construction. both electrical generating facilities are operating, Lake Norman will have the highest thermal loading from the discharge of once-through condenser cooling water of any comparable-sized lake in the United States. Lake Norman is generally typical of other dendritic reservoirs in the Carolinas Piedmont physiographic province. However, the lake receives relatively low quantities of pollutants cempared to some other lakes in the province; its trophic status is best characterized as oligo-mesotrophic (of excellent qualitj for body contact water sports but of low fishing potential). Water temperature on Lake Norman followed seasonal patterns typical of Maxi-Piedmont reservoirs. The lake began to thermally stratify during April. mum temperatures were measured during July and August and ranged from about 29 to 33 C in the surface waters and from about 9 to 14 C in the bottom waters. The lake began to cool during the fall and stratification decreased, resulting in relatively uniform temperatures from surface to bottom by the end of Novembi Minimum lake surface temperatures in the downlake areas were generally measurei during February each year iad ranged from about 2 to 8 C. D .s of the lake were determined primarily by local The thermal character" a so influenced locally by the operation of Lookout Shoal meteorology but wer' and Cowans Ford Hydroelectric Stations at the upper and lower end of the lake, respectively, and by Marshall Steam Station. Because Marshall utilizes cool
- bottom waters for condenser cooling, discharge temperatures during the summer were approximately the same as surface temperatures measured at locations out of the influence of the Marshall discharge. Maximum differences between surfa temperatures in the discharge cove and those in other areas of the lake were measured between October and March, with the thermal plume being observed over the largest area of the lake in January and February. During these latter months the plume was generally contained within the top 5 m of the water colun and had dissipated most of the excess heat within 10 km of the Marshall discht . structure.
The physicochemical characteristics of Lake Norman waters reflected the geolo! of the basin. The lake was characterized by slightly acidic pH values, low hardness, stable mineral composition, and generally low nutrient and trace me' concentrations. Variables relats to runoff, such as turbidity, iron, ortho-phosphate, and total phosphorus, were higher in the uplake areas with concen-trations decreasing downlake. Surface turbidity values were high during wint.
~
and spring with low values being observed during the summer and fall. Dissolved oxygen concentrations exhibited seasonal trends typical of Piedmont Carolina lakes with highest values occurring from December through April and lowest concentrations from July through October. Except for locations near t q Marshall discharge, dissolved oxygen concentrations in the surface waters wer b above 5.0 mg 1-1 throughout the year. Lcw dissolved oxygen concentrations existed in the bottom waters from August through October and were accompanied
~
iii
sunfish and perch. Uplake abundance of larval shad and sunfish was higher than at downlake locations. However, uplake crappie abundance was considerably lower than downlake. Larval perch were also more abundar.t in downlake collections. Newly hatched larval fish were usually more abundant in shoreline areas. Rapid dispersal from shoreline areas occurred as the larvae grew, and larger larvae were generally more abundant in the channel areas. No noticeable difference was observed in spawning times or duration of spawning throughout the lake, even though water temperatures near Marshall during early spring were slightly higher than ambient lake temperatures. Furthermore, no relationship was apparent between the number of larvae collected and the number of young-of-the-year fish present in August rotenone semples. In summary, the results of these seven years of studies by Duke Power Company indicate the i.ake Norman ecosystem prior to the operation of McGuire Nuclear Station is generally typical of other Carolina Piedmont reservoirs. Overall the principal trends identified were 1) low spatial variability, although slightly higher values of many variables were measured uplake compared to down-lake, and 2) seasonal factors generally outweighed spatial ones, with many variables exhibiting a spring pulse of measured values, occisionally followed by a lesser pulse in the fall. O O I vi
i CHAPTER 1, INTRODUCTION
. { } :.
J. E. H0GAN ~ PAGE BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
-0BJECTIVES'. . . . . . . . . . . . . . . . . . . .......... 2 REGIONAL CHARACTERISTICS . . . . . . , . . . . . . . . . . . . . . . 2 DEMOGRAPHY . . . . .... ......,......... ... 2 LAND USE , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 WATER USE ,,... ..... .............. ..... 3 TERRESTRIAL ECOLOGY ,.. ..................... 3 l
GE0 LOGY .. .... .... .....,, ............ 4 4 I METEOROLOGY ....... ..................... HYDROLOGY , ,... ..... ................... 5-LAKE NORMAN GENERATING FACILITIES ...... ............ 6 COWANS FORD HYDROELECTRIC STATION ................. 6
-MARSHALL STEAM STATION . . . . ....., ..,......... 6 -- MCGUIRE NUCLEAR STATION ...................... 7 l()_ PREVIOUS ENVIRONMENTAL STUDIES . . . ... ............. 8 DUKE POWER COMPANY (1976) .............. ...... 8 JENSEN.(1974) ....... .................... 9 WEISS ET AL. (1975) ... ............ ..... .. 9 THE'PRESENT ENVIRONMENTAL STUDY ........ .........., 9-LITERATURE CITED . . . . . . . . ............ ....., 10 0
1
BACKGROU!iD Duke Power Company is constructing McGuire NJclear Station on Lake Norman near e Cowans Ford Dam, 27 km north-northwest of Charlotte, NC (Fig.1-1 and 1-2). McGuire will have two generating units of 1180 PW net electrical design rating each. The station will utilize a cooling water intake drawing from the near-surface waters of Lake Norman. When water temperatures are warm, part of the once-through condenser cooling water will be supplied from a lower-level Lake Norman intake. The cooling water will be discharged back into Lake Norman, Two other generating facilities also withdraw water from the lake (Fig. 1-2). The first, Cowans Ford Hydroelectric Station, utilizes a submerged skimmer weir allowing passage of surface water downstream to Mountain Island Lake. The second, Marshall Steam Station, is a 1900 firl net electrical capability coal-fired generating station, about 23 km uplake by w6ter from McGuire. The once-through condenser cooling waters f rom Marshall are discharged to Lake Norman. Although Lake Norman is the largest reservoir within North Carolina, it is the smallest body of standing water in the United States on which two large steam stations with once-through cooling are located (0.32 FM net electrical per surface hectare at full pool). Therefore, concerns have been raised as to whether the thermal component of the discharges will cause detrimental effects to the biota of the lake. Duke Power Company (1975) demonstrated that the thermal component of tt.e cooling water discharge from Marshall Steam Station was such that the protection and propagation of a balanced indigenous aqL tic comunity in Lake Norman was assured; however, insufficient data were presented g in that demonstration to adequately evaluate the impact on Lake Norman of both Marshall and McGuire. OBJECTIVES The objectives of this study were to:
- 1. document the . ysical, chemical, and biological ct.aracteristics of Lake Norman and F,vantain Island Lake as these characteristics relate to the construction and/or operation of electric generating stations on Lake Normen, and
- 2. develop a database sufficient to assess the environmental impact of the operation of McGuire Nuclear Station and Marshall Steam Station on Lake Norman and Mountain Island Lake.
REGIONAL CHARACTERISTICS DEMOGRAPHY Lake Norman lies in the central area of the Carolinas' heavily populated Piedmo- t ( Fig.1 -1) . This area is characterizcd cy "apid population increase. Tr.e 1: egest nearby population centers, basca on the 1980 census results, are O 2
.. ~ . . . . - - , - - - . - - . - .- - -- - - - - .- .
l l Charlotte (314 447), Gastonia (47 333), Kanna olis-(34 420), Hickory (20 757), h': Statesville (18- 622), and Concord, NC (16 942 . Numerous small towns and connunities surround the lake, with populations ranging from a few hundred to several thousand people. LAN'D USE ! The area around Lake Norman is generally rural, non-farmland, and suburban. Principal farm crops are soybeans, wheat, and corn. Chickens, and milk and. beef cows are the main farm animals rai:;ed for profit (North Carolina Department of Agriculture 1971 as cited in Clay et al.1975). The industries of the ' area-are varied, but there are numerous textilo and apparel mills, and furni-ture and other wood product industries (Clay et al.1975). However, none of
-these industries are located on the shores of Lake Norman. Numerous single-family residences are located on the lake's shoreline. Apartments, condominiums, and housing subdivisions have also been constructed on the shoreline within the past few years. Several recreation areas, such as golf courses and Duke Power 1 State Park, are also. located on the shoreline.
WATER USE The towns of Mooresville, Davidson, and Huntersville, NC, obtain drinkinq e er from Lake Norman; they withdraw a mean of 0.13, 0.028, and 0.007 m3 s-1 respectively. Charlotte, NC, obtains its drinking water from Mountah Island Lake, downstream from Lake Norman (Fig.1-1), at a mean rate of 1,9 A s-1 O Lake Norman is also used extensively for recreational purposes. The mo:;t frequent uses are boating, fishing, and swimming. Estimates indicate that over 3.6 X 105 people used Lake Norman access areas in 1979 (Duke Power Compan/ 1981). Few point sources discharge wastes.to Lake Norman either directly or through
. tributaries. Duke Power Company's Marshall Steam Station discharges up to 64.4 m3.s-1-of industrial wastes resulting from ash pond overflow and once-through' cooling. . Duke Power also discharges uo to 0.0009 m3 + s-1 from_its Training and Technology Center. The town of-Catawba, NC, discharges municipal wastes into Lyle Creek at a rate up to 0.011 m 3 s-1, of which about 40% is industrial waste derived from hosiery dying operations (North Carolina Department of Natural Resources and Community Development 1976a).
Non-point _ sources of wastes entering Lake Norman are largely derived from runoff, particularly' from agricultural and forest lands, but also from construction ; activities. Residential developments on the lake shoreline generally have septic: tanks and-drain fields; a small amount of domestic waste may enter the lake. as a result of. inefficient operation of this form of waste treatment in clay soils. The lake also receives runoff from lawns and golf courses on the shoreline. TERRESTRIAL ECOLOGY h The major forests of the Lake Norman area are representative of the Oak-Hickory-Scrub Oak, or the Shortleaf Pine climax. Thirty to fifty percent of this area 3
it forested. Subclimu successional communities dominate in the region as a result of discontinued logging, agricultural, or residential uses. 'Bottomland and streamside communities are usually dominated by willows, sycamore, sweetgum, h and red naple (Clay et al.1975). Wildlife found in the Lake Norman area is typical of both the successional plant community and man-disturbed areas such as roads, buildings, farmland, and parks. Of species probably found in the area, about 35 are mammals, over 100 are birds, 35 are reptiles, and 23 are amphibians. Although not on a principal waterfowl flyway, Lake Norman is used by several species of migratory aquatic birds (Duke Power Company 1976). No substantiated observations of rare, endangered, or threatened plant or animal species in the Lake Norman area have been recorded. The southern bald eagle is the only such species which may occur near the McGuire site (Duke Power Company 1976). GE0 LOGY Lake Norman is in the rolling hills of the North Carolina Diedmont physiographic province (Fig.1-1). The underlying rock of Lake Norman is within the Piedmont geologic province, predominately in the Inner Piedmont belt but partially in the Charlotte belt in the area around Cowans Ford Dam. Rocks of the Charlotte belt are of plutonic and metamorphic origin; they consist of granites, diorites, and gabbros, with intrusions of felsic and mafic gneisses, and schists. Rocks of the Inner Piedmont belt are of metamorphic origin consisting almost entirely of mica gneisses and schists; these rocks contain lesser amounts of hornblende gneiss and granite (Clay et al.1975). Soils in the Lake Norman area are characterized by acidic, hiahly leached and weathered clay subsoils (Clay et al.1975). These soils generally have a loamy surface horizon, but they have slow percolation rates due to the clayey subsoil (Buol 1973 as citec in Clay et al.1975). METCOROLOGY Lake Norman is located about 100 km from the Appalachian Mountains and about 300 km from the Atlantic Ocean. The mountains and ocean moderate winter temperatures, but have little ef fect on summer temperatures. The mean air temperature in January is 5.4'C (range from -1.1 in 1977 to 12.0"C in 1950), with instantaneous temperatures seldom below -12 C. The mean July temperature is 25.9"C (range from 24.6 in 1947,1971, and 1975 to 28.0 C in 1977), with instantaneous temperatures seldom above 35'C. [Mean temperatures are based on the 1879 through 197', record at Charlotte, NC (National Oceanic and Atmospheric Administration 1980). Instantaneous temperatures are from Clay et al. (1975)]. Mean annual precipitation is 1.14 m (range fror 0.85 in 1950 to 1.58 m in 1975). [Mean precipitation is based on the 198 through 1979 record at Charlotte, NC (National Oceanic and Atr asheric idoinistration 1980. )]. The driest weather is typically during the fall. Summer rainfall comes princi- g pally from thundershowers interspersed bntwean occasional dry periods. Daily W 4
precipitation at the McGuire site is shown in App. Fig. 1.1-1. Approximately (3:
,,) 60 to 70% of total precipitation is lost from ground and surface water resources due to evapotranspiration. Mean annual evaporation of water from lakes of the region ranges from 0.96 to 1.02 m (Clay et al.1975).
Wind is generally from the southwest at a mean speed of 12.1 km h-l. During February,-September, and October the prevailing direction is northeasterly. [ Based on the 1950 through 1979 record of speed and 1963 through 1979 record
- of prevailing direction at Charlotte, NC (National Oceanic and Atmospheric Administration 1980)].
Sky conditions from sunrise to sunset are clear a mean of 111 days per year, partly cloudy 103 days, and cloudy 151 days. The clearest months are October and November; the cleudiest are December through March. [ Based on the 1949 through 1979 record at Charlotte, NC (National Oceanics and Atmospheric Administration 1980)]. Daily solar irradiance at the McGuire site is shown in App. Fig. 1.1-2. HYDROLOGY Lake Norman is in the Catawba River Drainage Basin (Fig. 1-3). The river begins at an elevation of about 640 m mean sea level (m.s.l.), 8 km southwest of Old Fort, NC. It flows easterly about 110 km through three impoundments to near Millersville, NC, where it turns and flows southerly another 10 km through a fourth impoundment into Lake Norman. The river continues southward 175 km'through six mare impoundments to Wateree Dam, where it becomes the O~ Wateree River. The Wateree. flows via the Congaree and Santee / Cooper Rivers to the Atlantic-Ocean. Lake. Norman was formed by Cowans Ford Dam, which was completed in 1963. The lake extends from Cowans Ford Dam about 54 km uplake to the tt ilwater of Lookout Shoals Dam, which has a norrnal tailwater elevation of 232.01 m m.s.l. Mountain Islar.d take extends to the tailwater of Cowans Ford Dam, about 24 km upstream from rauntain Island Dam. At full pool elevation of 231.65 m m.s.l ., Lake Norman has a surface area of 13156 ha, a volume of 1.3489 X 109 m 3 (Fig.1-4), a shoreli_ne of about _837 km, a mean depth of 10.25 m, ai.d a maximum depth of 36.5 m. The lake drains a waternhed of approximately 4640 km 2
. The average annual flow from Cowans Ford Dam and Hydroelectric Station is 75.6 m' s-1 The mean retention time within the lake is 267 d (Duke Power Company 1976, 1981).
The principal stream _ feeding Lake Norman is the Catawba Piver, which normally enters the lake through the Lookout. Shoals Hydroelectric Station turbines located 11 to 17 m below full pool of Luokout Shoals Reservoir which has a maximum depth of about 24 m. During flood conditions, the waters of Lookout Shoals Reservoir may spill over the dam crest at full pool elevation of 255.45 m m.s.1. Other streams feeding Lake Norman a e ncw completely or partially inundated at full pool ( App. Table 1.1-1). Those streams which had high average discharge before the area was flooded are Lyle Creek (2.8 m3 s-1). Mountain Cr'eek (1.5 m3 s-1), and Davidson Creek (1.3 m 3 s-1) (Wilder and Slack 1971). Prior to the construction of Cowans Ford Dam, the nearest recorded flows were l O from a gaging statior. 43.3 km uplake from the dam near Catawba, NC, with a drainage area of 3976 km 2
. The average flow past the gage for the 30-year l
5
period prior to 1962, when the gage was inundated, was 06 m s-1 The maximum and mir.imum flow rat +s were M12 ry s-1 en 14 August 1940, and 2 ry s-1 on 15 Septembcr 1957, respecti On 16 July 1916, the river reached maximum $ flood stage of 13.4 m at the mowha gage, and on 17 July 1916 the estimated flow at Cowans Ford Dam site was 5650 m' s-1 (Duke Power Company 1976). The largest flood since Cowans Ford Dam was built occurred on 15 March 1975 with late Norman surface level at 231.83 m m.s.1. and an approximate flow of 3 1100 m
- s-1 through the Dam and Hydroelectric Station (Duke Power Company 1981).
Flows recorded at Cowans Ford and lookout Shoals Dams fron June 1973 through December l')30 are given in Table 1-1. Lake !4orman attained full pool in April 1964. During a typical sumer surface elevation is about 231 m m.i..l. (12 600 ha surface area), and dring a typical winter lake elevation is about 229.5 m m.s.1. (11600 ha surface area). The lowest lake level to date (227.8 m m.s.l.) was recurded in late sumer 1970. Duke hwer Company is corr.litted to no more than a 15 f t (4.572 m) late drawdown (Duke Pc er Corcpeny 1976). Lake levels from June 1973 through December 1980 are shown in Fig. 1-5. LAKE fiORMAN GENERATir4G FACILITIES COWANS FORD HYDROELECTRIC STAT 10ft Cowans Ford Hydroelectric Station has f our generating units, each rated at 90-fM net electrical capat.;11ty at median uperating conditions. The station typically operates in a peaking capacity about four hours per day. A submerged weir in the forebsy allows passage of surface water from Lake Norman while retaining the water below elevaticn 221 m m.s.l. (Fia.1-6). The averace annual flow through Cowans ford Hydroelectric Station is 75.6 19 s-1 The Fedhral Energy Regulatory Commission (1958) requires a minimum everage daily flow of 8.6 0 s-l. Monthly mean flows from Cowans Ford Hydroelectric Station ar e 5 given 'n Table 1-1. Cowans Ford Dam, beside the hydroelectric station, has a spillway flow capacity of 5964 m? s-1 through 11 gates, each 10.7 m wide i>y 8.5 m high. Water seldom overtops the spiilway or is released through the gates. MAP 91ALL STEAM STATION Marshall Steam Station was recognind by tne electri'. utility industry as the most efficient steam station in the United States from 1966 through 1974. Its four generating units have a combined net electrical capability of 1900 fM: two units are rated at 320 MW and the other two units are 630 fM each. The two smaller units became operational in 1965 and 1966; the larger units began operating in 1969 and 1970. Marshall obtains cooling water from under a skimmer wall which retains 1Me water above 213.6 m m. s. l . ( Fig. 1-7) . The mcnimum rate of cooling water flow is 12 m3 s-1 for each of the smaller units and 20 m 3 s-1 for the larcer units. Winter and summer condenser cooling water design flows are 48 and 64 m s-1 3 g respectively. Temperature rice across the condenser (aT) during winter and W turrer was designed at 16.1 and 9.4'C, respectively. Monthly a verage winter 6
i P 4 (January through March) and summer (July through September) discharge tempera-Q- tures have not excteded 20.5 and 31.5'c ('able 1). The North Carolina ,
- National pollutant Discharge Elimination System (NPDES) permit for Marshall !
(North Carolina Department of Natural Resources and Community Development > 1976b) stipuletes that monthly average discharge temperatures are not to exceed ! 2 94*F (34.4 C) from July 1 to October 15, and 92*F (33.3*C) for the remainder l of the year. These permit limitations have not been exceeded. McGWRE NUCLEAR STATION l McGuire Nuclear Station will have two pressurized water generating units with , a net design rating of 1;80 MW eiictrical each. Units 1 and 2 are scheduled ; for commercial operation in September 1981 and June 1983, respectively. Condenser cooling water for McGuire will be drawn from Lake Norman through two l intake structures. Water will be drawn through an upper-level intake (Fig.1-8) located between 217.9 and 227.1 m m.s.l., at a maximum flow of 64.1 m3.s-1 per generating unit. Water will also be drawn from a lower-level intake (Fig.1-9) located between 199.3 and 204.2 m m.s.l., at a maximum flow of 28.3 m3 .s-1 per unit. Flow from the lower-level intake will displace an equal flow from ' the upper-level intake, resulting in the same maximum flow per unit 65 the upper. leu . i ntake alone could provide. Four pumps per unit will be able to supply cooling water to the steam condenser. Two pumps per unit will produce a flow of 40.4 m3.s-1; three pumps will produce 54.7 m3.s-1; and four pumps will produce 64.1 m3 s-1 O The quantity and source of cooling water will be determined by the temperature , of the intake water. During the months when the intake water temperatures are cool (approximately December through April), only two pumps per unit will be : operated. When temperatures become warmer (approximately June through October) four pumps per unit will be operated. Three pumps per unit will be used for the . short transitional periods. The upper-level intake will supply the entire cooling ! water needs except when the discharge water temperature approaches 35*C, Water will then be drawn from the lower-level intake to moderate temperatures such that average monthly discharges are maintained at or below 35*C. Total condenser cooling water flow per unit will remain 64.1 m .s-1 when the lower-level intake 3 is in operation. Thus, the lower-level intake will provide up to 44*; of the condenser cooling water flow when McGuire is operating at full capacity during months with the warmest intake water temperatures. Condenser cooling water from McGuire will be discharged into a canal from pipes located between 219.6 and 222.9 m m.s.l. The 1-km canal has a depth of about 12 m when Lake Norman is at full pool. The discharge flows over an earthen weir submerged at about 225 m m.s.l. Heated effluent from the canal will initially mix with the surface waters of the lake and then stabilize vertically-and spread over the lake surf ace, ultimately transferring most of its heat to the atmosphere (Colon and Leavitt 1973; Duke Power , Company 1976). The condenser cooling water AT is projected to reach a maximum l of 14.1 C during the winter months when intake temperatures are the coolest. '
/3 The AT is expected to be a maximum of 8.9'C during the warmest months. Mon t ' ' "
V average discharge temperatures are noi permitted to exceed 96 F (35.0 C) wi~ 7
-r U2,-- ~ .--,,-m-, ,,m..~~.,,,-,,,,m..-~,.,,....,,-.-. _-m.,~ ,,.m-,--,w-rm
the aseigned thamal mixing zone (North Carolina Department of Natural Resources an1 Comn.nity Development 1978). Furthermore, the temperaturet outside the $ mixirj zone (Fig. 1-2) are not permitted to exceed 5 F (2.7*C) above the naturt.1 wter temperature or a maximum of 90*F (32.2) measured as a 24-h averace 0.3 m below the water surface. l A.though McGuire was not operational during the time period this report covers,
- ,cveral activities which could affect the waters of Lakes Norman and Mountain Island were in progress during the preoperational period. Construction at the McGuire site began in April 1971 and maior earth moving activities were com-pleted prior to 1974. Major exceptions were dredging and draalining at the intake and discharge areas. These activities are sumarized in Table 1-2.
Wastewater from McGu). is treated in one of two systems and discharged to Mountain Island Lake, immediately downstream from Cowans Ford Dam. A waste-water collection basin receives only treated domestic wastes, overflow ' rom the standby nuclear service water pond, and runoff from the station yard; this treatment system began continuously discharging in September 1974 T he second, a Convertional Wastewater Treatment System treats all other non-radioactive waste and began discharging on a batch basis in May 1978. Tests of various systems at McGuire have resulted in water being discharced through the discharge canal to Lake Norman. The Emergency Core Cooling System obtained water from the standby nuclear service: water pond during 1978 and discharged it to the lake. The Condenser Cooling Water System hat continuously obtained water from the upper- and/or lower-lavel intake structures in Lake & Norman through one or more pumps since the mid-1970's. W PREVIOUS ENVIRONMENTAL STUDIES Numerous aquatic environmental studies have been conducted on Lake Norman and on the aajacent waters of lookout Shoals and Mountain Island Lakes. The most extensive studies were by Duke Power Company (1976), Jensen (1974), and Weiss et al. (1975). These studies will be only briefly described here, although more extensive summaries can be found in other chapters, as appropriate. DUKE p0WER COMPANY (1976) The purpose of the Environmental Report for the operating licenne stege of hcGuire Nuclear Station was to establish pre-operational baseline data, on which an assessment of the effects of McGuire Nuclear Station could be made. The aquatic environmental portion of the report covers chemical and biological data collected during 1973 and 1974. The report characterizes the water chemistry of Lakes Norman and Mountain Island with regard to nutrients, minerals, and metalt, as well as general water quality variables such as temperature, dissolved oxygen, pH, and alkalinity. The repot t also characterizes the aquatic ecology of the lake through an examination of density and composition of phytoplankton, periphyton, zooplankton, benthos, and fishes. Productivity and biomass of celected taxonomic groups were also examined. The potential effects on the aquatic ecesystem due to the operation of McGuire wert sessed, based on the monitoring program results. Species of importance g 8
i to the ecosystem were identified, and species-environment relationships were ! Q -investigated to aid in the assessment. No potential effects of the station's operation were assessed as being of significant detriment to the functioning of the pre-operational ecosystem or to any of its components. { 5 JENSEN (1974) Studies of thermal effects of the cooling water discharge of Marshall Steam ! Station Wre initiated in July 1965. Work during the first three years in- " volved W collection of physical and meteorological data for use in evaluating , the '.her.Cl behavior of the lake. In 1968 the research was expanded to include biolgiri. studies of thermal effects. Data pertaining to water temperatures, t various meteorological and water quality variables, and populations of fish, plankton, and benthic invertebrates were collected from 18 locations on Lake Norman in the vicinity of Marshall. The biological data involved population ' and diversity studies in the main body of the lake, and in the Marshall intake . and discharge coves. Studies on the effects of entrainment of planktonic t organisms, and their passage through the condenser system of the station were also conducted. ; WEISS et al. (1975) This study was conducted on the downlake end of Lake Norman from February 1973 through January 1974 as part of an overall program examining physical, chemical, and biological variables in five Catawba River impoundments. Variables that were measured include temperature, dissolved oxygen, nutrients, chlorophyll, i O and densities and taxonomic composition of phytoplankton and zooplankton. Data were treated statistically and the significance of interaction of environmental factors on the aquatic biota was postulated. THE PRESENT ENVIRONMENTAL STUDY Chemical and biological constituents of Lake Norman and downstream Mountain Island Lake were studied extensively by Duke Power Company from 1974 through : 1980. Data resulting from this study are the subject of the remaining chapters of this document. These data are in the form of microfiche in Volumes 3 and i
- 4. Only certain of these data were analyzed in detail and are discussed in Volumes 1 and 2. This restriction was due to changes in study' design, limi-tations of statistical techniques, and an attempt to keep the document concise ,
yet_ representative of data collected throughout the-study period. The study design reflected changes in McGuire's schedule, and actual and anti-- cipated regulatory requirements, as well as knowledge gained by continuing evaluations of the data collected throughout the study.- Because these changes were specific to individual constituents of the study, they are described fully in later chapters. - All locations on Lake Norman and Mountain Island Lake which l were monitored during- this _ study are shown in Fig.1-10 and 1-11, respectively, and described in Table 1-3. . l O 9 . , . _ _ _ _ _ . _ . _ __ _ . _ . _ , _ _ _ _ . _ ..~._._.~.__-. _ __. _
LITERATURE CITED Buol, S. W. (ed.). 1973. Soils of the southern states and Puerto Rico. Agricultural Experiment Station, No;th Carolina State University, Raleigh, NC. Clay, J. W. , D. M. Orr Jr. , and A. W. Stuart (ed. ) . 1975. North Carolina atlas: Portrait of a changing southern state. The University of North Carolina Press, Chapel Hill, NC. 331 p. Colon. F. P., and J. W. Leavitt. 1973. Lake Norman hydrethernal model study for Duke Power Conpany. Alden Research Laboratories, Worcester Polytechnic Institute, Holden. MA. Progress Report Number 1. 31 p. Duke Power Company. 1975. Marshall $ team Station 316(a) Denonstration. Duke Power Company. Charlotte, NC. 45 p. 1976. McGuire Nuclear Station. Units 1 and 2 Environmental Report. Operating License Stage. 6th. rev. Volunes 1 and ?. Duke Power Company. Charlotte. NC.
. 1981. Data Manual. Volume 1 Duke Power Company. Charlotte, NC.
Federal Energy Regulatory Comission. 1958 (as amended). Duke Power Company Catawba-Wateree Project No. 2232. Federal Energy Regulatory Comission. Washington. DC. Jensen. L. D. (ed.). 1974. Environmental responses to thermal discharges O from Marshall Steam Station. Lake Norman. North Carolina. Electric Power Research Institute. Cooling Water Discharge Research Project ( RP-49) Report No.11. Johns Hopkins University. Baltimore. MD. 235 p. National Oceanic and Atmospheric Administration. 1980. Local climatological data, annual sumary with comparative deta,1979. Charlotte, North Carolina. National Oceanic and Atmospheric Administration Environmental Data and Information Service. National Climatic Center, Asheville. NC. 3 p. North Carolina Department of Agriculture. 1971. North Carolina Agricultural Statistics. North Carolina Department of Agriculture. Raleich. NC. North Carolina Department of Natural Resources and Corruunity Development. 1976a. Water Quality Management Plan, Catawba River Basin, Sub-basin 32. Draft Report. North Carolina Department of Natural Resources and Community Development. Raleigh, NC.
. 1976b. National Pollutant Discharge Elimination System Permit No. NC ODD 4987, for Marshall Steam Station. North Carolina Department of Natural Resources and Community Development, Raleigh, NC.
_. 1978. National Pollutant Discharge Elimination Systen Permit
? NC OD24392, for McGuire Nuclear Station. North Carolina Department
- c. Natural Resources and Community Development, Raleigh, NC.
g 10 t
i I Weiss. C. M. , P. H. Campbell, T. P. Anderson, and S. L. Pfaender. 1975. The O lower Catawba lakes: Characterization of phyto and zooplankton l
. communities and their relationships to environmental factors. Department ;
of Environmental Sciences and Engineering, School of Public Health. ' University of North Carolina, Chapel Hill, NC. ESE Pub. No. 389. 396 p. Wilder H. B. and L. J. Slack. 1971. Summary of data on chemical quality of
- streams of North Carolina, 1943 67. United States Geological Survey Water-Supply Paper 1895-B. United States Government Printing Office, Washington, DC.
O , P i i LO I i 11 ;
.-,..m.. - . . . . _ _ , . _ . . _ . . . _ . _ . . _ _ _
Table 1-1. Avera9e monthly flow in Lake Nonnan and average monthly coolant variables of Marshall Steam Station, June 1973 through December 1980, h LAKE NORMAN FLOW MARSHALL COOLANT VARIABLES Lookout Shoals Cowans Ford" flow AT Discharge Temperature [ionth (m?.s-))_ 3 (m .s-I)_ (m3 .s-1) ( *CJ ('Q _ Jun. 1973 93 111 Jul. 63 89 Aug. 85 99 Sep. 61 81 Oct. 54 45 Nov. 45 56 Dec. 1973 90 74 4 Jan. 1974 118 96 Feb. 95 111 Mar. 78 103 Apr. 114 161 fiay 85 114 Jun. 80 58 Jul. 67 80 Aug. 108 94 Sep. 86 70 1 Oct. 60 92 36 9.2 28.4 Nov. 48 66 38 9.3 25.0 Dec. 1974 Jan. 1975 65 80 52 76 45 45 10.2 10.2 19.4 18.6 g Feb. 87 86 48 8.a 16.6 Mar. 120 220 38 8.4 17.7 Apr. 99 128 47 8.7 21.6 May 101 117 47 8,7 23.5 Jun. 119 176 49 8.5 25.1 Jul. 69 71 46 8.0 25.8 Au9 54 86 50 9.2 28.9 Sep. 76 72 45 9.0 30.3 Oct. 87 93 35 7.7 28.2 Nov. 95 109 30 8.2 24.4 Dec. 1975 60 96 31 8.3 18.9 Jan. 1976 93 90 41 8.9 15.5 Feb. 76 92 39 9.2 17.3 Mar. U7 41 38 8.9 20.5 Apr. 49 26 43 8.3 22.8 May 57* 26* 43 8.2 23.7 Jun. 123* 147* 41 7.5 24.9 Jul, 54 71 38 7.5 26.1 Aug. 29 33 37 7. 9 27.9 Sep. 40 35 38 8.2 29.4 Oct. 99 145 37 8.3 27.4 Nov. 54 112 42 10.0 22.4 Dec. 1975 83 91 41 9.1 17.2 Jan. 1977 63 125 39 8.3 12.6 Feb. 34 39 37 8.4 14.0 12
l t page 2 of 2 Table 1 1 (continued) l I LAKE NORMAN FLOW MARSHALL COOLANT VARIABLES Cowins ford flow AT ~ TKschargeTemperature Lookout Shoals Month (m3. s- 1 ). (m3.s-1) __ (m3 .s-2) (1 (*C) 94 40 34 8.2 17.8 l Mar. Apr. 97 89 28 7.9 20.9 ! 55 60 41 9.5 23.2 l May Jun, 49 69 45 9.5 24.7 l 75 37 7.8 24.3 ) Jul. 40 28.3 Aug. 35 54 51 9.6 62 33 48 8.5 30.2 j seu. 27.7 Oct. 42 41 48 8.2 ; 109 230 46 8.2 23.7 l 3 Nov. 76 107 44 8.5 18.2 Dec. 1977 14.5 Jan. 1978 94 64 44 9.1 i 140 40 7.7 11.6 ; feb. 66 14.3 Mar. 79 76 32 7.4 Apr. 61 62 29 9.3 19.9 87 90 31 9.1 21.8 : May 55 74 31 9.2 23.3
-Jun. 23,9-Jul. 38 47 38 8.8 l 69 93 51 8.8 25.5 Aug. ,
sep. 44 57 48 8.6 28.2 ; 30 36 47 8.2 28.1 Oct. 24.6 33 41 47 8.1
', 1978 54 90 47 9.2 21.0 98 93 45 8.5 15.8 Jan. 1979 14.4 67 65 43 9.0 ;
feb 7.1 16.3 104 168 41-Mar 86 148 30 7.8 20.6 ; Apr. 98 132 49 8.9 23 7 May-77 91 52 8.4 24.7 Jun. 96 52 8.8 27.3 ' Jul. 83 9.0 29.3 Aug. 51 70- 53 Sep. 108 116 51 7.9 30.3 1 103 159 51 7.9 27.4 Oct. 23.3 ' Nov. 107 11 5 43 8.0 81 118 44 8.8 19.4 Dec. 1979 17.5 Jan.1980 100 123 46- 9.5 62 62 46 10.0 16.6-Feb. 90 34 44 9.2 17.8 Mar. Apr. 130- 193 42 g.g 22.2
' 91 . 111 38 . c3.8 May 70 52 9.3 25.3 Jun. 70 27.7 Jul. 65 94 52 9.0 48 51 9.4 30.1 4Sep. 36' 27 51 9.2 31 .5 -
31 41 44 8.5 28.8 1 Oct. 36 23,8 36 36 34 9.2 55 10.6 20.8 Novl Dec 1980 31 43 O-- Dete < rom 30 Mex throueb 20 Juee 1976 were not inciuded in the evere9e eve to e computer malfuncticn on those days. ,
+No data 'are available prior to October 1974 for the Marshall coolant variables.
13 4
- a. - ._.______.;____________.____..________..______.-._
Table 1- 2. Dredging and draglining activities in the intale and discharge areas of McGuire Nuclear Station. O Date intake Area Discharge Area Aug 75 Siphoned intake water to discharge canal Sep. 75 No activity Oct. 75 Starttd dredging No activity Nov. 75 Dredged No activity Dec. 75 Dredged. Coffer dam breached Backnoe used to breach cofferdam to Draglined equalize canal and late water levels Jan. 76 Dredged and draglined No activity Feb. 76 Dredged vid draglined No activity Mar. 76 Dredged and draglined No activity Apr. 76 fio activity No activity May 70 No activity No activity Jun. 76 No activity No activity Jul. 76 Started dredging 12 July and Started pumping water from Nuclear started draglining 20 July Service Water Pond to discharge on 26 July Aug. 76 Dredged and draglined. F1nished Stopped pumping water from Nuclear draglining on 11 August Service Water Pond to discharge en 4 August Started draglining on 13 August Sep. 76 Dredged Draglined Oct. 76 Dredged Finished draglining on 15 October Nov. 76 Dredge was not used until 10 !Jovember. Draglined on g 11-12 flovember Dec. 76 Dredged Jan. 77 Dredged Feb. 77 Dredged Mar. 77 Finished dredging on 17 March O 14
O Tabie 1 oescriptioa or samPi4oo iocatio"s oo tetes norme" e"e nov"te4" island. Locations are shown in Figures 1-10 and 1-11, respectively. The depth listed is not necessarily the sample depth and is intended only as a guide fnr aid in determining relative depths arnong sampling locations. Mountain Island Lake Locations location 16.0 On Catawba River arm,10 m south-soutneast from the mouth of Jackson's Pond (Jackson's Pond is accessible by Lucia Access Area); depth of 4 m. Location 16.2 On Catawba River arm, at the mouth of McGuire Nuclear Station non-radioactive wastewater effluent channel; depth of 0.2 m. Location 16.5
~7IftTJiGackson's Pond (Jackson's Pond is accessCole by Lucia Access Area);
depth of 1 m. Location 277.5 On Catawba River arn,100 m northeast f rom the mouth of Riverbend Steam Station ash basin discharge; depth of 5 m. Lake Norman Locations location 1._0 . Torebay of Cowens Ford Hydroelectric Station at center gate of spillway. 30 m from the dam; in the McGuire mixing zone; depth of 36 m.
~ Location 1.2 -
fn the McGuire Nuclear Station upper-level intate area, 50 m from the center of the lakeside of the intale structure; in the McGuire mixing rone; depth of 15 m.
-Location 1.7 Forebay of Cowans Ford Hydroelectric Station at mid-channel, 65 n from the dem; in the McGuire mixing zone; depth of 33 m.
Location 2.0 - ~ On Catawba River arm, midway between Catawba River Channel Marker 1 and the northernmost point of the peninsula to the east-southeast; in the McGuire mixing zone; depth of 34 m. Location 3.0 6TRamsey Creek arm, midway between Ramsey Creek Channel Marker R-1 and the northernnost cove on peninsula to the southwest; in the McGuire mixing zone; depth of 22 m. O 15
Table 1-3 (continued). page 2 of 5 ~~ Location 3.8 In the McGuire Nuclear Station discharge canal, on the west side of the bridge; in the McGuire mixing zone; depth of 2 m. h Location 3.9 In the McGuire NuclearS.t5 tion discharge canal, on the northeast side of the bridge; in the McGuire mixing zone; depth of 12 m. Location 4.0
~0n Ramsey Creek arm, 50 m northeast from the mouth of the McGuire Nuclear Station discharge canai; in the McGuire mixing zone; depth of 8 m.
Location 4.3 ~ On Ramsey Creek arm, 665 m north-northeast from the mouth of the McGuire Nuclear Station discharge canal; in the McGuire mixing zone; depth of 22 m. Location 4.5 On Ramsey Creek arm, 150 m southwest of the midpoint of the line joinina the two large islands,1.3 km northeast from the mouth of McGuire Nuclear Station discharge canal; in the McGuire mixing zone; depth of 17 m. Location 5.0 On Ramsey Creek arm, midway between Ramsey Creek Channel Markers R-3 and R-4; in the McGuire mixing zone; depth of 24 m. Location 6.0 On Ramsey Creek arm, in the cove northwest of Ramsey Creek Acccss Area, 10 m northwest of the midpoint of a line connecting the points of the g mouth of the cove; in the McGuire mixing zone; depth of 8 m. Location 6.5 (in Ramsey Creek arm, in the cove northwest of Ramsey Creek Access Area, in the castern terminal branch of the cove; in the McGuire mixina zone; depth of 5 m.
~ Location 7.0 On Lucky Creek cove, at the intersection of 1) a line connectina Catawba River Channel Marker 2 and the top of a 525 kv transmission line tower and 2) a line connecting the point of the second peninsula west on the north side of the mouth of Lucky Creek cove and the point of the peninsula to the scuth-southeast; in the McGuire mixing zone; depth of 8 m.
Location 7.3 On Black's cove, at the intersection of 1) a line drawn west-northwest of Catawba River Channel Marker 2 and 2) a line drawn south from the center of the first cove from the mouth on i.he northern shoreline of Black's cove; in the McGuire mixing zone; depth of 8 m. Location 7.5
' r Catawba River arm,100 m west of CatawM River Channel Marker 2; in the 'cGuire mixing zone; depth of 30 n.
O 16
I i . Table 1-3(continued). page 3 of 5 l O toce14ee 7.5
~~~0n Catawba River arm, 750 m northwest of Catawba River Channel Marker 2; on the northern boundary of the McGuire mixing zone; depth of 25 m. l t ~ Location 8.0 ~
i On Catawba River arm, 250 m southeast of Catawba River Channel Marker 3; depth of 30 m. J
' Location Southwest8.5 of the Catawba River arm, in a narrow, unnamed cove, 1.1 km south of Catawba River Channel Marker 5; depth of 10 m.
4 Location 9.0 6n Davidson Creek arm at the intersection of 1) a line drawn southeast from Davidson Creek Channel Marker D 4 and 2) a line drawn perpendicular to the midpoint of the line connecting the midpoints of the two large islands . directly east and east-southeast, respectively, of Davidson Creek Channel ! Marker D-4; depth of 21 m. ,
" Location-9.5 Dn Davidson Creek arm, 375 m southwest of Torrence fork Channel Marker 1-2 near the town of Davidson municipal water intake; depth of 25 m.
Location 10.0 On Davidson Creek arm, third cove north-northeast of Davidson Creek Channel Marker D-7, 60 m north of the midpofoi. of the line connecting the O- points of the mouth of that cove; depth of 8 m. Location 10.5 On Davidson Creek arm, mid-channel 1.7 km north-northeast of Davidson Creek Channel Marker D-7; depth of 18 m. Location 11.0-
- Ori~ Catawba River arm, midway between Catawba River Channel Markers 9 and
. 10; death of 30 m.
Location 12.0 On Beaver Dam Creek arm, northwest of Catawba River Channel Marker 13. - one-third of the way into the southernmost large cove on the west side; depth of 6 m. . Location 13.0 On Catawbc River arm, midway between Catawba River Channel Markers 14 and .; 15; depth of 28 m. t Location 14.0 Marshall bi. cam Station discharge cove, directly north of Catawba River Channel Marker 15, 0.5 km r.orthwest from the mouth of the cove; depth of 9 m.
- O l
17
Table 1-3 (continued) Peue 4 of 5 Location 14.5 In the Marshall Steam Station discharge canal, southeast from the di d charge structure under the line barning of "no trespassing", depth of 5 m. Location 14.7 In Marshall Steam Station discharge cove first cove along northern shore. line from the mouth of the discharge cove; depth of 4 m.
-Location 15.0 dn Cataiba River arm, under fiorth Carolina Highway 150 bridge, ridaay between bridge abutments; depth of 30 m.
Location 15.2 On CatEba River arm. 0.6 km north from the middle of North Carolina Highway 150 bridge and 0.3 km east from l'arshall Steam Station stimer wall; depth of 30 m. Location 15.5 On Cataw'ba River arm, the first cove uplate from Marshall Steam Station i..+ake cove. 0.8 km west from Catawba Rivcr Channel Marker 17A along the southern shoreline; depth of 4 m.
~ Location 15.9 Cn Catawba River arm, mid-channel, 500 m southwest fron Cacawba River Channel Marker 18A; depth of 25 m.
Location 17.0 O On Catawba River arm. 300 m north of Catawb3 River Channel Marker 19; depth of 14 m. Location 17.5 On H K s~ Creek arm, within Duke Power State Park, under the bridge crossing Hicks Creek; depth of 0.3 m. Location 18.0 IOiarsfiall Steam Station intake cove,15 m from trie ash basin discharge structure; depth of 2 m. Location 19.0 East fFom the southeast end of Goat Island, which is near Catawba River Channel Marker 12, in an elongated, unnamed cove, the long axis of which lies on an east-west line; depth of 4 m. Location 25.0 On Catawba River arm, at the mouth of Lyle Creek; depth of 4 m.
' ~ ~ Location 25.2~-
0n Lyle Creek arm, mid-channel,1.0 km west fr';m the c uth of Lyle Creek; depth of 1 n. 1 0 18 I i
e Table 1-3 (continued) page 5 of 5 t O 1ocatio"255 On Lyle Creek arm, in the mouth of the first southern branch of Lyle Creek 1 0.2 km northeast from N. C. Highway 10 bridge crossing Lyle Creek; depth of I m. ; Location 34.0 , On Catawba River arm, 335 m northeast of Catawba River Channel Marker 13; depth of 21 m. . . _Loca t ion _ 4 5. 5 On CatTwba River arm, the first cove downlake from Marshall Steam Station ! discharge cove. 0.5 km northwest from Catawba River Channel Marker 15; ; depth of 12 m. Location 50.0 i On Catawba River arm, midway between Catawba River Channel Markers 16 and I 17; depth of 17 m. Location 60.0 In Marsh'all Steam Station intake cove, lakeside of Marshall intake ! structure 6 m from the intake; depth of 9.m. Location 65J On Catawba River arm, mid channel, 6.4 km downlake from Lookout Shoals - Dam and 0.6-km downlake from the Southern Railway Bridge; depth of 4 m. O location 68.0 East from the southern end of Long Island, which is 1.4 km southeast from County Road 1004 bridge across the Cat twba River arm in Catawba and-Iredell counties, in an elongated, unnamec cove, the long axis of which lies on a northeast-southeast line; depth of 4 m. i
~ Location 69.0- - -
On Catawba River arm, beneath the Buffalo Shoals Bridge (State Road 1004 ! in Catawba /Iredell Counties); depth of C m.
~ Location 72.0 On Catawba River arm, mid-channel 0.2 km uplake from the U. S. Highways 64 & 70 bridge crossing Lake Norman; depth of:4 m. -t Location 80.0 . _ )
On Catawba River arm, mid-channel. 0.2 km uplake from Interstate Highway 40 bridge crossing Lake Norman; depth of 2 m. l !~ 10 ; 19-I- . .
1 O
\ / \ /
n .- m l'* J. y, .
./ / .:
- e. (/y.
e U a* L *%.
.e C
i g"= E $ i L
- k s I
'{/ .u-t b .C .'/Y CA
(;; / s
~ /e, JJ m 'sp 8' n1 g'T '... , i:J ;
y
- v J m E
3 ..N. I t o. 5 -p T-1 v. i
-te .I -~ c ' 2 :
u 0
., !} e e % . } .,
r o
, -R -
r O. . l D,- -
. . ',] g , ', , -g.. ' "}
- ..tg i, N
c E
. %,. } {'
i r%,%s so. gs 'q .
) ; t:
v ic t a e N .. s ed nc . J. 96 .. ,
-1' l N e' et r- %"
P b ( g.
.'( *. UU a -
e. E % o 6 Nd f., p.
.. ,C l %(e.
{ V
'a e '.8%,
4J
8.. O l- vi t m .\ m Ud h,., . .I c . T -( ,0 s a. y 1 sy<*+ a tr.
st - a
. ., cc r.; \
a. L
~.1 CD e
L. O 20
l O f-s/ ([Lsfocoeit.cteic
- 4 cat inuit ttatie, g yc ..- .... . 1
'v, ,g , . f j[ \<
- varsnell 'h' v '
I . _ ~ ... C. steem 1 -
.g ~,.. .,' '. U, $ /N j,;',. 'Y . / *:j. f , Y, .' / $ l- hl if-o ?!!F ;pjgy 4 5
{3}.d@
.h y . . ,l. ~ ] , . , , @.s".71.t -
mi,e i M I Cncit r:, Comen1 FCrd Y W: lear \ H.creelt:teit $tatioe . . . . Statite \ O Figure 1-2, Lake flornan, tiorth Carolina, showing the thermal mixing zone f or McGuire fluclear Station, (tiorth Carolina Department of flatural Resources and Community Developrcnt 1978). 21 l tm nu umuu pluummi i . . . . . . . -
=
O 1enn. Area Mean Flow Volume
<QN L al e (ha) (m3.s-1) (r 10 m3)7 1 Wh+.erce 5 362 165 37.5 9 2 Cecir Creek 182 154 1.1 3 a o" 6 0.3 e a is % Statkon. 4 Fishing Creek 784 138 7.4 5 Wylie 4 912 116 34.8 /k 7 6 Min. Island 1128 76.5 7.1 6
7 Norman 13 156 75. 135.0 8 Lookout 338 65.01 3.3 N - 9 Hickory 1 401 57.3 15.7 10 Rhodhiss 1 043 48.1 8.4
#'s 11 James 2 721 18.4 35.6 b &L.
S' Catawba River i Drainage Dosin w sk.4[I M F 360Y 330 1 10 / 300 T>,
\ 9 y ? __
[~ 2703 240 2, 6 dd #, 210
- 5 P 180 I 4 - 150<S 3
120 e
) 2 g 90 3 r '
l 30 0 25 50 75 100 125 150 175 200 225250 275 300 325 350 375 C River Distance (km) Upstream from Wateree Dam y Figure 1-3. Schematic of the Catawba River system of dams and reservoirs. Area and volume are with respect to full pool and mean flow is with respect to outflow from that reservoir. Company 1976). (Duke Power g I___-_-____ 22
l O O Surface Area in Thossaads of Hactares 13 12 11 10 9 8 1 6 5 4 3 2 1 0 231 - i e i e i a i i a i i -:Za. 229 - - 229
} 227 -
227 3 225 - - 225 m 223 - 6,e - 223
- c. 221 - @m'4* -
221 2 219 - - 219
; 277 - -
217 S 415 215 213 - - 213 3 211 - - 211 E 209 - 209 E 207 - - 207 N w - 8 M5 - 205 203 - - 203
-* 201 -
201
~
i 199 - 199 197 - - 197 195 195 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 Storage Volure in Sillier.s of Cubic Meters Figure 1-4. Lake Nonaan area-volume curve.
.-e_, _
'2 232.0
231.5 - -' .~% " [- 231.0 -
~^".
m 230.5 -
"n 230.0 . ~~ ~ # % ,. . __.,.,
l u ~~ E 229.5 -
$ 229.0 S 228.5 -
228.0 - 1973 y 227.5 - u, . . . . . . . . . Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Date n3
, 232.0 ~
231.5 .,~' - __ -~-s .. m , . e 231.0 s
.- ~~~: ~ . .e . ; .' *,i '"~ 'r m z30.5 . :.a s, !,~~-- . N. - .e u
3 230.0 q.,._.-.,-
@ 229.5 229.0 -
C o 228.5
% 228.0 -
3974
$ 227.5 Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec.
Date page 1 of ~ Figure 1-5. Lake florman surface elevations, June 1973 through December 1930. Each dot repre,ents daily lake elevation. Full pool elevation is 231.65 m mean sea level (a.s.l.). O O O
11 ,i !llil,!1 .!{i i' li: ) i 1 I l'l'I i 4l l
.!.l i11llli1!lu 4
f _. o 2 _
'< c .,5 c e
e e g
,/,--
- s. " - 5 .
D 6 D P a . w' ' 7 7 9 . 9 . 1 v L_, 1 v o o N ' N ,
..n, .
t
'~ t c _ c ^. O - . O ~ . .- . / p e */ p S
e
- S ~ . '^ g u
g u A - A
\' ,% - .
l l
%._, ~N'-
f, u u , J e J t e t a . a
% . D . D n n u u ~ % J r-J f.-
y y
/' /- a a M M s.'w-r . s r p ~ A s A p
s ~~ -
's e,'f r r . , ' a a M M ' ^< f b ,, F e .,( b F
e
)
d c g,- - f 1 e u _ n - n n _. . a ~,- a i t l J J n o (c. 0 5. o, 5 0 5 0505 0505 050505 21 1 009 9 887 21 1 0 099 887 33 33322 222 33333 22 222 . 222222 2222 2222 2 22 222 5 1 _"
- E Ce,5x 8's *? -
0 m ; g = I* y~ e3 =>e w . e r u. g i F m"
,4 :1u :- ,:=
E l 11ltjl:j jj i;j. ~ l!:!Il:i; ;4.l!iii, ::!!;1!i.!i; 4
m 232.0 . 231.5 - d 231.0 - m 230.5 -
~~. ^:
A '- s. v
' _ 1s a= 230.0 - - ~ / w s j' 229.5 ~ _,,,'- ,e wf 'g 229.0 '
C o 228.5 .,
/ % 228.0 -/ $ 227.5 1977
[3 , , , , , . . , , . . . . Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Date m
- 232.0 7 231.5 , '3 231.0 -'
'% _/5 ~~'"'~.s $ 230.5 - ~~ ', "~ '
5 230.0 l' ',. . ' ' a - e 229.5 229.0 7^ C o 228.5
% 228.0 ~ $ 227.5 1978 g a , . i . . . . . . . . .
Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Date Figure 1-5. (continued). page 3 of a 9 9 9
5-'.^' ^ - - ma. eb h .iinha Asww-a JoAr4+L.44 hem 6 - ^ Ad iom s& A.W A h-4 h' --- e s >~ Q 4 0 4 as CM i 80 2 1-g
,,_s + + . ~ ~
k d '
@ Q l p ,/ m --h, a
v , 1,. 1
,! i i > >
1 i. ... (R . a - a H i 4' # f> I 4 i , !
'g d (,i d l l , <
t i f
- l. -
n, 4 -l s & <
-%- H 1,. @ ,
i 1- , { , l M M l t i a a i i
- {- (
, ..i .
"4 w s.,-an a i w i D.- >= ,
4 4 9 N Q Q ' l I g g I a - 3
*e p> l% . r ,. I j l' > ~
{ 'l. - l l </, . Q Q t.. /
.l
(' t
'(*
w
# t
- 1. .
l l , + s c 4-
.i I GD j b a
3 i \ C s s p s .e. t _5 _4^ M l I E 5.. f's O U i
-( , .c .c, e >e **J th "4 W , . . (
- I i e e e 9 l
^
[C W
. Q. m. C. 0 Q m. .
Q. @. O. m. C. m. O. - @. Q.@. "
- O @ M M N N *= m OO @ m M M N Q i
-e - s 'f ea. f*e N N N N fw M M M M M N N N N N L N 4) , ed N NNN N N N N N N N N tw N N N (e 7 .G ( *.l'S *u $J43 8,j) u0l*4 eAp[3 - (*l t t SJP30W).uolu A8[] *[ _ ;
i i t 27-
O Cowans Ford Dam rTop of Weir - 221.0m ;:
.. .ppet. . . . . .cj.:....
waters.. Epilimnion).,.,jlj
/- #. . #.l f .;.. 9fll. l 7...........................
1
., w am Q. .. -
c.... :;.gllihyf.w.......Metalimnion.:::::-
...... N.. <1:::::: / li!4ll ; ; ;; cool Lower Waters (Hypolimnion)llll //,3;!yq;i .
99.goyag....... ..*.... y . . ..i:!ss.
,r.: :.... ~ .:- . . . ..--ss s . - ......e.e.e...e.e twAwmidi,y /,3979n:[A a u .s m s u n g., g , a in r,x y, w x g,.n ,, , , ,m 7 ,_g g ,,m ,
Skimmer Lake Bottom h. t Weir Figure 1-6. Schematic of Cowans Ford Hydroelectric Station during a typical sumer. Elevations (Not drawn to scale)given are referenced to mean sea level. g 28
t Marshall 4 Steam Intake Canal r Heated Discharge Station Cold Water Used Water for Cooling p Skigsger Wall l
.: :O- b .* %'. . *,. ,'; t .* ,* .* . ' .*. *. ,* W..T Y. . . :.::.. :::s .,. :: u =. ;llX.a.v- . .. ~ . .,. . ..... 1:.... . = . . . . . . . . .. ..,... .. . . a _ a. :._. ...,. . . t.a x e n e , ,, , .. * *. . *, . * * , . s Condenser ,. . ,,,,,,,.... , . . ., ...g* * *. .'m. .r,ww . . . .f ,, , , ' take Bottorn ,,
Bottom of Skinmer Wall-213.6m 9.e _r ,7 % v' C n' I i Figure 1-7. Schematic of Marshall Steam Station intake and discharge con-figuration during a typical late spring or early fall. The l ' elevation given is referenced to mean sea level. (Not diawn to scale) 29
O ( Low Level Cooling Water Pipe i ( Traveling Screens I ( Trash Rack Slot -.- ( Pumps 1 A Top 234.7m m.s.l. 3 j j
- s. 1 i
,i U i
4
,- n Full Pool 231.65 m m.s.1 , /
Maximum Drawdown with l' .# McGuire Operating ' ' 227.08 m m,s.l. 2.44m l .D. m ,
\i ' Pipe to f ,
t Condensers [5 Dispersion Box >> I-- '> y ;
\, , e, '
Flow ' l y ( Bottom of Intake Channe1217.90 m m.s.l.g _,,__ . . . Cg/e ~s - (DispersionBox '-
-q *-
5.59m _L3.31m_ _- ,- -- . _ 4.86my,_ _ E.54m J. 8)n Figure 1-8. Schematic of McGuire Nuclear Station upper-level intake structure, g 30
O full Pool [LakeNorman 231.65m m.s.1 s- N '
'N < Nw , y \ sg Top of Dam I'% \ , , D _ }Fg '*?e, N
Y p ho l
's.,, ~~
1
- s. I l
"y s ,
f' Retaining Wall .,ITopofIntakb J h/ f - "I
- ' Structure f i l Spillway /','I i l
i
~.nE::g::E:: 'l g
N jg; , , Gates IntakeOpend 199.3 to 204.2m m.s.1. l not'dsp ,oa' l l Figure 1-9. Schematic of McGuire Nuclear Station lower-level intake structure on the Lake Norman side of Cowans ford Dam. (t40t drawn to scale) O 31 (
_ . _ _ . .~ -. _ . _ - _ . . _ _ _ _ _ _ _ _. _ _ _ . _ ,___ _ _ _ _ _ _ _ _ _ _ _ O 0k
- s. - .dr e ectr c Station
/ . ~@ . . .. .. : , . a .y :
g 26 2 ' f' cL. ss o .. i WP *)n' ;.Q,'..
?y '.., '.i .. *.:.*.. .I(I-; , . ' ' ' ..;.ye' V S;;.s ,h N .. , ; e '. ' f n.. ..a r,1,%j ., . ,.,,. <di. 6. . :s :. .. , j 35 6 .i ..........ev % ;.y.
26.0 ,
+
69.0- , [-
- 17 6 e6 o 7, - , ., !.'. ,., ).".,
.e ,_ g U . .;. c o . 'i . .!,),C. : * . . , ,'. . . . s, . . . . . r.: ; P - . . . ; . . ? '. . -u 's .v . ;./ j . (.. . t ...'syl.. '.o*,9 ,8 *ep.J . ....'uc .'. , . .i,.t, ' . . . , f .':, ..;
ed 'j' g.p . e
*i , /p ;
g
*i' . ' . . .; e .e : 4 9 ft$lf.. a, N, .' j. : 'l. ' ; ,'* '.;e) '.. .<...,t , ;;.<; . r .*. ,. . ; i, , t. . > ...., .. ( .-
- s. ,
yf .. S
'1 . i.;'; *s .. I' Ma Steam rshall o-shglt i h */ '
b-
' .(. . ' *y , ',;. Statim .
is -
"s ;~. ; .~4l '
u
.; e . L :s .i..~ .. '4 % .. , ;j; , , , , ,; .l . 5 as . ,
fe44 . . ..:., .... . . . .
.c ;4 ,,5 i.* . ... , ,c y1o,.. e 14 0 ,o[ j ..
s..
' ,' e' ,
k y .g ..,. c. lt s7 . :. M g
. u . . , ; .' .. "J"'I n; # / . . . . ., e c.. : s. .
p!.b. 1..~ i y
..,. - *?..<,.
- .:;;* m.
%s . . c.: . . ;., ..;:..:. se > , y. *, , .. '* :.- i . c ' 7 . . ;-g'j , o,r
- . ,, o ';, L: '.n>. n, 2. .e
. 4 . .. ?.. ".'
y.
- \, f <:., . }. h
- s. . '.
;,c ..c.-
o
"k,s
(' ,6.\I.Y '..3 ,'5 O. .
- ., . . v
- ,2,., ,a. :A? '
..//j ' 2Jj3 ....C;.,:..; ,
so s '
'b,' ..i ; .c R 9,6 % .e g :.
p, , (o y.
<.3.... .
e . dg
), :. . * .e, e 'a . . t> , ',,,,.'1 m h;k.k . j. -
Io '
. ,. ,, M ,'# ge ti g' '* ~
73 t) 1
' 2 3 4 m *.o " '" O r "?* b,; .1 . r. w '
1 y,j T ometers .%$b 14cGuire ! Cowans Ford . - hchar Hydroelectric StatiO" 5 v' Station t Figure 1 10. Lake No s n W W M # b 32
t
- e O
,$ ,y M
73 g Cawans for [f f %- Lab Norman f$, ,,,,,
' 16" "$,%@@$2*2 W.. a /g' f H,) irce1 ectr k's w
Station W' -. 7@ ti;%
*9 ' 'e Y:'U:l:
4{l q m. l
.h- (16.2 $m u$,'. ;w ,D .' c: 'vi% Ak 2
n.:t':,',; p dl ainM..
'44'd t/@.;-( . ?
gQ p . c.; ..r d v.': .: m .. s*a 1909'k* ' \d+3~- ... .gf Lucia ,,.6kh 16.2 g'6.0' *,,;.
- :r .a:Mic ., m ,' e %.~
dn c~. . y,,.f;r%.u % w....,~ . f $ $. . 31.. . dl:g;p;f
$T j.fl 6.j M b 'h f.R.e.,A.'l@,>.%..i s.
- v. < .
% %;n.nic.x: - 17.'., m.:c,!:- ., .r %:n .sp.pa b. . ..
16
.. in:.;ymi, M ..v&.
ig.5 ,? s .
~
3 *ff
. . . A ~! jft: r $
h.I:$y.. b
~kikfi'W)"6'w, ' d. n: ' . v.'.Y -
ih9. 5 %./'N x.nM-m%:::
- q. rno a,:R. at. ..
. .! ..,u v.*5 .y;, * ' * - ;*,v.6?jL 4 .3 .gff.i!.;' ~r: ' ':PQ;z*f..}.q). *5:;., ' W ?ll .dnkkY ,-; f,5 b . y':i.. h...t%,$i.Yd jf n, . $$.,'l'tk..h.0Y!
a .... .
.e $d' g'h,dT a pr+f'.V M' 'f 9b r ((q)' V E :i Mountain Island ';;e. - 9.g Hydroelectric % ,D, ,
Station T. ~p g,.h 4
.N M .* 0 1 2 3 Mit :& MM .. Y Kilometers N' . T ..'
6 Figure 1-11. Moantain Island Lake sampling Locations (e) 1974 through 1980. 33
l 1 1 i CHAPTER 2. THERMAL REGIMES - (- R. W.?CACCIA, M. C. GRIGGS, AND D. S. RIDDLE PAGE I INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 !
^
BA.KGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
-OrJECTIVE' . . . . . . . . . . .... . . . . . . . . . . . . 36 MATfkIALS AND METHODS ................. . . . 36 MONTHLY WATER TEMPERATURE MONITORING . . . . . . . . . . . . . . . . 36 SAMPLING LOCATIONS AND FREQUENCY , . . . . . . . . . . . . . . . 36 FIELD PROCEDURFS . . . ........ . . . .. . . .. . . . 36 CONTINUOUS WATER TEMPERATURE MONITORING . . . . . . . . . . . . . . .37 SAMPLING LOCATIONS AND FREQUENCY , . . . . . . . . . . . . . . . 3/
FIELD PROCEDURES . . . ................. . . . . 37 SUPPLEMENTAL DATA . ,...................... 37 DATA ANALYSES .., ........... . . . . . . . . . . . 37 HEAT CONTENT . . . . . . . . . . . . . . . . , . . . . . . . . 37 HYPOLIMNETIC WARMING RATES . . . . . . . . , . . . . . . . . . . . 38 TEMPERATURE DECAY ,....................... 38 EQUILIBRIUM TEMPERATURES . . . . . . . . . . . . . . . . . . . . . 38 RESULTS AND DISCUSSION , , . . . . . . . . . . . . . . . . . . . . . 38 THERMAL-CHARACTERISTICS .......... . . . . . . . . . . . 38 ,
-FACTORS INFLUENCING LAKE NORMAN THERMAL CHARACTERISTICS . . . . . . 40 LOCAL METEOROLOGY ..........,............. 40 HYDROLOGY ..... .. ............. . . . 41 MARSHALL STEAM STATION . . . . . . . . . . . . . . . . . 42 HEAT CONTENT . . . . . . . ..........,....... 42
SUMMARY
................,.............. 44 LITERATURE CITED , . . . . . . . . . . . . . . . . . . . . . . . . . 46 O
34
INTRODUCTION h l BACVGROUND Major changes in physical, chemical, and biological characteristics of a lake are often responses to seasonal changes in meteorology. In the Piedmont Carolinas, lakes follow a seasonal pattern of thermal stratification beginning in Apri! or May and continuing until overturn in October or November, A period of circulation in which the lake is mixed vertically begins in the fall and continues until the onset of stratification. During the circulation period, depending 0, weather conditions of that particular year, alternate periods of slight stratification and mixing are common (Weiss 1960). Lakes exhibiting thermal characteristics similar to those in the Piedmont Carolines can be classified as warm monomictic (Hutchinson 1957). In the Piedmont, lakes are essentially isothermal during winter (Decenber-March). As the spring season be.' is ( April), heat is added to the lake surface by the atmosphere dueino daylight hours. At night, the lake is unable to dissipate all the addeo heat to the atmosphere and warms isothermally as heat is mixed from the lake surface .. the bottom by wind action and con-vection (Edinger et al.1974). The lake thermally stratifies during periods of calm wir,. e heat is added to the lake surface. It then mixes down slowly, and S m n 3r layer forms. If the interface (thermocline) between the upper (epi 10 ad lower (hypolimnion) layers is stable, the lake remains strati-fied " m % ' the summer and the epilimnion thickens (Edinger et al .1974), The 1. ..x ' during the fall months (September-December). The stability of g the th w in decreases as convective mixing in the lake increases. Eventutily tr.e lake completely mixes (fall overturn) and cools isothermally until tne 3onal cycle begins again the foiiowing spring. Duke power Compar.y began me airing temperature on Lake Norman in August 1963, before the lake reached full pool in 1964. Between 1965 and 1971. Johns Hopkins University and management of Duke Power Company directed a thermal effects research project on Lake Norman (Jensen 1974). The focus of the study was Marshall Stean. Station, a four-unit fossil-fueled steam-electric generating station. Determination of the effects of a skimmer wall at Marshall on the thermal efiluent we among the most significant findings of the thermal research project. During st fied periods, Ap-il thrnugh October, the condenser cooling water was with from the hypolimnion of the lake. The resulting thermal discharge was a r .eiy the same temperature as receivina waters. As a result, cons'1 S mixing between the effluent and receiving water occurred in the discha sve and immediately downlike. During the winter months, the temaera-ture of the Marshall effluent was reduced rapidly in the discharge cove. The rapid reduction in temperature was probably due to Lake Norman bottom water flowing toward the discharge structure and mixing upward with the buoyant thermal plume, rather than rapid surface heat dissipation. However, the thermal plume generally remained buoyant as it entered Lake Norman, and was confined to the upper strata of the lake in the immediate discharge area. DJring the winter months when the Marshall thermal plume was expected to extend over the largest surface area of Lake Norman, the plume was slightly detectable uolake at Location 50.0 and downlake at Location 11.0 (Fig.1-12). 35
A study, from February 1973 through January 1974 by Weiss (1975), indicated that the seasonal heating and stratification patterns of Lake Norman were typical of other lakes in Piedmont North Carolina. The weir in front of Cowans Ford Hydroelectric Station intake was found to enhance stratification in the lake during the spring and summer by excluding hypolimnetic water from being-dischaged to Mountain Island Lake. Duke Power Company continued to monitor temperatures on Lake Norman after the Johns Hopkins University study (Jensen 1974) was completed. Thermal monitoring in conjunction with the water quality program (Chapter 3), and continuous temperature monitoring were conducted throughout this perdod. Data collected y prior to March 1974 have been reported by Duke Power Company (1976). OBJECTIVES
- The objectives of this study were to:
1) define the thermal characteristics in Lake Norman, both spatially and temporally,
- 2) identify those factors influencing the thermal characteristics in Lake Norman, 3) establish a thermal data base which may be used to assess the effects on
- Lake Norman from the operation of existing and future power generating D .v stations.
MATERIALS AND METHODS MONTHLY WATER TEMPERATURE MONITORING SAMPLING LOCATIONS AND TREQUENCY Water, temperatures were measured at monthly-intervals, between January 1974
-and December 1980, in conjunction-.with the water quality program (Chapter 3).
Locations and the periods each was monitored are tabulated in Table-2-1. Sample locations are presented in Fig.1-10 and 1-11 and described in Table 1-3. FIELD PROCEDURES , Temperatures at all locations except Location 16.0 were measured in situ using the thermistor sensor cn a Hydrolab Surveyor Model 60. Thermistor accuracy was
. +0.25 C (Hydrolab Corporation 1973). Calibration of the system was performed iirior to the collection of data and, in most cases, after data collection.
Temperatures were measured beginning at 1 m above the bottom. and thereafter at. one-meter intervals to the surface (0.3 m). At ' Location- 16.0, tempera tures were measured with a thennometer. l O 36
l CONTINUOUS WATER TEMPERATURE MONITORING SAMPLING LOCATIONS AND FREQUENCY Continuous water temperature monitoring was conducted between September 1967 and December 1980. For this study, however, data will be examined for only five locations from January 1976 through December 1980. Sample locations, depths, and tne period each location was monitored are shown in Table 2-2. _ FIELD PROCEDURES Water temperatures at each location were monitored with an arrangement of thermistors attached at various deaths to a floating buoy. The thermistors were wired to an instrument building on the shore which housed a Leeds and Northrup Speedomax strip chart recorder. Each monitoring system was accurate to +0.5 C. The chart recorders were calibrated according to the manufacturer's recommendations at the time of installation and at least every six months thereafter. Strip charts were collected weekly and daily readings for 0600 and 1800 hrs were recorded. SUPPLEMENTAL DATA Various meteorological variables were monitored at or near the McGuire Nuclear Station site, during the study period, to provide supportive data. Monitoring locations are given in Fig. 2-1, and periods of data collection, instrumentation, and methods of data reduct1on in Table 2-3. DATA ANALYSES All thermal data collected on Lake Norman from January 1976 through December 1980, as well as data at Location 1.0 from January 1971 through December 1976 are included in Appendix 2. For this study, we only consider that data collected between January 1976 and December 1979 with the exception of continuous water temperature monitoring data for 1980. HEAT CONTENT Heat content, defined for this study as the amount of heat above 0.0'C stored in the lake, was calculated from the monthly temperature monitoring data. Lake Norman was divided into four zones with the following locations representing each :one: (1) the Lower Lake Main Channel Area was represented by Locations 1.0, 2.0 and 7.5; (2) the Ramsey Creek Area by Locations 3.0, 4.5, and 5.0; (3) the Reference Area by Locations 8.0 and 11.0; and (4) the Upper Lake Main Channel Area by Locations 13.0 and 15.0. The assumption that the density of water remains at 1.0 g cm-3 and the fact that a calorie is equal to the amount of energy required to raise the temperature of I g of water by 1 C were used in the calculations. Data for all locations within a zone were averaged to obtain an average temperature for each depth sampled. The average temperature for each depth was averaged with the mean temperature for the depth above to obtain a tempe"ature representative of each one meter stratum in the water column. The representative temperature of each stratum was multiplied by the volume of that stratum to obtain a heat content for each stratum. The strata h 37
from 0 to 10 m and from 10 to 30 m were summed to obtain a heat content for each water mass in each zone. The heat contents for all of the zones were summed to obtain a _ heat content for the entire lake. A volume-weighted
- temperature for each water mass was calculated by dividing the. heat content by the volume of that water mass. For comparison, a theoretical hoat content value for the total lake was calculated by multiplying the total lake volume 'by the average monthly equilibrium temperature (Ryan-Harlemen 1973). All of the heat content calculations took into consideration the lake elevation recorded on the day of data collection.
HYp0LIMNETIC WARMING RAYES
-Hypolimnetic wanning rates were calculated from data collected with continuous c temperature monitors. A linear regression (Helwig and Council _1979) was cal--
culated based on data measured at 25 m between April 1 and August 31 each year.
-The warming rate was taken as the slope of the regression line.
TEMPERATURE DECAY Temperatures, measured monthly at the surface (0.3 m) and 5 m depths at locations in the Lake Norman main channel, were normalized to Location 1.0 (values recorded for Location 1.0 were subtracted from all other locations). The normalized temperatures were plotted against distance from the Marshall discharge structure. Plots were prepared for months when the Marshall thermal plume extended over the largest lake surface area (October through March). EQUILIBRIUM TEMPERATURES Equilibrium temperatures were calculated, based on the unheated windspeed function and the method developed by Ryan and Harlemen (1973), from data collected at Douglas Municipal Airport in Charlotte, N. C. Mean, maximum and minimum daily equilibrium temperatures were calculated for a 25-year period (1950-1974) prior to the study period. Daily equilibrium temperatures for-the 1975 througn 1979 study period were also calculated. l_ RESULTS AND DISCUSSION lI l THERMAL CHARACTERISTICS Trends in -temperature data collected during the study period,1975 through 1979, were similar to those previously reported for Lake Norman (Duke Power Company 1976; Jensen 1974; Weiss 1975). Lake Norman temperatures followed a , seasonal pattern typical of warm monomictic lakes (Hutchinson 1957). Thermal stratification typically began in Lake Norman during April and was l
- i. well established by June (Fig. 2-2-through 2-18). Vertical thermal gradients increased during the sumer as the lake surface warmed. Maximum surface temperatures (excluding the Marshall intake cove and Location 16), for each year, were generally measured during July or August and ranged from 27.5 to 32.9 C (Tables 4 and 2-5) . During August, when the maximum vertical thermal gradient for each year was measured, a metalimnion existed between 6 and 18 m i
O 38
at deep (3,20 m) sample locations (Fig. 2-12 through 2-18). Differences between surface and bottom temperatures at deep locations ranged from 6.3 to 18.5"C g during August. Lake temperatures began to decrea n in early fall. Stratification decreased gradually until the lake overturned in November. Cooling and mixing established relatively uniform temperatures from the lake surface to the bottom by December. The lake continued to cool during January and February. Minimum surface tempera-tures, in the lower areas of the lake, were typically measured during February of each year and ranged from 1.6 C to 8.6*C (Tables 2-4 and 2-5). Lake Norman bottom waters warmed much slower than surface waters after the lake thermally stratified in the spring (Fig. 2-19). A warming rate was calculated for the hypolimnion based upon data collected with continuous monitors at a depth of 25 m at Locations 1.7 and 7.6 between April 1 and August 31 of each year. Sufficient data for the calculations were only available during 1976, 1977, and 1979. Hypolimnetic warming rates of 0.75, 0.72, and 0.78 C month-1 were calculated for 1976,1977, ad 1979, respectively. These rates were similar to rates reported by Jensen (1974) of 0.73"C month-1 for Lake Norman bottom waters in the spring before a thermocline developed and 0.66'C4 month-1 after a thermocline was established. No continuously recorded data were available for the hypolimnion in the vicinity of Marshall. Seasonal trends in surface temperatures measured at Location 16.0 on Mountain Island Lake were similar to the trends discussed for locations on Lake Norman. Maximum values were measured at Location 16.0 between July and September each year and ranged from 26.7 to 28.00C. Minimum surface temperatures were measured during January or February each year and ranged between 2.3 and 8.0 C. Generally, surface temperatures measured at Location 16.0 were slightly cooler during the summer than surface temperatures measured at Location 1.0 (Cowans Ford Hydroelectric Station forebay). During the winter, surface temperatures measured at the two locations were approximately the same (Fig. 2 20). Year-to-year variations in Lake Norman temperatures were small during the summer throu0 h out the study period (Figures 2-21 through 2-25). The warmest temperatures were measured at most locations during the 1979 summer season. During the 1976/1977 and 1977/1978 winter seasons, Lake Norman temperatures measured during the first three weeks of February each year were below 4 C, the temperature of maximum density for water (Fig. 2-26). Inverse thermal stratification (bottom temperatures exceeded surface temperatures) was noted at most locations on the lake as surface temperatures decreased to 1.6 C at some locations. Variations in temperature between locations on Lake Norman were small, excluding i temperatures measured at locations near Marshall (Figs. 2-2 through 2-16). l Trends in temperature throughout the years at Locations 14.0 and 13.0 were usually warmer than temperatures at other locations. Location 14.0 is in the Marshall discharge cove and Location 13.0 is within 2 km of the Marshall discharge structure (Fig.1-10) . Temperature variations at both locations are primarily influenced by Marshall discharge temperatures and condenser cooling water flow rates. 39
, - - ~ - - .
O rac10as inetuencina taxE n0 aman TataM^t canancteaistics-LOCAL METEOROLOGY Thermal characteristics of Lake Norman were basically determined by -local meteorology. Annual and seasonal variations in the lakes water temperatures were the result of diurnal cycles in local meteorology. Data indicate that solar radiation was the dominant factor in determining these diurnal cycles. Temperatures measured with continuous monitors at Location 5,0 were plotted , to illustrate the diurnal variations in Lake Norman water temperatures (Fig. 2-27). Wind velocity during this 48 h period was low (Fig. 2-28); therefore, mixing due.to wind was minimal.- Lake surface temperatures began to increase at approximately 1000 each day and reached a maximum value at approximately 1800 (Fig. 2-27). Minimum values were recorded at approximately 0600 each day. Maximum solar radiation values were recorded between 1200 and 1400 each day indicating that increases in lake temperatures typically lagged solar t radiation increases by 4 to 6 hours due to the large heat capacity of water. At night, the surface temperatures cooled first to the temperature of layers immediately below the surface. This process continued until the. top 2 to 3 m of the lake was isothermal. Wind movement across surface waters also caused variations in Lake Norman water temperatures (Fig. 2-29). Reid (1961) reported wind has a substantial-influence on watw movement in a lake and may cause translational movement of.. surface water wnen wind force is sufficient to cause whitecaps. Wind-induced
-O circulation of water promotes a vertical transport of heat from the upper layers to the lower . layers; the epilimnetic layers lose heat while the hypolimnetic layers gain heat (Hutchinson 1957). In Lake Norman, the wind affected water temperatures at depths well below the lake surface-(Fig. 2-30).
Equilibrium temperatures provide a simple approach to heat exchange analyses - and are useful in examining long term trends in lake temperatures (Edinger - et al .1974). On a lake receiving no thermal- discharge, the mean monthly lake temperature should follow closely the mean monthly equilibrium temperature (Jensen 1974). The lowest mean daily _ equilibrium temperatures typically occurred d_uring January and February (Fig. 2-31). Mean values for-January and February rantred from 2.3 to 8.7 C._ The highest mean daily equilibrium temperatures for the 25 yr period typically occurred during July and August and-ranged fros 27.6 to 30.5 C. Daily equilibrium temperatures _for the period-varied considerably ranging from a minimum of -10.5 C to' a maximum of 40.2"C. , Daily equilibrium temperatures for the study period followed trends similar to the.25-yr mean daily values (Fig. 2-32 through 2-34). -Values for the 1975
.through 1979 period were generally within the range of the 25-yr maximum- -.and minimum the study per_iod equilibrium temperature.
were January The coldest and July-1977, and hottest months during respectively. Edinger et al. (1974) reported that lake surface temperatures typically lag equilibrium temperatures by one to four weeks. Lake Norman surface temperatures generally lagged equilibrium temperatures by two to three weeks, and followed
~O the seme seasoae' netteras es e9ei1ibr4vm temperet"ree- The coo 1est e"a ~8cmest 40
lake surface temperatures during the study period were measured during the 1977 winter and summer seasons which were also the periods when the maximum and g minimum equilibrium temperatures occurred.
. M0 LOGY Hydrological processes grtatly influence the thermal structure of a lake, though generally to a lesser extent than meteorology. Hydrology is governed by natural sources such as precipitation and stream inflow, as well as man-induced usages, ooth consumptive and non-consumptive. The operation of Lookout Shoals and Cowans Ford Hyaroelectric Stations, and Marshall Steam Station affected Lake Norman's thermal structure during the study period.
Operation of Cowans Ford Hydroelectric Station affected thermal stratification in Lake Norman in the forebay area during the study period. Stratification patterns in that area were also affected by the skimmer weir located in front of the station intakes. The top of the skimmer weir is located approximately 11 m below the Lake Norman full pool elevation. The weir prevents the colder bottom waters of Lake Norman from being withdrawn from the lake during periods of stratification. Temperatures measured with continuous monitors at Location 1.7 during two 24 h periods illus+. rate the effect of Cowans Ford Hydroelectric Station operations on thermal patterns in the forebay area (Fig. 2-35 and 2-36). Location 1.7 is approximately 65 m from the dam located on the lake side of the weir. Temperatures measured in the upper 14 m of the water column at Location 1.7 were immediately affected each time Cowans Ford began operation (Fig. 2-35 and 2-36). The maximum change in temperature was observed between 4 and 14 m in the water column. The cold deep waters mixed with upper warm waters, resulting in a net decrease in temperatures. Temperatures in the top 4 m of the water column were affected by Cowans Ford operation but changes were less than observed for the 4 to 14 m depths. Temperatures measured below 20 m were fairly stable when the station was operating due to the skinmer weir retaining the bottom waters. The increase in surface temperature (Fig. 2-36) was probably due to increases in solar radiation during daylight hours and not station operations. Temperatures at all depths oscillated slightly as Cowans Ford began operating. As station operations stopped, temperature oscillations were pronounced, especially at depths in the middle of the water column. Several hours af ter Cowans Ford ceased operation, temperatures returned to approximately the same temperaturc measured prior to operation of the station. Steam electric generating stations affect lake thermal structures not only by rejecting heat to the lake, but by displacing large volumes of water from one area to another. Stations which utilize bottom waters for condenser cooling purposes also displace large volumes of water from the lake bottom to the surface. When the lake is stratified, the hypolimnion is gradually reduced as the summer progresses. Temperatures at Location 15.0, in the vicinity of Marshall's skimmer wall, followed trends similar to those temperatures in other areas of Lake Norman (Fig. 2-2, 2-3, and 2-11). Bottom waters et Location 15.0 warmed only slightly faster than bottom waters at other locations. The thermal similarities illus-trated represent Marshall Steam Station operational period. h 41
3' o
,To realize the cummulative effect that Marshall had on the thermal character-istics of Lake. Norman, the average temperature profiles for the pre-operational period ' August -1963,1964, _and 1965, and the . operational _ period August 1970, ,
1971, and 1972, were plotted for Locations 1.0, 8,0, and 15.0 (Fig. 2-37). These years were chosen due to their similarities in meteorological conditions. The vertical thermal gradient in Lake Norman showed a distinct decrease from pre-operation to operation of Marshall. This change in the thermal gradient is very similar to that found on Lake Keowee after Oconee Nuclear Station became operational (Duke Power Company 1977). On both Lakes Keowee and Norman, the , variance in thermal gradient was primarily due to artificial mixing of the lake by.the utilization of hypolimnetic waters for condenser cooling purposes. MARSHALL STEAM STATION Waste heat is discharged at rates up to 1.710 2 kcalimonth-1 from Marshall when the station is operating at full capacity. This amount of heat could theoretically raise.the temperature of the volume of Lake Norman 1.3 C per month if = the entire amount were stored in the lake. Heat rejection rates (waste heat) averaged 0.93 1012 kcal month-1 during the study period (Fig. 2-38 through-2-40). Monthly mean heat rejection rates varied from a minimum of , 0.88 1012 kcal month-1 during 1976 to a maximum of 1.0 1012 kc?1 month-1 during 1979. Although daily mean heat rejection rates were erratic, tne monthly mean values' indicated Marshall operated at relatively consistent levels during the study period. O- The Mersheii thermal p,eme wes typ4celiy observed over the ,ergest eree during January and February (Fig. 2-41 through 2-45). Differences between surface temperatures measured during the study period at Location 14.0 and Location 1.0 vcried from a minimum of 3.8 C in October 1976 to ll.0*C in e January 1979. The most rapid temperature decays were observed between Locations 14.0 and 13.0_(Fig. 2-41 through 2-45). The majority of excess heat had_ dissipated before reachir.g Locction 11.0, approximately 6.0 km downlake from the Marshall discharge. Temperatures measured at 5 m depths indicated that most of the Mar: bali thermal plume was contained within the upper portion of the water
-column (Fig. 2-41 through 2-45). The maximum temperature difference between the 5 m depths at Location 14.0 and 1.0 was only 3.1 C observed in February 1977.
During spring and summer months (April through September) Marshall discharges , are of.approximately 1.he same temperature as receiving waters. As a result, the plume mixes deeper during these months versus the fall and winter periods when temperature _ and thus density differences tend to " float" the warmer water on the receivi_ng waters. HEAT CONTENT Heat content analyses are useful in the evaluation of seasonal changes, mixing, and heat storage in lakes (Derecki 1976; Duke Power Company 1977). Local
- meteorology is the dominant influence on lake heat content and is responsible .for.its cyclical nature. Other factors such as hydrology and thermal discharges j-are also important and may cause deviations in normal lake heat content cycles.
42
Annual cycles in Lake Norman total heat content were relatively consistent g during the study period (Fig. 2-46). Minimum values for each year were W observed in January or February and ranged from 4.8 1012 kcal (February 1978) to 1010 2 kcal (January 1975). Maximum values were typically observed in 1 July, August, or September and ranged from 28 1012 (July 1977) to 31 1012 kcal ( August 1979). Total lake heat content followed the same cyclical pattern as the theoretical heat content based on mean equilibrium temperatures (Fig. 2-46). Maximom values for the theoretical heat content were mucn greater than maximum actual heat content values. The Lake Norman epilimnion was affected more by meteorology than other strata of the lake because the thermocline impeded mixino and heat transfer to the hypolimnion. Lower strata were not able to absorb and store as much heat from the atmosphere as the upper layers and did not approach the theoretical heat content aftcr a stable thermocline was present in June. The cyclical pattern of the actual lake total heat content during 1975 and 1976 followed closely the monthly pattern exhibited by the theoretical lake heat content; however, during 1977,1978, and 1979 the lake heat content appeared to lag approximately one month behind the theorctical heat content (Fig. 2-46). The differences are probably due to a three-week shift in the Lake Norman sampling schedule ' rom the last week of each month to the first week of each month beginning in January 1977. Monthly mean equilibrium temperatures were used to calculate theoretical heat content and would be most representative of middle of the month values. Since Lake Norman tempera-tures lagged equilibrium temperatures by two to three weeks, temperatures collected at the end of the month would be reflective of equilibrium tempera-tures calculated for the middle of the month. Temperatures collected during g the first week of the month would be more nearly reflective of mean equilibrium temperatures colculated for the previouc month. Heat rejection rates from Marshall were a small percentage of the total lake heat content during most of the study period (Fig. 2-46). Heat rejected from Marshall contributed most to the heat load of Lake Norman during winter months { when the lake heat contents were lowest. Lake temperatures, especially in the area near Marshall, were probably affected more by heat rejected from Marshall during the winter than during other seasons, as evidenced by the actual lake heat content being greater than theoretical (Fig. 2-46). Volume-weighted temperatures (heat content of a layer of water divided by the volume of the layer) are useful when compring heat stored within various areas and depths in a lake. The Lake Norman volume-weighted temperatures were approxi-mately the same for the surface to 10 m and 10 m to bottom layers during winter months (December through March); however, as the lake stratified during the summer, the surface to 10 m values were much greater than 10 m to bottom values (Fig. 2-47). During the fall season the surface to 10 m layer cooled rapidly and was approximately the same as the 10 m to bottom values af ter lake overturn. The thermocline in Lake Norman inhibited mixing and heat transfer from surface to bottom waters. The volume-weighted temperatures for the 10 m to bottom layer lagged the volume-weighted temperature for the surface to 10 m layer during periods when the laxe was stratified. Heat storage within various areas of Lake Norman was evaluated by dividing tu lake into four zones: the Lower Lake Main Channel Area, the Ramsey Creek 43 1
. --- _ .~.--- .--- - - -.- _ - -
O Area, the Reference Area, and the Upper Lake Main Channel Area. Surface to 10 m and 10 m to bottom, volume-weighted temperatures were calculated for each zone.- Spatial variations between volume-weighted temperatures for the total water column in each zone were small (Fig. 2-48 and 2-49). The volume- , weighted temperatures calculated for the 10 m to bottom layers in each zone were also approximately the same. Surf ace to 10 m volume-weighted temperatures for each zone varied the most (fig. 2-50 through 2-54). The Upper Lake Main Channel Area volume-weighted temperature (surface- to 10 m) was - typically ' greater than the_ surface to 10 m volume-weighted temperature for other zones due to the Marshall thermal discharge. Differences were greatest during winter months, especially January and February. During months when the lake was stratified the surface to 10 m volume-weighted temperatures calculated for the Upper Lake Main Channel Area were approximately the same as values calculated ~for the other zones. This wasonal variation in surface to 10 m volume-weighted temperatures between the zones was due to the use of hypolimnetic condenser cooling water by Marshall when Lake Norman was stratified. Differences between surface to 10 m volume-weighted temperatures calculated for the other zones were small. Typically the surface to 10 m volume-weighted temperature - for the- Reference Area was slightly warmer than the Ramsey Creek Area and the , Lower Lake Main Channel Area.
SUMMARY
Temperatures on Lake Norman were monitored monthly, at various times during L (3 the period from 1974 through 1980, at 21 locations. These temperatures were U measured with a Hydrolab Surveyor Model 6-D. In conjunction- with the monthly sampling program, continuous hourly sampling was done at one location from l- 1971 through 1976 and at five other locations from 1976 through 1980. Recorders, attached to thermistor chains, measured temperature profiles in the water strata , at these locations. Temperature data collected on Lake Norman followed seasonal patterns typical i of warm monomictic lakes. The lake began to stratify during April. Maximum surface temperatures were measured during July and August and ranged from . 27.5 to 32.9#C. The lake began to cool during the fall and vertical strati- ' fication decreased. Overturn was typically complete by the end of November as L l relatively uniform temperatures were established from surface to bottom. Minimum lake surface temperatures, in the lower areas of the lake, were y generally measured during February each year and ranged from 1.6 to 8.6 C. Year to year ariations in Lake Norman temperatures were small during the l l
- summer season throughout the study period. During the 1976/1977 and 1977/1978 winter seasons, temperatures measured at most locations on Lake Norman were below 4 C for.a brief period in February and inverse stratification was observed in'the lake.
The thermal characteristics of Lake Norman were primarily determined by local meteorology and followed seasonal meteorological patterns. Annual and. seasonal variations in Lake Norman temperatures were the result of diurnal cycles in . local meteorology with solar radiation beina the dominant factor in determining
. by the lake surface during O diurnal cycles. More heat was general'-
d5ylight than was dissipated at night du 'ng spring and summer seasons resulting in a gradual warming up of the lake. During fall and winter, more heat was '- 44 l (- - - - _ - _ _ _ _ - . - - , - . . . -. ..
l l l typically dissipated to the ttmosphere from the lake surface than was absorbed, resulting in a net daily reduction in lake temperatures. Heat was transferred h in the water colun.n during diurnal cycles by convective mixing and wind action. Thermal characteristics of Lake Noman were also influenced by lake hydrology. Natural sources as well as the operation of Lookout Shoals and Cowane Ford Hydroelectric Stations and Marshall Steam Station affected lake hydrology. The operation of Cowans Ford during spring and summer seasons effected thermal stratification in the vicinity of the station to oepths of 20 m. Temperatures measured between 4 and 14 m were affected the most by station operations as mixing at those depths resulted in tne greatest changes in temperatures. Based on temperatures measured at Location 15.0, the Lake Norman hypolimnion in the vicinity of the Marshall intake canal was reduced only slightly, compared to other lake locations, during the summer due to the withdrawal of bottom waters by Maishall. Since Marshall utilizes bottom waters for condenser cooling water, discharge temperatures during the sumer were approximately the same as surface temperatures measured at locations out of the influence of the Marshall discharge. Maximum differences between surface temperatures in the discharge cove and those in other areas of the lake were measured between October and March when the lake was nearly isothermal. The Marshall thermal plume was typically observed over the largest area of Lake Norman during January and February. The largest surface temperature decays were me Lured between Locations 14.0 and 13.0. The majority of excess heat had dissipated before reaching Location 11.0. During the winter g the themal plume was buoyant and was generally contained within the top 5 m w of the water column. Discharge temperatures in the summer were approximately the same as surface receiving waters causing the plume to mix downward to slightly lower depth than in the winter. Annual cycles in Lake Norman total heat content were relatively consistent during the study period. The heat content for Lake Norman closely followed the annual cycles in equilibrium temperatures. Spatial variations between volume-weighted temperatures calculated for the total water column in each of the four Lake Norman zones were small. Surface to 10 m volume-weighted temperatures for the Upper Lake Main Channel Area were typically greater than the surface to 10 m volume-weighted temperature for the other zones due to the Marshall thermal discharge. Differences were greatest during winter months. During months when the lake was stratified, the surface to 10 m volume-weighted temperature calculated for the Upper Lake Main Channel Area was approximately the same as values calculated for the other zones. Typically the surface to 10 m volume-weighted temperature for the Reference Area was slightly warmer than surface to 10 m values calculated for the Ramsey Creek Area and the Lower Lake Main Channel Area. O 45
. _ _ __- _ _ _ _ _ _ - _ _ _ .. _ . _ . m ..m - =
LITERATURE CITED Derecki, J. A. 1976. Heat storage and advection in Lake Erie. -Water P.esources Researr,h. 12: 1144-1150. Duke. Power Company. 1976. McGuire Nuclear Station, Units 1 and 2. Envirnnmental 4 Report Operating License Stage. 6th rev. February 1976. Volumes 1 and
-2. Duke Power Company, Charlotte, NC.
l- . 1977, Oconee Nuclear Station Environmental Suumary Report 1971-1976. Volume 1. Duke Power Company, Charlotte, NC. Edinger, J. E., D. K. Brady, and J. C. Geyer. 1974. Natural water temperature responses, p.17-26. In Heat Exchange and Transport in the Environment.
- RP 49 Report No.14. ETectric Power Research Institute, Johns Hopkins University, Baltimore, MD. 125 p.
Helwig J. T. and K. A. Council (ed.). 1979. SAS Users Guide. SAS Institute Inc., Cary, NC. 494 pp.
- Hutchinson, G. E. 1957. A treatise on limnr'ogy. Volume 1. John Wiley and New York, NY. 1015 pp.
i Sons. HydrolabLCorporation. 1973. Instructions "or operating the Hydrolab-Surveyor Model 60 in situ water quality analyzer. Austin, TX. 146 pp. 5 Jensen, L.- D., D.- K. Brady. R. F. Gray, W. D. Adair, and J. J. Hains, 1974. O- Thermal and water quality characteristics of Lake Norman, p. 7-119. In L. D. Jensen (ed.) Environmental Responses to Thermal Discharges from
- Rarshall Steam Station, Lake Norman, North Carolina. Electric Power Research Institute, Cooling Water Discharge Research Project (RP-49)
Report No. 11. Johns Hapkins University, Baltimore, MD. 235 p. Reid, G. K. 1961. Ecology of inland waters and estuaries.' Reinhold Book Corporation, New York, NY. 375 pp. 7 . Ryan, P.cJ.'and D. R. F. Harleman. 1973. 1 analytical and experimental. 1 study of transient cooling pond behav.ar. Massachusetts Institute of Technology,-Cambridge, MA. 439 pp. Weiss, C. M. 1960. The Catawba River - limnological and water quality studies, June-October 1959. Duke Power Company, Charlotte, ;" 146 pp.
, P. H. Campbell,'T. P. Anderson, and S. L. Pfaender. 1975. The lower Catawba Lakes: Characterization of phyto- and zooplankton communities and their relationships to environmental factors. Department of Environmental Sciences and Engineering, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC. ESE Pub. No. 389. 396 p.
t l . C) 46
,-m-- -.---g , ~.e----m ,n , - . . - - m,e , -w , -- - - - - - . - --w ., . , w ,,-o,~-- e ~
e , -- .
Table 2-1. Monthly temperature monitorin9 conducted during the period from a January 1974 through December 1980. W Locations 1974 1975 1976 1977 1978 1979 1980 1.0 A A A A A A A 1.2 - - . Jul-Dec A A A 2.0 A A A A A A A 3.0 A A A A A A A 4.0 A A A A A A A 4.5 A A A A A A A 5.0 A A A A A A A 6.0 A A A A A A A 7.0 A - - - - - - 7.5 - A A A A A A 8.0 A A A A A A A 8.5 - - - - -
*iay-Dec A 9.0 A - - - -
A A 9.5 Mar-Dec A A 10.0 A A - - - - - 11.0 A -8 A , A A A 12 0 A - - - - - - d.0 A A A A A A A 14.0 A A A A A A A 14.5 Sep-Nov Feb-Dec - - - - - 15.0 A A A A A A A 15.9 - - - - Mar-Dec Jan-May - 16.0 A A A A A A A 17.0 Jan-Dct - - - - - - 17.5 Jul-Dec A - - - - - 18.0 A A - - - - - 34.0 - - - - Jun-Dec Jan-May - l 50.0 - - - - Jun-Dec Jan-May - 60.0 - - - - Jun-Dec Jan-May - A = dampled during all 12 months of that year. l
= Not sampled 47
O Table 2-2. Sample locations and depths for continuous water temperature monitoring during the study period. Location No. 1.0 1.7 4.3 5.0 7.6 9.5 - Date 1/71- 12/76- 8/78- /76- 1/76- 1/76-Effective 12/76 12/80 12/80 12/80 12/80 12/80 Depth (m) 0.3 0.3 0.3 0.3 0.3 0.3 s 3.0 2.0 2.0 2.0 2.0 2.0 6.1 4.0 4.0 40 4.0 4.0 9.1 6.0 6.0 6.0 6.0 6.0 , 12.2 8.0 8.0 8.0 8.0 8.0 15.2 10.0 10.0 10.0 10.0 10.0 18.3 12.0 12.0 12.0 12.0 12.0 21.3 14.0 14.0 14.0 14.0 14.0 24.4 16.0 16.0 16.0 20.0 10.0 27.4 20.0 18.0 18,0 25.0 18.0 30.5 25.0 20.0 20.0 30.0 20.0 Bottom Bottom Bottom Bottom Bottom Bottom 5 O 48
Table 2-3. Meteorological data collection information for the McGuire Nuclear Station Site. Monitored Variable Period of Record Beginning Ending Type Instrument Accuracy Method of Data Reduction h oTemperature 1-29-76 12-31-80 4 lead 1 0.5 F Avg. hourly to (@ 10 m) copper RTO nearest 0.5 F overtical tem- 1-29-76 12-31-80 4 lead 1 0.5 F Avg. hourly to perature grad- copper RTD nearest 0.5 F ient (27.7 m separation from 10 m sen-sor) Rainfall 1-26-76 12-31-80 Belfort 1 0.03" Hourly tota s weighing fron 0-6," to nearest 0.01" rain gage 10.06" model from 6-12" 5-780
- Dew Point 1-29-76 12-31-80 EG&G 1 0.5 F Avg. nourly to Temperature Model 110 nearest 0.5 F
(@l0 m) Hygrom-eter Solar 4-21-75 12-31-80 Belfort i 5% Total ly d'I Radiatico Model 53850 Pyrhelio-h graph 9-14-77 12-31-78 Eppley + 0.31 Total ly n~I Model Ty min-I to nearest 0.21 ly 8-48 1-1-78 12-31-80 Eppley + 0.31 Total ly h-I Model Ty min-l to nearest 0.11 ly 8-48 Pyrano-meter l i 1 O i l I 49
Table 2-3 (continued) Page 2 of 2 Monitored Period of Record Instrument i Method of Variable Beginning Ending Type Accuracy Data Reduction
- Wind Speed 1--29-76 12-31-80 Teledyne '~+0.5 mph 30 min. avg. pre-(low level 0 Geotech ceeding each hour 10 m) Series 40 to nearest 0.1 mph
- Wind Speed 1-29-/6 12-31-80 Packard- 10.5 mph 30 min. avg. pre-(High level Bell ceeding each hour 0 41 m) Model 101 to nearest 0.lmph Wind System __
- Wind 1-29-76 12-31-80 Packard- 15 30 min avg. pre-Direction Bell ceeding each hour (High level Model 101 to nearest 5 0 41 m) Wind System
- Wind 1-29-76 12-31-30 Teledyne ~~+5 30 min avg. pre-Direction Geotech teeding each hour (Low Level Series 40 to nearest 5 0 0 10 m) ,
4 O , i
- Located on Permanent Meteorological Tower at Mc uire Nuclear station.
Other variabl es were monit6 red within 1 km from lMcGuire. l ! 50
Table 2-4. Maximum and minimum surface water temperatures measured with continuous monitors on Lake Norman. h Minimum / Maximum Temperatures ( C) ' Location 19'il 1972 1973 1974 1975 1976 1977 1978 1979 5.0 ' 6.7 6.1 8.3 6.5 5.2 1.0 29.4 i 3 D 3T7 32.5 31.0 29.0 2.0 3.2 4.4 I'7 32.6 31.5 32.1
*- ' #O 4 "'
- 32.'3 30.1 5.0 1.6 1.6 3.7 5.0 31TJ
- 3T.T 31.8 5.7 1.9 3.2 4 .1 7.6 . , p
- 4.7 1.6 2.6 3.8 O
9.5 U 32.1
- 32.9 i l I
- Insufficient number of observations to warrant reporting.
l l O 51
O! , Table 2-5. - Maximum and minimum surface water temperatures measured monthly in Lake Norman. Surface Minimun/ Maximum Temperatures ( C) Location 1975. 1976 1977 1978 1979 l '. 0 7.7/28.9 6.1/28.5 2.5/28.1 3.5/29.1 5.0/29.2 1.2 .
*/ 28.1 3.5/28.7 5.0/29.3 2.0 _7.7/29.0 6.4/28.1 2.4/27.8 3.7/28.5 5.0/29.3 3.0 7.7/29.5 6.4/28.0 2.5/27.8 3.2/28.1 5.0/29.8:
4.0- 7.6/29.1 6.1/28.4 2.3/28.2 2.7/28.9 5.0/30.2 4.5- 7.6/29.1 6.4/28.0 2.7/28.0 2.8/29.0 5.0/30.1 5.0 7.5/29.6- 6.4/28.1 .2.4/28.3 2.9/28.5 4.3/30.9- ' 6.0 7.7/30.3 6.6/28.0 2.6/29.8 2.8/29.2 4.0/30.9 7.5 8.1/29.4 6.3/27.6 2.8/28.2 3.7/28.3 5.4/30.3 8.0 7.8/29.0 6.5/28.0 2.7/27.9 3.7/28.3 5.4/30.1 l 8.5 5.1/30.5
; 9.5 4.3/28.9 5.0/29.8 10.0 7.8/30.4 11.0 8.6/30.4 7.0/28.1 3.7/28, 9 4.3/27.7 6.8/30.8 -13.0 11.0/30.5 9.3/28.1 9.1/29.4 6.6/27.5 9.9/32.0 14.0 1 5.8/ 31 .2 14.6/30.5 12.3/30.6 10.2/27.6 13.6/30.6 14.5 16.4/30.3 <
15.0- 9.0/30.5 7.6/28.3 5.8/29.4 2.7/27.6 7.0/ 31 .4
.15.9- . */27.7 4.4/*
16.0 8.0/28.0. 15.7/27.2 2.3/27.2- 5.6/26.7 17.5 9.0/31.0 18.0 9.0/28.8 34.0 */ 27. 6 7.4/* 50.0 */27,6 7.6/* 60.0. */20.0- 5.1/* ; i
- Insufficient sample period to justify reporting.
l-t O l 52 l:
u /'\' 3 LAKE NORMAN s.
,.o k s tu. *o~o ., e t 4,
I \ N M l 1 s g/
/!
4,
.m _
s i.
"gg%g - 3.M-@dDh =&eam x ,. g g%g(Yp ~ .. ,. s [^' '
4' V' )
<n w Cowans For ff i b ili % ps 6f2.2,(- )
2/ [ McGuire Nuclear Station f k -) } --._
\ Permanent Met. Tower ) \
(Hign Level El . 271.3 m) \ k (Low Level El. 240.2 m) . 5 3 Catawba River 2 Figure 2-1. Location of the Pennanent Meteorological Tower at McGuire fluclear Station O O O
b c e D 2 I O v 0 o N 2 n o i t a t c
/ c o _
O L t _ a 9 7 9 l p 1 e h S g u o
. r h
t g 5 u 7 A 9 1 6 m 2 . o r f l ) s d u e
' J h r t u n s o a
( m e m O J n u i T e m t d s a r o f
) \ y C \ M a '
( s h ( . t e l p s \ r p o s A i e .
\ rC uN t
a , r r n _ a ea p n
. M mn e o t N t
n e _ ak e a b M L _ e F 2 5 2 e n r a u g O ' 8 J i F
" 0 0 0 1 2 3 W
4
Oq r i , , , 2-y
\ \ 26 8 6 10 12 14 8 0 22 2 24 22 20 h
i N N 16 - t
\ .
E \ N { 18 N / 1 \ ,% % / /
\\ i i $ 16 14 12 10 20 -
N
!"n '
8 12 T i \\ m 30 - Jan Feb Mar Apr May Jun Jul Aug Cep Oct Nov Dec Time (months) Figure 2-3. Mean temperature isopleths ( C) for data ntasured from 1975 through 1979 at Location 3.0, Lake Norman,flC. 9 9 e
O: O O
!!,hk ] 6 / \
6 8
\\\ \ - 10 12 14 16 20 '\24 \26 /
26 24
'22 20 13 16 } (f f ^ \ } j .
I 10 - Os_ , E f. o E - m 20 - , m 1 b 30 - l Jul Aeg Sep Oct Nov Dec Jan Feb Mar Apr May Jun t 5
- Time (months)
Figure 2-4. Mean temperature isopleths (C) fcr data measured from 1975 through 1979 at Location 4.0, Lake Norman, NC.
0 Y / t{ 8 6
/\ \12 \\18 \\ he \ 24 2 1 T2 10 \ 22 I \ $ \ i i i \ p 24 v 26 6 16 \
I 10 16 h' l .8 -j 14 1e -
/ 1 \s =u B
l - ./ h/ 5
, 20 -
u 30 - Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time (months) Figure 2-5. Mean temperature isopleths ( C) for data measured from 1975 through 1979 at Location 4.5, Lake Norman, NC. O O O w . --
O - O Oi W I 0 '
\ ?
I 18 16 1 10-8 10 12 j4 18 22 24
- 8 d- 16 2 20 l I
N .
- } .
N. 24 l 10 - f e / 1 18 8 6
\
20' 16 14 1'2 1 + 8 j ) i
\ j g 20 t
I 30 - Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr. May Jun Time (rronths) , Figure 2-6. Mean temperature isopleths-( C) for data measured from 1975 through 1979 at Location 5.0, Lake Norman, NC. .
n
\
N) x r/ j
\x\'Nxx+/ , z, 2 z' ,, , =
N
=
x4 y ~ x -
~
i 8 6 6 10 I 8 12 3 14
/
30
; I s \ . N N 9 R S i ,t E E 8 m e a Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time (months)
Figure 2-7. Mean temperature isopleths ( C) for data measured from 1975 through 1979 at Location 8.0, Lake Norman, NC. O O O
C>s a o 0-
\\ \ V / 2 I
c l 22 20 18 16 7 s 4 b x
" ~, \ ' \ $ l2 14 )
s l I -
?
c2 10 8 8 10 20 \ i
$ f ! \ ; \
k 30 - Oct tiov Dec l Mar Apr May Jun Jc' Aug Sep Jan Feb Time (months) Figure 2-8. Mean temperature isopleths ( C) for data measured from 1975 through 1979 at Location 11.0, Lake thorman,14C. I _+
0 1 c e D 2 l [ 1
/
I N v o 0 e 3 [ 8 1 n 1 o i t t 0 c a 2 O c o 2 L 2 . t N a p 9 e 7 S 9
/ 1 6 h 5 1 g
6 u 2 o r
/ g u
A h t 5 7 9
. 1 m ) o l
u s r f J h t n d o e m r ( u s a
% n e
m e m J u i T t a a O 2 d 1 . r o y f a ) M C ( s h t r e p l A p o s i 0 . 1 e . rC r uN a t a , 8 \ M r n ea p m mr eo 6 t N b n e e ak F ea ML [\8 9 2 n a e J r 0
/ -
0 1 0 i F u g 0 0 3 1 2 7 g- k$
l 3 O O N N 28 ' ~/g [ *' % \ V'
/
[ 18 16 l j t 12 I 6 IO g 0 / b l
% l 3 1 5
a a 20 - l 30 - Sep Oct Nov Dec Mar Apr May Juri Jul Aug Jan Feb Tire (months'; Figure 210. Mean temperature isopleths ("C) for data neasured f ron 1975 through ?979 at Location 14.0, Lake Noman, NC.
'N % --"
ON =e o y . W 7 p.no y e, r -
= ~ 0 - g~ e, C u N .,~N\ ~ \
O' e . E r w~ e, - e
/ -
y Q
/ > . $h //
C, a o a
/ t /j / /l/ -
r L so
- E S
l/ si g
.e= / c /
f ' / *^ ,
./ 5 u *a EE LP ~
Ne .
)
fs- I
~l i .s' w
O O S S 8 (unau) gda0 g3 u -
O O - , lo j, '
- Hby o , ,a' o ,!'j S ----Jun i,/ ,' >/p - - Jul , /o < ^
10 - ~ " ~ ^" 9
-++- S ep / ,- /
o' -
./
L., a' / ,/ 3 ~*~*- O c t ,- ,' s
,,*/
8 15 - 5* j .i.'
!/ $ 20 - ,
lr/
, // ' Ii 25 - / , , .!l i:
li! 30
' 'U 0 5 10 15 20 25 30 Temperature (OC)
Q 0 , i
\ l I o j i j fiov 5 -
j j, , l i ---- Dec j j - - Jan {l ,
. j" I ..- Feb - 10 1 -
C 1 ++- lia r o i l i ' -*+- A p r T, . j,; S 15 - i i i c jo ' < g . l . l 8 i i, , i. 20 - i l 1 'i o i 25 - i i l' i \ l o a 30 'I l' I'I ' ' ' 0 5 10 15 20 'i 5 30 Temperature (OC) Figure 2-12. Mean monthly temperature profiles for data measured from 1975 through 1979 at Location 2.0, Lake f4orman, f4C. (A
l l l i 10 Hay o ll #
), , !l l -/'; .5 - ----Jun ,' I ,
f,/
. _.Jul / ' d 10 - - - Aug ,- ' /h.
l 7' -++ S e p
/ ,/ / , ,/ l 3 -*-*- O c t ,i s ./
g )$ _
./ .'/' *i
.e j ,:V . j f? o 20 -
//
l l,p-25 l 30 O 5 10 15 20 25 30 Temperat;re (OC) 0 , ,
, h I 8 Nov i
5 kI o 8
---- Dec 1 - - Jan ;
I I s i - - Feb - 10 - 1 I< i
* -* *~ Ha r C . ) i 3 ! j" l -*-+- Ap r g - . i
- 1 1, I 15 - 5 t' O TF '.s ( 20 -
*)' ! l \ \ /
IO II 25 t 30 . 0 5 10 15 20 25 30 Temperature (OC) figure 2-13. Mean monthly temperature profiles for data measured from 1975 through 1979 at location 3.0, Lake Norman, NC. 65
O O o <. b l0 ll May ,, ,#,- /l/ 5 - ----Jun ,
. </,. - - Jul , ,A ' ,./ - - Aug ' ,i / .'
10 - -
/ ', /
g -*+ 5 e p ,' ,- k -*-*- O c t
,/ ,- , . -
B 15 , ./.' 5 j f,i / o 20 -
'7 #
25 30 10 15 20 25 30 O 5 Temperature (OC) Q 0 - 1
\ . g .- l j Nov l I l Dec 5-- ,
l l Jan l Feb l g 10 - l j< l e- Mar u . 1 3 } t -*+- Ap r b'
- } f, l - 15 -
i ,lg 5a *
\. ,,
I I,
$ 20 -
I
./ / I I
l ,< 25 - 30 5 10 15 20 25 30 0 Temperature (OC) figure 2-14. M;3n monthly temperature profiles for data measured from 1975 threngh 1979 at Location 5.0, Lah florman, f4C. ft
I i 0 . j' ,/ i May f ,- ',/ '
,/:#
5 -
----Jun ,
d' '.!
. .Jul , ',e' /
10 .
-"- Aug ,'
i \r .. j n ' / ,
,/
y', 'J ' ,.
-* + 5cp ,-
C b -+-e- O c t s'
/ $ ,' ,. ',/* /
15 -
,o , ,.
M
/ j fp g
a 20 - i l
.'../**f $ . '.l tW f /f , 4 25 - I ,h:
30 e i n / , , , 0 5 10 15 20 25 30 Temperature (00) O I o l f Nov
/ l I
I 5 -
- :i t 1 1 -- -- Dec '-
I I l
. io - - . Jan 7 10 -
f l, !
- - Feb ,1 -++- Ma r ! l
- API 15 -
i l 5 I
} " I \
t a i 20 - I ,, t
. i , h lI I
25 - I ! i
- l l '
t , s, , 8 > ,! 30 i i , 5 10 15 20 25 30 Figure 2-15. Temperature (DC) Mean monthly temperature profiles for data measured from 1975 g through 1979 at Location 8.0. Lake Norman, NC.
'- 67 i
l 0 , . . f i i j ,. . # i May j . 5 - ----Jun ,s
/ , ,/
dl ./ /
-. . .J ul - -"- Aug j ,g 10 -
7 -*+-Sep - ./
$ - H -Oct ' 's- 'ih y ,5 - ,.i,. ,.. . , -- ' * , **. (
J
$ l/
n 20 - tf'...
, ..c / /l l I j ./
li) 25 - jjj 1 30 0 5 10 15 20 25 30 O temnereture (oc) 0 . l tiov pl , 5 - : .I , l ---- Dec I i I. - -
- J an l
I !> j .- . . .- F e b
- 10 -
j l
-*+- Ma r 0 . s "' l l -**- Ap r 5
7j f 15 -
)
5 0 20 - 25 -
' i ' i i 30 0 5 10 15 20 25 30 Temperatt.re (oC)
Figure 2-16. Mean monthly temperature profiles for data measured from 1975 through 1979 at Location 11.0, Lake florman, f1C. 68
+ ' + - - , ~ . , , , _ _
0 May /
,r' 1 ,- i ../
5 -
----Jun i,' ' , . ('.' /'!.' . _.Jul ,/ ,
10 - - - Au9 ,/ ./ 7L -+-+- S e p
/
f*,- 3 ~*-*- O c t ,/ , / E, 15 -
,e ' , .. / f' g n' ./ /
5 $ ,/ 20 - h,,. i
-(' M 25 -
30 O 5 10 15 20 25 30 Temperature (OC) 0 e
, h / ,,. ' ' ,'
Nov 5 - ljl , l ---- Dec
- - Jan l l I - - Veb 'I '
10 -
} ) (,, j -++- M a r b l -+n- Ap r E -) I jo ' ' t 15 -
t 5m .
,4 s 20 -
j 25 - 30 0 5 10 15 20 25 30 Temperaturc (OC) Figure 2-17. Mean monthly temperature profiles for data measured from 1975 through 1979 at Location 13.0, Lake Norman, NC. 69 i e
O o May /
,. ,,' .if' 5 - ----Jun ./.'/ - - Jul < / . < - /
g 10 - ~~^"9 <
^ -+-+- S e p ,' ,'
u B ~*-*- O c t
.- ' /.e -[f.' .-
E 15 - ,r
- , i x
g ,! ;,. ,<.,i E 20 -
)l l p-I 25 -
30 ' ' ' ' ' O 5 10 15 20 25 30 Q Temperature (OC) 0 *,
/ / !
j I N ov.. 1 5 ( l - ~ ~ ~ Occ
.l l ,I - - Jan l I - - Feb - 10 -
I !-l\P i
. l C i -++- Ma r as *I l l l T, I, i o i -++- A p r d 15 -
i ,I 5 Et (i ii p i o - I
. i 20 -
t
)
i 25 - 30
- i i 0 5 10 15 20 25 30 Temperature (OC)
Figure 2-18. Mean monthly temperature profiles for data measured from 1975 through 1979 at Location 15.0, Lake Norman, NC. 70
35 - 30 - q 25 - 20 -
= E )'f 14 m h 15 -
f 25 m
-d 33 m 10 -
5-I I I I I I Jan I Feb I Mar I Apr 3 May I Jun 8 Jul Aug sep Oct nov Dec 3 Figure 2-19. Temperatures measured at location 1.7 by continuous monitors during 1979, Lake Norman NC.
* # G l . w--. m v- , - _
O \ 0 ' h.
- 0. 6 1 1 n
r 9 o t n n ' 7 a o o ' 9 r i i 1 i t t ' e a a r c c oo
' t r
L L i e I n t
' a ' w y
l
' b t
8 e
' 7 e f'
9 r i ' 1 g
' ;r i
r
- u d
I 0
" f i ) d s n ' h a t
7 n 0 7 o 9 ( n 1
' 1 s ' e n lf m o O <
f ' i T i t a c o I L
' t ' a ' d \ e ' r ' u .
6 sn
' 7 aa 9
1 em rro sl f
' e re uk , ' t a aL I .
en
't ro v
t n g.. i
' er co 5 at ' 7 fi 9 rn ' 1 uo I '
Sm i
' 0 2 - - - 2 5 0 5 0 e 0 5 0 r 3 2 2 1 1 u g
O GE E3n5ao i F
"~
i ;: '
0
'7.
o i b o
~
o e b o
'U*
o b o o bflf5 of a o o , 6 o , o o o , o o < 28' - e o I, o c o o o .o , n m ,, a o o o , a o l c o e, f - ,, 3 n , , , o r. I
., , . , 24 , -
rw e j j g <) <f 10 - fo '
.\
o
\
e
'a o o a \,a z
o o < o o j r
) "
10 12 ' " " ' ' 8 e 8 o o i i o o o. c -. - 3 o e_ , 5,- , u , , s n a o 10< [ 8 , -
, c 20 e o o f, lc , o .> n n o e e o ,e 11, e g 7, o i n c y , ,.
n e s .., \> < a o 16- - o n o o { o ^ c y 20 -
'7.6 y T - 'o < c-w s A , a , , o c e c k <
c o n n n o e A - s e, , a c o o c. 1, n
,\ 9 s --
3 y a c c o r, y r
# o a ry C , < n , q r c
o
' ' ' ~
30 - > Jan Feb Mar Apr May Jun Jul Aug Sep Oct Mov Dec Time (r'onths) Figure 2-21. Terperature isopleths (~C) for Location 8.0 ce Lake fioman during 1975. O 9 9
O O o h
r c' 'i c28. e 'o e O ' 6.4 m
o ) kAo v o ck o o o o fo> o I o c e o o o e o o o . o o o o - o o e o e o o o e o o o c- 26 o o o O o o o o a O o o s O o e O o o u o O - o n A
< o e <
o, a g O o o e gp -
" o f O o ' o 16 e o -
o o 10 o # o N c o o 4 o > o c I o o I 20 o 10 10 - o o o o o <. I ',' o 1 3 o o 8 o o
, c '
o 12 o o g , o o o n ,, 17: o o o o O 1, o o b o 4 o o o o a c t' o o o o e e o rl c e o e o r) o o o O o o }g o o -I o o ' g o o o o ko o e 0 o U e o o o e a o o o e
'\ oo -
o 16 7 o o o o
' n o 3
o e ' o r 20 - o o a o o o n - f o o o , m , o 0 e a
' o f e o <
o .. e o o o - o O o * o ; o O O O o o O , I e o o m o c a , c, o 06.3 o 30 - Sep Oct Nov Dec Jun Jul Aug Feb Mar Apr May Jan Tire (mon +hs) Figure 2-22. Temperature isopleths ("C) for Location 8.0 on Lake Morran during 1976.
u o 8 0 ' 2./ t \ a c 2 7.9 a o o e o e L v. o ; o o o O O o ' f {): i o o ' o o < > 4 t o 14 o o c o o , o : I o o . { o o o o o c c a
)
4 ; j
\ 2\o o l ' , 0 ?
6 2a c 6 o e t
\ o o ,
a o o
}on '
10 E e 16 . o c o 22 a o o - o 1 10 - , . O o q o o o o o o 10 o - N - o o g , j o o , O O O o O o O O u U o ') r N ' e - r' o o O v a 0
- i o O o O q O o r - O p 9 O a {c '
O 3 r a o in l' G ) c O o o n o a . j o - t3.C 0 - a
'I 1 o o r o o a s o o o o o na b a o e '
20 - , ' - I o a o y e o j o o o 0 o , i o o o a o o o c a
- o o o o n e o a o o o o o . .
n a o n o o o o o o 7e o o a <> o o r, n
,3 0 - o o a o o o , c ., o r 0 0 o o C o ;
o o O o o o O c e o o 30 - Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Time (months) Figure 2-23. Temperature isopleths (*C) for Location 8.0 cr. Lake 7:oman during 1977. O O O
__..--.~.~_-----....--,_,~-~.,.-.--._~.-...x. - g . _ . . . - - . . - . _ _ , . . . . - .
. i U ~ N 62 e- - o OO O f' O O OOO O O O O O O O O O O O OO O O eoO ou O
O M OOOO O O O O o D 0 0 0 oeO O D eO + 0 o e 's o o o ..i o
+J V
O O O O O V 'O O O P G p p o 'i O ( OO O O O OO O eo0
'h a
N A . OO OOOOOO OO O - 0 0 J r. . O O O O oO Do OD0 0 k m ( 'l Cn o s 4, 4
- s OOO OO( O , D ~ OC O yC , O O OOOO 2 OO O O O O O C CO N @ .
,} } **' N &
N
^ 0 "3 e .g 7 E. fc 4J J OOD000 r- 0 (
- o o;or oooO- Op oooOoooooOoO l . 6 w
O C 2 bd 7 H r 4 OO O OO O O O O > O O J OOOOOO O O OOOO OO OO OO O
/ a to O CO v " # o
[ tt N g J g L OO O O J O O O O O OO QO OO OO O O O OO O OO OOO O n l U t O o
,-- v # r E , < 5 ~u ooooooOO O O O OD O O OOO OO OO OOO OO O OoOo Q) . a .
O ' _-D--- A
~ 0 E L 3 +2 0OOOO OOO OO OOOOOO O OOOO O OO O OOO , [
G)
- e _. &
9 9
.C3 2 -{-
gO OOOOOOOOOOO _=7 OOOOOOO OO OOOOOO w-O OOO Cm O . l . 4 , N I N c 0 O.
, L 3
O OO O O OO O 0 0 6 OO O OO O O O O OO O O O O O O i r- O eo O fo m , o
- t I
[ o o o l M N n l
- (s.nalaul) yada0 76 l
l _ - . - - ,. , ,. .,_ . . . - . - .. _ _ _ , _ . , _ . . _ . _ _ , . _ _ _ _ _ , . . , . . _ . _ . _ . , . , _ _ = - . . . - . _ _ __ . _ . , , , , . . . . _
% O_ ~_
e-
% y O % N *e C c C' ~ - i JO - c s y . , .m O ^ - 'O O OO O O O O C C O O C- D ; t- e c ; i, O c- t s t eO N- __
4d 0 N 0 O t O ;- C C C C C c C . O t 13 , n_ g W (' M m O O O C CO O C' C C O CD C ,. ; O C D r O C D o O O r N *" c N k
- ch 3 D O st M C 20 0 0 0 0 ' O O O' C D J O o o O O O , C O C O D O rp CO '
/~ e .n O
N k - M r
- 3 e
to j g 7 C to
*' e C )
O OO O OO ( O O O O O O C O O , O O O O OC D O O M O O ( w bh O
/ C $ cb a
e
'O O OO O O O / O O s i OO '
O O O O O c0 0 r O eO O c c
~
e 8
+s ro u
O to d
=g L 'O n OO C a oO c p oo occ 0 O O C O O O O O C 0 O O + / -
o
/"
a E
?,, ' O O O DO OO C O O C pO O O O O O O OO O O O OO O O e-O to # e g / ,
L 3
' ^ ~ D C 'pM5 O O O O O OO O C O O O O O 1 O o oo to a L W
D B Y w
- o. w s/ x w w >~
O OO O O O O C O O O O O O O O O O O oO CO O O D C n c .
. m e_ _
t C ' #D g L \ w . - - 3
/ cn . n O O OO O ooo ; eO c0 0 O eo O poe O e O OoO e i u.
o o o o e N t<> t f l (5004M)qada0 77
0 4 5 10 - i' o Ik 15 - 0 3 8 bl un 0 20 - o 25 - o
-+-+ Februa ry 1977 30 -
February 1978 35 30 5 10 15 20 25 O O \s Temperature (oC) figura 2-26. Mean temperaturcs measured at Location 1.7 by continuous monitors during february 1977 and 197fi (meen of twice daily values) on Lake florman. 78
100 Expected Values, 9c- . .......... 90-85-- { S0- h 9 75 - 3 70 - e 65 - S 60 - a
- 55 1 50 -
I 45 - 3 40 - 0 35 - l t 30 - l 25
$ ,I y 20 f ~
15 l 10 l ! I 5- l ' ' ' 0 0700 1100 1500 1900 2300 0300 0700 1100 1500 1900 2300 0300 0700 Time (hours) 40 35 - 30 - g
. - _ m E
e
~gN y*%,~
3m 20 - g w e ohc c a %% 15 #pC'O --0" 0- ' C 0 0 0 ON [5
.3m a a. 6m - ~ 2m ,_o.10m 5
u 4m .. _ ?4m 0 0700 1100 1500 1900 2300 0300 0700 1100 1500 1900 2300 0300 0700 Time (hours) g fioure 2 27. Incident solar radiation dat a (Eppley flodel fl- W "yr."greter ) collected near the tru!re site and ter nerature vari 3bilit y at Location 5.0 on Lair wi un iron 1 throanh 3 June 1978 79 i
l O 6 7
? .E 3 3 -
E
/ .
a ' ' ' ' ' ' ' ' ' ' ' ; 0 0700 1100 1500 1900 2300 0300 0700 1100 1500 1900 2300 0300 0700 i Time (hours) 40
~
O 30 - G E 25 - E B 2 20 s > w f a
*- 15 - .h a
i 4 10 - 5 0 0700 1100 1500 1900 2300 0300 0700 1100 1500 1900 2300 0300 0700-Time (hours) figure 2-28. Lower level wind speed and e'- temperature data c<,11ected from pd- the licGuire site from 1 throuob 3 June 1970. 80
i l 15 0 12 - t O 6 9 - m W W a m
?
6 -
- =
3
, , , , , , ;
- i i 1 0400 0800 1200 1600 2000 2400 0400 08 3 1200 1600 2000 2400 Time (hoursj
- ~ ~ - - -
40 e 0.3m . 16m
% 10m .... 2 5m 35 -
12m 33m _ . . . _ 14m 30 - u o 25 _^
~
1
*--' W ^ e s e 1_ .1 1 . .# 1 1 z, /l l3
{ f
,o g " 20 t-# : \
{g f ,, . ! %,.......'^'.. . . _ = - =-
..._..........- ~.' - " - " ' . . . , . . . - . . . . . -
y 15 -
...\ - - ~~~' .......___..,___.........<'\ _ > , - --~_'
10 - 5 0 040L 0800 1200 1600 2000 2400 0400 0800 1200 1600 2000 2400 g Time (hours) W figure 2-29. llind speed data tron the t'c,uire r site and water temperature variability at Loca:. ion 1.7 on 16 and 17 June 1979 81
o G V , 2-4 - 6-8 - 10 - 12 - 0 14 - B E 16 - i5
-o 18 -
t 20 - 22 - 24 - 26 - 28 - 30 - i 6/16/79 0600 hr. 32 -
.. . 6/18/79 ~' ' ' 1000 hr,
- , i i i 5 10 15 20 25 0
n Temperature (OC) Figure 2-30. Water temperature variability at Location 1.7 on Lake horman before and i f ter periods of hiah winds of up to 9.0 m. set-I . l l 82 1
45 40 -
~
yl l 30 - 1
^
25 - 1
!h '
h} ' 20 - Maximurr, O 15 l k D , S O 10 - hi Mean h k* 0 pt[ kb l
-5 '
d Minimum
-15 I ' I I I I I I I I !
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Tine (months) Figure 2-31. Twenty-five year (1950 through 1974) daily minimum, nean, and maximum equilibrium temperatures calculated from data from Douglas Municipal Airport, Charlotte, NC, using the unheated model of Ryan and Harleman (1973). O O O
f 1 40 35 -
/
30 - I 25 - Y i b r
\ '- ll \'
B \ E 20 ~ ' ffk lU
$ j y 15 -
j I f J li 10 - 4l 5 ll 1 e l S s l l / f EI b (I 1975
/
0 l b
-5 - ' i ' -10 Jan Feb e
Mar Apr May Jun Jul Aug ' sep ' Oct Nov Dec 40 O 35 - i 30 - l, d fl 'd o~
?
25 - I If J IV i h , '1 B .
- 20 l
Ih.j g 15 - l l s j ).. { 10 - jl j A I
. 5 5 j i $ 1 ( }
9 , 1976 l
-5 - ~
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time (months) Figure 2-32. Daily mean equilibrium temperatures for 1975 and 1976 calculated from data from Douglas Municipal Airport, Charlotte, NC. using the unheated model of Ryan and Harleman (1973). 84
40 35 - h S 30 - l h 25 - ' tj d / lj ' N)Nl ' I [ 5 20 - I J
! f l )/ [
15 -
.;! ! [ j I
y' ~ I ; 1 5 5 - A 1977 I /
-5 - -10 ' ' ' ' ' '-+ ' ' ' ' '
Jan Feb Mar Apr May Jun 'ul Aug Sep Oct Nov Dec 40 35 - h 30 - I I I I b 25 - l' I 20 - I! !
!l" ' f E
g 15 - l
\ %d
[
,i i{.f{
g l1
$ j (
g 10 - s .( / . l} p / 5 l{
\ ! j 'h 9 k j % 78
(
-5 -
f fi i i i i i i i i l i i 1
- -10 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time (months)
Figure 2-33. Daily mean equilibrium temperatures for 1977 and 1978 calculated from data from Douglas Municipal Airport, Charlotte, NC, using the h unheated model of Ryan and Harleman (1973). l 85 I
l O 35 I 30 - ; C If L hlf l )hj k ( fl !' g 25 - jl s 4 ! ( _ 'd 20 - a h f
-o lf 'N S
15 I Iff k l ! f- 0l n t
.o 10 +
(r ){ 1979 f w L' i )
.0 .
I V 'l l ( i i 5
' ' ' ' ' ' ' ' ' ' I -10 Jan Feb Mar Apr May Jun Jul- Aug Sep Oct Nov Dec -
Time' (nonths) f f l T'
._ O . rioure 2 34. oe4,y meen enu4,4hr4um temperetures rer ie79 ceico,eted trem aete-from Douglas Municipal Airport, Charlotte NC. using the unheated model of Ryan and Harlemen (1973).
86
600 500 - g i m-m 400 - w e e CD
% 300 -
5 m o -2 200 - c.
- D 100 -
0
' ' ' i e i 0600 0900 1200 1500 1800 2100 2400 0300' 0600 Tine (hours) 30 25 :. .
h p = N == W,.,,,7 / i =^
= =^- i y 20 5: : : : - O ^ ; g N_e : :_ %
c T N
~
B --**'... -"~'~~""....-..-...-.....-....--~' ~.. m 15 . . . _ . . . . . . . . . i u . e
- e. ..-.-.-...._._..._.....-.~._.-.....---.....-.-.-.-.--.-.-..
B - - - - - - - - - - - s __--_-___ , , _ . _ _ _ - _ - - - - _ _ _ . 10 -
-*-6 .3m 12 m 5 - -*-+- 4 m - - - 14 m -* 8 m - - - - 20 m -+-+-10 m ---- 2 5 m 33 m ,
i ' 0 i _0600 0900- 1200 1500 1800 2100 2400 0300 -0600 Time (hours) figure'2-35. Cou ns Ford hy P o distharge er.d water terperature variability at Location 1.7 on 14 end 15 slune 1979. O 87
Y 600 C, 500 - i M M 400 - E s.- a> E"
.2 300 -
u
.m Q
o 4; 200 - x 100 - 0 ' 0300 0600 0900 (200 1500 1800 2100 2400 0300 Tiiae(hours) 30 ~1 . . y/f 2 ;
^w# 1 -1 +- & *'l S ~^ %u ~c, -
0, 25 ~
--e A_ - ~ :
f
-e A
C
.. - ..~.. ..... -..- ..- ..-.. .. . .....-.......-
G ~. . ...,.."... o-20 - e L n g -.........s...-~~........-.-..~...,.-~...._... u 15 - N. . . . . . - . .....-------~~~-------- - - - - E s 10 -
-*-6 . 3 m - 12 m 5 - -+-+- 2 m i - - 14 m H 4 m - - 20 m -e-+10 m ----- 2 5 m 33 m 0 '
0300 0600 0900 1200 1500 1800 2100 2400 0300 l
,q Time (hours) %)
Figure 2-36. Cowans Ford Hydro discharge and temperature variability at Location 1.7
- l. on 20 and 21 August 1970, t
88 t
.j \[? .* a dfi (s' 1
i n //
,j /4 ,/ / / /Nt /1
- s k a
/ l' l i e
f '/ I / .l ,f ,
' ~ / i /
j
/ / / / /' (
t c t " t l
,a / * ~
- ,'.. /
' ,j f $ ,' ,/ $
l
\
t ,/, ' l l '
,a l l
i
/ , ! ,, 1
- n. , , !
/
l 1 lo' et en I.C
- t j i,
f/
.- 4 t w 0, n _:
j Locat!'" 2.3 l > a o r 3 .. o ;a e n :- , n , 3 4.- , c m ro n4 - en %...,,g . Figure 2-37. Vertical temperature profiles at Locations 1.0, 8.0, and 15.0 on La;e Norman. The profile at Location 8.0 for the operational period was detemined from 1970 and IPl data only. (August 1962-1965, pre-operational period ; August 1970-1972, operational period t. :) O O O
. (3 V
6-
,1975 n 5 ,
e "o . ! .. ,
! 1 4 -k ; ! ,, (4 i
( I N j 6 3 . I( h' 'l t {
)
0 ' O I I l
, i f r $ -i i
l U6' s '
'2 ci:
2 - l f 1
' ' ' I I I i i , i i 0 ~
7-t, Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time (months) 6 _ 1976
^
o "o 5 - i" - l I I
. 4 - % _( [) f i , $ 3 _ ' I i y f '
I
/
l Oh I 7 h i g, I < g
/ I) t* k ,
I f '
] l l
f g }' 2 - r l 1 O. OL
' ' ' i i e V - dan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec I
Time (months) Figure 2-38. Mean daily heat rejection rates from Marshall Steam Station for 1975 and 1976. 90
6 1977
~
S 5 - ? -
\ l
)4 u j I (ik i l ( kl y j ,l h 1 l is 3 J l i I 3 b ( , l t f. E 2 l
, , , q lf % 5 {
y l. P l - l j 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Tine (nonths) g 6 1978
^
c 5 - 3 ' I 9 $ f\ s a 8 l k,if ' 7 ' l
\ I i p
- ,o i g' 3 i ij ll f
, t j g
{l ~ l (
$e - I j ( 0 1 0 l i g l \\
1 0 - Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time (months) Mcan daily heat rejection rates from Marshall Steain Station for h Figur e 2-39. 1977 and 1978. 91
----- ~ -
i O 6
- 1979 ~ -
5 R S. .
)
I i
,f b \ t 7 ,f ff '
[l I, ' 7 f
,f 1 1 l - ' )
3 l g f }
' , 1 8
I
/
y , T ,
" 2 - .!:I , l(lh I j _
1-ae" res se, ' ,,, May Jun Jul Au9 Sep Oct NOV Dec O Time (months) O Steam Stati0n f0r
" S"" - " ";,e; oe41y oeet re;ec140, ,,,,, ,,,, ,,,,,,,,
92
- - - - - - _ _ ___ ~%-m
J
-l i I I I I i 110 14.0' 1 3.0 .0 8.0 7.5 2.01.0 .-
Sample Locations 9 - a ,
;V 8 -- ~
Surface (0.3m) decay Jan.
.u v7 -
s g ... ..... Fe b .
' 6 -
E s Mar.
\
q 5 - s - 0 Oct, s
=:sE' 4 ._ 's s : Nov..
N
- y. '~
s Dec.. g- 3 -
's, .g 2 - ) . %~~~ ~~ ~ ~ - -~~____ -0 -
p _- .
. )0 -. 5 10 15 20 25 I l. -l i I I I 10 -
14.0 13.0 11.0 8.07.5 2.0 1.0 4 Sample Locations W 9 - n P8
. ,o 5m decay Jan. ,.7 - .. ____ Feb, t6 - -*-e- Ma r .
t: E5 .
-e 0 Oct, w , b1 4 - -e-e- Nov.
1 1 2 Dec.
-5 3 -
i :p L R -2 - _N _ c:
) . N_ '
- _j
= " ~ '. _ i 0 - ^ ~ ~ -
1 O' 5 10 15 20 25 Distance from Marshall _ Discharge (km) 9. Figure 2-41. Temperature decay at surface (0.3 m) and 5 m depths measured on Lake Norman during 1975. The plotted values were normalized to the temperatures at Location 1.0. 93 .
,y..- - -~,,-.--.w. -.-...i - , p .e- m --p .-- aws a g
I l .I l l O. ,0 14.o is.o n.o I 8.o 7.5 I I 2.01.0
- ., Sampie Locations 9 -
u'8 o 7 \ . Surface (0.3m) decay Jan. 87 t g t,
' ... ..... Fe b . *6 - * ; Mar.
g -- 5 -
\\ +--+ Oc t .
E4
-a .. s \ -e--*- No v . \ *- De c . 'b 3 - \
gg _. ~~~~ s S 1 t. g -
-q~ _: _._-_.
0 . _. _- _3_. _ - x 1 , 0 5 10 15 20 25 o 30
.i .,I.o i3.0 l- 1 n.o Sample Locations I
8.0 7.s I I 2.0 3.0 I 9
^
3u8 - g7 _ Sm decay Jan. C
.g ------ Fe b .
m 6 - t -*-*- Ma r . E5 -
-o--e- Oc t .
v
- b . 4' --
u-
* ~~*'NOV' m
Dec.- D3
& ~ +*2 - - - - - - ~ ~ ~ ~ - - 'w
- .v.-
m '~ %g %-----.- . 2g
+_ g-1 . .
0 -5 10 15 20 25 { Figure 2-42. Distance from Marshall Discharge (km) Temperature decay at surface (0.3 m) and 5 m depths measured on l t Lake Norman during 1976. I. L t. 94
( I I I. I I I I I4.0 13.0
-10 .\
11.0 Sample Locations 8.0 7.5 2.01.0 g 9 -- g ,
,. s v 8 - \ ~ \ Surface (0.3m) decay '$'7 -
s s Jan. g . .. ..... F e b . g6 w
'N : Mar, -s ; 5 -
s s _ 0 Oct. s.
,E ., 4 s -
Nov, s s M Dec.
.ba 3 -
s e s,
- 2 -
) _
w 0 -
.x -- - v
_)0 5 10 15 20 25 l l- 1 I I I l "I 10 14.0 13.0' 11.0 8.0 7.5 2.0 1.0 - L Sample Locations 9 -
^
v 3 8 - g7 . 5m decay- Jan.
,C .. ___.Feb. .y 6 - % -*-+- Ma r .
E 15 -
-o-e- Oc t .
E .4 - -o e Nov,
$a - -1 i Dec.
E3 - -
~' s '
M gg ..
~,'
h ',
'1 - 7/ " &% ,_ _ m ~m
p .0 -
== .. z _ h ----. _
7- x '
-1 '
l 0- 5 10 15 20 25-L Distance from Marshall Discharge (km) Figure 2 43 g Temperature decay at surface (0.3 m) and 5 m depths measured on Lake Norman during 1977.
.95
.. - . ~ . . - . - -_ -.. - .- . - . . _ - . . -l l l l l l 1 -10 14.0 13.0 11.0 8.0 7.5 2.01.0 Sample Locations 9 -
C'8 0 Surface (0.3m) decay
]7a , - Jan.
g s
. .. ....- Fe b .
d' w 6 \ -*--+- Ma r ,
&5 - ' -o--e- Oc t .
E4
- s
+ -e- N o v . %
- Dec.
t3 - y2 - N s %
~ ,
1 0 -
-- - ~ 5w .1 , , , i 0 5 10 15 20 25 l l l I I I I 14.0 13.0 11.0 8.0 7.5 2.0 ',1.0 10 Sample Locations 9
U L8 - g7 _ Sm decay Jan, y ------ F e b .
.( 6 -*-e- Ma r .
E5 -
-o-e- Oc t .
E : : Nov. 4 - 3 g 2 Dec. c; 3 - 4 -2 - 1
,6 - ~2 ~ -~__.---
0 - L 3 0 5 10 15 20 25 Distance from Marshall Discharge (km) Figure 2-44. Temperature decay at surface (0.3 m) and 5 m depths measured on Lake Norman during 1978.
- 96
6 l l 1 I I i 13.0 11.0 8.0 7.5 2.01.0 10 14.'E
\ Sample locations 9 - \,
G8 C
~ \ \ Surface (0.3m) decay Jan.
87 - j g . .. ..... F e b . y' 6 -
's -
Mar.
+ \
q 5 - s e Oc s,' , .e g 4 s, -c : No a G
% ~, -
s i a Dec. .s 7 u 3 . 4 s' i y 1 2 s ' 02 ~
~
m '~~ ,
~
I
%~- =" "__" -A-x 0 -
1 _ 0 5 10 15 20 l l 1 I I I l'~ _14.0 13.0 11.0 8.0 7.5 2.0 1.0 Sample Locations h 9-U -
.o . 8 g7 _
Sm decay Jan. g _____7eb. -~ 56 g n : Har. E5 -
-o-o- Oc t .
E -e---o- No v . a 4 -
% = 1 Dec.
E3 - ~ s P '~~. 3 2 - A % 0 - Y '^ - N ~~ F
-1 O 5 10 15 20 25 Distance from Marshall Discharge (km)
F1gure 2 45 Temperature decay at surface (0.3 m) and 5 m depths measured on Lake Norman during 1979. 97
O O O
-e-e- Marshall Heat Rejection : Theoretical Lake Heat Content ~ ^ ^
alculated Lake Heat Content g
"o 7 -
hi ' 8 9 g 3 - 4, 3 l E - h ( ' U i 8 I E > 4
% 2 -
N
\
c { ' ee -
\ . .
4 0 5 1 E k ' 4 3 8 - r u 0 1978 1979 1975 1976 1977 Time (months) Figure 2-46. Heat content considerations for Lake florman for the period 1975 through 1979. The theoretical heat content is based on monthly mean equilibrium temperatures.
35 c0 10m to bottom vol-weighted temps C-0 Equilibrium Temperatures c : 0 to 10 m vol-weighted temps i e- total lake vol-weighted temps 30 - i 25 - { 20 - I e i h \ 1 8 5 ( \ 15 -
/ , <t, 4,
1 f L f A l c d 5 - t I I . I E I I I i 1 I I I I E E t I iiiI E t f f I 2 i9 i f e I A l 3 l l E f G t t i O A iiiI 3 I i R 1975 1976 1977 1978 1979 Time (n.onths) Figure 2-47. Lake floman volume-weighted temperatures and monthly mean equilibrium temperatures. O O O
O-V - 35 Reference Area 0 to 10m depths 30 - total lake % 10m to bottom % 25 - E 20 - B e 2 r E 15 - > 5 l- ' < { 10 - f </
~.
I I ' 0^' ' ' ' ' ' ' 1975 1976 1977 1978 1979 Time (months) 35 Lower Lake Main Channel Area 0 to 10m depths total lake % 30 - 10m to bottom % _ 25 - l P
* - 20 -
h // > %L . i I l R 15 - H A 10 - f 1 p 5 - 0- 'I ' I" '"""I^ ''''I'''''' 1975 1976 1977 1978 1979 O Figure 2-48 Time (mo"the) Volume-weighted temperatures for the Reference Area and the Lower Lake Main Channel Area of Lake florman during the study period. 100
1 l 35 Upper Lake Main 0 to 10 m depths O Channel Area 30 - total lake 2 10m to bottom : 25 - G l 20 - L E I 3 > I ) ( E 15 - E I B 10 J, ' V s I f 5 ,
" ^ ' ' ' ^^^ ' ' '
0 ' 1975 1976 1977 1978 1979 Time (months)
- 35 O
Ramsey Creek Area Oto 10 m depths 30 - "
- 10m to bottom : :
_ 25 - P g 20 -
- 3
}
i I) - a 5 3 \ i
\ l '
10 - *
'f 5
0-1975 1976 1977 1978 1979 figure 2-49. Time (months) Volume-weighted temperatures for the Upper Lake Main Channel g Area and the Ramsey Creek Area of Lake Norman during the study period. 101
EE e e e 30 Values for a month are from left to right from:
- 1) Reference Area l
- 2) Lower Lake Main Channel 25 3) Upper Lake Main Channel
- 4) Ramsey Creek Area 20 - I
^
v i E 3 15 - l 1 E N E 5 l 10 - I I J 5 0 Jul Aug Sep Oct flov Dec Feb Mar Apr May Jun Jan Time (months) Figure 2-50. Volume-weighted temperature (0.3 to 10 m depth) comparisons for four zones in Lake fiorman in 1975.
30-Values for a month are from left to right from:
- 1) Reference Area ,
- 2) Lower Lake Main Channel 25
- 3) Upper Lake Main Channel
- 4) Ramsey Creek Area 20 -
^
u
? I 3 15 -
c ~ $ S 5 10 - 5 0 Oct Ncv Dec Jan Feb Mar Apr May Jun Jul Aug Sep Time (months) Figure 2-51. Volume-weighted temperature (0.3 to 10 m depth) comparisons for four zones in Lake Norman in 1976. O O O
O O iO . 30 Val'ues for'a month are from left to right from:
- 1) Reference Area i
2)' Lower Lake Main Channel 5
- 3) Upper Lake Main Channel
- 4) Ramsey Creek Area .
20 - G , E l 3 15 - . 2 - ! - E- : 2 g s s i L 10 - t i i 4 5 - s t 0 Jan Feb Mar Apr May Jun- Jul Aug Sep Oct flov Dec Time (months) Figure 2-52. Volume-weighted temperature (0.3 to 10 m depth) comparisons for four zones in Lake f.orman in 1977.
- 4 >
?
30 Values for a month are from left to right from:
- 1) Reference Area
- 2) Lower Lake Main Channel 25
- 3) Upper Lake Main Channel
- 4) Ramsey Creek Area 1
20 - G E 3 15 - 2 E; E m 5 s 10 -
,I 5 -
0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct tiov Dec Time (months) Figure 2-53. Volume-weighted temperature (0.3 to 10 m depth) comparisons for four zones in Lake florman in 1978. O O O e
O O- O. 8 30 Values for.a month'are from.left to right.from:
- 1) Reference Area-
- 2) Lower Lake Main Channel 5
- 3) Upper Lake Main Channel
- 4) Ramsey Creek' Area
, 20 - C O I I E ,. 3m 15 -
~ 5 S e 3
10 - w 5 - I O Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Time (months) Figure 2-54. Volume-weighted temperature (0.3 to 10 m depth) comparisons for four zones in Lake Norman in 1979.
l I i CHAPTER 3. WATER CHEMISTRY O a. C. PERxiNS AND T. t. WniSENAN1 PAGE INIR000CTION . . . . . ... .......... . . . . . 108 BACKGROUND . . . . . . . . ...............,. 108 OBJECTIVES . . . . . . . ..... . . . . . . . . . . . . . 108 MATERIALS AND METHODS . . ........ . . . . . . . . . . . 109 SAMPLING LOCATIONS AND FREQUENCY . . . . . . . . . . . . . . 109 FIELD PROCEDURES . . . .................. . . . 109 LABORATORY PROCEDURES . ......... . . . . . . . . . 109 DATA ANALYSES . . .. ...... . . . . . . . . . . . 110 RESULTS AND DISCUSSION . . . . ... . . . . . . . . . 111 GENERAL WATER QUALITY VARIABLES ........ . . . . . . . . 111 DISSOLVED OXYGEN . . . ................ . . 111 ALKALINITY AND pH . .. ........ . . . . . . . . . 112 TURBIDITY , . . . . . ........ . . . . . 132 MINERAL COMPOSITION .. ................ . . . 113 IRON AND MANGANESE . , .. ... . . . . . . . . . . 113 SPECIF?C CONDUCTANCE . . . . . . . . . . . . . . . . . . 114 AQUATIC NUTRIENTS . . . ... . . . . . . . . . 114 INORGANIC NITROGEN . .... ...... . . . . .. . . 114 PHOSPHORUS . . . . . . . . .... . . . . . . . . . 115 SILICA . . . . . . . . .,.. ..... . . . . . . . . 115 TRACE METALS . . . .. ..... . . . . . . . . 115
SUMMARY
. . . . . .. .. . . . . . . . . . . 116 LITERATURE CITED . . . .. ... . . . . . . . . . . 118 O
107
INTRODUCTION BACKGROUND The physicochemical characteristics of Lake Norman have been documented by previous water quality studies (Bowling and Flowe 1977; Duke Power Company 1976; Jensen et al.1974; Weiss et al.1975). These studies showed similar trends for the variables measur ed. The general water quality of Lake Norman reflected the lithology of the Catawba River drainage area. Waters associated with the metamorphic rock types described in Chapter 1 are typically low in dissolved solids. The mineral composition of Lake Norman was dominated by sodium, calcium, and bicarbonate. Total hardness concentrations ranged from 8 to 15 mg-CACO 3 t-1 (Duke Power Company 1976). Alkalinity values in Lake Norman were generally less than 25 mg-CACO 3 t-1 These hardness and alkalinity values indicate that Lake Norman is a soft water lake. Fluctuations in pH were connon due to the low buffering capacity of Lake Norman water with pH values fluctuating between 6 and 8 pH units (Duke Power Company 1976; Jensen et al. 1974). Because oxygen solubility increases with decreasing temperatures, the highest dissolved oxygen (D0) levels were observed during the winter moaths. The DO concentrations were generally at or above 7.0 mg t-1 throughout the weter column from December through April (Duke Power Conpany 1976; Jensen et al. 1974; Weiss et al. 1975). Hypolimnetic oxygen depletion began in June with anoxic conditions being observed in the bottom waters from August through October. Inorganic nitrogen levels were dominated from December through May by nitrate plus nitrite while annonia was the dominant nitrogen form from August through g November in the bottom waters. Nitrate plus nitrite concentrations ranged from less than 0.010 to 0.50 mg-N t-1 with an average annual mean of approximately 0.25 mg-N t-1 Ammonia concentrations for Lake Norman waters ranged from less than 0.010 to 3.1 mg-ft t-1 with an average annual mean of approximately 0.20 mg-N t-1 (Duke Power Company 1976; Jensen et al. 1974). Orthophosphate concentrations for Lake Norman waters were around 0.005 mg-P t-1 Silica, a nutrient source for dictoms, ranged frou 5 to 8 mg-Si t-1 before impoundment (1939 to 1961). Following impoundment of Lake Norman silica values ranged from 2 to 5 mg-Si t-1 in the surface waters. The decrease in silita concentrations was attributed, in part, to diatom uptake (Duke Power Company 1976). 00]ECyIvrs The objectives of this study were to:
- 1) document the physicochemical characteristics of Lake Norman with regard to dissolved oxygen, pH, alkalinity, turbidity, conductivity, minerals, nutrients, and trace metals,
- 2) describe the vertical and horirntal dist.ribution of the physicochemical variables, and
- 3) examine seasonal trends of the physicochemical variables. 9 103 f
J
1 MATERIALS AND METHODS l SAMPLING LOCATIONS AND FREQUENCY Twenty eight locations were sampled on Lake Norman from 1974 through 1980 (Table 3-1; Fig.1-10 and 1-11). This report summarizes data collected from January 1975 through December 1979 at the following locations: 1.0, 1.2, 2.0, 3.0, 4.0, 4. 5, 5.0, 6.0, 7. 5, 8.0, 11.0, 13.0, 14.0, 15.0, and 16.0. All samples were collected monthly during this period except for trace metals which were generally collected on a quarterly basis. During 1975 and 1976, sampling was performed the last week of the month. Beginning in 1977, sampling was performed the first week of the month. This sampling change caused an apparent shift of approximately one month in the temporal trends of some variables.
, FIELD PROCEDURES Profile measurements of 00, pH, and specific conductance were obtained in-situ using a Hydrolab Surveyor Model 6D water quality analyzer, in September 1976, s oxidation-reduction potential was added to the in-situ measurements. Profiles were taken at one meter intervals from surface (0.3 m) to 1 m off the bottom, at all locations, throughout the five year study period (1975 through 1979). The manufacturer's recommended calibration procedures were perfonned before each monthly sampling. Methods for the measurements of these variables arc presented in Table 3-2.
LAB 0PsATORY PROCEDURE 5 A diaphram pump was used to collect the samples analyzed in the laboratory. Beginning in June 1977, duplicate samples were collected near the surface end ' bottom of all locations. Prior to July 1978 profile samples were collected at all locations for nutrient and mineral analyses. Examination of past data indicated that more locations than necessary were being sampled to adequately characterize the water column. Beginning in July 1978 the number of locations profiled were decreased to the following: 2.0, 3.0, 5.0, 8.0, and 15.0. Samples for nutrient determination were collected in acid washed linear poly-ethlene bottles, stored on ice, and returned to the laboratory. Samples for metal analyses except mercury, were collected in acid washed linear polyethylene bottles containing 0.5% HNO3 as a preservative. Mercury samples were preserved with 1% HNO 3, The analytical methods, references, and preservatives for each variable are listed in Table 3-2. All analytical methods were approved by the USEPA (1974). New techniques for various analyses were employed to lower the analytical detection limits and increase laboratory efficiency. The method changes are documented in Table 3-2. The limit of determination (Currie 1968) and detection limits, which were determined on the majority of variables, are also documented in Table 3-2. The precision and accuracy of the data were affirmed in accordance with the procedures outlined by the USEPA (1972). O 109
DATA ANALYSES All physicochemical data collected from 1974 through 1980 is given in Appendices 3.1 through 3.34 This chapter will discuss data collected from 1975 through 1979. In summarizing the large amount of data collected from 1975 through 1979 the following locations were grouped into specific areas: Ramsey Creek area (Locations 3.0, 4.5, and 5.0), Lower Main Channel area (Locations 1.0, 2.0, and 7.5), Reference area (Locations 8.0 and 11.0), Upper Main Channel area (Locations 13.0 and 15.0). The following areas consist of only one location: McGuire Intake area (Location 1.2), McGuire Discharge area (Location 4.0) and Catawba River area (Location 16.0). These groupings were based primarily on the geographic areas of each location. However, the variability (standard deviation, maximum, mini-mum, median) of each location within an area also aided in grouping the locations. Unless defined differently in the remainder of the chapter, " surface waters" refers to depths from 0.3 to 2 m and " bottom waters" to depths equal to or greater than 20 m. In discussing seasonal variability the following monthly divisions were made: winter (December through February), spring (March through May), summer (June through August), and fall (September through November). Bicarbonate values were calculated from alkalinity values using Ham's (1970) factor. Carbon dioxide concentrations were derived from Stumm and Morgan (1970) as follows: pK3 r [6.572-0.01 (TEMP)] 2,04 x 104 30ND [(1+(5.73 x 10-3 /C0!fD))] pK2 = [10.614-0.012 (TEMP)] 8.18 x 10-3 /COND g [(1+(5.73 X 10-3 iCOND))] l [H] = 10-pH K3 = 10-PK I K2 = 10-Pk 1 mmole C0 2-1 [H]
= ALKALINITY (
24 ) K(Igf) 3 l l mg CO 2 t-1 = (44) mmole CO 2 t-1 Where TEMP = C ! COND = unho cm-1 ALKALINITY = meq rl Water chemistry and meteorological data were sub.iected to Pearson's correlation analysis (Helwig and Council 1970) . sly results with p _ 0.05 were considered statistically significant. Standard deviation is denoted by ' " O 110
;OL RESul1S AND Discussion GENERAL-WATER QUALITY VARIABLES DISSOLVED OXYGEN Previous aquatic chemistry studies performed on Lake Norman reported D0 trends similar to those observed during this study (Table 3-3). Dissolved oxygen concentrations-in Lake Norman _followed seasonal natterns typical of other Piedmont Carolina reservoirs (Bowling and Flowe 1977; Katnik et al.1974). Lowest D0 values generally occurred from July through October and the highest concentrations occurred in February and March (Figure 3-1). Deplet 9n of D0 concentrations in Lake Norman begen in late April with reoxygenation of the water column occurring by late November. The mean 00 concentration in Lake Norman was 7.7 mg.t-1 (s =
3.6) with-annual means ranging from 7.3 mg t-1 (s = 3.3) in 1975 to 7.8 mg t-1 (s = 3.5) in 1978. Dissolved oxygen concentrations in Lake Norman generally 4 ranged from 8.' 0 to 12.0 mg t-1 from December through April (Fig. 3-2 through 3-5) indicating thorough vertical mixing throughout this period. . As thermal stratification developed in May, a redaction in D0 concentrations was observed in the bottom waters. Thermal density gradients forming in the water column limited-mixing of epilimnetic and hypolimnetic waters. Thus, as biological respiration and chemical oxidation continued, D0 concentrations in the bottom waters steadily-decreased throughout the summer and early fall. Fron July'through es October, bottom D0 values were less than 5.0 mg t 1 (Fig. 3-2 through 3-5). By November, destratification was well underway and only the bottom 1 to 3 m of the_ deepest locations exhibited low 00 concentrations (Fig. 3-6). From 1975 through 1979 all locations except 13.0 and 14.0 exhibited 00 concen-trations -above 5.0 mg.t-1 at the surface. From July through October, Location 14.0 exhibited surface D0 concentrations generally at or-below 5.0 mg t-3 The hypolimnetic water used for condenser cooling at Marsnall Steam Station was the reason for the low D0 at Location 14.0. Only during September 1977 and October 1978 were surface D0 values at Location 13.0 below' 5.0 mg t-1 A concentration of 4.1 mg t-1 and 4.2 mgtt-1 was recorded in September 1977 and October 1978, respectively, at Location 13.0. In October 1978, as a result of Marshall Steam Station using hypolimnetic water for condenser cooling, the D0 at Locations 13.0 and 14.0 was between 3.0 and 4.5 mg t-1 from 0.3 through 8 meters (Fig. 3-7). The. decrease in D0 in this general area of Lake Norman was also documented by ,' Jensen et al . (1974)._ Dissolved oxygen concentrations at Locations 11.0 and 15.0 may also have been affected in October of 1978 by Marshall's operation (Fig. 3-7). However, fish and other oxygen requiring organisms were probably not adversely
'affected by the oxygen levels observed at these locations.
l - L Mean D0 values for July through September were calculated over the study period ' to give an overall indication of 00 trends in the vicinity of McGuire Nuclear Station. Locations in the vicinity of McGuire exhibited D0 concentrations above
- 5. 0 mg t-1 from surface (0.3 meters) through 5.0 meters (Fig. 3-8 through' 3-10).
Due to biological respiration and chemical oxidation depleting D0, the concen-trations at 10 m were at or below 5.0 mg.t-1 from July through August O <r49 . a-c "rev9h 3 '.0). 111
Oxygen saturation values in the surface waters of Lake Norman ranged from 50 to 140% with 42% ci the' values being greater than 100I. From Fiarch through August the surface waters exhibited oxygen saturation values generally around 100% (Fig. 3-11). Bottom water oxygen saturation values ranged from 0 to 110% with less g than 1% of the values being above 100%. Maximum surface to bottom differences occurred from July through October (Fig. 3-11) and reflected natural processes of photosynthesis and reaeration in the surface waters and respiration and oxidation in the bottom waters. ALKALINITY AND pH Alkalinity values, an indicator of bicarbonate concentrations (Wetzel 1975), indicate that Lake Norman is a soft water lake. Annual neans ranaed from 11 mg-CACO 3 t-1 (s = 2) in 1978 to 14 mg-CACO 3 t-1 (s = 5) in 1975. These alkalinity values were similar to those reported in Lake Wylie (Katnik et al.1974) and in Lakes Keowee and Hartwell (Duke Power Company 1977 The range of alkalinity values in Lake Norman was from 5 to 40 mg-CaC0y t-l.overall mean of 12 with an mg-CaC0r t-1 (s = 3.0). Little difference was >bserved between surface and bottom alkalinity values (Fig. 3-12) except during the fall, when bottom alkalinities increased. This increase in alkalinity concertrations was due to carbon dioxide reacting with carbonates of calcium and magnesium to form bicarbonate (Hutchinson 1957). This increase in bicarbonate concentration during the fall generally kept the bottom pH from f alling below 6.0 (Fig. 3-12). Due primarily to the geology or the area, pH values in Lake Norman waters are slightly acidic. The mean pH value from 1975 through 1979 was 6.7 (s = 0.5), with annual means ranging from 6.5 (s = 0.6) in 1975 to 6.7 (s = 0.4) in 1977. Highest surface pH values generally occurred during the summer when photo- a synthetic activity was greatest. This trend was observed in Lake Wylie, (Katnik W et al.1974) and in Lakes Keowee and Hartwell (Duke Power Company 1977). Surface to bottom pH differences were greatest durir.g the summer (Figure 3-12). This was attributed to in increase in carbon dioxide, due to biological respiration, lowering the pH of the bottom waters (Fig. 3-13). Photosynthetic activity and reaeration of the surface waters prevented any major increase in carbon dioxide in the surface waters. Alkalinity and pd values exhibited small spatial differences (Tables 3-4 and 3-5). Previous aquatic studies on Lake Norman reportec similar alkalinity and pH values (Table 3-3). ! TURBIDITY i Turbidity, an indicator of suspended particulate matter, ranged from 1 to 74 NTU l in Lake Norman. The overall mean turbidity value was 17 NTU (s = 14), with annual means ranging from 12 NTU (s = 12) in 1977 to 21 NTU (s = 15) in 1975. The higher surface turbidity values were generally obstrved during the winter and early spring, while July through October were periods of low surface turbidities (Fig. 3-14). Historical records (1941-1970) indicated that the mean wind speed was lowest from July through October ranging from 10.6 to 11.4 K H-1 Highest wind speed values were recorded from January through April (12.8 to 15.2 K H~l) (NOAA 1979). Surface turbidity values correlated with wind speed (r = 0.51) indicating that wind speed directly influences turbidity values in Lake Norman. Due to inputs from the Catawba River and Marshall Steam Station's use of hypolimnetic water, highest turbidity values occurred in the Upper Main Channel 112
~
. l area l(Fig. 3-15 and Tables 3-4 and 3-5). - Turbidity values decreased with
' Q;- distance from the Upper Main Channel area uue to settling of particulate matter (Fig. 3-16). MINERAL COMPOSITION Mineral composition and variability in Lake Norman were attributed to geological processes involving the physical and chemical weathering of soils and bedrock outcroppings. Feldspars (orthoclase [K(AlSi 380 )], plagioclase [Ca,Na(AlSi 380 )]), quartz (SiO2), and olivine [M92, Fe2(Si0u)] are the most commonly occurring minerals. Surface waters associated with such metamorphoric rock types are generally characterized by low solute concentrations (Hem 1970). Lake Norman waters, exhibiting a hardness value of approximately 10 mg-Caco r t-1 exempli fied 3 soft water lake. Bicarbonate was the major ion in Lake Norman. Sodium, chloride, calcium, magnesium. -silica and potassium were also abundant constituents in Lake Norman waters-(Fig. 3-17). Minor constituents included aluminum, iron, and manganese. Except for aluminum, no substantial yearly variability was observed in the concen-trations of the various minerals. Aluminum concentrations appeared to be associated with increases in turbidity. However, no statistical correlation was observed between aluminum and turbidity, probably due to aluminum being analyzed from quarterly composites.
- IRCN AND MANGANESE O Both-iron and manganese possess similar chemical properties and are important as micronutrients of freshwater flora and fauna (Wetzel 1975) and as indicators of oxidation-reduction processes. Although the concentrations of these micronutrients were generally low throughout most of the year, the concentration of iron and manganese during the fall are higher in the bottom waters.
The mean. iron concentration for the study period 1975 through 1979 was 0.8 mg t-1 (s = 0.9), with annual means ranging from 0.5 mg t-1 (s = 0.7) in 1976 to 1.1 mg rl (s = 0.9) in 1975. Iron values in the bottom waters were consistently-high in the fall of each year (Fig. 3-18) and were associated with anoxic conditions. . The anoxic conditions were conducive for iron ions to diffuse from the sediment into the water column (Wetzel 1975). Due to increased runoff associated with rainfall, surface iron values were highest during the winter and spring (Fig. 3-18). The . correlation of iron with turbidity (r = 0.72) in the surface waters
-indicates the major source of iron in the surface waters was from suspended partic-ulate matter.
The Piedmont soils of the Catawba River drainage basin are a major source of iron to Lake Norman. This was reflected in the higher iron concentrations L in the Upper Main Channel area of Lake Norman (Fig. 3-19, Tables 3-4 and 3-5). l' - The deep water areas of Lake Norman consisting of the-Upper Main Channel area, the Reference area, and the Lower Main Channel area exhibited two annual peaks in iron concentrations (Fig. 3-19). Iron concentrations peaked during spring due to increased suspended particulate matter in surface waters. A fall peak was observed due to diffusion of iron from the sediments into the bottom waters. l
=
Generally, the other lake areas showed high iron concentrations only in the [ spring (Fig. 19). t l 113
- - . _ _ _ __ _ - _. .._~ ._- _ - _ - _ ._-_
The mean manganese concentration for the study period 1975 throuqb 1979 was 0.21 mg t-1 (s = 0.54 ith annual means ranging from 0.16 mg t-1 (s = 0.53) a in 1978 to 0.27 mg.t-{ w(s = 0.67) in 1977.Surface concentrations of man- W ganese were low throughout the study period (Fig. 3-18). Manejanese concen-trations were highest in the bottom waters usually from SeMuber through November (Fig. 3-18) when the reducing conditions were conducive for manganese ions to diffuse from the sediment into the water column. Manganese concentrations in the bottom waters correlated inversely with dissolved oxygen concentrations (r = -0.58) and oxidation-reduction potential (r = -0.75). Manganese concen-trations were highest during the fall in the Upper Main Channel area, Reference area, and Lower Main Channel area (Fig. 3-20). The Ramsey Creek area to a lesser degree also exhibited this increese in manganese concentration in the fall (Fig. 3-20). SPECIFIC CONDUCTANCE Specific conductance is an indicator of ionized substances in tresh water (Wetzel 1975). The mean specific conductance value for the study period 1975 through 1979 was 40 umho cm-1 (s = 6), with specific conductance values ranging from 23 to 92 umho cm-1 Surface conductivities generally ranged from 40 to 45 pmho cm-1 (Fig. 3-21) indicating uniform mixing of dissolved solids in the surface water throughout each year. Previous studies on Lake Norman reported similar conductivity values (Table 3-3). Surface to bottom differences in conductivity rere greatest during October and November (Fig. 3-21) as a result of ions (iron and manganese, etc.) diffusing from the sediment into the water column. Inputs of dissolved substances into the Catawba River (surface runoff, etc.) were indicated by the relatively high specific conductance values recorded at the Catawba River area (Fig. 3-22, Tables 3-4 and 3-5). $ AQUATIC NUTRIENTS l INORGANIC NITROCEN The mean nitrate plus nitrite concentration for the study period 1975 through 1979 was 0.26 mg-N t-1 (s = 0.14), with annual means ranging from 0.22 mg-N cl
- (s = 0.11) in 1976 to 0.34 mg-N t-1 (s = 0.17) in 1978. Maximum concentrations l of nitrate plus nitrite generally occurred in winter and spring in both the sur-face and bottom waters (Fig. 3-23). These high concentrations were associated with oxidizing conditions that prevailed in Lake Norman waters during this part of the year (Wetzel 1975). During the summer, nitrate plus nitrite concentrations decreased frcm the high spring values due to nutrient utilization (Chapter 4) l which was accompanied by increased rates of bacterial decomposition and lower l redox potential. The minimum nitrate plus nitrite concentrations occurred during l late summer and fall (Fig. 3-23) and were associated with reducing conditions in Lake Norman bottom waters. These trends were also observed in Lakes Keowee and l Hartwell (Duke Power Company 1977). Little spatial variability in nitrate plus l
nitrite was observed in the study area (Fig. 3-24, Tables 3-4 and 3-5). l l The mean ammonia concentration for the study was 0.15 mg-N cl ( s = 0.17 ) . l The highest values occurred when reducing conditions were prevalent and dissolved oxyaen concentrations were low, generally in October and November. Ammonia con-centrations in the bottom waters were inversely correlated with oxidation-reduction 114
- - -. - - --- . - ~ _ - . . - - - - ..- -.- -- - -.- -7 . potential (r = -0.62). Little spatial variability of ammonia concentrations was observed in the study area (Fig. 3-25, Tables 3-4 and 3-5). As with nitrate plus nitrite concentrations, ammonia concentrations in Lake Norman were similar to '
those reported in' previous studies performed on Lake Norman (Table 3-3). PHOSPHORUS- _ The temporal trends of total phosphorus and orthophosphate were similar. However, because 64% of orthophosphate values were less than the analytical detection. limit of 0.005 mg-P t-7, no significant corre'ation with total phosphorus concentrations was obcerved. The mean total phosphorus concen-tration for the study period 1975 through 1979 was 0.020 mg-P.t-1, with annual
- means ranging from 0.013 mg-P t-1 (s = 0.009) in 1977 to 0.028 mg-P t-1 (s =
0.016)- in 1975. Surface concentrations of total phosphorus were highest in the spring-and were associated with runoff. Concentrations of total phosphorus in the bottom waters were high in the spring and in the fall (Fig. 3-23). The higher total phosphorus concentrations in the spring corresponded to periods of increased turbidity due to rainfall. The hypolimnetic increases in total phosphorus in the. fall were due to the release of orthophosphate from the sediment during anoxic conditions (Golterman 1975). The greatest variability in total phosphorus concentrations was exhibited in the_ Catawba River area (Figure.3-26). Hi0hest total phosphorus concen-trations in Lake Norman were observed in the Upper Main Channel area (Fig. 3-26), with a gradual-decrease in total phosphorus with distance-downstream i.e.a the Upper Main Channel area. Total phosphorus concentrations exhibited
-O tremos s4 miler to et"er reservoirs <0vue eower compenx 1977; weiss et el. 197s).
with total phosphorus concentrations in Lake Norman gradually decreasing over the years (Table 3-3). Reasons for this apparent decrease in Lake Norman may be due to the adsorption of phosphorus onto suspended clay particles which' settle to the lake sediment (Golterman 1973). SILICA The mean silica concentration for the study period 1975 through 1979 was 3.6 mg-Si t-1 (s = 0.5), with annual means ranging from 3.2 mg-Si 2-1 (s = 0.5) in 1977 to 4.1. mg-Si t-1 ( s = 1. 0) - i n 1979. Highest silica concentrations were generally _ observed from 0ctober _through February (Fig. 3-27). These higher silica concentrations were probably due to a decrease in diatom productivity since- silica is removed from lake waters during the development of diatom popu-lations (Hutchinson 1957). Due to the assimilation of silica by diatoms
~
(Chapter 4) and subsequent sedimentation of diatoms (Wetzel 1975), silica con-centrations decreased during spring (Fig. 3-27) and remained low in the surface
- waters until destratification of the water column in October and November.
Silica concentrations exhibited similar spatial trends throughout the sampling l areas of-Lake Norman (Fig. 3-28, Tables 3-4 and 3-5). L L TRACE METAL.S_ i Copper, cadmium, mercury, zinc, and lead were mo.11tored to evaluate the trends
'in trace element concentrations of Lake Norman. Trace metals are those metals .O with concentret4 ens esoelli not exceeeie9 i m2 t-1 caeb4e 1974).
115 ,
*e =.-we, , , - w.-r :---r-- , - -
- e ym,- o w -- w r- .w y ---e-r v--evy-- r--
Copper concentrations ranged from 1.0 to 16 og. rl with an overall nean of 3.4 ug rl . Copper concentrations were genera!1y highest during the summer (Fig. 3-29) and were attributed to hypolimnetic increases associated with $ dissolution of copper-containing organic compounds fron the sediment (Hutchinson 1957). Cadmium concentrations ranged from 0.1 to 0.5 ug rl with an overall mcan of 0.2 ug cl. Generally, where seasonal changes were detected, highest concentrations were observed during the spring and summer (Fig. 3-29). The seasonal trends observed for copper and cadmium were typical of the trends observed in Lake Keowee (Duke Power Company 1977). Mercury concentrations ranged from less than 0.1 to 0.3 og F l. Generally, mercury conccutrations were below the analytical detection limit of 0.1 ug r . Mercury concentrations in fresh water are usually inorganically complexed and removed from the active mcecury cycle (Schindler and Alberts 1977). Mercury may be adsorbed by clays, sands, or oxides in sediment resulting in relatively low levels of mercury in water (Schindler and Alberts 1977). Zinc concentrations ranged from 1.0 to 48 ug rl with an overall mean of 11 pg cl. Where seasonal changes were detected, the highest concentrations of zinc usually occurred during the fall and/or winter (Fig. 3-29). Lead concen-trations ranging from less than 2.0 to 3.8 ug El showed little variability over the period 1977 throuch 1979.
SUMMARY
In characterizing Lake Norman waters 28 locations were sampled from 1974 through 1980. period. Thirteen of these locations were sampled monthly during the entire study Some analytical methods were changed and/or updated during the period g to obtain lower detection limits and increase 1: 'ratory efficiency. The physicochemical characteristics of Lake Norman waters reflect the lithology of the basin. Lake Norman waters were characterized by slightly acidic pH values, inw hardness, stable mineral composition, and generally low nutrient and trace metal concentrations. Little yearly variation existed in variables analyzed from 1975 throuah 1979. Previous I ake Norman studies manifested similar variation in the physicocilemical variables observed during this study. Spatial characteristics in Lake Norman indicated that variables related to runoff (turbidity, iron, ortho-phosphate, and total phosphorus) were highest in the Upper Main Channel area with concentrations decreasing with distance downstream from this area. Greater.t over-all variability was generally observed in the Catawba River area. Surface turbidity values were highest during winter and Dring with lowest values being observed during the summer and fall. Dissolved oxygen concentrations exhibited seasonal trends typical of Piedmont Carolina waters with highest values occurring from December through April and lowest concentrations from July through October. Except for Locations 13.0 and 14.0 near the Marshall Steam Station discharge, dissolved oxygen concentrations in the surface waters were above 5.0 mg 2-I throughout the year Generally, Location 14.0 exhibited surface DO concentrations at or below 5.0 mn '-! from l July through October. Only during Seotenber 1977 and October 1978 were surface ! D0 values less than 5.0 mg rl at Loca tion 13.0 Anaxic conditions existed I in the hypolimnion from August through October and were accompanied by increased l concentrations of anmonia, iron, manganese, and alkalinity. Following destrati-I fica tion in Nove,ter, all ef fects o# m ner anoxia had dissipated frcm Lale Norman 116 l l
h - waters. - Nutrient concentrations exhibited seasonal cycles typical of Piedmont Carolina waters with ni_trate plus nitrite being the dominant nitrogen species 1- in winter and springt and ammonia the_ dominant species during fall. Total '
' phosphorus concentrations: exhibited seasonal trends similar to turbidity-with -highest concentrations being observed in the winter and spring and-lowest concentration in the summer. - Cadmium, copper, lead, mercury, and zinc were monitored to assess trends in the trace metals concentration in Lake Norman - Changes in cadmium, zinc and - copper concentrations were associated with changes in oxidation-reduction - potential. Lead and mercury concentrations showed'little variability with concentrations _ generally at the analytical detection limit.
e lO O L 117
1 i LITERATURE CITED Bowling. T. J. and M. B. Flowe, 1977. Chemical characteristics of piedmont O lakes. Workshop in Aquatic Ecology in the Southeast. October 14, 1977. Augusta, GA. Currie L. A. 1968. Limits for qualitative detection and quantitative detection. Analytical Chemistry Vol 40. Duke Power Company. 1976. McGuire Nuclear Station, Units 1 dnd 2. Environmental Report, Operating License Stage. Eth rev. Volume 1. Duke Power r.empany, Charlotte, f4C. _. 1977. Water Chemistry, p. 115-192. la Duke Power Company. Oconee Nuclear Station environmental sumary report ~1971-1976. 'ialume 1. Duke Power Company, Charlotte, NC. 405 p. Golteeman, H. L. 1973. Vertical movement of phosphate in freshwater, p. 509-538. In: E. J. Griffith (ed.). Environmental nhosphorus handbook. John Wiley hnd Sons, fiY. 718 p.
. 1975. Chemistry, p. 39-80. In: B. A. Whitton (ed.). River ecology, University of California, Berkeley TA. 7ES p.
Helwig, J. T. and K. A Cou' '.,(ed.). 1979. SAS user':, guide 1979 edition. SAS Insti m'.e incorporated, Raleigh, NC. 494 p. Pam, J. D. 15 Study and interpretation of the chemical characteristics of naturai ;er. Geological Survey Water-Supply Paper 1473. V. S. Government i 'intina Offica, W?shington, DC. 363 p. Hutchinson, G. E. 1957. A treatise on limnology. Vol. 1. John Wiley and Sons, New York, NY. 1015 n. Hyd..,,ab Corporation. 1973. Instructions for operating the Hydrolab Surveyor Model 60 in-situ water quality analyzer. Austin, TX. 146 p. Jensen, L. D., D. K. Bredy, R. F. Gray, W. D. Adair, end J. J. daines. 1974. Thermal an: cter quality characteristics of Lake Norman, p. 7-119. In: L. D. Jensen (ed.). Environmental responses to thermal discharges from Marshall Steam Station, Lake Nont.ao, North Carolina. Electric Power Research Project (RP-49) Report No.11. Johns Hopkins University, Baltimore, MD. 235 p. Katnik, K. E., E. A. Hollend, and W. J. Krzyanowski. 1974. Field baseline water quality studies, p. 27-211. In: D. W. Anderson, G. W. Wadley, and B. G. Johnson (ed.). A baseline predictive environmental investigation of Lake Wylie. Industrial Bio-Test Laboratories, Northbrook, IL. 357 p. National Oceanic and Atmospheric Administration. 1979. local climatological data 1979, Cherlotte, NC. National Clinctic Center, Asheville, NC. Orion Research Incorporated, 1970. Instruction manual. Cambridge, MA. h Rubin. A. J. 1974 Aqueous-environt. ental thenistry of metais. Ann Arbor Science, Ann Arbor, M1. 390 p. 118
i i Schindler, J. E. and J. J. Alberts. 1977. Pehavior of mercury, chromium, and i I raduium in aquatic systems. V. S. Department of Commerce, Springfield, ! VA. 61 p. . i Stumm, W. and J. J, Morgsn. 1970. Aquatic chemistry. New Yort, NY. 563 p. [ t Technicon Industrial Systems. 197?. Operr. tion manual for the Technicon ) Aut0 analyzer 11 System, lechnical Publication No. TAl-0170-20. l' Tarrytorn, NY. United States Environmental Protection Agency.1972. Handbook for analytical quality control in water and wastewater laboratories. Technology Transfer, , Cincinnati, OH. ; I
. 1974. Methods of chemical analysis of water and wastes. Office !
Technology Transfer, Washington, DC. 293 p. } t Weiss, C, M., P. H. Campbell, T. P. Anderson, and S. L. Pfat.'ider. 1975. The l lower Catawba lakes: Characterization of phyto- and zoopiankton communities and their relationships to environmental factors. Department of Environ-mental Sciences and Engineering School of Palic Health, University of , North Carolina, Chapel Hill, NC. ESE Publication No. 389. 396 p. Wetzel, R. G. 1975. Limnology. W. B. Saunders, Philadelphia, PA. 743 p. O ; i t F O 119 i
, . . _ _ . _ _ . _ - . . , . . . _ . . . _ -...._.,._.-,J._m___...___._ . _ . . . . . . . _ _ _ . _ _ _ . _ _ . . _ . . _ . . . . . _ _ . . . .-.._ -.. --._ . .
l l Table 3-1. Locations sampled and types of variables analysed f rr,m 1974 ' through 1980. Locctions 1974 1975 1976 1977 1978 1979 1980 1.0 *>** *"5 *"4 ***4 ***4 *"4 *"4 1.2 0000 0000 0000 7770 ***0 ***0 ***0 2.0 ***7 ***4 ***4 ***2 ***0 ***0 ***0 3.0 ***7 ***4 ***4 ***4 ***0 ***0 ***0 l 4.0 *"7 ***4 ***4 ***2 ***4 ***4 ***4 4.5 **** ***5 ***3 ***4 ***0 ***0 ***0 5.0 ***7 *"4 ***4 *"2 *"O *"O *"O 6.0 ***7 ***4 ***4 ***2 ***0 ***0 ***0 l 7.0 ***7 0000 0000 0000 0000 0000 0000 7.5 0000 ***4 ***4 ***2 ***0 ***0 ***O 8.0 **** ***5 ***4 ***4 ***4 ***4 ***4 9.0 ***7 0000 0000 0000 0000 0000 0000 9.5 0000 0000 0000 0000 *000 *S50 ***0 10.0 ***7 ***4 0000 0000 0000 0000 0000 11.0 ***7 ***4 ***4 ***2 ***0 ***0 ***0 12.0 ***7 0000 0000 0000 0000 0000 0000 13.0 ***7 ***4 ***4 ***2 ***0 ***0 ***0 14.0 **** ***5 ***4 ***4 ***4 ***4 ***4 14.5 3222 ***4 0000 0000 0000 0000 0000 15.0 ***7 ***4 ***4 ***4 ***0 ***0 ***0 15.9 0000 0000 0000 0000 7772 5551 0000 16.0 17.0
***7 1119 ***4 0000 ***4 0000 ***4 0000 ***0 0000 ***0 0000 ***0 0000 h
17.5 6666 ***5 0000 0000 0000 0000 0000 18.0 **** ***5 0000 0000 0000 0000 0000 34.0 0000 0000 0000 0000 7770 5550 0000 50.0 0000 0000 0000 0000 7000 5000 0000 60.0 0000 0000 0000 0000 7000 5000 0000 Each digit in the four digit code represents a different group of variables sampled that year as follows: 'st digit - physical variables; 2nd digit - nutrients; 3rd digit - minerals; 4th digit - trace metals. The value of a digit represents the number of times that group of variables was sempled at l a location during that year. A number is shown even if only one of the ! variables of a group was sampled. An asterist (*) indicates a group of variables were sample; more than nine times in a year. For more detail, see Appendix 3.1 thraugh 3.34. l l O l 120 l
O O O Table 3-2. Analytical methods for chemical and physical constituents measured on Lake tioman from 1975 through 1980.
+t%d * , - r.r i_, d.- cres*m t_4 e_a -_ r.t-e..v t w--~
4 +- 4 t " r ~_.~ t - n e_i_,s_e s i e* -C,r - Alkal mity, tt f ai Electe ric titratics t s 1/1975 17/; MC 4'C 3 1-I* r4 ef 5.1 1/11*5 '/1974 0.5! FT f' 6 m 1 ^ j Aiumi v Atmic e Merrtiee/MGAl 0 2 m-1 ttonic eM erttiee/ M I 8/ ' 7 'M-10/1 ** 3 1/1775-7/19?6 4*C kvwM e Actceated rPate t 0. "M m 1.1 ' 3
- N9 m " 1 5/1'4774 7/190C Aut'matedsal}cytet*/ f/1976 4/i?'T nitrepruss1d*
0 7 7 ,c 1 Atoraic shorptica/%At t/1975 12/f**? f.5? t'"O 3 O 11 p1 1 , Ca t= i p OM e -1" C R ~; l'; Calc b.n 8tmic ebsorttica/CA I 1/1975-12/IUS 0.55 F*:"y Seccific ie" efectree ? t/1975 6/1975 **r f' 3 % 1'g C M oei de 7/1276-17/1 M C 7 m .1'3 Automated ferrigeridc 3 C.5* N y C6M T A W-f
- thrc%,4 Atomic absetrtitm/FGil 1/i 975 17/1m 9 gfg 1*
C eat %c t a- * .
- m i f i c Te==pe-sto r e cons ** sated ni ckel j p 973,y p 3,q g ,,
+1ettrode 9.5* W 3 C 7 ei , 1.0 &c f ^
Attw"ic absception/w-# 1/1975-i?/I *7 Cert er 0.1 e -l GI1-f /T975 72/10*O C. 5Y MS 3 Irr*. disselved Ator ic ahorrtice/CA I s I/f975 12/t:53^ C . 5*PC 3 01*01 " 7 m- I' Atecic e55certiee/DM N re Irra, total 1/1W5-12/1'50 0 5* W 3 7 e ; L7 a ' z teed At m ic atserttion/ MA' , , 1/1575 12/10#5 0.5" W 3 9 m7 re t ,'* C . C ? n~t F W esium Atv ic absorttic-/DC ?%N ~, 1/1975-17/i W 0.55 W 3 0. "?
- t - '"
w a rrganese Attr-ic ahorrtica/CA i Flar*iess at ric s M orrtteal 1/1975-12/1 W I CT VO 3 0.f MI^ *? , Wrc.urr. retal U.5% W y M M ', E 9 7 f -' Atoric absorptica/wGA? T/f"75 17/I P f.i r 6 e t 4*; 0. tv5 -m 41 a . nm *: . ' ' Automted cavir wdactical 1/1975 YE/1'rm
*itrate * *6itrite Or t evute Aut mated ascerbic acid j g yg gg.gg g.c g p*5 @ .1'I W W I'1 eekution cSta-ide g ,, g ,7 n, 0, f da t ion- redx t
- en retet ial Siirer-sil eiectrode,ver 9p pg a 7p.pg ated 1/1FS-1?D C In- it T & T'1
- Ou ve**. di cci ved Tear *rsture cea:*]l l
l refer-cenc*ic cel Te-Ter 4ture ctw ?*asa ted gla ss 7 p pg,77p q pg t z, r4 el ec t ro fe - tresglf3tr d'q*stica f cil ee,*d 0 'r-4 ,w. .p t 'I O rg , r . E% W %s.tetti 1/l c 75-12 / W'* 4 ", . tv &Jt',P59ted 4 W0rbic acid rew t w I/lF5 I'/I'UTO 0.55
- C3 p 1 ^
. gy .t'I-Ictasst r Etce'ic eM erttica/El l -1 1/1F54;/1w et 7 7 m _p .1 r ? m 9.T' Auto-atad m lyesificate t I 5'Iira 1/ tWS 5 D P" Gy N y c3 .1 l c g ,,., y 5 dM Att**ic absorrtica/00 !
I 1/1 F 5 12/t r0 la-tttu 0.1 %* 7 "-: ** a t ure Tr c~ nt er t ur-crie t ect i
,w 6sen tr $ adit,1 1/1r$-Sf t?'*t a-t 1 r".
f tua'di!r ' erN:1~+rtr ic tc$idity' f / t rS -12/1 W vintmn u Pq a.7 m t-' ,
.,t W n-c m~ rt t-e T / 7 7 t ?m9 y a.t'l q. *
- O* ' t ets.~ rtim /"'
ca tie ,e u< t
- t e . tes trei 4~.te r** t sam ttfvity is eiv<-
- W .t on i 11H t ay l kit of metr w e4* ion w r* }cf' .s t, ca tar -Med 3 $ 9 ', ' s". I17 '* I 9a bf SCMb T3M fImI4; r eep , }}']
**t I $I'I $ f] db" I { ?*** *e ' f t e r** * *'t j
- l. - - - - -
Table 3-3. Mean (X) and range (R) ccicentradcas of water quality variables sampled during th'ee studies. performed on Lake Norman. All variables were measured in the vicinity of Location 2.0 and/or 7'.5 near the surface (0-2 m). g LAKE NORMAN STUDIES Jensen et al. Duke Power Co. Duke Power Co. 1 974 1976 1981 VARIABLES (7/68-11/71) (1/75-11/79) (8/73-6/ 74J ! Dissolved Oxygen (mg 1 1) Y 9.1 9.0 9.4 R 7.4-11.5 6.7-12.1 5.7-12.8 pH Y 6.8 7.1 7.1 R 6.2-7.3 6.7-8.0 5.8-8.6 Alkalinity (mg-CACO 1 l) Y 12 11 11 3 R 5-18 9-14 5-28 t Turbidity'a Y 5 12 11 R 3-12 5-24 2-60 Conductivity (u mho cm.1) Y 38 40 40 R 27-60 29-49 28-49 0.16 0.19 0.22 Nitrate-Nitrite-N (mr-N 1 1) YR 0.010-0.43 0.020-0.34 0.021-0.56 Anmonia-N (mg-N 1 1)a Y 0.11 0.035 0.13 R 0.01-0.27 0.014-0.15 0.006-0.82 h Total Phosphorus (mg-P l~ )a Y 0.053 0.022 0.013 R <0.053-0.086 0.010-0.031 <0.005-0.051 Silica (mg-Si 1 1) 1 3.1 4.3 3.4 R 2.0-4.2 3,5-5.7 2.0-4.8 Iron (mg*1 1) Y 0.7 0.4 0.5 R 0.2-2.3 0.02-1.1 0.1-2.4
- Jackson Turbidity Units were reported in the Jensen (1974) and Duke Power Company (1976) studies. Nephelometric Turbidity Units were reported in the Duke Power Company (1981) study.
a Analytical methodologies may have cor.tributed to observed differences in these variables. Percent transmittance was converted to JTU values in the Jensen (1974) study while a turbidimeter was used in the Duke Power Company (1976,1981) studies. Amonia concentrations were determined colorimetrically in the Jensen et al. (1974) and Duke Power Company (1981) studies while a gas diffusion electrode was used in the Duke Power Company (1976) study. The analytical detection limit for total phosphorus was 0.053 ma-P 1-1 in the Jensen et al. 1974 study. O 122
- - )
i a l I i i ,
- i. i 1
!' i i a , a
- O! !
m *
' * -". y, i
2 N ,. + $ 1 9o o. s.. C, ', 4
% -t. - ~. R e o o ~o oa ,. n . - a i
- . " ". e. . ~. ~. .- *. f.2.E'+a '
..,. ..e. -,8. -
i n 0 8f 4 so dee~~ 4 4
- d- d e ar ag a . 6 - c. - a e d-S 6 6 6 ~ d a 4 9 L
- l T
1- W 'd v d a d + d, . 3 i 1- t , s .
'8 i 4
- M3 e * ** e 8
. .s. ~- ~ o ,-f *,, e %* - .Ea~ , i a * ,.~ - -
m . g . =g . g a =g . g . g . .,g..sa , r M a[
.s j g e n o .
m g , . ) w e.
- e. u - o. 8 .. -
- a .
. 9 . . ** d de !
4 e a o o o o > e ' '
? .o . o o :
m *
- l-
, * $ G E , o , e C ;
I, U , d ~$. ~.'J 3 ;': g6. g* .$* 36 !
.t g.g. m e g . g.g,g. 3 a , *' 4g.g . *' ". > J)
EuJ 9 = * ~ e e R $ "a **8 9b ~ o -.o
. *a g4 n
i 72 ,$ uw
$
- d Y Y j v L-om I- . +
5 u g ~ y >, 1'I d' ? .m. o m .e n a n a $ a 4 N. * ~ M N * . a
- i. e .o. a. m -ma i.
i mo a I*s
- y. 9- m N. e .
. 2 . 3 . N . . .
o e
- o. .
- o o
. a.
m
.. * ..4
- ws oM "f ; *" * "* '*
C g. 9
*g g "" o
- u 3
o o t
.i.l o a o- u* ;
, -- o > cr: .c
-s ., a.
e n m e 3 r, 4
- o .. e a' 9 m k, m .m i.
9.- o, e ce f
- i g.g.-
L-{; ~ m
. = ". *. .o '. ?:
e
~ g g.g * ". g. $ *gig. s *' .g.4.g. ." . *. N"8* *9
- g.-" ,
mc N. A ;s B
$ .$ $ Wo 7t* . .: 6 e " .O 6 6 6, a, Y tCNw" i *y . . m o -f N .
- 3 8 u . ~ . R
' 87. m 5
- N o.
'-i 2 q a * ~
o SiEI o a a ikIa *i"Y*i~i@Y m o u - o , s-2 s-a- ~ o -
. n *o s $.
- sc.
}. o o.
e C .o- o -o os os o-my a WC
- E> = ~
a $$ - e u ~* ? O ? N* **
- R ? E Gm W* o-I-
j.e=~~ a
~ .f .
U" gg.g.u. '
- a. e g g.
C* i Nm_ fg py{ 4 s. j . - . - . eu ei u
.ee g#ss'.' g *.gg$ ~ . in -
o . W* 6u%,
~ - o a 6gl .
e .-
-- - o o o e @ .C U Di "a"K " * " " 8" # """# " * * * * * " " " " " * " " . , @3 n b 2 .c i- i . .
i- . ,. WW - m $ 7 4 I' . ?
- o. -n E'
- I'
. - - . , ., 2 ,- - .-
I e h - .e .* -
;: _ E 0_ y3 *- { 3 T i m- 6 g - -
j
- G $ $ ,,' ~
e T
- I s; 9 A .- 5 - 3 -
r 1. ()
.o i $' '; I Y i . .x g C ~ -
- "1 1 .
=Z - -
H c. E = .- o , e. .z e - o f f i t O , l I 123 1
l 1 1 ll l e b a t 4 n 2 7 7 i k 5 2 6 9 1d hd gU e e e r C a e 31 1 2 57 4 7 2 2 00 3 7 6 9 0 50 70 0 0 1 0 4 7 2 6 02 8 9 2 5 6 3 1 - 0 - 0 - 54 04 05 ul yr - 1 1 3 oc rn eA 6 2 1 8 - 0 001 05 05 3 - 1 - 3 - 1
- 0 s 2 4 1 0 0 hi m 0 4 6 4 4 0 0 0 0 1
t a 3 2 1 6 0 t 2 . R 0 5 0 0 0 0 2 0 0 0 5o < < 1 7 n 9 1 e r me ow 1 r fp a 7 5 5 e ne 6 0 0 1 ir 2 . de aA 4 0 2 ed M 3 1 0 0 8 5 5 6 4 4 2 0 5 1 5 l ps l 11 57 3 7 8 3 0 4 4 7 mr re 1 6
- 0. 5 7 0 1 6 ae en wn 6 3 2 1 - 05 6 0. 5 - - -
1 5 - st oa 0 5 1 6 2 4 0 00 00 00 3 0 3 1 01 e Lh 8 0 0 0 0 5 1 0 6 0 sm C 0 5 2 0 0 0 0 2 0 0 0 e < < 1 l0 b2 a in ra ah a 3 i vt e 1 4 4 5 r 7 9 0 0 ds a 2 es 4 0 0 50 00 8 3 5 6 3 t e e 2 9 4 2 0 7 0 2 cl c 71 57 2 7 9 3 - 0 - 0 84 e n 1 - - 74 34 95 1 1 5 l s e - ea r 5 31 34 - 08 06 0 050 05 03 - 0 1 - 2 - 1 - 0 se e 0 4 7 3 9 0 0 0 0 1 r f 2 5 1 1 6 0 fA e 0 5 0 0 0 0 o R 2 0 1 0 0
)
Ra ( e sa r 9 e 6 gy a 5 3 4 nd ne i r 7 0 0 1 8 . 9 au aA 6 0 1 70 40 3 5 rt M 1 0 4 2 0 0 8 6 s 3 83 l 1 1 57 4 7 8 2 0 - d ne re en 6' 2 3 - 5 00 0kh - 3 - 75 95 1 1
- 0 4
ah t pn pa 0 4 1 9 3 4 4 8 0 0 3 0 C '0 0 0 1 1 1
) Uh 0 8 2 1 6 0 2
- f. n C 0 5 0 0 0 0 2 0 1 0 0 (i
ss nn ao ei mta 1R R 1R iR 1R R iR iR iR IR IR iR iR iR
) ) c mo l ) )
1
)
1 ) - 0 1 1 ) - 1 1 2s i - - 1 N - -
>l 1 I - - P
( o . ' m 9 1 i g 3 c- m ) P g mr oa m O (- l - n t v ( C o g ( ) ) ) a h N ,i o t n C m - - ( s 1
- )
i 1 ot e - ) u e f i u 1 1 1 Ba g g U 1 y ( t g
- e r -
m T i t o i ) 1 g g x ( i y r m a h S O ( t t ( h p - 1 g ( m ( n y .t i i p s g 5 d t y v J ,
,n s o 1
m 3
- e v
i t .i r. - o h ( m g
. ( m u s e
n i t e a h P n n .i e e l i d c t i p a ( u s n o l i u a n o l c e a l s a b d r o h a n
.i b s r n .i c n g ,mOr k
e t t t o g a i H l u o i o l l n T D P A T C i , i r a a a
T S I C M P C>
O ,, c)
# o 1 e
N 9 s' N g < a e, c C.)
\ i 3 p <a j . .
2i 4 ,e < , e R
*t p l '
2' n : 7 C1 g j, , a _es < :
- f. a Q As -
, [ .C ~4D D G 4s 1 <
s. W N1,g', o g to tw O e=* [
,e' n i l f 4 <
n t u
,f sY',c' n %
9,2 <
< c 1 S w ~
o a -
~'% m 8 C)
R.' W f o d N je
/
- 4 9
' C ,a, < n : .e <nN ag}~, -a O 4$ # 4Y w
s e
.O p
w fr3 L y da C *
, y G) @
UN NN ~ $ C C C m ON
< U d e % ', n # C C ex C) see 4 % Cn M f ,
a >i M Y * *= dM, '
- OG a
([ ,& W w a-R n > m O O .ue
' W VT .N g % ,
o @C
, .e r=
I g " e
< C Cl d , e ec ,C!
1 < y.glsf - n, < '& N i * - t i a i e n e i 1 l t 9 9 C. C. 9 ~' L E 1' ? "
- 5 m
,k' fN E () hh4{h5k$ k O ,
l 125
j ll' , ll ll)l1!l! lI I. c e 9 D g 7 9 1 h T v o n N u o 7 r h t t 7 t c 7 O 9 1 1 n
. o t
p r
- I f e
f5 S 0 0 2 g
. n t
u i o A t 8 a i c 1 o y L r l u t J s a 5 m h o t n
)
t 1 t o - o e M e 1 B i n - u g J 7 1
.I
( n 8 s y e i a u M l 0 a 1 v n
. e r g r p y A x 2
1 o d
. e .
r v9 i a l 7 M o9 s1 s i h d g b u i e no F ar 0 eh 1 Mt e n 2 a - J 3 0 0 0 e 1 2 M r u g i 1 rE F e n
- l l ll Illl I l l l l lll
, ! it t : l 1 ? ,
O 9 7
. 9 -
c 1 e C h e 9 g - v u o v r o h N t 8 7 7 9
. 1 t
c m O 7 o r 5 . f t 0
, p 5 5 1
5 n o 8 0 . i g t g u a a A c o 1 L t y a y l u ) J s m h I 5[ o t n t o 1 O t - o M g B , a n J ( 7 g s
- e u
y l g M a a v 0 1 n e g r y 3 p x A o 2 \ 1 d e
. v r l 1
a o P s s i d t b e n F a 1 0 *e t n 3 i a - J 3 D 0 0 e _ 1 2 r u g i 5oO F O M
OT 5. O Ch w g th a> n n eJ e- f O "
~ - ..
W C) (t e o e p ... s ee8 (7 cc
. U ~
3 5
~% 90 m
_ _ c O m _ g r G) 9 3
% w f0 L .
M O
~
S $
'~ O e
g' .O
- p r
__~ g
.e.
4 e e L c. m 3 7 CIb i '"" 1 l 1 O O o e N k (w) wida0 0 128 l E
i O h
. g i
c e u D o r
' 8 h t
v 7 7
< 7 r
o N 9
/6 1 m
o t
. r t f c
5 O
*3 I 0 5
3 0 . 1 t t p e n S i o t a c g o u L A t 3 f a
)
y l i l u - J s I O m h - 8 5 o t g f n m o t ( 0 e M 8 i n s u J e 7 u l a v y e M a n 8( e g y 9 x r. o i p d A e 2 1 v l o s r s i a i M d n9 a7 b e9 t e M1 F 9 . 5 3 n. e a e J r
- - - u 0 0 0 o 1 2 i F
ed eyo O CD _
. , l 111 I tI t lll
Locations 12 7.5 8 11 13 14 15 a 1 . . C
\
k 10 -
? 8 C
5 /
/ l U
o " E 7
'0 -
8 7
/ Bottom 30 - / - 0. ~ - . . . . . .....i,,,,, , ,
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Distance From Cowars Ford Dam (km) Figure 3-6. Mean dissolved oxygen values (ng 1-l) at main channel locations on Lake Norman during the latter stages of estratification d during November 1977 through 1979. O O O
ll l O Y 5 2 g n
~ ~ ~
i r 6 . 4 u 2 d 5 n
. 3 2
a m r 2 l o 2 f e 1 k I ,
- 2 a L
5 N 4
. 0 2
n o 9 s 1 n 3 I , o i 8 t 1 a c
. 7 o 1 l l
6
) e 1
m n ( k n a
. 5 m h 5 1 a c D
4 d n
. r i 1 o a F
m s m o t
. 3 1
n a t s t w n o o a o B . 2 C O i 1 t
- 1 )
1 m a c o I r - o 1 F 1 l 1 e g c n m
. 0 a (
1 t s i s D e
. 9 u
l a 8 v n 7 e g y . 8 s x8
. 6 o7 9
d1 e
. 5 vr l e ob 5
7 a r-6 0
. 4 . 3 ic so st DO 7 . 2 -
3 r-e
. 1 r
2 , u g 1 , ~ 0 0 0 0 i 2 3 F O M
' l ll) l1 l
n . c.3 m z ,,, A s'\ v l+
.t, *tt .nq ~.
t " 1 ,. p
. o ; ..._.._ -f D. * ,
g D.
- f..._ ,
n. l T
," J f; '
tykf* C
~
n 5m 10 m
/ )!g mn) * .. W.. -Ii i
n
.,4 1 .n ... - ?, a h. hh g ..s. m 1;- "k':
Figure 3-8. Mean dissolved oxygen concentrations (ng-1-1) at 0.3, 2. 5, and 10 m during July (1975-1979) at locations in the vicinity of McGuire Nuclear Station (f!MS). 9 9 9
,!ll \lil! ! ;l; 1 - 1lli l il lli 4! l 1 5 ~
7 9 O 1 t s u g . u) AS ! g! ? i n( rn - uo - di
= m t 2 0 1 . ma t . 0S hw 1 Rs. -
da r l'
,N @s ne ? v ![- al s . .
a ,cu [ .
- - q5 @
?;
I T 5N
,e , -- , 2 r i $' . f,W'tNB]
t W' n T~yG-t ' 3.-G
,u c
0M t e ah t
)
I f
' o l - y gt oi
( n i sc i $ ni ov O i t e ah rt t nn ei c m ns w on 3 m co i 0 , 5 - nt
-. , - ea gc i.M.jj . . .-. f!4 ._
yo xl
)-
f k.9 I ,- o d a t 1 ,
. = . u e v9
[ ' n'q:> . . v.i TD 4 p T. l 7 o9 s1 s e .J i h df- ' "' fjb . d g J ; .;- f u
)
c l T no
' ar eh Mt ~
9 3 e _ r u 1 g i
, i } F O - l' I > !- l' ,
0.3 m 2m k '
},
d . .. p[{ g* . s N --- 3/s.l'~~ sq w!
,4 s
sA
- ir - *e _
C 5 Ok_, , j d ** ~ -~
.o r igore 3-10.
tiean dissolved oxygen concentrations (ng-1 I) at 0.3, 2, 5, and 10 m during September 1975 through 1979 at lecations in the vicinity of the McGuire Muclear Statior (ft'IS). e 9 O
O l 120 Surface (0 - 2m)6
+ Bottom (>20 m)o 100 --
o._ -4 e -
.2 80 -
5 m 63" O Yu
.t 40 -
8 1 o 20 0 J F M A M J J A S 0 N D Months Figure 3-11. Mean rarcent dissolved oxygen saturation in the surface and bottom waters of Lake Norman from 1975 through 1979. l l O 135 l - - , _
O
,A ) .o - ,\
c C o
- . +(~, \y, .,e
[ : t, N .
. , e 'E a
eh /
.'E en 8 ., .a;
- ! . z' .. xE . .
m
- d' .. t .
JS g( .m SE e . m ,C s . g
.o o ; . . w ,,4 - ,
4 R~
. o . o
(' o m h 9 .
"P . ,' 0 4% mE a p - . 9 - 7 4 e'.s . .
a
- m
. 8 $ . - p . ,
o o y a b e'* ph . . a y ,. . a - o v y y .I $ .
"x. . ,E g 7 q n's w .~}
a "m s,e i
. w d, .
e 4, 5 . , 4 " *3 O $ C m Ar ' s q - o e . o .- s c q* : : N . : se E / : "! Es Nw
~
d,p . 48 eu i
' 8-c .C . * $ p 42 . g g w 1/l R, . m f '
n N
~* "
ES N
- 'N '
- s8 r<
N j
,e xt w ,e n ,e - ,: . , , %,a 7 .
- a. E v
A .
- M ti w $ '
w-49 M . , e . , L I I I i ipo ! ! ! ! ! a
- - - . e : g, e e .. e .c>
m (51)wn Hd patput)S) pd ( _g [(-(;.t ) f.4 7 /44uitetty 9 136
( 's O Surface (0 2m) 35 CO 2' pH o 28 -
~ ~W ~ _ ~
7
". 73 E
n . g 8 9 6 7
=--2-_
A
- =
0 0 N A M b b _d s h h f Morths 35 Bottom (2 20m)
._/ C0 7, 8
28
^ ~ -
21 - F i 8
~
14 -
\ -
6
/
7 l M M $ $d S 0 $$ 0 Manths Figure 3-13. Mean carbon dioxide and pH values in the surface and bottom waters of Lake Norman from 1975 through 1979. l 0 l 137
N O o O) i m e ,a o # CD e a en O L MR .C
+'
e '% ,
% a m w k /'d' @
g\ , 1< , g o L p a t O M d' ,
\,. , e 9 t )< na h /j .
w~ L s a w ' n V
,d' n c % es I - c < q *
\
' N 2 W 4 m ? =d ,M v a \ .. . .C. ' b D D T ' O D
k\ a *:
% tn . m o
en : i S *D es @ 9 E > a g al < st & il E
/
( . a .'-O e < , m O 3
<NK .
O
~
g q , , m
-A .x S' ~
9 W< ' % . 4 < r s
? '
- i b, K M 0
' M *> L ! ! ! 1 f I ! !
3 ERCPSCOMRD '
,[
u._ (01%) #3tf!Cahl 1 0 138 i
lJj j1 ,I l O 9 7 9 1
. . , h m .s . o g
u r A. . . h
. ~
t w I-\. . . 5 7 9
. . 1 '/ . /. / .
0 m
^, . . ,. . e o r
9 f
.E w.
5 n
. s w v v y , . , , . ,, . a s m % r . . ,N . J *e o k .- N t /, . .
s e k y.. .
. /.
8 9 L a
. \ . ,
7 f
- s. ,- o
. s . .m V w V' * , , . s N. m , . . a e
D r
~ a / %.. 5 d .- . . 6 e - j. 5 t
c O
"*n - - . . ' w - -. c, , m A e . e l I . . J %
e s
. . , m , / . J h - ,/ . M n . 7 i \. . .
i
- m. l/ , /
s.
, )
U
.3 m.. ~
e.8mJ - f T N (
~ w W v , < , x , , s e
u
. . l . . P a / . / R v ~
t
- p. y t
-l +
r -
, ,~ r . , .
i d i u
- ~ %.. J e
e b r u t A, . *e % t
* . n a /. N % e .r . f. .. , / e F
1 b 5
/ .
v ,s g , p , 1
. . 3 \
f . . e l r u
. , g i - v r < . ,. ., . F O -we .\ lllll[1\ 1,jl l 1jl ill:fl' Itl, 1 . ( fIil ,
5 ., 1 s s 5 2 y r a 9 A 4 u 2 r b e 3 F 2 g n 2 i 0 2 r 5 u
. 1 d a ,
l 2 s n 0 i o 2 t a
. 9 1
c 3 1 l o 8 l 1 e n 0 . 7 n 5 1 a h
)
c 6 m 1 ( k in 5 m a 1 a r D d n m . 4 r a o 1 o m t F r t o s o 3 N s n
/ 0 B
1 n a e o w c 3 - k i 1 2 C t 1 1 a a m L L c o
. 1 1
F c o r e
)
i n 9 0 n !!
. e 1 t T s N i
(
. 9 D s
e 8 u . l a9 v7
- . 7 y1 9
0 8> 2 t i h
. d g - 6 i u bo . 5 rr uh tt 5
7 . 4 n7 a7 e9 M1
. 3 / .
5 1 2 6 1 3
. 1 2 s e
r 1 i 0 G 0 0 0 u 1 2 3 g i F 7~ 5g G z.
1 't
\ Al 21 Fe 2^. Fe 1; Pn 1 Mn 1 r
Al Si0, SiO 72-2 20:: hl 7'l / CL 12; 7 K 18'; g 4; 6% 8%
\ Ca 9" Mg 33' H0 3 Mg Si HCO 32%
3 Al 17 Fe 3% 1975 Mn ir,% 1976 S10 2 ha 6 23% CL O '6 cu h 31% 6% HCO 3 At 3~ AL 1% Fe 31 Fe 2% Mn 1Y Mn 1% 1977 SiO SiO N# Na 2 2 C1 14; 14y _ 5% K b.' 9% 9% c Ca 7% 28% 7% 31% HCO HCO 3 3 1978 1979 [] Figure 3-17. !iean mineral composition (%) in Lake Nornan 1975 through 1979 expr essed at pe ent of total concei.t.'ation (nm 1-l) of A1, Fe, 111 SiO , HCO ' ' ' '#' ' "" '~ 2 3
O
.C f o, .o y 'a 3* 3 C n e o S s e q *s 3 - - $
y
/ ,e '} .9 s . .
h .,e D , 1*
,= y'n Sj e 2 :p] w e 's p, r -
I, 1.:s;
\ m/s .,
0 o, 9 p= 8 y n . b n , er,wh , , , f Q C m\
. a '^fi * ' Ce 9[
[; . n
- m.s n , - \N . ., 3 ,e 3m j ., i y
h=
,e ' ,/
3
-= ' 6 (l et fi 1 C
e=%
\ t w
- 3 ,
q; b
~g a '
W [ $ a h < o ( n [ g*
'\ NI, w ' - ~~
4 o . ., S g ff M e C n . ,~ - 2 +.. ., 2 sg
&/ ,' , u .-
91, 5 "
% .. g ak ti=
p u .- ee W M <
, uw Cw o n om . --6e uu**
dN. . . o , o g 3
'l.
va ,,ee
- m a [N -
ne ec g-me y 4 %e 71 @ j- n E * $ g ,,
. g y , c- .y d E C
- oo
- f. - "
m$ r-
~_
d
.' gj V
Ny 5 . _ %s,' _T. o eg
. ms gr M,
yg[ . ,g N (A,e ,e
=g s.g 4< - m ' *
- 7 aQ - - ? m
'n Q
i , - , i 3
* , , D o .- 9 0 4 o.
c, O, a
" 4 - N ~ - 9 Cn
- g. l * ) WOJJ 'LL" ,
(l* ' M @ gg(,9 g g O 142
1i i l I 9 7 O 9 1 h g u o
,o 1ii\ ..l.
e
. ,.,a.
h t r 5
- m. , .
7
, 9 1
m o
, . . r f ".\ .
l 'j . , n n y . . a
/
4
. m
_ o
- v. , -
, . i > t , s 1 ,, ._
e
;t . s . k a
r . . L
. s' . .
l'
. f -/ . .
o
. . s s s/ 4 . . ., . . a e
r
~ ,
a o I e d e 7, , . , t c
/, . .x ,f.
l e e s O
, , , . . n , 2 , .1 e , . .s %. >
i 3 . , ~~ l )
~ . m l a
m s . 1
. I g m. , g . .
m
, y ,. . ( . s' /. s ,~ , ,
n o _ ,, . , i
, 1 , . . . t .s/ .
J a
. . , r ~ . . ,,, t m . n j; . . /, .. - e c
n m . ,
. < + . e .x . .
o c
, > e 1 e ,
- m. , V. ,%
o n
. s r .. i /,
x'\ . n s . . _ / j .
. a v.. . j . .
M e
. ~ .\ . ,
F 9
,s' . . ,? , ;
1 i ( . . 3 e
/.//; . r u
g 1t i
~ e ,
u a 7 1 e F O _
-bw 1 ,1, llr!l1 11l il 1 1l 1, l i l
.a-.-_. - _m_ 4a__ _ ___a ._ m __J .A J -.e. A s --__m_.,a4 s, . . - - , _ - - . ._J a JL.- *.x ,
4 O r, g
- e. .
e.. -
- h p{
o, o
, a C m . -1 E o.. r - t' o t* r /- , :-
j
. , i J
H
~
9
~
9 -
. s ~ a a s s.~. @ m 4 m Ne \
3 .* 9- C' e,e e i a'
- O 4
\'s E l y %
3 . T
. i a 2 e, . p
[ s i o J v e<
-- )+ ,, =
e v ; Qj - m 4 e D d
, o. , )* q < g. .-
c
- i t
- 8: 9 *.
- O li "
= ~ e &* - - o e E i {
y 4 It~ $ " i i I
/p 5 h h < - 4." , v e
l { e E Y Ur g s q;. a . .
.O e-E. & \..'\ 'fre ,
m .-
,, a }v @
L In &
. ,, 4J C
the j
- a U vs n l R< C
* . a. o 3 /
6 U
' g. ~
4< t _ a__ o e, '
- et e - -
e_ ,, J. - m ,
+ . - -
o e em
-1
{ g
&* j g%
mQ 3a , i, I m-
.* .b . ,
i e c ele to L* k
, , . . qi. bO t - # a l 2 Y , , i e C C i
t 4 < . "- N k ( , \ Rc#
. 4 i ~,. W ., I" y .
y
,k . . "- . . - c. . .- m c, m e N O
q S
- n. .
i G1 f k. L i (Ii ' O 2 e a < < - f ** P' d O 144
O O O 0 - g 45 t 45 10 - T
- i I
45 r 45 5, o 40 4o a 20 - 52
\
30 - e 55 f/\ t i e s : i i t i e , e I Jan. Feb. N r. Apr. %y Juie July Aug. Sept. Oct. Nov. Dec. Nnths Figure 3-21. iban specific conductance values (umbe-cn'I) observed in Lake Norman from 19?? through 1979.
m o I
.> \ 3 9
m f r n 8
/
0 a A m t p A j/ a 5 A r o s l f m r J ., e 4 > '. e
*r k y a Ae ?
2 ( s/ , A , L
. D . f W , =
a t
/
y s, 0 5 4 o s a o _ t A e s M w s a - n , r A r t [ R
= .
J 4 5 a C = f J h d 0 t w, j .
- e t
. a # v . A c e
s
/ 0 m t
s p l A . $ v . F e 1 a , s 84 4 4 w . n h
,t w e , c s C e .,
6 5 s e t n i g/ w 4 , A! ) m R A . . D - Mt ,\ *
\g o / . g c-V . F A
t O o y
" J p . i h A
e d a m s % -c s e c e y 7 Ct u w J h s ( t t S a I D J esn s D [ 4 e R u w 9 I
# U
(- . A l
. 0 *ec , a M v A
I R A t At = . A J 3
/..
N . J F c n e t u r s 0 6 s s % *s o a n , a f R J .e, t { t f t. y/x . a A
- I 5
d c u n9 M 9 o7 c9 F P 1 f A . 5 c
._J e
s M w 5 4 e v c A f A f@ i i u
'# s .
M co f J *a J er lo
, 0 E a ph \ R I # st g . 4 J
Q . A A
~ . 0 E n5 - I .
W a7 Rf A .
$ e9 rt ss M1
[ e A w t J
- 0 6
s s % t a e
- y m,J .
M f a
. J w 2 2
m . M R / 3 J P A K
# e F
r u
.s's 1
J g i e ,.c F s % w 4 a y 3 e uam
O ca 9 . a N a ',D',A < z $
'u,9<
s~ { 2- o '" o < a; - - . - YO h < w Io h i w C
^
g .
,g [ ^* uk 4 . ,R oI -a N /, .E e3 - b' y) ,E ^
EE N l I Et 6/ < y I L 1: (4 1 m as . . a2 . e
/
C d 7'/ : :w y $.: :a iu 4 zp " .d
/ . ,e / - ,e y o
E E u g tg <
- o
- t E U < w w g g 1 q +1 )
o . o a
=-}
- m n :
n n 20.
. ',f . . . w . % Vd CN ,N < , rg Ch g .
o '.mn ," g
- a ji sy - o g'< ro c 1; 8 u +Q, 4- .- m /- ,e 'M < ,*
h e
< . ,E .- . e so m k ./ ' d ' 4 C.Ch h <
r r g '-* v . w .
< w w r-I '
as < ,
. o _m . ,. r, ,_ m= < ,b k/ -
3 ::
# s
- I d a
! I I l l l 9 "i *. *. N 9 e p, g g g a = ** * *
- o
- a. a d a a & 2
( g. t N f=) anat a + a;esus ( g.d bs) snaoudsowe tuoi 0 147
IO 1,0 " t er?p maim v mw+1 mae a gertsperg AyR A 1.0 1. 0 utit su!M (MA**n attA ( a74WEA #1vFB aptA 9I ' O. 0 - D. 9 - 89 . 0 - 0A - 0.4 - 5e - 0.6 -
- n. 4 -
0.6 - 0.4 - 9.4 -
- s / \,
S.,
- 0. ? k O.2 -
G.2 - s h s
.I F W 4 4 J J A 4 0 4
- M
~
19 9 A W J J A % 0 W P F W 4 m J 1 A 5 0 W tt J F m A et J J 4 > 0 m p
.,m*.
m - ~~
>-4 4
CD t9 o 1. o eacct.f a' teT4F7 Aitte m-MYet prtr%nwa Asfa tapssrY rprm esva 08 - 0.9 ' O.s - O6 - 0.6 C6
/ V v ., .
J
- as a w
- J # $ o e 3 J F m A 4 J J A 5 D W ft f F # 4
- J V A
- O es 0
~.s. - - ~ , , , .
Figure 3-24. Mean nitrate pius nitrite concentrations (ag-1-l) in selected areas of Lake Norman from 1975 through 1979. e 9 9
O -
~3 0
L-E
* ,a a +J N !
- m. @
E' ; N
.e s r:. *. a . , , o ' ' h +
y a 3 C g , O
.. 9 . O a
f g j
-
- _/ O r
Z r_ s _(-s ' "
.o - .6_-- ---
QJ 6 I 5 % b i; ,f . . .x s o o e g to 9 - J l i l '
- 8 '
, . c ) o / , ; .; - , ; i% c = -
- e D
}
N Q
, e . . L f N p - ,
k @
.- / I U 8 't _.
i e e
--. c. e t 5
it e- U
. 4 < ,
Q
. T - *
- C3
/
J V 1 e
\. . ,
e\ p - - , ,C. O ; e c
~
N b e w t Os
,8 " f " ^ < ? h r /p I :-. , , ,
e' l
\
2 i C f: - =
" \s h - 4 . . m C \ > - - ,h - ,
y l .o-a < , I i < y M
. g b o e d C a \ + C 9 < s a .c/ ,e <-
O y c e\ < , j! ;[ s's, O r . . u e
-+ , j7 a e o e e Nw < .m c
['e 3 l c 0 e < o h 9
* ,/ , *.
w
- m- ,
a :: h y' g k < 4 ,
. $ - 3 Cb CN a ; x. . e &T . , , \, ,
i
- h / i g
- v c i .
* 'l a' # c , 6 e m /. l N # i ! X l 9 im Q \ L e -
3 Ch
& n < a .e u.
O
- i e c c 149
A C O Va ! , e O L 3 - 4
! e. s s ./* e i ~'~~ ,
C E o z
\
v & a
,7 . . Q \ 0 p * ' ~'
a g j / % [' J, I
< o o
e - _ t l + r c
. . . e o o -
e
- , m
- . A
' ' I t j
g a 3
)n C.
E } M y e < e y < < U m x, . . a - n c f q < = a = 0 i \
. d 6 s } T , ~f 3 6 -) e a e a a ! 'e E '~
g , , . . c J'
' 4
- g>.*
p < A g a s I #
~ j .
a " 4 = = , h g o N x e 2 E N . . s : a a a , , s s . =
, , . r r
- l G
g
-L -
O
' .r 5 ? ' l . , a l '
c E / -
=
L p N ' u
" e l . . . . O O
C I 3 o ' e <
% 2 i
e
% F 3 2 E. U / , , s o e c. a e '/ m < = 3 . n L t' 's ; . s as b
l 4 g ' '
< - - OO l
S g 2 A o-t i 2 8
- s E "Q
e e e o c k
, 3 , p ,/, - o i co , & a L a- - e n - L O .C h ! ,/ pp e
- 1
- *N s t 7 - \ . c in m5 E 9 - * - ~
om
.O < ,
I , E-4 h s f f
. .- < - e s'
e < r y s . h a < , g B g g 5 C)
,' e 8 s . - -,
[ 150
QJ ? 4 e
@ O m f,
o
- s.
. s "3 R @ , O 4 s r < m N 3
- O 4 0 w h TY N O b 0,,
~. p'f a n .- o cn xs 4,4 ,
s t . AE 5
- , e u ? +
- C
'
- O a- ,
e ( .
, +
e O C) Y o ka g - e
\,a '
h a fI e - , I E I
- 7
/ ,
r
** v o
e . mu C@
)f - a OD o '"M@ DC , , Lw a *. '
- C 9 El d .
n ~* O U *"E o / - x C t d .
- OQ E
Uu x
< w C U & y ee. % or y o
si m.e .C e a o O
- C C.)
m C D < w g> *e
. , w ~ ~
sti
,o a
a
~
4 e i i L 9 9 D. 9 0,
. ~ o .-
(g,k I$ E I[I$ l'\ tv - 151
,e . r. ,,
- ' FUR %t ! * : P.AtNe t Af ? 4 0FFt9Fw'T Apla
.e 134A *La tE fTAM=tt amEA FAvswea sign gara .o s, u-a* , . e " w-* .t- % eC
- j. 4E
[ u . u . i eC SC j 9 m a w 90 I J a s e a n J F m 4 W J J A 5 o u D 5 * *
- 0C J $ 5 0 4 5 J F 4 A u j J A ' O 4 9
. r. ..~ --~ --~
m N s? ea
&ctt;st tw*ar? 4*r A se-ser<t !s t 5 ?5 ust y AHs snesty ( pe rs anta e
6C % f l
.1 m, y/ e r, s -
4 ,. ,' . m % ,
~+%y,v n-
- o. - n-
. . ,3 . . , . .. . . . . . ,-,2 2 . . . . .
- s. .. %~
Figure 3-28. Iiean silica concentrations (og-1-1) in selected areas of Lake florman from 1975 tnrough 1979. e S #
O O O 25 50 10 Copper Location 1.0 Zinc Location 1.0 Cadmium Location 1.0 40 08 20 30 , 15 1 06 _ po 04 10 _
~
10 0.2 5 _ WI SP SU F A WI SP SU F A WI SP SU F A WI SP SU F A W! SP SU F A Wt SP SU F A WI SP SU FA Wi SP SU F A W3 SP SU F A 1978 1979 1977 1978 1979 1977 1978 1979 1977 Seasons Seasons Seasons G u 50 25 Zinc
- 1. 0 Cadmium Location 8.0 Cooper Location 8.0 Location 8.0 20 40 _
08 15 30 _ 06 20 10 04 - 5 10 ~ 02 ' N WI SP $U F A WI SP SU F A W1 SP SU F A WI SP SU F A We SP SU F A W1 SP SU F A Wi SP SJ F A W1 SP SU F A WI SP Sb F A 1978 1979 1977 1978 1979 1977 1978 1979 1977 Seasons Seasons Seasons firure 3-29. Mean seasonal variation of ccdnium, cooper, and zinc (nq-1-I) from 1977 through 1979 at Lccation 1.0 and Location 8.0. U3shed lines indicate nissing data.
CHAPTER 4. PHYTOPLANKTON O M. S. RODRIGUEZ PAut INTRODUCTION . . . . . . . .. . .......... . . . . . 155 BACKGROUND . . . . . . . . . . . . ... . . . . . . . . . . . . 155 OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . 156 MATERIALS AND METHODS . ........... . . . . . . . . . . 156 SAMPLING LOCATIONS, FREQUENCY, AND PARAMETERS . . . . . . . . . 156 PROCEDURES . . . . . .. ............. . . . . . . . . 158 STANDING CROP , ...... ... . . . . . . . . . . . 158 FIELD PROCEDURES , . . . . ...... . . . . . . . . . 158 MEASUREMENT OF LIGHT PENETRATION . . . . . . . . . . . 158 SAMPLE COLLECTION .. .. . . . . . . . . . . . . . . 158 LABORATORY PROCEDURES .. .. . . . . . . . . . . . . . 159 POPULATION SAMPLES . . ..... . . . . . . . . . 159 CHLOROPHYLL SAMPLES ... . . . . . . . . . . 160 CALCULATIONS . . . . . . . . . . . . . . . . . . . . . 160 PRODUCTIVITY . 160 (]) ...... . . . . . . . . . . . . . . . FIELD PROCEDURES . . .. ......... . . . . . . . 160 LABORATORY PROCEDURES ..... . . . . . . . . . . . . . 161 CALCULATIONS . ...... ....,, . . . . . . . . 162 DATA ANALYSES ., ....... . . . . . . . . . . . . . . . 163 RESULTS AND DISCUSSION . . .,....... . . . . . . . . . . 163 COMMUNITY COMPOSITION ,... . . . . . . . . . . . . . 163 LOCATION COMPARISONS . . . .... . . . . . , , . 163 TOTAL ABUNDANCE . ........ . . . . . , . . . . . . . 163 TAXONOMIC COMPOSITION . . . . . . . . . . . . . . 164 TEMPORAL VARIATION . . . ... ...... . . . . . . . . . . 165 TOTAL ABUNDANCE ...... .... . . . . . . . . 165 SFASONAL VARIATION . . . . ... . . . . . . . . . . . . . 165 YEAR TO YEAR VARIATION . ... . . . . , . . . . . . . . . 166 SEASONAL SUCCESSIONS AND ASSOCIATIONS . . . . . . 166 VERTICAL DISTRIBUTION ... .. . . . . . . . , . . 167 PRIMARY PRODUCTIVITY . ., . . . . . . 168 GROWTH AND REPRODUCTION 169 (]) RESPIRATION, IMMIGRATION, AND EMIGRATION
. 172
SUMMARY
. ... . .. . . . . . . 174 LITERATURE CITED . . .. ... . . . . . 177 k_______ -. _ _ _ . _ _ .
INTRODUCTION QACKGROUNO O Phytoplankton are the algae of open water communities. In relatively large, deep. impounded systems phytoplankton generally account for the majority of autotrophic production (Wetzel 1975). This is particularly true of systems such as Lake Norman, where fluctuating water levels and unfavorable substrates limit the developmerst of extensive macrophytic and periphytic communities. In Lake Keowee, a South Carolina piedmont reservoir of relatively similar size, retention time, and substrate type, phytoplankton accounted for approxi-mately 98% of primary production (Rodgers 1974). Thus the phytoplankton in Lake Norman may represent the major autochthonous source of organic matter for consumption by the heterotrophic component of the community. Weiss et cl (1975) described the mean standing crops and primary productivity of Lake Norman as similar to other impoundments on the Catawba River system. The phytoplankton community in the vicinity of McGuire thclear Station was described as low in abundance, highly diverse, and dominated by green algae and diatoms. Maximum abundance occurred in mid sunner. Menhinick and Jensen (1974) found similar results in the vicinity of Marshall Steam Station, with the exception that seasonal maxima appeared to occur during the colder months (November through March). Weiss and Kuenzler (1976) classified Lake Norman as oligo-mesotrophic, while the U. S. Environmental Protection Agency (1975) described it as slightly eutrophic. Several studies have examined the effects of the operation of electric generating facilities on the phytoplankton of southern reservoirs. The major identifiable g impacts appear to be related to water movement. Among steam stations with sur-face-water intakes, studies indicate that homogeneity among sampling locations increased during station operation (Weiss and Anderson 1978). Among steam stations with hypolimnetic intakes, the primary impact of station operation appeared to be a dilution of epilimnetic populations ir, the discharge area with less abundant hypolimnetic populations (Duke Power Company 1977; Smith et al .1974). The hydromechanics of the condenser cooling water system of Plant Allen, a steam station located on Lake Wylie, North Carolina, resulted in the redistribution of algal populations from one area to an area with typically dissimilar populations, and in addition created increased retention times of phytoplankton in an eddy upstream of the discharge (Wilde and Paulishen 1974; Weiss et al.1975). The observation of steam station effects which were directly attributable to increased temperatures was generally limited to studies of phytoplankton entrained through condenser cooling water systems. Gurtz and Weiss (1972), studying phytoplankton entrained at Plant Allen, observed depressed produc-tivity following condenser passage, with the degree of depression related to initial temperature and the magnitude of temperature increase. However, delayed growth stimulation appeared to occur following the initial decrease in productivity. Knight (1973), work'ng at the same location, documented decreased abundance and diversity in entrained populations. Entrainment studies at Oconee lluclear Station, Lake Keowee, South Carolina (Duke Power Company 1977) and at Marshall Steam Station, Lake Norman, llorth Carolina (Smith et al.1974) indicated that condenser passage had very little impact g on phytoplankton populations. 155
) 0 08JEc11vts The cbjectives of this study were to:
- 1. document the taxonomic composition of the Lake Norman phytoplankton community,
- 2. describe seasonal patterns in phytoplankton abundance and taxonomic composition in Lake Norman,
- 3. examine vertical and horizontal distribution patterns in Lake Norman phytoplankton populations, and
- 4. characterize rates of production by phytoplankton in Lake Norman under various environmental conditions.
MATERIALS AND METHODS SAMPLING LOCATIONS, FREQUENCY, AND PARAMETERS Sampling of the phytoplankton populations and chlorophyll concentrations of Lake Norman was initiated in July 1973. The sampling history of each location is documented in Table 4-1. Locations are described in Table 1-3 and mapped in Fig.1-10 and 1-11. Population and chlorophyll data from 1973 are presented O ead e4scussed in the McGeire N clear stetioa envireemeatel nePort (Oeke eower Company 1976). From Janodry 1974 through February 1975 duplicate euphotic zone composite popu-lation and chlorophyll samples were collected monthly at Locations 2.0, 3.0, 4.5,
- 5. 0, 6. 0, 7. 5, 10. 0, 11. 0, 13. 0, and 16. 0. In addition, duplicate population and chlorophyll samples were collected monthly at the depth of 1% light penetration, at one-half that depth, at the surface, and at 1 m above lake bottoa at Locations 1.0, 4.0, and 8.0, to examine the vertical distribution J the phytoplankton. "
Locations 1.0 through 7.5 were selectcd to characterize the area of the lake predicted to be within the thermal influence of McGuire Nuclear Station, while Locations 8.0,10.0,11.0, and 13.0 were intended to act as references and to characterize the northern are: of Lake Norman. Location 16.0 was chosen to monitor populations downstream of Cowans Ford Dam in Mountain Island Lake. Euphotic zone composite population samples were collected weekly at Locations 1.0, 4.0, and 8.0 from April 1974 through September 1975 to document short-term variability in the phytoplankton community. Long-term variability in phyto-plankton biomass was examined through the monthly collection of chlorophyll samples f rom March 1975 through Septem%r 1977. From March 1975 through November 1975 duplicate euphotic zone composite chlorophyll samples were collected at Locations 1. 0, 2. 0, 3. 0, 4. 0, 4. 5, 5. 0, 6. 0, 7. 5, 8. 0, 10. 0, 11. 0, 13. 0, and 16. 0. Sampling at Locations 7.5,10.0,11.0, and 13.0 was discontinued in December 1975, and sampling at Location 6.0 was discontinued in October 1976. Justification for deleting these locations from the monitoring program is contained in the McGuire Nuclear Station Environmental Report (Duke Power Company 1976). In March 156
i 1977, sampling at Locations 1.0, 3.0, 5.0, and 8.0 was modified such that dupli-cate chlorophyll samples were collected at thret or more discrete depths within the water column, in order to obtain further data on vertical distribution of the phytopln kton. This program continued thrcugh September 1977. Beginning in October 1977, and continuing through December 1980, duplicate euphotic zone composite population and chlorophyll samples were collected monthly at Locations 1. 0, 1. 2, 2. 0, 3. 0, 3. 9, 4. 0, 4. 5, 5. 0, 8. 0, a nd 16. 0. In addition, duplicate discrete samples were collected at a depth 1 m above the lake bottom at Locations 1.0, 3.0, and 8.0. Location 1.2 was added to the monitoring program to characterize populations in the immediate area of McGuire's upper-level intake, while Location 3.9 was added to characterize populations in the.McGuire discharge canal. Two studies were conducted to examine rates of primary production. Fr om February 1974 through January 1975 primary productivity was measured monthly at Locations 1.0 and 4.0, and from January 1978 through Janua y 1979 primary productivity was measured weekly at Locations 3.0 and 8.0 and quarterly at Loca+. ion 1.0. Quarterly measurements continued at Locations 1.0, 3.0, and 8.0 from 1979 through 1980. The initial study was conducted to characterize pro-ductivity in the vicinity of the McGuire intake and discharge. The second study, beginning in 1978, was conducted to compare productivity at a reference location (8.0) to locations within the area of McGuire's projected thermal influence (3.0 and 1.0). Population and chlorophyll samples were collected at discrete depths in conjunction with the more recent productivity study. Because of the potential impact of McGuire's operation on the operation of & Marshall Steam Station (located approximately 23 km uplake from Cowans Ford W Dam on th Norvan), an additional sampling program was initiated at locations in ta vicinity of Marshall Steam Station. From June 1978 through May 1979 population and chlorophyll samples were collected twice each month at locations ll.C, 13 0,14.0,15.0,15. 9, 34.0, 50.0, and 60. 0. Beginning in June 1979 aad continuing through December 1980, population and chlorophyll samples were collected monthly at Locations 11.0,13.0, and 34.0. Samples collected at ail locations except 14.0 consisted of either euphotic zone com-posites, or corposites of water collected at 0.3 and 5.0 m. Surface samples were cc11ected at Location 14.0. The Results and Discussion section of this chapter is based primarily on the following data: monthly chlorophyll data from Locations 1.0 through 8.0, 1975 through 1979; monthly chlorophyll data from Locations 11.0 through 60.0, June 1978 through May 1979; monthly population data from Locations 1.0 through 8.0, January 1978 through December 1979; monthly population data from Locations 11.0 through 60.0, June 1978 through May 1979; monthly population data from Locations 1.0 through 13.0, March 1974 through February 1975; weekly chlorophyli and popu-lation data from Locations 3.0 and 8.0, January 1978 through January 1979; and weekly productivity data from Locations 3.0 and 8.0, January 1978 through January 1979. All data are included in Appendix 4. O 157 j { _ - __ d
- - - , . -~ - - . .. . _ - - - ..- .- -
PROCEDURES STANDING CR0P-Field Procedures-Measurement Of Light Penetration
- Light penetration was measured at each location on 'each sample date. From 1974 through 1976 a Montedoro-Whitney Model LMD-8A solar illuminance meter was used to detennine the depth to which 1% of incident solar radiation penetrated. In February 1977, the Montedoro-Whitney meter was replaced with 1
a _ Kahlsico Model 268WA310 radiometric submarine photometer, preferable to the Montedoro-Whitney meter because it measures energy flux (uWatts m-2) within
~t he photosynthetically active range of the spectrum (400 to 700-nm), while-the Montedoro-Whitney measures the sum of visible and infrared radiation.
Meters such as the'Kahlsico Model 268WA310 and the Licor Model LI-185A thus provide more accurate-estimates of the amount of solar-radiation actually available for photosynthesis. In September 1978 a Licor Model L1-185A light _ meter equipped to-measure both energy flux and quantum flux (uE m-2 3-1) in the 400 to 700 nm rangc *ame the primary instrument for measuring light penetration. From February 1977 - h December 1980, vertical profiles of light intensity
- _ at depth intervah were measured at each location on each sampling date.
, Sample Collection
, All standing crop samples were collected with a van Dorn or Kemmerer water bottle or with a graduated cylinder (surface samples only). Euphotic zone composite-semples were prepared by compositing 1-t samples from the depth of 1% light penetration, from one-half that depth and from the surface.
The change'in instrumentation used for the measurement of light penetration from a Montedoro-Whitney Model. LMD-8A to a Kahlsico Model 268WA310 and sub-
- sequently' to a Licor Model LI-185A probably resulted in the measurement of greater depths of 1% light penetration af ter January 1977. The Montedoro-E Whitney meter measures solar radiation in both the infrared and visible ranges, while the Kahlsico and Licor meters measure in the visible range only. Because infrared radiation is absorbed in water more rapidly than visible light, the Montedorc-Whitney meter probably underestimated the depths of 1% light penetration compared -to the Kahlsico and Licor meters. The collection of composite standing crop samples based on the 1% depths estimated by any of the three instrurents should result in similar estimates of standing crop whenever the mixed depth i
is greater than the Kahlsico/Licor 1% depth,-since only the mixed layer muld be subsampled. Based on 1978 data (Figures 4-30a and 4-30b), che mixed depth exceeded the K5hlsico/Licor 1% depth from September through mid-May. Thus, during this period the use of any of the three meters should produce similar standing crop estimates. However, during the period from mid-May through August, the Montedoro-Whitney. meter most frequently estimated the 17 depth to be slightly -less than the mixed depth, while the Kahlsico/Licor estimates were most frequently slightly greater than the mixed depth. Composites based on the Montedoro-Whitney estimates thus most frequently represented the epilimnion 158
only, while composites based on the Kahlsico/Licor 1% depths frequently contained one subsample from the upper metalimnion. Chlorophyll profiles obtained in 1978 (Appendix Tables 4.2-4 and 4.2-5) indicate that the standing h crop of the upper metalimnion averaged about 45% less than that of the epilimnion. Thus, the inclusion of one metalimnetic subsample in the composite would produce a mean standing crop estimate approximately 18% less than that which would be obtained using epilimnetic subsamples only. The error inherent in the phyto-plankton enumeration technique was approximately 20% (Lund et al.1958). Subsamples for the analysis of standing crop parameters were withdrawn from the composite sample. Samples collected to characterize discrete depths were withdrawn directly from the van Dorn or Kemmerer bottle. Population samples consisted of a known volume of water, generally 950 m1, measured in a graduated cylinder and transferred to a glass French square bottle, to which 10 to 20 ml M3 preservative (Meyer 1971) was imediately added. In January and February 1974, a surfactant was also added to the samples. However, the surfactant inter-acted with the sample and preservative such that the saLples could not be analyzed. To prepare samples for the analysis of chlorophyll a. content, a kaown volume of water, generally 250 to 500 ml, was filtered in the field through a 47 mm glass fiber filter. Approximately 1 ml of a saturated magnesium carbonate solution wa. added to each sample during filtration to prevent acidification and decomposition of chlorophyll a_ to phaeopigments (Strickland and Parsons 1972). Filters were stored in darkened centrifuge tubes on ice. Laboratory Procedures Population Samples h Population samples were allowed to settle undisturbed at a rate of at least 4 h.cm4 of container height (Weber 1973). Supernatants were aspirated off and discarded, and the settling proces: was repeated in smaller containers until a final known volume of approximately 5 ml was obtained. Samples were diluted when the settling process resulted in a very dense concentration of phytoplankton or other suspended matter. Subsamples of each concentrate were pipetted into Palmer-Maloney counting cells (Palmer and Maloney 1954) for observation at 500X under phase-contrast illumi-nation. A minimum of 100 phytoplankton units (Lund et al.1958) were identified and enumerated for each subsample, over a known area of the counting cell. Prior to 1977, phytoplankton units were defined as follows: for diatoms (Bacillariophyceae) one cell was counted as one unit; for all other classes, one unit was defined as one cell for unicellular species, one colony for colonial species, and one 18-um length for filamentous species. Beginning in 1977, each entire filament of a non-diatom species was defined as one unit. To aid comparability, all data collected prior to 1977 and reported here have been converted to the latter definition of units as follows. Where possible, a mean filament length for each non-diatom filamentous species was obtained, and divided by 18 to obtain a mean number of 18-um lengths per filament. Each count of a filamentous species made prior to '.977 was then divided by the mean number of 18-am lengths per filament to obtain an approximate number of filaments observed. Where it was not possible to obtain a mean filament length (cs when a species was very rarely observe::), the mean number of 18-um !engths recorded g per filament observation was obtained and used in the same manner as mean number of 18-um lengths per filament above. 159 f f
- - - - -. - - . - . . . - _ - - - . _ ~ - . - . - .
(~ Taxonomic identifications were carried out to the lowest practicable _ taxon, generally species or Major taxonomic references included Bourrelly
;(1968, 1972). Cocke.(genus.1967), Eddy (1930), Huber-Pestalozzi (1941,1968), Hustedt (1930), Kim (1967), Patrick and Reimer (1966), Prescott (;962), Weber (1971),
and Whitford and Schumacher (1973). Dr. Larry A.; Whitford and- Dr. Charles W. Reimer:were retained as taxonomic consultants and confirmed identifications
. of many of the species observed.
From 1 to 30 cells of each species were measured with an ocular micrometer, depending on frequency of. occurrence, and the mean cell dimensions for_each species were calculated. For colonial species, the number of cells per colony was also recorded for several colonies, and the mean number of cells per colony was calculated. For filamentous species, filament dimensions _rather than cell dimensions were measured. Chlorophyll Samples Filters containing chlorophyll samples were ground with a tissue grinder in a known volume of_90% acetone and stored on ice for a least 15_h. Samples were
- centrifuged to remove filter fragments and analyzed on a Turner Model 111 fluorometer and/or a Coleman _Model 124 double-beam spectrophotometer. _ FolloWng an initial reading samples were acidified with oxalic or hydrochloric acid, and reread to obtain estimates of true chlorophyll a and phaeopigment concentrations (Strickland and Parsons 1972). Samples were not acidified prior to 1975.
Calculations The depth to which-1% of subsurface incident light penetrated was recorded directly from the Montedoro-Whitney _ light meter, and was estimated from submarine photometer data by regressing the log of percent subsurface light penetration against depth. -Chlorophyll a concentrations were calculated according to Strickland and Parsons (1972). Phytoplankton counts were converted to standing crop estimates expressed as numerical densities. Estimates of standing crop expressed-as biovolume were also obtained, by converting the mean dimensions '(length,. width, depth)- for each species to an approximate biovolume per cell or filament, using volume-formulae for' appropriate geometric solids. For non-filamentous colonial species, the mean number of cells per colony was multiplied by the biovolume _ per cell to obtain a colonial biovolume estimate. _However, all diatom bio-volumes were expressed on a per-cell basis. Standing crop _ estimates expressed as biovolume were obtained by multiplying.mean biovolume per phytoplankton
~
unit by the species numerical density. Standing crop estimates for March 1974 through February 1975 were based on mean unit dimensions for that period. Estimates reported for October 1977 through December 1980 were based primarily on mean unit dimensions recorded from October 1977 through September 1978. PRODUCTIVITY Field Procedures During the initial study period (February 1974 through January 1975) primary productivity and rates of community metabolism were estimated monthly by 160
measuring changes in the dissolved oxygen concentration of samples incubated at 4 to 6 depths for a midday period of 3 to 6 h. Depths were chosen on the & basis of light penetration, measured with a Montedoro-Whitney solar illuminance W meter. Samples were collected with opaque Kemmerer or van Dorn bottles. From cch depth, 2 sets of 2 transparent and 1 opaque 300-ml B00 bottles were filled. One transparent bottle from each set was retained for the determination of initial dissolved oxygen concentration, while the remaining samples were incubated at the depth of collection. All oxygen samples were preserved in the field with
? ml MnS0u solution and 2 ml alkali-iodide-azide solution ( APHA et al .19711 Two series of dawn-to-dusk 4-h incubations were conducted, in May and October 1974.
From January 1978 through January 1979 productivity was estimated weekly using both the oxygen method described above and the 14C method. Samples were col-lected and incubated at 0. 3, 1. 0, 2. 0, 3. 0. 6. 0, a nd 10. 0 m. Incubation depths were chosen on the basis of light profile data collected on Lake Norman for the months November 1974 to November 1975, and represented the approximate mean depths of 100, 50, 25,10, I and <0.1" light penetration. Samples were incubated for 2 to 4 h at midday. An Epple/ pyranometer, Model 8-48, con-tinuously recorded incident light. Vertical profiles of light intensity in terms of both energy and quanta were obtained during each incubation with submarine photometers. Vertical profiles of temperature, pH, dissolved oxygen, and conductivity were obtained during each incubation with a Hydrolab Model 6D (Chapter 3). At each depth, sets of initial, transparent, and opaque 300-ml BOD bottles were filled to examine changes in dissolved oxygen. In addition, sets of transparent and opaque 300-nl BOD bottles were filled and spiked with approximately 5 uti 1"C as sodium bicarbonate. Samples for the analysis of alkalinity, chlorophyll, $ and nutrients were also collected at each depth. Population samples wre col-lected at 0. 3, 3.0, 6.0, and 25. 0 m. Duplicate samples were collt.ced approxi-mately every other sampling period. Laboratory Procedures Following incubation and fi>ation, oxygen samoles were titrated using the azida modification of the iodomett ic method to determine initial and final concen-trations of dissolved oxyge.1. Changes in oxygen concentratiors in transpa ent < and opaque bottles were used to calculate gruss and net productivity and respiration (APHA et al.1976). A 1-ml subsample was Wthdrawn from each 14C-spiked sample and analyzed using liquid scintillation (this procedure wasspectrophotometry totocheck not carried out prior August true activity of the g1 Cike the1978). To determine uptake for each sample, a 100-ml subsample was filtered onto a membrane filter, and then rinsed with 0.002 N hydrochloric acid to remove excess inorganic 1"C. Filters were assayed for radioactivity using a liquid scintillation counter (Beckman Model LS-9000). Filtrates were purged of inorganic 1": and subjected to flash evaporation, in order to concentrate 1"C-labeled dissolved organics released by the plankton during incubation. A subsample of the concentrated filtrate was aralyzed using the liquid scintillation counter. O 161
Alkalinities were determined by titration to a pre-determined pH end-point O ( APHA et al.197f). The appropriate end-point pH was determined for each sampling period by examinirq the inflection points of the curves resulting from at least two full titrations. The initial pH of each alkalinity sample w s recorded. _ Samples collected for the analysis of total phosphorus, orthophos-phate, Kjr.idahl nitrogen, ammonia, and nitrate plus nitrite were analyzed as describec in Chapter 3. Calculations Estimates of carbon uptake (mg C m-3.h-1) from samples spiked with 1"C were calculated based on APHA et al. (1976). Opaque bottle rates were subtracted from clear bottle rates. Total available inorganic carbon and carbon dioxide con-centrations were calculated based on tield measurements of conductivity and temperature, and laboratory measurements of pH and alkalinity, utilizing equations derived from Stumm and Morgan (1970). Carbon uptake rates on an areal basis (mg C m-2 h-1) were obtained by plotting volumetric uptake
- (mg C m-3 h-1) vs. depth (m), and integrating the resulting curve by planimetry.
Daily areal uptake rates (mg C m-2 d-1) were caiculated by multiplying hourly areal rates by the ratio of total ly d-1 to ly h-1 (mean during incubation period). Mean values for mg C m-2.d-l for each week were calculated utilizing a weekly mean value for ly-d-1 Annualestimatesofproduction(gCm-2yr-1) were obtained by adding the weekly mean areal uptake rates (mg C m-2 d-1) for 52 consecutive weeks, and multiplying by 7 d wk-1 O ^ssimi'atio" dividing uptake ratios C rates(mymgmgchla-1n-1)werecalculatedateachdepthbyM)
, C m-3 h by the mea (mg m-3), determined spectrophotometrically. For depths <3 m, values for the -
mean concentration of chlorophyll a. in the upper 3 m were used; for 6 and 10 m, values for chlorophyll a_ concentrations at those discrete depths were used. Photosynthetic efficiencies, or carbon uptake rates per unit chlorophyll a. per unit light, were determined from the slope of the initial lintar portion of the plot of photosynthesis (mg C mg chl a-1 h-1) vs. light intensity (pE m-2 3-1), for each set of incubations. Mean ITght intensity at each incubation depth was determined by subtracting reflected light from total incoming solar radiation during the incubation, converting to photosynthetically active radiation (Vollenweider 1974), multiplying by the percent penetration of subsurf ace incident light to the desired depth, ano dividing by the length o' the incubation period. Where necessary, data expressed as ly h-1 were canverted to uE m-2 s-1 using a conversion f actor of 52 8L (pE m-2 s-1)/(ly h-1) (Harris 1978). Respiration rates could not be calculated directly from l'C data. However, apparent respiration rates (R') were back-calculated, based on uptake rates and changes in algal standing crop expressed as carbon. Algal biovolumes were converted to estimates of algal carbon utilizing the equations of Strathmann (1967). The year was divided into several time periods, generally 2 to 5 weeks in length, during which concentrations of algal carbon in the mixed layer of the lake either consistently increased, decreased, or remained stable. For each time period, the net change in algal carbon in the mixed layer was cal-culated and divided by the number of days in the time period, lhis quantity was then subtracted from the mean daily areal photosynthetic rate during the G time period, resulting in cn estimate of daily apparent re,piration in the mixed layer per squarr meter of lake surface. This estimate was in turn 162
1 divided by the 000th in meters of the mixed la to obtain an estimate of apparent respiration as mg C mh-1 ger, and by 24 h d-1, g DATA ANALYSES Differences in total abundance among locations were examined on an annual basis with Friedman's analysis (Conover 1971), using mcan aburdance values for each month at each location. Separate analyses were performed on chlorophyll a, algal density, and algal biovolume data. Differences in mean annual taronomic composition among locations were examined with cluster analysis (Helwig and Council 1979). Variables used in this analysis were the mean annual densities at each location of six taxonomic groups: the Chlorophyceae Bacillariophyceae, Cryptophyceae, Myxophyceae. Dinophyceae, and a group representing all other classes combined. In order to examine the relative importance of temporal variation in taxonomic composition as compared to spatial variation, a cluster analysis was performed using as variables the percent composition of the five major taxonomic classes plus unidentified algae, observed at each location during each month in 1978. Individual linear regressions (Helwig and Council 1979) were used to examine the relatienships between the following parameters: maximum assimilation ratio -( ARmax) and teuiperature; ARmax and total phosphorus; ARmax and total nitrogen; and log of the percent subsurface light penetration and depth. RESULTS AND DISCUSSION COMMUNITY COMPOSITION O Ten classes and 306 species and varieties of phytoplankton were observed in samples collected from Lake Norman from October 1977 through December 1979 (Table 4-2). Distribution of species in classes was as follows: Chlorophyceae, 148 species and varieties; Bacillariophyceae, 54; Chrysophyceae, 27; Haptophyceae, 1; Xanthophyceae, 3; Cryptophyceae, 9; Hyxophyceae, 28; Euglenophyceae, 13; Dinophycece, 22; and Chloromoncdophyceae,1. In terms of abundance, the major classes were the Bacillariophyceae, dominant from late fall through uid spring; the Cryptophyceae, abundant in late spring; and the Chlorophyceae and Dinaphyceae, dominant from summer through mid fall. The Myxophyceae were an important cun-ponent of warm weather populations in 1974, but did not recur in large numbers in 1978 or 1979. The Haptophyceae and Chrysophyceae were ocr tionally abundant at isolated locations for short periods of time. Genera wh; consistently constituted a significant part of the total density or biovolume were Nannochloris, Rhodomonas, Melosira, and peridinium. Community composition was similar to that observed for Lake Norman by Weiss et al. (1975). LOCATION COMPARISONS TOTAL ABUNDANCE Differences in total abundance amor.g downlake locations (1.0 through 8.0) were examined utilizing Friedman's analysis on 5 years of monthly chlorophyll data g 163
. . . = _ - - - - - - - - . . - - _ _ - . _ . _ 01 -(1975 through 1979). Significant location cifferences were detected only in 1976 and 1979 (Table 4-3). During these y;ars, location 8.0 exhibited highest mean annual chlorophyll concentrations- (4.6 mg.m-3 in 1976 and 3.6 mg m-3 in 1979). These values did not exceed those observed at other downlake locations by more than 1.5 mg m-3, indicating that, although at times significant, varia-tion among downlake locations was not substantial on an annual basis (Fig. 4-1). Lakewide differences _in total abundance were examined utilizing chlorophyll data from uplake as well as downlake locations, collected from January through December 1974 and from June 1978 through May-1979. Significant differences among locations were detected, with uplake locations and Location 8.0 consistently exhibiting higher chlorophyll concentrations than downlake locations (Fig. 4-2 and Table 4-3). Location 34.0, the highest ranked location hibited a mean annual chlorophyll concentration of 5.0, mg m'{the-lowest while Table 4-3), ex-ranked location (1.2) exhibited a mean annual chlorophyll concentration of 2.0 mg.m-3, indicating that differences between uplake and downlake locations were somewhat more substantial than among downlake. locations alone (Fig. 4-2). Higher chlorophyll concentrations uplake may have been the result of the higher con-centrations of total phosphorus observed uplake (Chapter 3). Friedman's analysis of biovolume data utilizing chlorophyll data (Table 4-3)However, . generallyanalysis supported the data of density results obtained did-not produce consistent relationships among various areas of the lake (Table 4-3). Q TAXONOMIC COMPOSITION
;Mean annual densities of the major taxonomic classes (Chlorophyceae, Bacillariophyceae, Cryptophyceae,.Myxophyceae, Dinophyceae; all other classes were combined and treated as one variable) were compared among locations using cluster analysis (Helwig and Council 1979). Cluster enalysis of downlake locations -(1.0 through 8.0) and Location 16.0 (Fig. 4-3) revealed that differences in -
taxonomic composition among locations did occur, largely as the result of differ-ences. in the mean annual densities of diatoms, which were highest at Location 8.0 and lowest at locations in the Ramsey Creek area. This was particularly evident in 1979, when diatom densities were as much as 50% lower at Ramsey Creek locations than at Location 8.0-(Table 4-4). Cluster analysis based on the mean annual densities observed at all locations, uplake included, from June 1978 through May 1979, revealed that uplake locations maintained higher mean annual densities of diatoms, and lower densities of green algae and unidenti id or minor classes of algae, than did downlake locations-(Table 4-4 and Fig. 4-4). Diatom densities at Locations 1.0, 2.0, and 8.0 were _approximately 2(N lower, and at Ramsey Creek locations, approximately 50% lower than those observed uplake. Densities of green algae were approxi-mately 30% lower uplake than downlake, while densities of unidentified or-minor classes of algae were 60% lower uplake than downlake (Table 4-4). A cluster analysis based on data from Merch 1974 through February 1975 isolated uplake locations 11.0 and 13.0, which at that time exhibited higher mean annual densities of cryptophytes and diatoms than exhibited at other locations (Fig.4-5). 164
1 Examination of the mean class densities associated with each cluster (Table 4-4) indicates that, although detectable, taxonomic differences among downlake a locations were not substantial, and in fact, were generally outweighed by W similarities in temporal variation among locations (Fig. 4-6). As the foliowing discussions of temporal variation and vertical distribution are based on data collected at downlake locations (1.0,1.2, 2.0, 3.0, 3.9, 4.0, 4.5, 5.0, 8.0), the phytoplankton communities of specific geographical areas will be discussed independently only where instances of large differences in taxonomic composition or total abundance occurred. TEMPORAL VARIATION TOTAL ABUNDANCE Seasonal variation Mean concentrations of chlorophyll, averaged over a 5-year period and over 7 locations (1.0, 2.0, 3.0, 4.0, 4.5, 5.0, 8. 0), yielded little evidence of consistent patterns in the seasonal vat iation of phytoplankton abundance (Fig. 4-7). The minimum monthly mean chlorophyll concentration was 2.6 mg m-3 in May; the maximum was 5.0 in September. Seasonal trends in mean biovolume data averaged over 2 years (1978 and 1979) for Ramsey Creek (Locations 3.0, 4.5, 5.0) and main channel locations (1.0, 2.0, 8.0) were somewhat more substantial (Fig. 4-8 and 4-5); however, maxima did not exceed minima by more than an order of magnitude. This is typical of other oligo- to mesotrophic piedmont Carolina reservoirs (Duke Power Company 1977; Weiss and Anderson 1578; Wilde and Paulishen 1974). In contrast, Wetzel (1975) stated that seasonal biomass maxima of algal populations in temperate lekes h are generally on the order of 1000 times higher than seasonal minima. Patterns in seasonal variation were, however, discernible in biovolume and chlorophyll data wqen examined on a yearly basis, rather than utilizing 5-year mean data. Chlorophyll concentrations in surface waters, which ranged from 0.3 to 18.0 mg m-3, and biovolume, which ranged from <100 to 4000 mm3 m-3 attained mid summer maxima in 1974 (Duke Power Company 1976). This pattern persisted in chlorophyll concentrations at main channel locations through 1976; no consistent seasonal trends were exhibited at Ramsey Creek or McGuire discharge locations during this period (Fig. 4-10). More recently (1978 and 1979), bimodal maxima were observed to occur in spring ar.c in late sum.mer to early fall (Fig. 4-10 and 4-11). Among all downlake locations the fall peak in biovolume varied from 1000 to 2500 mm 3m-3 in 1978 and 1979. The size of the spring peak, composed primarily of Melosira italica, was much more variable, ranging from 800 to 1300 mm 3 m-3 in 1978, and from non-existent at Ramsey Creek locations to 3800 mm3 m-3 at nain channel locations in 1979. Unlike Ramsey Creek locations, McGuire discharge locations (3.9 and 4.0) did exhibit a small spring peak in 1979, presumably due to the displacement of water from main channel locations through the McGuire Condenser Cooling Water System. No consistent seasonal trends were observed in algal density, which ranged in yrface waters from 300 to 5000 units ml-1 Density, however, is not generally l '>nsidered a good estimate of total algal biomass, due to extreme variation in cell size among species. g 165
h Year to yeer variation Mean annual calorophyll concentrations exhibited a net downward trend from 1974 through 1979 (Table 4-5 and Fig. 4-1). Valuesfordownlakelocations(1.0 through 8.0) ranged from 5.7 to 7.0 mg m-3 in 1975, and from 2.1 to 3.6 mg m-3
- in 1979. The decline in chlorophyll a, concentrations was potentially due to the decrease in mean annual concentrations Cf total phosphorus which occurred during'this time period (Chapter 3).
SEASONAL SUCCESSIONS AND ASSOCIATIONS
- The se6sonal variation of each of the major taxonomic classes was quite con'-
sistent in 1978 and 1979. Densities of the Chlorophyceae peaked in late summer-early fall (Fig. 4-12 and 4-13), attaining maxima up to 2200 units ml-1 (200 mm 3 m-3 of biovolume). The major components of the green algal comunity
- during the period of-peak abundance were small coccoid greens, such as Nannochloris spp. and Chlorella spp,, and several very small members of the genus Cosmarium (less than 10 pm in length). The green algal community was quite diverse during late summer-early fall, with an average of 23 taxa observed at each location, the majority being members.of the Chlorococcales. In-1974, - the Chlorophyceae peaked in mid summer (Fig. 4-14 and 4-15). Dominants were similar to those observed in'1978 and 1979.
Diatoms reached peak abundance during the spring (Fig. 4-12 and a-13). Densitits as high as 3800 units ml-1 and.biovolumes up to 3700 mm3 m-3 were observed in
- March 1979. The primary components of the spring diatom pulse were Melosira italica and M. italica var, tenuissima, the year-round dominants of the diatom population. ~During the period of maximum diatom abundance, rarely more than 11 taxa of diatoms were observed at each location. Other diatoms common during this period were Asterionella formosa, Melosira distans and M. distans var.
alpigena, Nitzschia agnita, Rhizosolenia spp., and Stepha'nodTscus spp. As previously stated, no peak in diatom abundance occurred at Ramsey Creek
- locations in 1979. In-1974, both spring and fall peaks were observed in diatom <
abundance (F'g. 4-14 and 4-15). While both peaks were of similar size in terms of density, he- fall peak was much larger in terms of biovolume. The spring peak was dominated by Melosira distans var. alpigena, although M. italica was also important; the fall peak was dominated by M. italica and M. italica var, tenuissima. Diatom abundances were lower in Ramsey Creek than in the main , channel. De'nsities of the Cryptophyceae peaked in May in 1978 and 1979 (Fig. 4-12.and 4-13), exhibiting a maximum density of 1500-units.ml-1 and a maximum biovolume . of 300 mm3 m-3 Rhodomonas minuta was by far the most abundant of the crypto-phytes, both at peak density and throughout the year. On the average, five taxa were observed at each location during peak abundance, generally including several species of the genus Cryptomonas as well as Rhodomonas minuta, in 1974, seasonal trends in the abundance of the cryptophytes'were variable, although minima were generally observed during the summer (Fig. 4-14 and 4-15).
. . Dinoflagellate populations peaked in late summer-early fall in=1978 and 1979 (Fig. 4-12 and 4-13). Densities did not exceed 150 units.ml-1; the maximum observed biovolume was 2200 mm3 m-3 The genus Peridinium constituted the major part of the dinoflagellate community. The most frequently observed species I
166 p - ry - y --r-a-w,- - +c-~--w --w--t-
, er p -s > 9,yn a -+w ,
i I during peak abundance were Peridinium defiandrei. P. inconspituum, P. lomnickii, P_. wisconsinense, and Glenodinium oynnodinium. observed at each location during peak abundance. A June peak in dinoflagellate
~ -
On the average, six species were g abundance was observed at downlake locations (1.0 through 8.0) in 1974 (Fig. 4-14 and 4-15), Blue-green algae were not observed in substantial numbers in 1978 and 1979. Maximum observed density during this period was 200 units mi-1 and maximum biovolume approximately 100 mm3 m-3 One exception occurred, in September and October 1978, when densities at location 16.0 reached 900 units ml-1 and biovolume reached 300 m3 m-3, due to high densities of Oscillatoria lemermannii . Blue-greens present at other locations peaked in late summer-early fall (Fig. 4-12 and 4-13). In contrast, in 1974 blue-green algae constituted an important component of the total algal comunitz, attaining peak densities and biovolumes of 400 units ml-1 and 800 mm3 m , respectively. Peak abundance was observed in May and July (Fig. 4-14 and 4-15). Anabaena wisconsinense accounted for the majority of blue-green algal biovolume. Other species of Anabaena, as well as Anacystis cyanea, were also present. On the average, four taxa were observed at each location during peak abundance in 1974. Based on the above discussions of relative and total abundance, the recent seasonal variation of the phytoplankton assemblage of Lake Norman in the vicinity of McGuire Nuclear Station can be characterized as follows. Relatively low mid winter biovolumes were dominated by diatoms, primarily Melosira italica. Diatoms, Rhodomonas minuta (a cryptophyte) and small coccoid green aQae consti-tuted most of the mid winter algal density. A spring peak in biovolume consisted mostly of Melosira italica, which declined rapidly at the onset of thermal 3 stratification. Rhodomonas minuta then increased to dominate algal densities W in May and June, although diatoms and green algae remained important constituents of the relatively low total biovolume and density. The mid summer community was dominated by small green algae in terms of density and by dinoflagellates in terms of biovolume. Both of these classes increased to a late summer-early fall peak, with the dinoflagellates constituting the major percentage of the late summer-early fall peak in total algal biovolume. The diversity in terms of number of species observed at a location reached an annual maximum averaging about 60 specics per location during this period (October), as opposed to a late winter-sprinq minimum of about 30 species per~ location. The small population of blue-greens, primarily Anabaena spp. , other unidentified filamentous blue-green algae, Anacystis spp., and Agmenellum quadriduplicatum also appeared during the late summer-fall perioc. As overturn progressed, biovolume declined and the typical winter association of diatoms (Melosira italica), cryptophytes (Rhodomonas minuta), and small coccoid greens reappeared. VERTICAL DISTRIBUTION The vertical distribution of phytoplankton is a function of turbulence within the mixed layer, the mixing regime of the lake, and the ability of the algae to influence their position in the water column utilizing flagella or the regulation of buoyancy via gas vacuoles. During the period in which Lake Norman was not stratified, vertical distribution of algae based on chlorophyll profiles ranged from uniform to patchy, with no particular water stratum consistently maintaining higher or lower chlorophyll concentrations than any other stratum (Fig. 4-16 and 4-17). The major dominant during this period was a diatom, g 167 J
Melosira italica, which is subject to relatively rapid sinking (Lund 1954).
'h Thus, it is expected that algal distribution would be fairly un1 form during periods of high turbulence, and patchy when mixing rates decreased.
From the onset of- stratification in the spring to the initiation of overturn in the fall, populations tended to be increasingly concentrated in the epilimnion (Fig. 4-16 and 4-17). Chlorophyll concentrations in the epilimnion were as much as 30 to 40 times greater than hypolimnetic concentrations, which declined to as low as 0.1 mg m-3 ( App. Table 4.2-6). Dinoflagellates and very small green algae dominated the epilimnetic populations during the stratified period. Dinoflagellates are large, but motile (possessing flagella) and thus during periods of low
- turbulence are able to regulate their position in the water column. Small green algae rossess large surface to volume ratios and therefore require pro-portionally less turbulence to prevent loss to the metalimnion. Thus, both major summer dominants were adapted to remain within the upper mixed layer-exposed to the light required for autotrophic growth.
The Chlorophyceae were well-distributed in the water column during non-stratified periods, but were heavily concentrated in the epilimnion during the stratified period (Fig. 4-18). The Bacillariophyceae, consisting primarily of Melosira italica, were patchily to uniformly distributed during the non-stratified period. However, during the stratified period-diatoms showed a marked peak =- density in the hypolimnion. Densities in the hypolimnion declined rapidly whe.r. exia was attained (Fig. 4-19). During periods of peak abundance,- cryptophytes were concentrated in surface
'O waters, even when the lake was not stratified. In addition, vertical distri-bution within the upper 6 m was quite patchy (Fig. 4-20). During periods of peak abundance and throughout the stratified period, the motility of the major genera was apparently an important factor in regulating distribution of the Cryptophyceae.
Dinoflagellate peak abundances were restricted to epilimnetic waters. Vertical distribution within the upper mixed layer was quite patchy, and dinoflagellates were frequently concentrated at the surface or at 3 m (Fig. 4-21). Vertical migration by dinoflagellates in apparent response to light intens' y has been documented by Harris (1978). Blue-green algae were not a significant component of the algal population in 1978-79. However, vertical distribution data from 1974-75 (Fig. 4-22) indicate that the Myxophyceae were generally confined-to surface waters, with maximum abundance _at the surface or in the mid-euphotic zone. The dominant species was Anabaena wisconsinense, which possesses gas vacuoles. Gas vacuules can provide buoyancy to blue-green algal cells, allowing them to remain within the upper mixed layer, Vertical distribution data from 1978-79 (Fig. 4-23) also illustrate the concentration of blue-green algae in surface waters, particularly in the top 3 m. PRIMARY PRODUCTIVITY Measurements made utilizing the transparent-opaque bottle dissolved oxygen O technique ( App. Tables 4.3-1, 2, 3, 8, 9, and 10) provided results too variable to give significant information on the primary productivity of Lake Norman. Thus, all the results in this section are based on the MC uptake method (App. Table 4.3-4 and 4.3-5) . 168
Annual production in Lake Norman was approximately 110 g C m-2.yr-1 at Location 3 and 130 g C.m-2 yr-1 at Location 8. Carbon fixation rates on a daily basis ranged from 8 to 860 mg C m-2 d-1, averaging 210 mg C m-2.d-l at Location 3 and 330 mg C m-2.d-l at Location 8. Daily fixation rates varied $ seasonally (Fig. 4-24), peaking in midsummer and declining to a midwinter minimum. Utilizing Wetzel's (1975) system of classification, lower Lake Norman can be classified as oligo-mesotrophic, based on measurements of mean primary productivity, phytoplankton biovolume and biomass, chlorophyll a, dominant classes of phytoplankton, light extinction coefficients, total organic carbon, total phosphorus, total nitrogen, and total inorganic solids. This classifi-cation agrees with that proposed for Lake Norman by Weiss and Kuenzler (1976). Maximum carbon uptde rates in the water column (Pmax) ranged from 0.3 to 28.0 mg C m-3 h-1 (Fig. 4-25). Other than very low values observed in January and February 1978, no seasonal trends were evident. Maximum assimilation ratios in the water column ( ARmax) ranged from 0.2 to 12.0 mg C mg chla-1 h-1 (Fig. 4-26). Assimi.ation ratios gradually increased from a winter minimum to a mid May-early June maximum of approximately 12 mg C mg chla-1 h-l. Values fluctuated from 4 to 8 mg C ng chla-1 h-1 through mid August, then stabilized at approxi-ma tely 3 mg C mg c hl a.- 1 h-l . Beginning in late September, assimilation ratios gradually increased to a peak of approximately 6 mg C.m h-l., then declined to a winter mean of 3 to 4 mg C mg chla-1 a.-l.g chla-1 increase An Isolated in the assimilation ratio was observed at LocatTon 8 on December 6. Phytoplankton standing crops at a given location are the result of reproduction, death, immigration and emigration. Reproduction and growth rates are, in turn, controlled by light, temperature, nutrient availability and the specific g physiological characteristic; of the dominant algae in the population, while death rates are controlled by the above factors plus predation. Immigration to, and emigration from the population may occur horizontally due to the movement of water through the reservoir, and vertically due to sinking or turbulent resuspension of the phytoplankton. Algal motility or buoyancy may also con-tribute to immigration and emigration. GROWTH AND REPRODUCTION Photosynthetic rates were measured in-situ in Lake Norman via the 14C method, to approximate the potential for algal growth and reproduction. Information about the factors controlling photosynthetic rates can be gained by examining three characteristics of the curve which results when photcsynthetic rates are plotted against light intensity (Fig. 4-27). This plot is generally referred to as a P vs. I curve. Because community photosynthesis is dependent on the amount of chlorophyll initially present in the populations, photosynthetic rates are here expressed as assimilation ratios (mg C mg chla The first useful char-acteristic of the P vs. I curve is the initial-1 slope, h-1). which is basically a measure of how efficiently the phytoplankton community was utilizing light. Photosynthetic efficiency in this light-limited, linear portion of the curve is a function of adaptation to light, and of the taxonomic composition of the population, in that different taxonomic groups contain different ratios of the various photosynthetic and accessory pigments. Species adapted to low light O 169 f t
g intensities generally utilize low light very efficiently but become light-U saturated at relatively low light intensities. Thus, high values for the initial slope of thc P vs. I curve are thought to correspond to adaptation to low light intensities. The initial slope of the P vs. I cerve is generally considered to be temperature-independent (Harris 1978). Photosynthetic efficiencies for Lake Norman, corresponding to the initial slope of the P vs. I curve, evidenced no clear seasonal trends, with the exception of relatively high efficiencies during May (Fig. 4-28). The lack of seasonal trends indicates that no consistent pattern of scasonal adaptation to light was occurring. Rather, fluctuations in efficiency may have been the result of short-term adaptation, which can occur following two or three cloudy days (Harris 1978). In addition, the seasonal succession of various taxonomic classes (Fig. 4-18, 19, 20, 21, 23) did not appear to consistently affect photosynthetic efficiencies. Thus, differing ratios of photosynthetic and accessory pigments did not appear to have a signifi-cant impact. The apparently high photosynthetic efficiencies observed in May were probably artifacts of the operational characteristics of the Montedoro-Whitney licht meter, which was used on May 3,10, and 17,1978. The Montedoro-Whitney meter measures solar radiation in the infrared as well as visible range. Because infrared light is absorbed in the water column much more rapidly thaa visible light, the Montedoro-Whitney meter overestimates the rapidity of light extinction in the water column as compared to the Licor and Kahlsico meters, which measure only visible light. Estimates of light intensity in the water column based on Montedoro-Whitney measurements of light extinction would thus be low compared to those based on Licor or Kahlsico measurements, resulting in the observation of apparently high photosynthetic efficiencies. g The second characteristic of the photosynthesis vs. light intensity curve is Pqax ( AR m ax, when expressed per unit chlorophyll a_), the light-saturated rate M photosynthesis. This is the value at which increasing light intensities fail to cause an increase in photosynthetic rates. By c.efinition, ARmax is not light dependent (Fig. 4-27). Temperature and nutrient availability affect the maximum rate of photosynthesis achieved under light-saturating conditions (Harris 1978). In Lake Norman maximum assimilation ratios generally followed a seasonal pattern (Fig. 4-26), incra sing from a late winter minimum to a maximum in late May-early June. A second, smaller peak was observed in mid October-early November, followed by a decline, then a sharp peak December 6, af ter which assimilation ratios dropped immediately to pre-peak levels. As previously stated, AR max is basically a function of temperature and nutrient availability. A linear regression of photosynthetic rate (ARnax) with tempera-ture indicated that, from December through May, temperature could account for approximately 72% cf the variation in photosynthetic rates. From January through early April, nutrients were irregularly abundant, declining rapidly following the onset of stratification (Fig. 4-29). During the period from December through May, a linear regression of ARmax with total phosphorus revealed that total phosphorus could account for only 315 of the variation in ARmax; nitrogen accounted for less than that. Following the onset of stratification, any relationship between temperature and ARmax was masked, as temperature continued to increase and ARmax underwent a gradual decline. This tends to indicate that ARmax was no longer temperature limited, but rather was related to nutrient availability. The changes in concentrations of phosphorus and nitrogen support the idea of nutrient limitation: concentrations of total o phosphorus dropped to virtually undetectable levels following the onset of O stratification, while concentrations of inorganic nitrogen declined gradually 170
throughout the summer (Fig, 4-29). Nutrient concentrations remained low until the completion of destratification in the late fall. The increase in ARmax & during destratification indicates that mixing of nutrient-depleted waters with W the relatively nutrient-rich waters of the hypolimnion increased nutrient availability, which in turn increased ARmax. Additional evidence lies in the fact that destratification progressed fairly uniformly until about October 18 (Fig. 4-30). A stable thermocline developed at about 22 m, and this was not disrupted at Location 3.0 (a shallower location) until November 1 and at Location 8.0 until November 22. Within 2 weeks following this final disruption, an increase in ARmax occurred at both locations, followed by a subsequent decline. Ample evidence exists in the literature to indicate that, within the context of light and temperature regimes, phosphorus availability generally sets the limits to algal production in lakes (Schindler 1971; Deevey 1972; Schindler and Fee 1974). In Lake Norman, the molar concentrations of total nitrogen and phosphorus during the summer indicate that phosphorus was deficient relative to nitrogen for algal production. Molar ratios of total nitrogen to total phosphorus ranged from 55:1 to 300:1 during the stratified period. Ratios greater than 16:1 suggest phosphorus limitation of production (Redfield et al. 1963). The simultaneous decline in mean annual concentrations of chlorophyli a_ and total phosphorus over a 5-year period, as well as the observation of higher chlorophyll a, concentrations at uplake locations, where concentrations of total phosphorus were higher, provide additional evidence of the importance of phos-phorus in regulating algal production in Lake Norman. The third characteristic of the photosynthesis vs. light intensity curve is Ik, the saturating light intensity, or the light intensity at the intersection of the line defining the initial slope of the curve, and the horizontal line g intersect..g Pmax (Fig. 4-27). The value of ik is dependent on both the initial slope of the curve and on P max. Thus, i k is dependent on temperature, nutrient availability, light adaptation, and pigment ratios within the popu-la tion. No clear pattern emerged in examining the saturation light intensity values from Lake Norman (Fig. 4-31). . Thus, it appears that the max' ium growth rates of algae incubated in situ on Lake Norman were dependent on temperature during late winter and early spring. During M'c remainder of the year nutrient availability, specifically phosphorus availab'l'.ty, appeared to restrict the maximum photosynthetic rates attained by the paytoplankton. Low light limitation of the entire water column, in which subsurface incident intensities were insufficient for algae to attain light-saturated photosynthetic rates, was eocountered only three times, on January 25, March 8, and December 20, 1978. This is based on. Harris' (1978) contention that saturating light intensities generally approximate 120 uE m-2.sa. The estimation of photosynthetic rates from in-situ suspension of phytoplankton provides a means of assessing the relative importance of temperature, nutrient availability, and light adaptation in regulating photosynthetic rates. However, this method provides no means to assess the effects of vertical circulation which would occur under natural conditions. Phytoplankton, particularly those with no means of controlling their position within the water column (algae without flagella, gas vacuoles, or other means of buoyancy regulation), are subject to sinking and turbulent resuspension, and are generally uniformly distributed within a well-mixed water column. One primary impact of vertical 171 l
circulation is to influence the degree to which the phytoplankton are exposed to light. -_The relationships between tha depth of maximum circulation, daily incident light, and the degree of light penetration in the watei column regulate the relative rates of production and respiration in the algal community.
.Sverdrup (1953) quantified these relationships, creating the concept of the critical depth, defined as the theoretical maximum depth to which an algal community can circulate and still increase in abundance due to growth, under given conditions of incident light and light penetration. Thus, if circulation occurs to a depth exceeding the critical depth, respiration will exceed pro-duction, preventing increases in algal abundance due to growth.
Based on the calculation of theoretical critical depths (Sverdrup 1953; parsons et al._1969) for Lake Norman, it appears likely that -light and mixing, in addition to temperature, were important factors in controlling increases in algal abundance from late fall through early spring. Critical depths from January through mid March 1978 and from mid November 1978 through January 1979 averaged less than 5 m (Table 4-7), while circulation was occurring from lake surface to bottom. Mean lake depth is greater than 10 m; at Locations 3.0 and 8.0, the maximum depths are generally greater than 25 m. The dominant algae during this period were diatoms, which do not have flagella or gas vacuoles as means of regulating their position in the water column. Critical depths began to increase during the spring, as incident solar radiation increased and light extinction coefficients of the water decreased. Also at-this_ time, thermal stratification began to develop, substantially decreasing the maximum depth of circulation. Critical depths rapidly exceeded the depths
=
of maximum circulation of epilimnetic algae (Table 4-7), making it very unlikely that light was limiting increases in algal abundance. This situation persisted throughout the stratified period (May through October). In addition, the dominant algae during the stratified period were either flagellated (primarily dinoflagellates) and thus were potentially able to regulate their position at optimal light intensities, or were very small (small green algae) .with increased surface to volume ratios to decrease sinking rates, further reducing the likelihood of growth limitation due to low light intensities. During the stratified period, the potential also existed for a reduction of algal productivity due to photoinhibition, caused by excessively high light intensities in the epilimnion. Mild photoinhil'ition was frequently observed at the surface in Lake Norman (Fig. 4-32), but presumably natural populations did not-remain suspended at the surface long enough to andergo substantial photoinhibition, due to circulation within the epilimnion. Thus photoinhibition was not considered to be an important factor limiting algal populations. RESPIRATION, IMMIGRATION AND EMIGRATION The rate of change of the phytoplankton standing crop is controlled not only by reproduction and growth, but by respiration, death, grazing, immigration and emigration. Neither phytoplankton-loss rates nor immigration were measured in this study; however, values for apparent respiration (R') were back-calculated based on changes in the standing crop of algal carbon in the mixed layer, and ' on mean daily photocynthetic rates. These values were then used to examine whether photosynthesis and respiration alone could potentially account for
-O observed changes in algal standing crop during any particular time period, 172
or whether it was necessary to postulate the substantial occurrence of grazing, immigration or emigration. Ratios of apparent respiration (R') to maximum g photosynthetic rate (Pmax, mean for the time period of interest) were compared W to published values for the ratio of measured respiration (R) to Pmax. Measured ranges of R:Pmax for specific taxonomic classes were obtained from Harris (1978). If the calculated R':Pmrx values for Lake florman fell within the range of published values for R:' max, it was assumed that photosynthesis and respiration alone .ould potentially account for observed changes in standing crop. As a second means of checkin1 the comparability of R':Pmax values with those measured by other workers, values for Lake Norman were superimposed in a plot published by Harris (1978) of R:Pmax values vs. the ratio of euphotic depth to mixed depth (Fig.4-33). Where R':P m ax values wera close to those predicted by Harris' plot, it was again assumed that photosynthesis and respiration could reasonably have accounted for observed changes in standing crop. During the period in which the lake was stratified, virtually all changes in algal standing crop as measured by algal carbon could potentially be accounted for by the relative rates of photosynthesis and respiration (Table 4-8). However, when the lake was isothermal, R': P m ratios were uniformly less (0.01-0.02) thanthosemeasuredbyotherworkxers for diatoms (0.04-0.08), which were dominant during the isothermal r driod. This suggests that populations were increasing too rapidly or were not declining rapidly enough to be explained by the intrinsic photosynthetic and respiratory rates reported in the literature for populations of diatoms. Two potential explanations for this are 1) that passive immigration may have been an important factor in increasing algal abundance, and 2) that Melosira italica, the dominant algal species during the isothermal period, may possess a characteristic R:pmax ratio somewhat '.ower than that reported for other diatoms. h Evidence for the importance of non-metabolic factors such as passive inmigration and emigration in the population dynamics of Melosira italica was reported by Lund (1954,1955) and Knoechel and Kalff (1978). They observed that, due to the ccmparatively rapid sinking rate of M. italica, the major Tactors affecting its abundance were tui ulence and the mixing regime of the water column. Lund (1954, 1955) observed that Melosira filaments sank rapidly durino periods of low turbulence, sinking all the way to the sediments at the onset of stratification. He presented evidence that cells could remain alive and viable on the sediments for periods of up to 3 years, even under anaerobic conditions. When turbulence was sufficient, for example, during destratification, filaments were apparently resuspended in the water column, sometimes in gre-t numbers. In Lake Norman, I some of the trends in abundance of M. italica app ared to be associated with water l movement and changes in turbulence. A rapid decline in abundance was observed l in mid April (Fig. 4-34), following the appearance of initial, unstable thermal i stratification. Although this decline could potentially be explained by
- relative r ites of photosynthesis and respiration (Table 4-8), its close correlatian with the beginnings of stratification and with a simultane us decline 1n turbidity, plus observations from the literature, tend to indicate that Melosira cells were merely sinking rather than dying. This idea is also supported by the observation of what appeared to be a viable population of Melosira italica in the hypolimnion (Fig. 4-34). (Silica limitation did not appear to be a factor in the mid April decline in abundance, as concentrations of silica did not f all below 2.1 mg.1-1; silica is potentially limiting to the i
growth of M. i talica a t concentrations < 0.8 nc 1-1 (Lund 1955).) The reap-pearance of Melosira as a dominant in late f all-early winter corresponded to g 173 f
. - - - - - - - - - - - - - - - - - - - -- ~ -
1 pd the mid and final stages of destrattiica+ ion, with the largest increases occurring in bottom waters imediatel/ following the completion of overturn , (Fig. 4-34); this suggests that resuspension was occurring. Passive immigration ' may also have contributed to the rapid increase in abundance in early spring. l l Peak abundance occurred earlier at uplake locations in 1979 (no data is available -
, from 1978 at uplake locations), sugcesting that an influx of cells from uplake
- may have contributed to peaks observed at downlaka locations (Fig. 4-35).
However, mo , reset.rchers noting early spring increases in populations of
; Melosira att ibuted thew peaks to increasing solar radiation (Lund 1955; !
Pechlaner 1970; 7.nocck and Kalff 1978). It is likely that the spring peak i in Lake Norman, also, was primarily the result of a growth response to incrcasing , light intensities, This is particularly evident at uplake locations, where inputs , from the Catawba Rher did not appear to be substantial (Fig. 4-35). However, the j spring peak cannat be entirely explained by relative rates of photosynthesis and respiration unless it is assumed that the R:pg3x value characteristic of Melosira t
,ilalich is somewhat lower than for other diatoms. Assuming that passive inni-gration was not substantial, the very eccurrence of a spring peak in abundance at a time when critical depths were exceeded by mixing depths suggests a reduced i R:P ax ratio. Evidence for very low respiration rates also lies in the obser-i vatTon of an apparently)v;able the phctosynthetic zone during the periodpopulation in w&ich the lake of !ielosira was thermally in the hypolimn stratified. Other workers have observed large populations of various species
, of Melosira on the sedimerts (Lund 1954; Stockner and Lund 1970) or in the
~
l hypoTiiiinion below the photosynthetic tone (Talling 1957), ; To summarize, the relationship rf relative rates of photosynthesis and O respiration to the standing crop of phytoplankton was dependent on the mixing regime of the lake. During the period in which Lake Norman was stratified, i photosynthesis and espiration could account for all of the observed changes in standing crop, although predation, immigration and emigration were un-doubtedly occurring to some extent. Maximum photosynthetic ratet and standing crops were in turn apparently limited by the availability of phosphorus. During 4 the period in which the 15ke was isotnermal, maximum photosynthetic rates of populations suspended in-sitt. were p'imarily a function of tempereture; however, light was apparently an important fMar in controlling the growth rates of natural populations subject to circulation to bottom oepths. Changes in standing crop during the isothermal period were apparently the result not only of growth and respiration, but of sinking and resuspension and a possible influx of cells from uplake.
SUMMARY
e The phytopiankton community was sampled monthly on Lake Norman from March 1974 through- February 1975 and from October 1977 through December 1980. Population-samples were examined at 500X to determine species composition and abundance. Chlorophyll- sampii s were collected monthly from January 1974 through December
-1980. Samples were filtered, extracted in acetone, and analyzed fluorometrically.
and/or spectrophotometrically. Primary productivity was estimated monthly in i 4 1974 using the in-situ transparent-opaque bottle dissolved oxygen technique,
- and weekly in 1978, using the in-situ transparent-opaque bottle D C assimilation
; technique. Samples to characterize the vertical distribution of the phytoplankton Q were collected at 10 depths weekly in 1978.
174 L
. _ _ ~ . ~ . - _ - - - - - ._........_~..,_., ,._-,_-..-.., _- _._ - __,. m ___ _ . _ .. - _ . . -
The major taxonomic classes of phytoplankton observed in Lake florman were the Bacillario syceae (dominant throughout the winter), the Cryptophyceae a (abundant in late spring), and the Chlorophyceae and Dinophyceae (dominant in W sumer and fall). Ten classes and over 300 species and varieties of phyto-plankton were observed, with f4annochloris, Rhodomonas Melosira and Peridinium
~
among the most abundant genera. Oue-green' aliae Triyxophyceae) were~reWtTveTy abundant in 1974 but not in 1978 or 1979. Phytoplankton abundance, as measured by mean annual concentrations of chloro-phyll ,a, was f airly consistent among locations south of the Davidson Creek-Catawba River cMnnel confluence, but increased at uplake location 3. Differences 1.1 taxonomic composition among locations were evident primarily in the distri-bution of diatoms, which were most abundant uplake and least abundant at Pamsey Creek locations. liowever, differences in taxonomic composition among downlake locations were generally outweighed by similarities in temporal variation. Temporal variation in phytoplankton abundance was comparatively low, chlorophyll annual maxima rarely exceeded minima by more than an order of magnitude. During the early phase of the study (1974-1976) annual maxima were ubserved in mid summer. In 1978 and 1979 a bimodal tendency was observed, with small peaks in early spring and late summer-early fall. Mean annual chlorophyll a, concentrations declined throughout the study period, possibly due to declining concentrations of total phosphorus. Seasonal succession of phytoplankton exhibited a consistent pattern in 1978 and 1979. Diatoms, small cryptophytes, and small coccoid green algae dominated late fall and winter populations. Diatoms peaked in early spring, followed by cryptophytes in late spring. Small green algae and dinoflagellates dominated populations during the stratified period. Both greens and dinoflagellates $ attained maximum abundance in late summer-early fall. Phytoplankton were patchily to uniformly distributed in the water column during t;a isothermal period. During the stratified period, phytoplankton were heavily concentrated in the epilimnion, with epilimnetic chlorophyll concen-traticns as much as 30 to 40 times higher than hypolimnetic concentrations. Cryptophytes, dinoflagellates and blue-green algae attained maximum densities in surface or epilimnetic waters. Green algae were fairly uniformly distributed in the water column during the isothermal period, but were concentrated in the epilimnion during the stn !ified period. The vertical distribution of diatoms was patchy to uniform durug the isothermal period, but diatoms maintained higher densities in the hypolimnion than in the epilimnion during the stratified period. Lake Norman is probably ',est characterized as olino-nesotrophic. from 1975 through 1979 chlorophyll
- concentrations in st . ace waters ranged from 0.3 to 18.0 mg m-3, and biovolumi i:ng.d rom <100 to 4000 mm3 m-3 Annual primary production in the downlake area was measured at approximately 120 g C m-2 yrd.
Daily primary production peaked in mid summer, coinciding with maximum day length, light intensity, and light penetration Production rates of the phytoplankton community were apparently regulated primarily by light and temperature during the isothermal period, and by the availability of phosphorus during the stratified period. O 175
Much of the variation among locations in total abundance and in taxonoinic i composition was based on the relative distribution of diatoms, prirmrily Melosira italica, as was most of the variation between years in the size of the s;7Tng peak in abundance. In addition, temporal and vertical variation during the isothermal period were largely a function of the distribution of Melosira italica. The distributict. and abundance of Melosira italica were ~~ functions not only gf growth and respiration, but of water movement and turbulence. . i i 1 O I f O 176
LITERATURE CITED American Public Health Association. American Water Works Association, and 9 Water Pollution Control federation. 1971. Standard methods for the examination of water and wastewater,13th edition. American Public , Health Association, Wa hington, DC. 874 p.
. 1976. Standard methods for the examination of water and wastewater, 14th edition. American Public Health Association, Washington, DC. 1193 p.
Bourrelly, P. 1968. Les algues d' eau douce Tome !!: les algues jaunes et brunes. Editior.s N. Boubee & Cie, Paris, France. 438 p.
. 1972. Les algues d' eau douce Tome 1: les algues vertes. Editions N. Boubee & Cie, Paris, France. 572 p.
Cocke, E. C. 1967. The Myxophyceae of florth Carolina. Edwards Brothers. Inc., Ann Arbor, M1. 206 p. Conover W. J. 1971. Practical nonparametric statistics. John Wiley and Sons Inc. New York, NY. 462 p. Deevey, E. S., Jr. 1972. Biogeochemistry of lakes: major substances,
- p. 14-20. In G. E. Likens (ed.) Nutrients and eutrophication: the limiting-nutrient controversy. American Society of Limnology and Oceanography, Inc. Publications Office, Aller Press, Inc. Lawrence, KS.
Duke Power Company. 1976. McGuire Nuclear Station, Units 1 and 2, Environmental Report, Operating License Stage. 6th rev. Volume 2. Duke Power Company, h Charlotte, NC.
. 1977. Phytoplankton, p. 193-239. In Duke Power Company. Oconee Nuclear Station environmental summary report 1971-1976. Volume 1. Duke Power Company, Charlotte, NC. 405 p.
Eddy, S. 1930. The fresh-water armored or thecate dinoflagellates. Trans. Am. Microscopical Soc. 49(4): 277-321. Gurtz, M. E. and C. M. Weiss. 1972. Field investigations of the response of phytoplankton to thermal stress. Department of Environmental Sciences and Engineering, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC. ESE Pub. No. 321. 152 p. Harris, G. P. 1978. Photos,9 iesis, productivity and growth: the physiological ecciogy of phytoplankton, \rch. Hydrobiol. Beih Frgebn. Limnol. 10(l_IV): 1-171. l Helwig, J T. and K. A. Council (ed.). 1979. SAS user's guide,1979 editica. SAS Institute, Inc., Raleigh, NC. 494 p. Huber-Pestalozzi, G. 1 941. Das phytoplankton des susswassers 2. Teil 1. Halfte. E. Schweizerbart' sche Verlagsbuchhandlung, Stuttgart, Germany, 365 p. a W l 177 l
1968. Das phytoplankton des susswassers 3. Teil 2. Auflage. E. O . S&eizerbart' sche Verlagsbuchhandlung, Stuttgart, Germany. 322 p. Hustedt, F. 1930. Die susswasser-flora mitteleuropas Heft 10: Bacillariophyta (Diatomae). Verlag von Gustav Fischer, Jena, German Democratic Republic. 466 p. Kim. Y. C. 1967. The Desmidiaceae and Mesotaeniaceae in North Carolina. Ph.D. Thesis. North Carolina State University at Raleigh. Raleigh, NC. 126 p. Knight, R. L. 1973. Entrainment and thermal shoci ef fects on phytoplankton numbers and diversity. Department of Environmental Sciences and Engineering, School of Public Health, University of Not th Carolina at Chapel Hill, Chapel Hill, NC. ESE Pub. No. 336. 73 p. Knoethel, R. and J. Kalff, 1978. An in situ study of the productivity and population dynamics of five fre.hwater planktonic diatom species. Limnol, and Oceanogr. 23(2): 195 218. Lund, J. W. G. 1954. The seasonal cycle of the plankton diatom Melosira italica (Ehr.) Kutz. subsp. subarctica 0. Mull. J. Ecol. 42:W TTs.
~. 1955. Further observations on the seasonal cycle of Melosira italica (Ehr.) Kutz subsp. subarctica 0. Mull. J. Ecol. 43: 90W02.
O teed, J. W. G. , C. K4Piin9, eee E. O. te Cren. 1958. The 4 everted m4croscoPe method of estimatir.g algal numbers and the statistical basis of estimations by counting. Hydrobiologia 11: 143-170. Menhinick, E. F. and L. D. Jensen. 1974. Plankton populations, p. 120-138. In L. D. Jensen (ed.) Environmental responses to thermal discharges from Pirshall Steam Station, Lake Norman, North Carolina. Electric Power Research Institute, Cooling Water Discharge Research Project (RP-49) Report No. 11. Johns Hopkins University, Baltimore, MD. 235 p. Meyer, R. L. 1971. A study of phytoplankton dynamics in Lake Fayetteville as a means of assessing water quality. Arkansas Water Resources Research Center University of Arkansas, Fayetteville, AR. Publication No.10, 58 p. Palmer, C. M. and T. E. Maloney. 1954. A new counting slide for nannoplankton. Am. Soc. Limnol, and Oceanogr. Special Publication No. 21. 6 p. Parsons T. R. , K. Stephens, and R. J. LeBrasseur. 1969. Production studies in che Strait of Georgia. Part 1. Primary production under the Fraser River Plume February to May,1967. J. exp. mar. Biol. Ecol. 3: 27-38. Patrick, R. and C. W. Reimer. 1966. The diatoms of the United States Vol. I The Academy of Natural Sciences of Philadelphia, Philadelphia, PA. 688 e Pechlaner, R. 1970. The phytoplankton spring outburst and its conditions in Q LakeErken(Sweden). Limnol, and Oceanogr.15(1): 113-130. 178
I Prescott, G. W. 1962. Algae of the western Great Lakes area. Wm. C. Brown Company Publishers, Dubuque, IA. 977 p. g Redfield A. C., B. H. Ketchum and T. A. Richards. 1963. The influence of organisms on the com;)osition of sea-water, p. 26-77. In M. H. Hill (ed.) The sea: ideas and observations on progress in the study of the seas. Vol. 2. The composition of sea-water; comparative and descriptive oceanography. Interscience. New York, fly. Rodgers, J. H., Jr. 1974. Thermal effects on primary productivity of phyto-plankton, 9eriphyton, and macrophytes in Lake Keowee, South Carolina. M. Sc. Thesis. Clemson University, Clemson, SC. 88 p. Schindler, D. W. 1971. Carbon, nitrogen, and phosphorus and the eutrophication of freshwater lakes. J. Phycol. 7: 321-329. Schindler, D. W. and E. J. Fee. 1974. Experimental Lakes Area: whole-late experiments in eutrophication. J. Fish. Res. Board Can. 31: 937-953. Smith, R. A., A. S. Brooks, and L. D. Jensen. 1974. Primary production,
- p. 139-147. In L. D. Jensen (ed.) Environmental responses to thermal discharges friini Marshall Steam Station, Lake fiorman, North Carolina.
Electric Power Research Institute, Cooling Water Discharge Research Project (RP-49) Report No.11. Johns Nopkins University, Baltimore, MD. 235 p. Stockner, J. G. and J. W. G. Lund. 1970. Live algae in postglacial lake deposits. Limnol. and Oceanogr. 15(1): 41-58. h Strathmann, R. R. 1967. Esti , ting the organic carbon content of phy*oplankton from cell volume or plasma volume. Limnol. and Oceanogr. 12(3): ' 418. Strickland, J. D. H. and T. R. Parsons. 1972. A practical handbook of seawater analysis. Fisheries Research Board of Ccnada, Ottawa, Canada. 310 p. Stumm, d. and J. J. Morgan. 1970. Aquatic chemistry. Wiley-interscience, New York, NY. 583 p. Sverdrup, H. U. 1953. On conditions for the vernal blooming of phytoplankton. J. Cons. Int. Explor. Mer. 18: 287-295. Talling, J. F. 1957. Some observations on the stratification of Lake Victoria. Limnol. and Oceanogr. 3: 213-221. United States Environmental Protection Agency. 1975. Report on Lake Norman, Catawba, Iredell, Lincoln, and Mecklenberg Counties, North Carolina. Pacific Northwest Environmental Research Laboratory, National Environmental Research Center, Corvallis, OR. 17 p. Vollenweider, R. A. 1974. A manual on methods for measuring primary productivity in aquatic environments, 2nd ed. Blackwell Scientific Publications, Oxford, Englar.d. 225 p. O 179
i i
'A Weber. C. I. 1971. A guide to the common diatoms at water pollution surveillance f V system stations. V. S. Environmental Protection Agency. Cincinnati OH.
98 p. l Weber. C. 1. (ed.). 1973. Biological field and laboratory methods for l measuring the quality of surface waters and effluents. U. S. Environmental i Protection Agency. Cincinnati OH. ' [ Weiss, C. M. and T. P. Anderson. 1978. Belews Lake: A summary of a seven f year study ( August 1970-June 1977) to assess environmental effects of a coal-fired power plant on a cooling pond. Department of Environmental i Sciences and Engineering. School of Public Health. University of North ; Carolina at Chapel Hill. Chapel Hill. NC. ESE No. 475. 138 p.
' Weiss. C. M. , P. H. Campbell. T. P.- Anderson, and S. L. Pfaender. 1975. The- -
lower Catawba lakes: Characterization of phyto- and zooplankton coninunities and their relationships to environmental far. tors. Department of Environmental Sciences and Engineering. School of Public Health. University of North ; Carolina at Chapel Hill, Chapel Hill. NC. ESE Pub. No. 389. 396 p. Weiss. C. M. and E. J. Kuenzler. 1976. The trophic state of North Carolina i lakes. Water Resources Research Institute of The University of North Carolina. Raleigh. NC. 224 p. Wetzel, R. G. - - 1975. Limnology. W. B. Saunders Company. Philadelphia. PA. 743 p. [ ' O Whitfore. '. 4. ane c. J. Schumecher. Sparks Press. Raleigh NC. 324 p. 1973. A menual of fresh-weter aisee. Wilde. E. W. and T. A. Paulishen. 1974. Phytoplankton study p. 277-357. , In Industrial Bio-Test Laboratories. Inc._ A baseline / predictive environ-ihental investigation of Lake Wylie. Industrial Bio-Test Laboratories. Inc. Northbrook. IL. t l I i l O 180 l
-.-,-..-_--_..,--_,,-.-.__.--.,--,---~.__.--._,,,_,-_,._.,__...,-...,__,--....,,_,-__-__.....,..m. ,.,,-.,...-m_-.-r .
L' l l W el l l l ! 6
! e' ; . _ _.- _ w.--..._ w - . .-. .
l F. I s. 3 t e_ , . . _ , . _ + . _, ,. . .._4__,. - . . ., . j ( , { >l "
. I ,
I E! l
~4_ . _
I I J._. ..-__w. l I
-. ; _4_
i
.d' _4 4_ . _, _ -c .j ;.
3e ,.l t - Ei B i 3li W' { l t 7$...9....-. _ 9 _.4. ._ 4 . _ , _..._._4..f+...._ : pi a i 4 . l , f_._._. _ _ . . _ . . _ _ . - --_ - J
** ll. -9.. .d_+
l
.~.-4 l .
j _ _ 4 _. . -4. . l
.-4 -.
C l I f 1 l
~._.I_..+;_......,__1 + d' g "1 .._. . _ i ._ . .. ts t.
C- l l I
,. I I A 1 e-['
_4_ __ .Ji __a .__.--J j._ .f . ._
~]
3 i . 8 e - l s- i e n. . . _. % . 9 el N r,
,, i a
[- j m , _. ,%__._4____.4_.4_._, j o; e. ! i j 6 m 1- E 0g E ei l l 1 l j
- * - - - ~4__4--..-9_.. --p_.-, . , _
g N k h N 3 V $ k-Y e
- - - - __I ~ ,t ~__ i o .r. ] v e .. j e Wl *l _;_
g - -
, , i
.- y y! y i a a!
< l l ,
4 4 - J . 1
@ l' 8 m .I p,
7 i y g p. g , i , p ..i__j lI , 0 , -4
,. ; e. _
- co el e, ,,
a ,s /j # 3s c. 1 mm gi' a r= I. K L L ve' y 0 Q) i .s,>4g
- 1 y- ~ es tw
.e. y cl l es y vi to E
- 4-t g g gj l k 3 3-I I M - f---4i t--
e
.e
- u
..e. c m l l _%4-
-p .Q ,, f -
E e. > V O p 9 -. f. w .c _ 4_ _ _ _ . - V lV V WI V' V 2 in $ c
- l .U
. 1 a .o asn -- --
- e. D
.d Ch o e. aj
.i nl s ca c.,-
- 3. E t
- g p C- g K g g m
w -+--+I t ci s. ~ s , g y m j c. .a .Oi5 -- -
-y gg .I T
.c m !; c; y ev
* .t : L ;
o g
- o. O ,
.__-_._-4_._,-..-_ . . _ __._.;.,
y .. ,
- n I e 9 re r ej r e i .L_t... . _ _ __3: )
- r. rt :
n - . m _t_ ._.a _ v _. I I l t
, p . .. -m t. L 0 -m h* ..
o , r,, . r. :., l;e ;o
- r - L 4,.
0 * * '
- 4 s,, t f., %. , ;k
.' .c R r gw ,T a
N g g, i f. f
- s "gg s. - c- s r. s .a -
m 4 'tLt .,%~ .. tdv .r H A " t5
.ws r. h C lA1 il' t -[- .I ;
f - M r. 4 *m~g c =
*- e
{ e
=
t. g -l. p. M, , S i;1, it.: .ta ly ;3 ., iv ,, .
- m. . h ~L - t .t. ---
g
+ ,. w t, yd. d b p== p= **e-11 *-
o r r , U_ . p-. pe s I 6*}@ i h 4'
.c. 1 s
4J % Od 4 I i c. .' Y . M h '::i+72T ft ;F j .. % O t n wv: . t y r.h'r ww;,;,;;/;&. r ::.a[k : h,;th o. < ;; Ta n. .r pr'. w W' 1. g17 r .a. .s v "! x : u, e,
- .p # c c p e ~ e - o- & <3 s ,_
.,Je "- aj w r 3 t.,-tc
- r. 4%* a .g l .y..oV Wc. s . - ;.;;...r % L. m.
.eyr 4 ,r c U.; r v t.r 2. ;Ls,J f,
_b._?L - , . .&+o*5m. - .-_r 5 - s c- I- y4" 6 - 6 v "7? fi 7.E 7 ?e,1 *P
~r E N E "EDY . v $ n l Plf K 6 .~. $ l*e; " l'. vv . e.,,r- o. . v. ,
h .m.=. ,. c u. ,6 u, e 3 . ~u. .s . eu ,c. t, < i
. . .,,.'.4 .. L.. x . w v ~ow b < sQ: - twe v .
e ~ 7 w > > h.m .e. I w w w o w a. !.- % c. o u n u a i
- w e 7 su m tv 181 i
O is
. . .. . .. \ .s sg +
gae a n (~ $ . t4 % f T/ } v'k J M'V g, _ gY' g
\r \
L '4 a g4 p '. i s. p e. -b, 6 w, e . u. ( i +p 3 i .a p o'O % 3,o e
@ gd y, p. \@ [ h . ,
4 q$[' .
' 9 @ . <41 pM' v 0,.e, ) '11 i I
U,+ t. ) 3 g(T'- M
- pe% \*g u*
d b- t t 1 @ $,' A N il i #
\d g
4,e
- ps yst0' q t Y d D I s'kyy- r l
w+pf ,t ji
%g' 4dJ ,
I ,e' ge5 e s
\\ d p . e 't.dE.I- .p,,. el* ' e * ,,4 g#
t St
> $ g ",g %#j ' .
g4 ',yv g.'pf ]9p, 4 L,e ,f- i 4 y '*ty O F - .
' ' , -,e e-s ,, e4*+ nej '
pdkg Y,,g W pe,k,4 #g, 4-y * . 4 g o' '3.N" I,i/1gpY-Qv0,e" o* ,$ W y1 ,
/g MS,.1 g A.c < ..- v;.. .
Wg,. g*4&y s* , , - 4 en /, . ,
- 4 3 # 4r's *. ,,# 0 y, *
' . ~gv9fsy t _93'\,#..gi ,MI .Tge# F p \ og g gu' O f. # ,
k,%,:(('$ p?,fg*b '*(*@' , # [s'* AQ . 1$ **[Ay 3 y'@ s 8V, pvp.*.s o'pYu- NcP., @ , 0,,. y y@ i-V .- C',q (. g- - ' st 14 e A g $d \ s' Dr h y e U y4 (> y'\f\go ,t i,e . p,i',s .'.rl, , .e + ,
, te 4.'i# et$ p vstO*. f' dpq,* ,e ,o 3 . j ',y J's p g p' \. ' 1 o .,k, t l g, $ ., g ,o ~
T 1 ,- . .
<% og , /
- F *
$ e g eS e i g
e / p R i . 4 a , -.
. 8% % ~
4 eb g .y J ,W, e n \ _ , a
<W /
4 g' y
- 4 A
I % I M
%L 4 *,1 4 !w u #4 ut '*s b kO sej s e t *98I Il Dl two bs t og kD a (Jtpd )(gg Nh 4 t %w=J t )
g g 44 g gs O gi 4 4 03rJM ir omg
- ) I u g a
di w n a e u ste
%n oQ m g t, g pH b lf u u. Ma A$ .3 ~ %d O bQi W) nus eIidA n i \$ o 4
t l ( t3b ,wHbn 3 0 -N a ng 4 171 4*D4 h v4 ) 1 u, m'IN g4('g 4 t' a O hD J*We d e...u3 (ty I U u ,tu.. tP* pa 4'k W, t i) DJ A ay tw6 i M wg) 4. 4 Wi+4d[+g34 kjQ hI I NIt o n3r iun _ _ ,, I'vi (N t at 9 yet k9k - 9 48% ,O] Jy, g'i,(( w(u t t s nr u ug Ma wDo a HI 4*b 'e uen w yun 4*L% G ebetg*1h 'g w mngtnp)eg 6 n 4f 5 a I m jg j p
%M"([* *Wu b up w wi4e p pi + u 'f g III ivi s gu ea gg g Mi 3 e go;yne n,9 p%p$i ~75 ite uv a jp g g g
t rl*% . guy yt 9gs 3 a *t g # U ( Jm )) t, ,h j 3 s f, "p* ou a nt
*_H_D hk t )d*g,,=yo,"g ,,P v (p %g hMah 4 d s W
41 1
% n )e r a t_
a #au b e i yt v* s J i {[i J n utge%q n>4 gi t M M8 04
'1 M upv 4 gg %
NM um7i ih w.nm N 4 n g ge e
% w st ap 4
6 h*1q@t p6,yJ4a qJ h "s o,g g 0 (2 0)*w. 5 ' 3 db A 4 (% v .up
%# dwgJg f y9c A
( a.**aflpGt "ifg th ' W) 4 te i no Ob'E W) 1 i w+4p+>1) Ob vtt v %11 tt h(54tI 1*% n N i.ag.u*tch"t i A8 l itt stu *\ I' D 4+ue idginE Es t y6 til Av e h 4 3 N )}t'e s n= nt; pu e pn S I # 5 % o n3 s# h i ed g 04 e nte n ( 9u% . ) uta4, tM edis r v vP W_ sMtgi'si gy glg wp py p.(a+Ob+w 4p g VJ ea a h)g i 0 ,4 ei e_ _ N a_n s _pu u g (t t) mu wyg ng ,9p )s UW)2 1 h g ggg nu!O Weitwt aug w @t Vil inire'h'u 4 O f d g M 43 % a d l 721 c (vm gin tg@%gedps,4 s~ M, 4%d Nwn l u% anH(rne &v (n1 u v b s u y g H5i dh c W ( '4 tu% su e,t w 4 3)4r.,0j fi *F*on ~w atot oa) t y qa.mthi INk A M]) . (1 d aw n )e a pg K%s wt tw bub ei i;o@ii,wb g w ae bu e.__ p 4) ie ata ne os u gnNyd ~ 3 w;s i gt ti e wTiluti Wgg s syne ;;s T r o b g ue n wo po n) )(d'_ Evb E(iWeda T%st (b l *"** yggl'a%pu, e -a e g;spa.p M u.%uts t e w ili giWi wh '4t h a il oe s_ iGi es.tyy dspi ui 9 s t (S ba lbnC g e wgb eyg, 4peg ,9 O aiv ( 'b w)u aa A sC f 4os d
'p bu.a.nu_ Wgd 6ngj )g_ oM'y ggte 4 pa pt i ]3 a
m.d Q pi ' equ )) w n p#g (w M6W aq) (4 g gg gg pA pai e tmtung p a M ,JJ y g Ot S M pg , p.m gug n @i ~girp.au g jgg )]' g blI m9d 04 ngua u.p,g_e1ng'g w gg ;s yu n yeM y g id g di e gi nMk3ki i l# e gi ct n u.%v64 agl4 4si,vi;^i g' gm U) MENab f(6 g&i 3y st a w 3 us.B;i3 s y a i Ch g iW 6: gggg m%ys F W uw on pO de N upo s gNn33 l [JkE"i k p gg u ivrytgu by 3 gid nim;id 1 g nn 29 gp o u 9 gg 4g g-w o ~iYsn.3sig;;n G ; n tnf .s4 yntog g sin;ca a ]w} st i g 'g wii gub o W ggy 9,p,uyn tan n u tw a 9 yp pp q3 g g C 9 q y 3) 9 (;,(g%n e e' nyg3 >da ;Ugsi g3 ~L' [wC b w9 g gyd g 5 g EjTin,s
'ig;5 g G ' q; M ab u g gw lvpo ut u(gj;ig t
wtg s gy
+ Wok Q g,ntw . yu o pgivgg]g qg pppa3.a jag ii gy g g (f nu g p in geg a p gsg gd g
pp
- u ) og,% ((u ,n ;g {;w ;is ]f g
,d E,~i,1m.m) c irsu 4g pgg-g hnp}q**9a b=1 v 4 4
4 {$gb
%2[y{a g ep %We~ )dih g m+4 i @] ;so g(B g
g- gn tT yf ;g 3 ag fpp nfig 3 g iep;qQq T y )) g n/ /-
I o d # 3" 1 31 G2s :rri7s t4 r S2: n,48in u0 N57r, e r 5. % ? 4n;ne?mMuAetN;u3 s.n7s t m o :w
; 5 4
- 0Os*Mi3ew377se7 u e s t 7 1 t
m 2 i ' *
* ' f a '4 f 8 13t '<" l s'ii 2r o m 'm o
1 r E' f e m g d e a e P v s~
- r t s -
e 3
- r, s
- m b r , t ~ ) h o m. M t ,
t. 1 a s e
,e e e+ + 4+ +
t e e . e z w r.
, s e .s t
- y a c y iw..
. n
- a t
'e <
s
~
w%e te5 n
- 0J a
o f e t v e'id==
)
e , r e t e 5' f f milc w
. ) % "- t' c1* ( te '";i s, a t l . e) 'a n to )s h*T t
e i
% s +
r oe t e ta c*! c t s
- w e
t a_d
- ck a )w .
s,
- t e
- sr
*( 'mia sdt - = a s .a t r e
sGe e N t= #
=:
a c e., ( nem u>*(h t ?e2a r rt r4 = e t,
=w 1* ic t
if
- t. )* -. 7+ r o
s" F-h m, 's *t em
'p) o t ea su,W'h. at -
- s t ie j y; e, w s** *" tW ei e t .
s d w 2 p~ ,.*
- F e e- r te ea c r y .
cy 5 . . t e t n (atse y; - t, Ct 1,I %tern
- a 2M 64%s t W q s
8~ i r R e t J a& ie- e 3+ irw - > tb J%+At
- ai ) -
mtyt e
- o 1 h na .
u t + A f ( t, I 2 a s e i ~ I m- *gs qg el ms(s r-t.
. )t n ~
g(w
- (
ms e4 r g.0 tg% w. c 3* r mT
. emeo-n +e gt . . saue. )t.
e pn a Glegt .r %
- o " e=
' ' a e ( eg?a I' #
t sMntW*I e
-a r
1
- e. *L v.**ai~ms+=
r - - gu v - aI s
- a' G.
- t ei
- i .t - =" y e e t l
a gr f s , d* *r
.s .
{( et e .W e 0 . t. eoe t ea 'qet mg n s, eCs es
- wmf ye t
w eeNmNat e t ( nf. rv - g f' n ns ( t 5 i. .s ttah h a h ewty " i,t yts, elvi ee n4 . ' n uf s2- M rsw i n ee,eti t
- e n_h e s ' - t e r t y'a* et t t %. ysn t, r**d
( 6 l t fi yrr*, s Le -.*e I a "* l yre e tui ee iTi i pppmriIys [*Te e - -
.::waW +itQt5 vnp M *r.n*
t t w
- y. - 5 s n. n gn +yy. 0eOgO_0CgrbwhmTi e .r t. o < hs
,t pt Nt* n e-g"&8 s ~
e_ ess+ r s e + LP*GrfP nf t.- y$e5 ia tE a
- vsupp%wrswaefemmr%M"*.9t' l n p* et y 3t r r ei e
s
- pS + e t l u it ar , , ,a }L .
L e._ i **
- _i*' m i"_M eo 1a. ~ __
F 0c$MGcg"a[,8 o t 8
- t cM O rd en pa en ma um l r oo 2
=
vN y o oI 4 9n%9;no92$HM 4 2 7 65ya49?2I3mM:uN%ns?J ae7 337
, 3 9
i e a3 SOT bk e 0
;p5 5 n %a7 s $ ist s : 'o n 6c? e ' ?S7 1 4 36577U? 7 4zW s e5 4 7a6m7?1i 3ss;2: %7 1 5t T a v 7 1 t
I 5 43?
?*nf 3 62T ?
nL lo, s a ic e en w mo B t re t n d9 s w e s n7 w a9 t n fw a 5 w 1 as . a b s e xr G se ae S
)
x t d h tb Gs
- t s a t e .
t nem t . e t S t sI m s) w e " e s w;
- e ig s oc n n "0 %w e t e n
e t, c" +n J
'1 *~ o v
s s t w"
- w dn l s pp m
d kD e t e I i3 w, i. f S r n a" f 4 . n ei1 e t r3 S n s
- t. oa id*.
Ti f e y s t
- 1. *<
G n ah =h g r s e l g ou
=
a r t e e ( cM e"Mt r a s m e
. n . it o.
n r 6 t e n
~ c.
it e s w" t T c r. r e -
) e n
w i t t s. er we r 1 .) m. gtUs r ( n ie* t w iw e aa
*e t .
e g 5 oo it a( ne e, u s t ct ef
. n t
st c. t S ot ne Dt t e t r (c )$it mh r t r s a ef . c i 'a) t e e ?w ee
)d' . te 1
r
- e. si r s"i '
yh ht i a r 4jt
.u sM?s) a w vse gf i u D r
e t.v 9l t e . 5.
** ier t ot v s e
ae*u"ts e icM+ s in t..;
=.
(t mrs ah t s, e e e go it r r < o !.
- y. )
P 7 pest_ia c e M.r*3 c i - s
*r n io g~ it )
ipg uawa s u a iJ?M e v S t
-a N t, "
- eh w
P - va ll
- bi *i es g u X w *' *t = *h
= w +
4 ue+ w* r e 7 oi )
.r . e egc s e y- cs E: (
o "J' wn = a n.f *- oC oe
- s g(
n s[w u i _t7o p Mf f 9 rt fa ? e I, ae a d , " G
- 5, ene**z* ji % n i* &e2*r%)m e m a s hvi. w me te# 'f ?m y
a e ** u b D n s g 8.s, 8 t t j
. " sms s T t t t 21 4 r E
A E r E e
* " s y;
5 5. i a th sw tisye ( t i e r ey r t s r *. t i a e e **. E M = r% , H 2 n i. a( l e I et Wjfaeh2" f pr e r gad.i t n c, 3_wn i T s-e tr ? ! e&' ( t-
%,xyD. + fab C
M*oUe. 'L' *Sibn~3':t
*
- i
- rS eb e
- H P
O h p s '? m, o '5 e thBir
- D eMf eie i, t ueaw.c s
f p o e e t. e* p , e.c-nsuir %E r e. p e reT. '? 2 '.2s
' ti ea ,
- 1*
* " mM u , 'dn EEu v
r , s t. e ppw;ts 19 g g e p y
- e~K rmQr.g w t r ya f m
gtS o r l o R gswiUc n r y1 e s t, s E1f l i. r
'k '* ers, Pg- ".u
- ti t O t i '
bt 0 n e 8 s . s n. fs - s* t ac a "t'L A- _." I9f t +
. o_ic _o (_it 6Pl' Lc (Lv LLC LC m 12 .( M ;. _ i .[r LLt(hh09!'6i!E P ,
O " A_a L A A[ss p{C t t A . . _ r ( S( kc TO C
.oN t
1i ll lil(jllii(<fl f [ll(lj;fl1l I ll I
I 4 f o ,_ O 2 7 . a 5t ) 7 q7
- a. 9 3 % 7, fr$7 g. s1 3 3 g3 g77 uy ma7 1 94$ u, rsNwMrqn2y3p;;3 5
n45nyyo.h 73 mvmgwiaO g p :, b r 7 3 p n,y y9 v *c3-g y
. 7 4 p, eg ..
1 A3 1 7 a2 p 4 J - v t 1 i ; I
' m3: t 3 1 4 t y
a t. P .n
?
t e, c s ir y s e t c le 4 u, e
, t i
r i h
~r q
1 a r e r e **w ft . r- ) r e e r d n
.c 1 o e t a w
e
) )
S n GM p o 1* . w r o. . aet a ig
)
4 h t e. ia. r e M n e u W it e c w r,
- e 's G r )
s i t e
- n ( t n e ) s > ,+ c, - c g u C( c r h( r t
t er i m g
; h t M e. e t e , g r
b e c izr w s Gs t
- M t t t . i rt 6, it f e iy .t t e a s y a e t m t p m u mr m( t t r - I' i f
e u'aw t t gs ea ) lat zi f s eg i. o a e ( r h 5 q g a is q mj . ti teth.n si s ut RJ p *r MR nt s yui es es taa *
- e t
( erCs e sM s a w r a t u se e. t e
- t n e
s 4 4 t $nI odr e
) w s, E gc n a E ) aa v 1* s r g 'a t
e t. e( e ec (l'p e
. a a ' 14 a s is r t) t e a t e ,- a vr.a n j e t
n 4
],
y n e . N** ew+r u uw !t atv e'1 Wr l4 o e :q%3ErTes e rw e mtao rtmes iz v o1 eh
-v .)t s r - rt o eu n s
- v N at st y v ah n .f
( r ar. .i r F h t t ner e~ n iaw ns t
- ee,te s ["_awt aei E
- a*
l( ( ae.i s r me te Q p *vas s r r e rt a d avaa( vh r v aAfh t r gi a
- y~ t C
%s(u'Daeri ea u gei pe 'tat tss 5
u
- n t
*C aa w v " a 3- t
- e A *'8 a
it 4 asa. mn rt rt is a it f ee y a ' f r. 8 te r b ** ri e
$ tet atA qi t d nsa Ia aM t
a s I'to$ t' e mH b es s a aav tG a M snIia
= h t
- r. 8C pt e
T7e r fw*u* - r .*
= e r
n .eaft su m %l a t t m a i N r" *
.le* Iea
- e h
t pC i *bi@nua, l o
%tla Twt l . a c
o a a_ + T. ' ta r w h.ys p pms y"1 !? a*f.g' a eaaa pu n% ru m iL s e f e e I' a rI 8 t , e~ C e v st ni t e i e e t t a pt t' o ' . r'a b m t ! e pa t
- y r
s mJ -
"f a*neuei 5 8'
- s r
0In
- r_
zpshtCsJdlg't vsi w m ~y"i . " e f r +e C(E e t- EI G N " _L1M M _h au a ! t
. .r F
1 M E" t _N s S ~% u 5 s s_m _ne ' 5_t 5 5b5 3{fta e ([.rhseF W C ( Co9D9OD, 9 a p
*, 5 4 55 Ss97 7 s i n *1 %u s 77?2 22I 7 i286M'R 2 4
1 I Mct M4 I 18 fi I' n.4 P55736: 7 271 52t 9 546mi3 1 1 25 4 90M7424%972966nm3i1 1 661 ? 2e 75
? t232tne1 4 1 i z 0591 1 2 90I Se c3 4 1o v
e h - t h a it d o e
% S C Y M n ) g n z r o n a
cs G l t n ah
%r h m e
t n
- c. ri e
h P t) it m h s e a t D e m.: e5 w u e c t
- e4 e b r u 1 54 >
e it igr ** e ( C H 1 3 N t S m de gc ) t M t ie * ,MC r a s= *
- n S
)
lga t t e ( ) e t e= r. 4 )re C
) a v
r d s s e tdM. t s s e ae o a h' z N9 e ed C o i
, qsG al a d a a e d o . ew pt ir t ,t r t tc t )
_ u (C io h ee ir es e n e hb oeG b
) gs e gsnge rG t
r e) s t e e u t e e. i.urn in t i oh s a n et t r n p r c r- C s w s r e ; r e f it u qC _ i s u) r h e e. t t ea st h si c nt 5 4 g hl 4; e epmdTw* r oy n g
%ew o eri ( sS r e s - eh g t c c r
( g fe ao s eie s. ) e g
,(
t t + = t - Sit
*; g e s, ( ee i t a p e(A t aE + e n 4.r g .it e RCh r
e s 9 ; t Cta a) Wl T TMH 1+a o c gi as a gt m r
. y) rsl e
ryic)it a t e e39im r
- 0 eg$
o ( JS S
]
1c_ e, pts ( gw a w e~) st mi c a a w.( .h ire iei s i fpc a er e t w t g 2'. u( e t ic na) eapw . ra .t m a ut eu a ws( tLutf'h8 fi ( H tt t eT ev g
.S r a !i C" .ta +
a . g e e v ( f a t ; 4 e=G)se. a u( aaf h
- r. cF t t t
- o a t ( ta f vv t a c aM t e .h c a e r
.
- a f
. g (N . nn efa -
ei3 p A r 4 . t) a ayOsiwtse
.s( cs {
rin.et - l r s a a( ne gn f
. a a. t t is s t v a E'G . a r ( e$
r u e - h v e t 2 a v eT rt 1 ge es e e .v a a A c_v ( wt t e f t, 2 ay
- t (ae vt s
w j ea s eae y Gii r et m e.M.
. u) c i_r*es.Nrg+ n e s s +
r C
> ie s sf 4 ue Te4nes -4bteim r m r_a1 4 s a ud a .c u tMii gn*yM ri r t
eI v R s yr eu*"t_i t tg i s.a . r . wh t m c. - ri t
,ye n e .fea .r u m
I ei s . iI ~ a r % M R(nm*t o' t.s r t u ja t r a e t i ct o a . R A
- t. iT~
e e o a mt one
- eas c d o y ' p. cc a pe s t s rr a os
- te l
eeers v f*ta ppt 5 e p aR e r t 0 - .m3mp.$ e t uep*_wii r e a *t A L t . a e a .i p ,prN a r c; .y %isnU iU p
- pl
}
l
.s elayews n n p" r e (
b r m Od*iU<u[T d Js e e O d l-ae r t t r L+ t s ct t l l t M_e at p l' A AN E A tRM ee C t 5 L 5 s_5 5 7 { 5 5 { S 5 5 $ s p S L55-{L55FI LiTL{iT_t. t-a e S & T rt .;' ,
- .LT Taw
. r # A 7A A((Ih O
pgw
O
,I & m,- & mw. oe u, - .. n - c .. i < - -M 8ri r Mg)no mm .c #. g# Js.es. .*sn gie e e ,pPn ,e%~ g. o. r u 4. =
4p ou*-
- 9. M' t$ ev- *h 8
-9 8 i 4 t~$ .=
a e( g g
.r.-
(* '.9 O _ 4 gp Ps
.- ..,t m u s, m'.t' ,.. g,)-g1m +r.. .- p ,a 3 L = v,.w nme-es~ s- < - -
ra > , 4 m p e ,s, s 19 eqa,s e r, e ,% eg,e. - m m t. v ep of se it . +=g m - .. tw t f' t " h -
,m-Q S
h CD e' n'O . .e -
=&
t
.e1 l
( ./ w e
& gr ' -
- q e. . .
-w e o { M ~=
es ( . b e 1* a m LJ y a 4 &,
.h -
er
% i a C 'f'* W L h (
e=+8F .@ S ed 68 h h i 4 Ve o t' C C th + %m s W e C a a*. e ** * ( a - 6 t' se. l L tg .2 +-t, p om > s 6
- 8. N + s 4% a a. vo w
+.4 -
w n A tt , t s 6 t-( a, es-g e q e ~ e u c.i 6 .,.4 & e += 6
=
gri w v
+
e :- - , av-9 <.
- ae4 .A +
E, , . po o - 6 . - o .,
-. # , #. -. oo .r .. ... M, i y -g r p -(i' 3 -
- c. , u y (a 6 w' i #
A dat -I t 4 4 O 4 b b 4 0 44.8 () 4 * .* :' $ ( 7I T x= {,4 - e
- O O es 4 - ei 8--p dl D t - l ;* *& ' $- w 6 &
M 4 i. . k u G **6 ts
.stdee e n- -:*. - ,r r s -. .e e; a- -= *. v s s =e .- ,. m ~ s e# .e.
w .s y 6 pi i e o$ n y.,.- - - .y e.
~ s ~ . .1. ,+
e t vs. e. - .s,q ,.- - .s .s , +,s ,-w
- ,iu v e,o - +6,-C 4.# e4 4 e- t!* *t ia . i-
- 4.a o
C g h,(-- >19 6 0 .
- r uti ,6.-ii +el4 4
p+ e-'la 4 4.l 4 '
% *I 4 Sie 4 C ' .s
- 6 6 4
+
T ' I
- I. b l w** e.I.. P', <-
, '1 r r,1n*'. i s, + 1 a((
8,.,.
- c. -. .-
4J 6
- a. >pJ, .$ - , m 6 e.s+ '
D.
. . . e er p gs us m ot ei6 ui - 1io t )
fa r
; 4,! q- . a ( e- *,t 4 4 &f4# g- f i e w l t i vi % @MI ,? %. @ e
- 1 i Li 6 # =l@ 444 ! w, '*9 4l 4 6- i+
e- 4 V e seJ <*g- D ( * - ' s ne ' **: g .(t b:4* 6
!+
v .r.. 1,. g r-
- 4'6 9 4 g 6 e C6a.
%i. ==. Ca .
W.
+ <e6 ca . e . = i .+o; m' r.,=. 4(-
s.g. v . t- i , wi g e. ir .t. c- a - oi4 e o. v e y c # -iet 7 W1 .I 0
%. , . . Ca $ e:t4g i.= *I E' -=-' I *. l q .s t8ti $e6!w 4 & :
l 4' l *e i ( %* tv. ? Oi4 e W 6, f ta -
,e--
i I 1
! es **U (iel b p e- s-
- tiCit yt 1 Ci- y-i+
r- < ,e 5 e.e j
-l6 i,c ;e w 5 p y u e- i - t or .*4e-9 t Ch,
- 6 e ..- 1* t . > . .-3 4' i s
; 4 i ., e p+ ic p-i .
y s. * - v - .p
+* # '
t t .4 +<-.+ 1 U$ g e e e #. a . i
&g&ie-l <- ut* ,i.
e jJ v - , st 6 D 1 e' 6- j 6,4 (81 e- a t e i e, I e dt , 9 la t ,s a iF si .- Av+. e o b 4 g4 e tw a e i e i .. 5<
, <4 a:, ,
6 ! Lo D'. u.i.44., j
~MLt.. e . 6 , ., -
4 . - . . .
- eo mI , a i p .4 16 t.kski .'t M b + s is #6 ..a w 4 i
eg 64 et 64 54 e6 a u vassait h,7 Q J }. pio tc v. e, s. 46 a - a e L k= e- ko il*~ ) g.s ( i . O t. g ft= ** W em Ol uw m .-- e --~ na' nn u(
@o F4 W Pg n.m . s, e,
FP. 4[69
- e. ~ .%. >=. . C. e W ~-f 4e-pn mvm g(> 'Q m
o 2.= u e o. fV Gt en S i D$ 09fg 4 m n .gv~w,x.9.-A.,, m- ., o A ,*4
- Y'.% - -
o L:' p,a r,rai,,-w A. N g/n g u
~
ap. 5" ) gg $g Sh' M td'l8* 9'*'- F* Ul"C C
- r r> ei
+
- t i
#'D b
hm ( -p e 5 w
~
C
- l. .
N -
-1 **h. ++c- new y, 5 ( .n-a v ae Tj -h p e + - c.
3 s - QI vi
- a ed e o 3 m ,s s k C w ts Cj c
t 6 saa 4.s C C o o o ce s a ,o. s a
- # .- ( o( e, e m t., xc ., u- s 4 -: p\k .ee c- + wl y e -e ed es *(q e QA .m k a
w b $5h 6 6 fr ,r s,> g. e,.t* $ O C - b. # 6 v .# +*- L c-t* a e et * + o .- o o s s -f +
-s . > .e. a _ri e s -
2 o: we a a,. . e w , - 6 .i , U e.' 3, b a' - pi
, e,.A > -i - v ;
t 4 4 .m a i 4. 8"'. 4
-& 4 .
C 6
.O .#
( *
, e ey + c= 4. i -
ec k, o ao ga f.ad v- ec. d 4 6 (b ui i 6i OW, m(,,.4'.~e$L g,J 9
-- a -* . 4+ r> % u - , ed*l m cm .9 -
v e -e v u. k, ek 4 -ee . & & 9 n== .e r. W .: 4' j L ! L 4.s . @ e'j**l
- it !-
@ 8tH ' 4 .L w ow - r m' t 4. . .
r - c , u, -- m ., .t .s. w: -l m .s ; .,3 m, o$ 3 ,n. +,s_a . sa . . s#,, as . p<
- t. 5 m..i ., na .
5 - , s.oa
.+tm .y m i :
s 6. e #>. +-t ,< 4. i . o. w 9. wt - $ +
,. .: .s y' = .c a +. +c 43 s m a sa o e *t*-- o oy ;6i..n.s -i s -gr +- vi 4i. <: - e v* t*. w p '-i T:" .m a 6 s e eie.- t # -- . * - 3 .s m 4
[. h O
^
OID os)bweA= 6 % %lO ,k d( ga e,.D s f. t,as tP9% u-<. I gio,,1
-1* { . ,e , s.o,2 ; e, - .c wi 1 , ,a # : ( .,o. u-2 48.% #aea c 1 *ve + 4, iv;;.m s.,
5 m, a:o
~.. u.- < ,.- +. , 3 .a r . i o.:
w c ,ea a m i ye :- s 4, i-s t<4s .i+1 M M e.a . m :; ;, _ -.+ it ( f. 4 .;
. .c. l c , .- 2 m e< -w u g. . ,,ic; a + .. lin ei .is - j- . s u. . i u c, 6e - i < <s, o '; 't.age l. =.. v.m.jq -ft, p w i p g .; ..m a et n -
o a , .* d_l',. t b,, @ q D c m .1,
.- d '% . . q,i+,c. , L( k 4(.. .;c, 3 r Q % e,o k 'd- ;m - ,s i +
j J w'
'em.,si,s - 14 -d ..,y, Iaf,$I#. .,t. 4, ;: # @ 8!$ r gJI O h (de .* O Y .I O . +4 -.ej "*; - M +k' f e0*l"o.
GP k' [ I & T T
.a 4 % W i, .= l% .t' 4
Q .a.
.c A ,g i f'r a e u u u-se. -* 41 F-- a .s ,+. -a a 3 3 +is. a ~a so + w b s a+-
s t p r. o: . c ea s l;e o a .L uW , sn O 2 < 104 1 1
Page 4 of 4 i Table 4-2. (conti nued) Biovoice,sm3 P voiril L, em4 .n 13752 PT = LIE Futtfeto-Kaos 40911 { -nc msta-ase f ary 327e9 P spp. t hrence rg ??925 CMLCSX*VtT2FHYCI AE Goapste+r sep Diesing 5653 i i W L7 O O O
4 9 1 I i L Table 4-3. Results of Friedman's' analysis of phytoplankton total abundanca data,1974 through 1979. i i
- i 1 i 4
i l I San annual valua of mawn of abirianr+ . [ tap and tx)* ton ranked locatieas "easv-e of Phnths included Top reWfhation botte ranked locat%~ ! in analysis T y 2 critical. Locations ranked by mean rad 7 abundance . + 11, 13, 8. 4.5, 7.5. 5, 10, 6, 2. 1, 16. 4, 3 7.5 e7 #* 5.4 rg e-1 ! Chlorophyll a 1/74-12/74 29.5 21.0 1/75-12/75 18.8 21.0 N50 3.4 eg-m-3 i 12.6 8, 5, 2, 3. 4.5, 4,1 4.6 sq.m-1 1/76-12/ 76 18.5
- 1/77 12/77 6.0 12.6 450 l
1/78-12/78 13.9 15.5 M50 2.2 mge-* i 15.5 8, 2. 1, 1.2. 5, 3. 4.5. 4. 3.9 3.6 nr;-m-1 i 1/79-12/79 2P.7
- 6/78-5/79 53.0 -25.0 34, 15.9. 11. 15. 8, 50. 13. 1, 2. 4.5. 5. 16, 5 0 #9'"~, I*U "9'".,
i
- 3. 3.9. 4. 1.2 '
.t co I cn 56.5 21.0 13, 11, 8. 10, 5. 3. 4.5, 2. 1. 7 5, 6. 4, 16 1433 units mi-8 n57 units 1 Dansity 3/74-2/ 75 1854 units-mi-! 1252 units -e;-'
- 1/78-12/78 . 24.1 15.5 8. 5. 3, 2. 4.5. 1.2. 3.9. 4. 1 16 4 15.5 8. ?, 1. 3, 1.2. 5. 4.5, 3.9. 4 2
- 13 unit mi-8 1760 unitsel-' i
- 1/19-12/79 6/78-5/79 26.5 25.0 8. 15.9. 2. 15, 5, 3, 16. 1. 4.5. 1.2, 50, 13 1926 units.a
- 1-1 1351 units.=1-8 j 31, 11, 3.9. 4
- - I j Biovoltre 3/74-2/75 _ 9.3 19.7 ftSD 4 1/78-12/78 11.2 15.5 M50 1/79-12/79 4.2 15.5 ftSD
{ 6/78-5/79 37.8 25.0 15.9. 15. 13, 11, 34, 8. 50. 2, 1, 16, 4.5. 5' 1249 mi n-5 561 mt m ' i 9 q 3.1.2. 3.9. 4 1 NSD a significant differences amng locations at the 0.05 level of significance. i i 4 I 4 l I i r I
Table 4-4 . . tits of cluster analyses comparing sampling locations r>n the basis of mean annual densities of major taxonomic [ elasses of phytoplankton. 1 Maxinun Class densities (units-m1*I): Locations ' standardized used in Sampling distance w. thin Chlorophyceaej Bacillarighyceae Cryptephyceae Dineph crae 'Other Ctesses analysis dates locations in (1uster Tang Kean Tarry "grogvceag Nan danga a cluster Maan Hean gny_ ..m ame Mean _ a je 1, 1.2, 2, 1/78 throug5 1.64 1, 3.9, 4, 1.2, 2, 3 547 514-577 505 456-517 307 252-354 20 19-27 21 17-25 213 201-231 3, 3.9, 4, 12/78, 4.5, 5 584 577-590 466 455-477 341 301-331 16 16-29 27 25-27 295 273-317 mnthly 4.5, 5, 8, 8 602 - 6 31 - 378 - 30 - 19 - 246 - 16 16 571 - 661 - 2 32 - 1 31 - 25 - 296 - 1, 1,2, 2, 1/79 arough 1.45 1, 2, 1.2 649 639-661 707 623-759 314 230-362 18 1 -21 23 20-25 291 299-297 3, 3.9. 4, 12/79, 3.9, 4, 16 508 483-539 616 566-690 275 189-332 13 10-14 18 13-25 252 211-287 monthly 4.5, 5, 8, 3, 5, 4.5 640 605-665 464 455-474 326 312-334 17 15-18 28 26-30 311 294-323 16 8 647 - 929 - 472 - 21 - 19 - 316 -
~
co N 1, 1.2, 2, 6/78 through 0.91 1, 2, 16, 8 409 331-446 644 625-799 277 232-334 46 19-119 23 19-26 276 258-291 3, 3.9, 4, 5/79, 13, 50, 15, 15.9 299 263-338 790 746-PA9 300 261-337 37 23-44 13 9-19 119 91-150 monthly 4.5, 5, 8, 11, 34, 14, 69 275 157-339 9~2 837-975 188 129-270 24 3-38 17 3-27 89 55-120 16, 11, 13, 60 114 - 724 - 156 - 6 - 0 - 67 - 14, 15, 1.2, 3.9, 4. 3, 4.5, 395 355-444 442 341-525 270 226-323 20 14-24 22 17-26 270 233-322 E 15.9, 34, 50, 60, 69 1,2,3,4, 3/74 throuah 0.96 1, 4.5, 5.2.3,7.5 288 226-310 323 292-F1 255 223-290 56 51-66 28 21-37 **5 *21-74 4.5, 5. 6, 2/75, 4 10, S 334 318-354 424 407-433 270 225-142 62 43-77 27 22-35 *41 *36-44 mcnthly 7.5, d. 10, 6 31 9 - 250 - 126 - 50 - 54 -
*72 -
11, 13, 16 16 229 - 375 - 147 - 37 - 16 -
*25 -
11, 13 366 341-337 460 ?'" 462 4 71 457-4P5 52 51-52 24 21-27 *55 *59-60
- Values rs' resent densities of unidentified algae in this analys s.
9 O O
~ _
J e. l i f f a h " f ( r
@. D m ee b ew ry em th r: m l
' E ( e,t '
- e. N r+,e @ 4 (- J
-m O
e - ii C- N @ s ,e N a i er r ., 1 e- ( ,, m N. e m gw - )m o N m et ,m. W e **E fS,l A f j .L
. .- l. i h. I i so *% d 6
i t.D e O f gv u Ln $1r .- j+Y. e%u
#N i.e re N - b-e b
r bm m m
.4 m ea m, s w i
e', C W sc o M Dm in @c N Nfa
*^
fg C e ,7 f h.
*r P *O
$ 'Av ou g i me a -N .. ., ( s. ' g *n -N N ., -.
*N **
- e-
*' I
- 80 '
D 4t D "40 va h C! C. f = tt "N' *"'e'- c t i 0 w"* U 'C' f' ** ou N em e~- no en w eo N
.e.4 .
o . i {
- O
. C r i l o .ht
- W
# W
( q , en k i i !' VU v.-
- I O
Y 3 7 O sr. N 69 m e4 h N W en-tj eu ts tu. cc e. e.
- e. W N h> m -. o m.
.s-W ml l W ~ to to e
e oN
- tu h
e-ev e
-~ ~
fa w ** wm i y ; og a @@ EEI U k r=~ O'- m m . a C W 3@ (} @ O (C N MN MM M Mth m V f O , I*; fn c. i ! .' y , *.0 { to
-. r, of e-20
- e. Q L
N N- e,q Nm g ; no. _ Lg 6. W *- 7
->f I,
V C et . C* o- m n re. l5 l t t #U
.C C f3 el k t
c
*- t.
[ rw wN (L or N w -c N. t-M t% 19. D Or.e e e, of N Nr NO" 'ts
,6 t.
VO** D 6; { r l. 9 ' CU e c
- Q U
V an 3 9 u ( r
'1 t
i
- f0 th N I I
' fQ gf* I **. W & '{ , . @ m, M6 N. ** N. N fC. t ' O. ,
j, m. O. **. W. wr. ==. W CC. #. 4D. et. @.
. . y
' fa - ., C; C O N @ @ w *# M er e e N w O O to e O w c
, & .C R l+w < w e. N en et .o m e= m e. e m WQ * % ej v - CMC m- -
f' 1 o m t t CO a C ~ l ".p
+
3 er "v@c' W e ce co m@eco -. N. o. v m. co. m. e. N. N. w e , co. m. e . .. ooo ^$'r l O N > m 4, w OMNeQQ CMNmOO O N w w O ** - p f r0 ** * *- c dl au e DI r 4 'j F fU fO , 80 W U Ei E i 2 vn a de s I C y m L I. 9 M. N. w. N. w. @. N. e. @. N. e. #. @. ~. @. N. @. M. W C. W. e. N. W. r " C b L ##MMNN m @ M #9e. N m e m f9 N N @NwemNm e
- r 1
! 80 i'3 g M,g e i C- E f 1
..to U V G' 3
E m @C
- b i -
l
* ,O .)
i m a-
%j =t in c N ce p.
N** twANN e m e N c: rh e,. A N >- t%e% et m @ N tri ch N es N N r% t% w - cr%esF *tew es e% m c * !, R7 Ot 8UI @ C55 7@ e-* *= e e. C ** e s O Ch th CF (h Ch Ch
== e= m m w e== (T' Ch w *th @ea =e- it. Ch w & Ch w r .C.h @ O Os .-pw ti t l e 4 4 $ C Q w ,I
- i. -
p so
}
l 80 3e , , H t i l
^J [
- l. k c t C ,l O
+i w I em '
dC ; w *f W t.t. 4 sa c: A l r
)
I
}
t i I L 188 i o. l ______u_.- . _ _ . _ _ . . _ _ _ . - . . _ _ _ _ . _ . . . . _ , _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . . _ _ . . _ _ _ - . .
, q .,,:', r 133 .i ?221 ?57 62 6a ?
W712t?*i1 i21'S)m o ?
; +
533: 2T 6 2' 8 2-2*
- 5' V <, T ,D"26M?7 i 1
5 ' 7 ?9i575W?U;+
$ 2 '5 2 ' 2 W?
fl1 Z? 9'
? -' 1' 7 ' 4 ** * ' Tti "
s 4 d 11+gg w c:b 21 1' 1 l' ? 8 7 e 3, 9 1 ti; , ,, 4,;y3_ d 3 531 1 5i1357 M71 2?34 ?E - : ? 1 35 54 5t 531,60
- 7 ?7' E 127 5~4T 2 5?1 71 1 tT
~ 2 1 e35" t 32* %7'l':i
- 67 27 626E7 !
7 35MM75:1 32 3 L1M2 2F 'c
' 62 T'I (
C 5' 1' 5'
? ' * ' d S5 , O 3 8 T ,
A , ,% 1 f - ,7 , o y e 3 ,, 2a 7 3.L 1' S2232 5,1 921 0i 7 35i.1. 9 6 6 7 ~ . 2 51 : c. 2 1 S' 1 % : s.593,".14 95 r ,r ti,,. , 1 1 ;: 5522211 -
' - ?. f 7?1 i2233 32243* .5122 i2 I22I1 '
1 a .?m e y ,y . u . e n g a - g a J * , P (
, 7 , m , ,2 8997 438 U4v.!! 1e uW7' lc ;t i522n561 7 5 't 57 4 8il S1 s ; 2, 10al r )r
- T 7MMto5 I1:li19 15391 3
! V4'A' rd 3 1 1 ?
l a 5? ' t 3 * :
?5M:5?M1utir11 i
5 a 01! 7-t . E a lg e ,m.,9,w l i m r At t , 7 N o - on, g27 , ib ig , 3375'.2537.*.?5, 4
- 5. 641903 5' e
2' e r; 3, . .3?21 4 3 5 6 c a. 5 e6B8 086' '1 t m47793142122333 11722R ST 209' 1t255E4134 11 g7'2 3II * 6' T k , >1 ! 4 2' 3 T.
?
a IN'r, , yq,,,s, L A r , (
. ,(
G. 8 *; 315t c35a657 1 175?i e!!174t t is7 936462112 2523 222a724 9T 5t 1554)1 3? 3 M33l51 a4 64 1 * * '? 2? hi7 ' 1 'A* *'i wb d g n a } 0 d s. 3 ,* t s ,9 .,. . 8 33 !3?4 3 u, sg 537 5 37 37 7 '4 a
- r. 4 81 t e. g.
- - S7?11 0N 7 n -
f12531 - - 4 472 - Mo t - - PN - 5c n r, 1 1 1 i *
- 1 11 1 1 i 1 I i
o (
)
t ,w a &c c t 1 o % vq L t t a a
, ny M1_
si e h ,g t e e, 2 2 0 99 51 00 730562 85144 M00000 '. - 9 D 27 90 2* 7 2 7 ?.5 4 5 5. f $. 1I 000 317 422 0f 0 - 55 55 00 1 3 21 00 1 7 17
' C 21 51 02G(G Mj2 t2 O
d hW.,. n c, , i o g
}
( M. t e t c .n. + u ; d 6935555641 3713259s? 1 61 451 ;' 1 o
- gi h 5733C. 727,20497 2, e, 5 - .
1 r g 1 0111 - - 1223 235' 552 1 055653a345533221 3565647 4)2 u p g o r 1 11 r
. t e
d9 n7 ,g ' U a9 - y 1
}
e t , _ h. i 0 } g 4 v1 7 4353
. - - 02081O402683655204837 0 6. ?. 35031 t. 0 3 0 3. i. 4 41 M
i t y s'
- e.
2 0323 57 1 1E ??L123Fa1 17a30484 7 2t12J3 4 1 11 1 11 11 1 1 1I1 11i 1 1l 5 a 37 ? 1111 1 to7?i 1 1 1 f, cr
- du ua .m.
on ) ra pJ ,g~d - yh - rg ,C j,, 23 3I 60M4301 A i% a933(27" MlE 153 9s%N1 8 l7 i63t 7213 82 21 6l 7 32 - - 5I51 *:3 3 - ' au C I 1 22I1 2: t1 8' 6S58%81 - 71$ 5F 1 4 7 - 8 ( L29MI'9473
- v mo i r
*;(, 1 +
rh , Pt i
~
g ' i 2 3
- 247 32
- 2 g
- 5 - - FS322f
=
I 5 ', 1 rIt 4 3? ! ? M n1 <B6A 7
. i 223% 2 4
n:554U'3 1 1 %4i1 :62-2
* ;: 4 3
1 arr il .
.7, 72g f 7
2os - 6 5d 327 0 .
- c," 4 3
4 .
.,, J
. a < t e w h.$*i 95?2?51 9521 4 3771 1 52 1? :122a5296235336297a1 3 1 12 1 23 } 22 1tg31 2 gn7, g7 _ l t e 1 ? 1 22 1 ' 4 1 w O b g e b r r 1 q' s t _ a t w o e Fe 6 e % w 9 a w % a T
~
f' a 1 i 1 A
+
- J, 4
- W-m , - ~ -O e., m m nn ,, ~- ..
g g- g t- o,** w
.N ** w +
h
#w .n s,u tv e.
oeu A
- e. , e m.em ,,. m m .e. . A. -.,,,,,e,..
. , o , 5- oo.- .=.ri,n..- c.a v ,- .- ..e_ s re g . % e-.N . ~N o, .. N P.- - - = .li a i e
att 2M EP~::;t we t S g p; * :: c; g e n w - ~s gu' ~ r ~ ~ n g ~e r - , g g a n ~ p m a: sem m a n g ,- .,. o N
, %g_ , g , ,
N w s.e % s a - - ---- ! w G29' i o. 4-N ..n. e +
**
- P. O. e. sh. . 4. m. m . m. #. . m. e. m. N. O. e . to. th. p e. h. e. @. m. e. es. N. N. O ce e. e. at, m. M. t* en. m. r l n 6 N. O. 8= . .
- e. e- - - N N w N m m N e w e e e +' *s -
- t to M m e .N m - ew N N m w en w e ,ri w es m m e e - #
!- O . Q
, _ vAKa - $-
! Oi. b T4 e ,* mA @ n i@ 885 a. - @ m e. N (3er o Aa+.r'=a l ss g$ N e d ew-9 h.c m o en e en M w e6 & *tY et MNh f') .. N e-. tT.imsdn as w w in w p. e u3
@ th et. N s-- tw # es i N .ti N t-N tv o.' owt>e toe- ew en ** N met at ue. g Mo'** A " *-A seAe,-e.ms so es w = =
- b .O.e.e.4. ewneA e===- t l g g $. !
I e. se [ ed -- l } l ** . ey 4 eeaI
. i 4
a, ' m . N. s,*=g e. O. m, ** . .w . 8. #. m. O. to. O. O. P. O. m. m. m. N. F 95< .4.** W+ P. W. en. W. W. e=. em. W #.O. @.Ch. P. O. O. GU O N.*av. A. *. e y
- a m m m e , ee N 0 - o o o e4 N N M e as et # N M h m E 8 9 se-e6<m.4 4Q tt; i
- m. b e @ # e 8% sam e N.es N m EP O
** e= P.
m s=ute e e4 es W - (I i l, i
-E.
MD i N - i
? ,
et se w ase skm 6e. g; e pe. #% e g,o sc w goNee.m , yo,,g o-et.oN wg an g e*>cm mN e6etoc M *. N m e- tv ** enNce aoN- m e ewrs w -44mN.ooer mM tonemineN*.=eee*en 2 m .e ki a e-eA ev .= e. e N tv n a ( . 1 . f I $ ! N ta e e@ N w.** ** m tt e m O - rp - cc N OA g' C f, e g g AGO *= II'at,( $ ea.In N 6 4 4P O M 68 0 4 P- fw s- D tM 4 Cr GC P. 49 4 8 5'% 97 0 ee - - a m W N 4,9 5 ** I 4 km N f e- - u .- - em h. . +~ ~ c e 4 I
**d l a W f I. I 4, .
I .2 i .a . I I wY e m '
*F E. . .n. .. . . #.
e., e#
. , , a. .n. . . e.. . .. m , w #. . N. * .o ..n.. .. . E. .
COOOQOC e~n~
,o o - .. . o.w m. m.. ~ . . + . . ,. . . e,,
O - == C QC
.w m.m.
O i O Oe- Q O O O -O O OO e= 0 Nm
.O 4.4. 4:-
o e 4> f 4 i h8 '- E i w r f, , 5 , e.~* E ! s .! A
'* 4 l ei -
m ;
*2 ,
7 e. w. . o. =. N. e. N, w. t.. O O == N
. . N.==e.N N. N.isto.erc.Me.wc.@
N
- e. e, ~. N. =. ~. w. m. e. e. m. .n. e. #. ~. . . . m. N. e. o. w. o. w.
h * @ O == c w @ b W etM en e en @ in - M N w w m @ - ..9 es e f*) m et m f't I h, b
> F j- I ) i = 1 .e.e.m. e. l f o.m.w. e. e N. e. m. o. e. m. m. e. e. m. oe,~.e.m. .e.N.m.e.o.N.m.m.e.m.e.s. . .w e.
- f. t
.m ~ .n . m c. o . o ~ o - ~ # N ., u. ~ m ., e N r N -
- q. N.e OF#PW -o N O er w .
. m su N.m pe N s ** se e- ** e* = = * * * * ** e== - N == e em s= - ee ** ** N - i e ce ** e* e- ji e " U ;
s ! i e- - F ; + V *
- 0 }
3 7 t i "c
- w. 1 g i
a N.,e w P&;Oomme*OMM-Am+g N i . S e w -OPA E - e Ee awe
- MNO
+-om mtom ~ * - m e **eOw em * - s o m e m ^ + - * >,z - -- . w.m n 4 o y '0 -m. +-4ee a= w aJ4 8"% ye INg N ao a= M Fw 8') **n et f*a Q 43 ** # M @ @ @ Q w @ 40 *%iw w 5 ,.
l ' S [ l; V y - -j-
- r e #
3- . menesone ++t.> ,, e N e m e - ~ .a em we net4 ni eaN fm uNAne= d.n8. o. e.; a.en= p smare...u,.he*dks et st am& #Qc . p u . N sw *e*m'det et e @ c m @ M N i , e f a . t- ?h fY l g a E N f"'ttw4O N e*aa==*.*ea O fy ep 4 @ O') e e e p '"1 .w @4 N ._ ) J'
.Q
{ k
;}}