12 Case Study: Discrimination of Factors Affecting Unionid Mussel Distribution in the ClinchRiver,Virginia, U.S.A. John H. VanHassel INTRODUCTION The unionid mussel fauna of the upper Clinch River has been well documentedasone of the best remainingepicenters of the highly diverse and endemicCumberlandian fauna. Surveysof 1912–1913byOrtmann (1918) established the presence of 49 mussel species in the upper Clinch River. Stansbery(1973) addedanadditionalsix speciestothisnumberbasedoncollections between1963 and 1971, for ahistoricaltotal of 55 species.Subsequent surveys were performed during 1972–1975byBates and Dennis (1978),during 1978–1983 by Ahlstedt (1984),and in 1988 and 1994 by Ahlstedt and Tuberville (1997).According to the latter authors, the recent mussel fauna of the upper Clinch River consists of 39 species, to which one species,the tan riffleshell ( Epioblasma florentinawalkeri), wassubsequentlyadded.Ofthese 40 species,10(25%) are currently federally listed as endangered. These surveys indicate that as manyas15(27%)ofthe 55 previously knownspecies may be extirpated from the upper Clinch River. In addition, some of the remainingspecies appeartohave undergone recent declines in abundance (Dennis 1987; Bruenderman and Neves 1993; Ahlstedt and Tuberville 1997).Thissituation parallels theglobaltrend of mussel imperilmentdescribed in Chapter 1ofthisvolume. Specific causes for the faunal decline in the Clinch River have not been identified, but thereare anumber of likely candidates. Active logging occurred throughout thewatershed from 1870 to 1920 (Masnik1974),withconcomitantsedimentation impacts. Extensivehabitat modification andpollutant loadingfromcoalminingactivities hasoccurred within theregion (Helfrich et al. 1986), particularlyfrompoorly regulated activities priorto enactmentofthe Surface MiningControland Reclamation Actof1977.Additionalpotential sources of impact wereassociated with development of industry and agriculture during the past century (Ahlstedt 1983; Neves 1984); however,the proportion of forested land in the watershed has actually increased, from 42% in 1941; to 57% in 1960; and to 67% in 2000 (Masnik 1974; Hampson et al. 2000), with acorresponding decrease in land devoted to agriculture. Additional important considerations were the impactsofpredators such as muskrats (Neves and Odom 1989)and the loss or decline of fish host species for the parasitic glochidial mussel life stage (Ahlstedt 1983; Neves 1984). Thus, the extirpation of mussel species in the Clinch River has multiple suspected causes, but quantitative elucidation of specificcausesisrelatively rare. 4284X—CHAPTER 12—17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 311 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) Studies aimed at establishing cause and effect of impacts on freshwater mussels are faced with several difficulties. Thecharacteristic patchy distribution of mussels necessitates careful consider- ation of sampling design andlevelofefforttoenable quantitative differentiation of important measured characteristics (i.e., mussel individuals,populations, or assemblages; Strayer and Smith 2003). If these differencesare not quantified, then the relative influence of various environ- mental factorscan only be speculative. Another limitation is that few structural and functional measures of mussel condition have been sufficiently developed to enable dependable impact assessments (see Chapter2). Population density and size demographicsare mostoften used. Growthiscommonly measured in studies on fish and avariety of other aquatic organisms but is problematic for use with freshwater mussels. The slow growth rate of most species requiresimpracticably long studyperiods in order to obtain quantifiable increases in growth. Physiologicalmeasures provide ameans of real-time documentation of mussel condition; however,few have been widelyused on mussels(see Chapter 10 for further discussion). Cellulolytic enzymeactivity in mussels and Asian clams ( Corbicula fluminea)was successfully applied in someprevious site-specific Clinch River studies (Farris et al. 1988, 1991), but there is a need for additional physiological measures or assays. Athird difficultyencountered whenconducting impact assessments on freshwater mussels is that manytypesofdesired measures (e.g., physiological indices, bioaccumulation, and toxicity tests) require that the collected organisms be sacrificed (Naimoand Monroe 1999). This is often undesirable whendealing with rare and imperiled species. If it can be scientifically justified, the use of surrogatespecies (e.g., common unionids, otherbivalves, or otherinvertebrates) should be considered. Previous Clinch River impact studies by Cherryetal. (1991) found that Asian clams were useful surrogates for toxicity testing of unionids. Theobjective of this study was to quantify factorsaffecting freshwater mussel distribution and abundance within Virginiawaters of the Clinch River.Toachieve this objective, differences in mussel species occurrence, abundance, age distribution, physiologicalcondition, and contaminant body burdenwere measured at selected impact and reference locations. Sampling sites were care- fully selected for good mussel habitat based on previous surveys in order to minimize the influence of natural habitat characteristicsindefining sources of impact. The biological information was evaluated in conjunction with concurrent, site-specificmeasurementsofwater andsediment quality,habitat quality, riparian land use, point andnonpoint dischargeoccurrence, andfish host availability. Theenvironmentalmeasures rangedinscalefrommicrohabitat to watershed leveltoprovide acomprehensive assessment of potentialfactors influencingClinchRiver mussel distribution. METHODS S AMPLING L OCATIONS The Clinch River arises in extremesouthwest Virginia, flowing into Tennessee and joiningwith the Powell River before entering the Tennessee River. The Virginiawatersofthe Clinch River lie within the steep-sloped Ridge and Valley and Cumberland Plateau physiographic provincesofthe central AppalachianMountains. Theaverage gradient of the upper, free-flowing portion of the river covering188 miles (302 km)fromits source near Tazewell,Virginia, to Norris Reservoirin Tennessee is 9.3 ft./mi (1.8 m/km)(Masnik 1974). The river is characterized by extensive pool- riffle development, including several islands and braided-channel segments. There are also large beds of water willow ( Justicia americana)alongthe shorelines and shallows throughout the upper Clinch River.The geology of the region is dominated by exposedlimestone and dolomite forma- tions, which produce acarbonate-rich system with pH in the range of 7. 5–8.5. Sampling stationswereestablishedat12mainstemClinchRiver locations(Figure 12.1), includingtwo sites at Station 6(leftchanneland rightchannel).Stations were selected at Freshwater BivalveEcotoxicology312 4284X—CHAPTER 12—17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) representative riffle-run habitats encompassing the length of the Virginiaportion of the river, which extends approximately146 river miles (235 km) from the Tennessee line to the confluence of the North Fork and South Fork Clinch River headwatersnear Tazewell, Virginia. Sampling stations were selected to provide amix of known productive mussel sites with sites that would provide agradient of potential impacts from large communities and point sourcedischargers. An additional sampling station was established on the Guest River at Coeburn,Virginia, 6.3 river miles (10.1 km) upstreamofits confluence with the Clinch River at Clinch River Mile (CRM) 244.1. The Guest River,asthe largest Virginiatributary to the Clinch River,represented amajor potential influence on water quality. Two sampling stations were established at CRM 267.3 in order to evaluatepossibleeffects from AmericanElectric Power’s Clinch River electric-gener- ating plant, which is the largest industrial facility in the study area. Station 6L was located within the planteffluent mixing zone, while Station6Rwas located on the opposite bank outside of effluent influence. Each of the13samplingstations was characterized by quantitative measurements of flow direction, maximumhours of direct summer sunlight,bankheight, bank slope, bank stability (percentage of bare soil), riparian vegetation height, riffle length, upstream pool length, downstream pool length, width-to-depth ratio (average width divided by average thalweg depth of runs and pools), riffle-to-riffle ratio (average distance betweenriffles divided by average width), and bend- to-bend ratio (average distancebetweenbends divided by average width). Substrate composition was determined from four 1-m 2 quadrat samplesfor mussels. Bedrock (solidslabs) and boulders (greater than 256 mm)were measured by hand. Thevolumeof Clinchport Dungannon St.Paul Cleveland Tazewell Richlands 15 MI Virginia N 327.5Pounding Mill1 319.5Cedar Bluff2 244.1Guest River Mile 6.3TGR 213.1Clinchport11 235.1Dungannon10 249.7Burton’sFord9 256.4St.Paul8 264.1Carterton7 267.3Tractor-Crossing (Left &Right)6 269.5Hackney5 307.7VanDyke4 312.5Raven3 Clinch River Mile Name Station Number Kentucky Clinch River Guest River 1 2 3 4 5 6 7 8 9 10 11 TGR Tennessee FIGURE 12.1 Map of the Clinch River, Virginia, showing site locations for sampling of freshwater mussels and associated environmental measures. Case Study: Discrimination of Factors Affecting Unionid Mussel Distribution 313 4284X—CHAPTER 12—17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) rubble(64–256mm) in thequadrat wasmeasured by displacement.Gravel(2–64 mm),sand (0.0625–2 mm), silt (0.0039–0.0625 mm), and clay(less than 0.0039 mm) were excavated using aportablesand dredge, with gravel and sand volumes measured by displacement in the field, and siltand clay samples returned to the laboratory for volume determination by centrifugation at 750 rpm for three minutes. Substrate penetrability was determined using three randomly spaced measurements within each quadrat using asliding hammeronahalf-inch (1.27 cm) steel rod. Depth (cm) of penetration of the rod into the substrate was measured following three standardized hammer strikes. Surface and bottom current velocity (cm/sec) was measured within each quadrat using a Mead Instruments flow meter. Riparian land use influencingeach site was determined by areview of USGS7.5-minute topographic maps and aerial photographs of the Clinch River watershed,from which the drai- nage area, percentage of forest,percentage of agricultural, percentage of urban, and percentage of mine landswerecalculatedfor 10-mile (16km) segments upstreamofeachsampling location. Discharge permit holders were reviewed to determine the number of major NPDES discharges and sewage treatment plants within the 10-mile (16 km) segment,including tribu- taries, upstreamofeach site. W ATER C HEMISTRY Water sampleswere collectedateach samplingstation on eightoccasions betweenOctober1991 and 1994. Field water column measurements includedpH, temperature, conductivity, alkalinity, hardness, and turbidity. Samples were preserved for laboratory analysisofammonia, total organic carbon, phosphate, and nitrate–nitrite. Sedimentinterstitial water samples were obtained by pushinga45-cm length of 5-cm ID PVC pipe with aPVC cone attached to the bottom end into the substrate. Aseries of 0.3-cmholes around the circumference of the pipe near the bottomend were positioned at adepthof5–8 cm belowthe substrate surface.Interstitial water that flowed into the pipe was sampled using apolypropylene hand pump that was used to initially flush the pipe and then collect two 500-mL samples. One of these samples was immediately preserved using 1:1 HNO 3 for total recoverable metals analysis. The secondsample was placed on ice, filtered (0.45 m m) within eight hours, and preserved using1:1 HNO 3 .Fieldmeasurements on the interstitial water includedpHand conductivity. Laboratory analysisofammonia, total organic carbon, phosphate, nitrate–nitrite, and dissolved and total recoverable concentrations of Al, As, Cu, Mn, Ni, Pb, and Zn were performed according to USEPA (1979).Metal concentrations were determined usinganAES–ICP analyzer. This list of parameters was selected based on their significance in previous Clinch River sampling efforts. S EDIMENT C HEMISTRY Sedimentsampleswere collected twice at each sampling station—in July and Octoberof1992. Samples were 500-mL composites collected from depositional sediments using astainless steel scoopand were preserved intact on ice forlaboratoryanalysisaccordingtoUSEPA (1986) methods.Inthe laboratory, samples were driedfor 48 hoursat50–608 C, sieved through12 mesh (1.4 mm) to remove debris and large particles, then milled and sieved through 100 mesh (0.15 mm). Theportion of each sample to be analyzed for aluminum (as Al 2 O 3 )was digestedinaclosed PFAvessel usinghydrofluoric,hydrochloric, nitric,and boric acidsinamicrowave digestion system.Another portion of each sample to be analyzed for As, Cu, Mn, Ni, Pb, and Zn was similarly digested using hydrochloric and nitricacids. The vesselpressureswere not allowed to exceed 95 psi during the digestion. Metal concentrations were measured usinganAES–ICP analyzer. Freshwater BivalveEcotoxicology314 4284X—CHAPTER 12—17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) F AUNAL S URVEYS Unionidmussel assemblagesateach sampling station weresurveyed quantitatively betweenJuly 1992 and October1994 usingastratified (by habitat) randomsampling approach. At each site, four 1-m 2 quadrats were placed randomly in riffle/run habitat and excavatedbyhand and by a portablesand dredge to adepth of 30 cm or bedrock. Gravel and sand substrates were rinsed through twonested sieves (19 and5mm). Allunionidswereremoved,identifiedtospecies, measured for length (mm) with digitalcalipers, and returnedtothe river. Thepercentageof each species that was age fiveorless was determined from age-length keys developed for the Clinch River by Scott (1994).Other bivalves(C. fluminea and fingernail clams)were counted and returnedtothe river. Qualitative searcheswere also conducted at each samplingstationbysnorkelingand the collection of fresh shellsfrommiddens.These surveys were performedtwo to three timesat each site betweenOctober 1991 and 1994, with atotal levelofeffort of approximately4.5 man- hours per site. Unionids wereidentified to species,counted, and returnedtothe river. Fish were surveyed at each sampling station on one occasion betweenJuly 1992 and October 1994 by backpack electrofishing.All speciescollectedin45minutes of sampling time were identifiedtospecies,counted, examined for deformities(such as erodedfins,lesions,external tumors,and glochidia infestation), and then returned to the river. Index of Biotic Integrity(IBI) scores were calculated for each sampling station using the 12 metrics adapted to the Clinch River by Angermeier and Smogor(1993). T ISSUE M EASUREMENTS Tissue metal residueswere measured in C. fluminea collectedfrom each of the 13 sampling stations on three occasions per site between October1991 and July 1992. Each sample consisted of 10 clams of approximately10mminlength, which were placed on ice and returnedtothe laboratory for digestion and metals analysis according to ASTM Method D4638-86. In the laboratory,clam soft tissueswere excised from the shells, oven-dried at 608 Cfor 24 hours,and individually weighedand digestedin1:1 HNO 3 .Concentrations of Al, As, Cu, Mn, Ni,Pb, and Zn were determined usingan AES–ICP analyzer. Cellulolytic enzymeactivity (Farris et al. 1989)was measured on C. fluminea collected from each of the 13 sampling stations on aminimum of six occasions per site betweenOctober1991 and September 1995. Each sample consisted of 10 clams of approximately10mminlength. Soft tissues were excised in thefield, placed in individualvials,and put on dryice fortransporttothe laboratory. One-gram samplesofeach clam were homogenized in aphosphate buffer (0.15 Mat pH 6.1) at awet masstobuffer ratio of 0.02g/mL. Samples werecentrifugedfor 15 minutes at 15,000 rpm,supernatantswere decanted for endo/exocellulase analysis, and the pellets were recoveredfor dry mass measurements. Corbicula fluminea collected from each site were also measured for tissue levels of cholin- esterase activity on two occasionsper site in 1995. Ten clams per site were collected, frozen for transporttothe laboratory, and analyzed as described by Fleming et al. (1995). T OXICITY T ESTING In situ toxicity tests were performed at each of the 13 sites on two occasions in 1995. On each occasion,130 juvenile Villosairis ranging from 9to15mmlength obtained from Station 1were measured (mm) using digital calipers and randomly placed in 13 mesh bags (10 per bag) with approximately 1.5-mm openings. The bags were attached to rocks in riffle/run areas at each site for approximately six weeks. Endpoints were survival and growth (shell length). The toxicity of accumulated silt (0.0039–0.0625 mm particle size) at each site was tested in the laboratory using a24-houracute toxicity test usingthe freshwater rotifer Brachionus calyciflorus Case Study: Discrimination of Factors Affecting Unionid Mussel Distribution 315 4284X—CHAPTER 12—17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) (Snell and Persoone 1989). Silt was collectedfrom each site by placing 14! 14! 2.5 cm 3 slotted plastic boxes in depositional areas of each site for approximately six weeks on two occasionsin 1995. Collected silt was transferred to plastic bags, placed on ice, and returned to the laboratory for toxicity testing withinseven days. Testing consisted of the placement of 50 cysts of B. calyciflorus in 30-mL polyethylene vials containing 5mLofsilt and 20 mL of moderatelyhard reconstituted water (USEPA1990). Replicate vials were prepared for each silt sample. The vials were placed in a rack and rotated at arate of two times perhour at atemperature of 258 Cand a16:8, light:dark photoperiod. Hatching success of each replicate was determined after 24 hours. D ATA A NALYSIS Statistical analysis of studyresultsfor water andsedimentchemistry, habitat andlanduse measurements,faunalsurveys,and toxicitytests wasperformed usingSpearmancorrelation analysis andthe distribution-freeWilcoxon testfor one-way layoutsonSAS (SAS Institute 1985). Amultivariate analysisofthe relationship betweenmussel structuralindices and measured environmental variables was performed using canonicalcorrespondence analysis (Ter Braak 1986). RESULTS S AMPLING L OCATIONS The13samplinglocationsexhibited distinct differences in anumberofinstream habitat and riparian land use characteristics. Themostobvious difference was an increase in mean annual stream flowfrom187 cfs(5.3 m 3 /sec) at Station1–1637 cfs (46.3 m 3 /sec) at Station11. The mean annual flow of the Guest River at StationTGR was 144 cfs (4.1 m 3 /sec). Mean riffle/run current velocities during summer sampling events weretypicallyinthe range of 28–37 cm/sec, with alow of 8cm/secatStation4and ahigh of 65 cm/secatStation5.The Clinch River flows in general from northeast to southwest; however,the direction of flowatindividual sampling sites varied greatly. As aresult, with the additional influence of riparian vegetation height, the hours of summersunlight received by individual sites varied from as little as 2–3 hours at sites with north–south orientations (e.g., Stations 5and 9) to 8–10 hours at sites with east–west orientations (e.g., Stations 2and 6). Theheight of streamside vegetation varied greatlybetweensites, and from bank to bank at individual sites, ranging from grasses to mature hardwoods. Thebank height at most sites during normal summerflows rangedfrom 1to3mbefore leveling out onto the flood plain, with the exception of sharp reliefpresent on one side each of Stations 6, 7, 8, and 10. Banks at all sites were fairly steep (30–408 )with the exceptionofStation2,which had a very gradual left-bank slopeofabout 108 .Bank erosion was minimal at mostsites, with alimited amount of exposedsoil observed at Stations 2, 5, 6, 9, 11, and TGR. Theriffle-run areas that represented the primary sampling locations at each site ranged from 14 minlengthatStation 4toanextensive 180-msegment at Station11. Thecorresponding upstream anddownstream pools tended to increase in lengthfromupriver sitestodownriver sites, ranging from 40 matStation 1to275 matStation10. Width-to-depth ratios were fairly uniform,ranging from 26.3 m/m at Station 4to43.4m/m at the Guest River site. Riffle-to-riffle ratios varied by afactor of five betweensites (from 1.6 m/m at Station 11 to 5.6 m/m at Station 4). Bend-to-bend ratios wereall in therange of 14–20m/m except Stations 4(29.6 m/m) and11 (37.2 m/m). Substrate composition from quadrat samples of the riffle/run areas of each site was dominated by the sand/gravel/rubble fractions (Table 12.1). Silt was greater than 5% at Stations 4and 11, while the boulder fraction represented greater than 5% at Stations 4, 6L, and TGR. The rubble fraction was exceptionally high at Station9.Substrate penetration, measured by standard hits to asteel rod, was typically 9–14 cm. The exceptions were Station TGR, where the sediment was most Freshwater BivalveEcotoxicology316 4284X—CHAPTER 12—17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) compacted (mean penetration of 5.3 cm), and Station1,where the sediment was mostunconsoli- dated (mean penetration of 16.5 cm). Riparian land use influencing each site was arbitrarily defined as the drainage area to the stream segment 10 miles (16 km) upstream of each sampling site. This drainage area ranged from 8740 to 15,500 ha forall sitesexceptStation1,for which thearea was3550ha. Within these areas, forestland made up the greatest percentage, ranging from 64 to 77%; the exceptions were Stations 1(46%), 5(52%), and 9(57%). Most of the remainingarea (9.7–54.9%) was agricultural, particu- larly grazing land.The percentage of land apportioned to urban development was typicallylow, ranging from 1to3%except at Stations 3(8%), 4(6%), and TGR (7%). Likewise, mined land comprised fairly low percentages of the land use at each site, typicallylessthan 1% except at Stations 3(3%), 4(3%), 6(2%), 7(2%), and TGR (2%). W ATER C HEMISTRY Clinch River water chemistry did not vary significantly betweensampling sites for most measured parameters, either in the water column or sediment interstitial water (Table 12.2). Noteworthy resultsincludedamean phosphate concentration of 1.4 mg/L (interstitial water)atStation1, which was substantially higher than measurements at the other sites, and significantly ( a Z 0.05) higher than concentrations at Stations 3, 8, 9, 10,and 11. TheGuest River site had significantly higher concentrations of dissolvedmanganese (mean Z 1240 m g/L) than all othersampling stations.Other parameters showingstatisticallysignificantdifferences betweensites included conductivity, hardness, and total recoverable zinc, all of which were higher at the Guest River site than most or all othersites. S EDIMENT C HEMISTRY Two rounds of sediment sampling revealedfew notable differences betweensampling stations (Table 12.2). The Guest River site exhibited higherconcentrations of manganese, nickel, lead, and zinc than the Clinch River mainstem sites. Stations 4and 11 also had fairly high levels of zinc, while Station 4also exhibited ahigher concentration of sediment aluminumthan the other sites. Comparison of these concentrations to screening values for determining possibleareas of sediment TABLE 12.1 Clinch River QuadratSubstrate Composition (MeanPercent by Volume) Station Boulder ( O 256 mm) Rubble (64–256 mm) Gravel (2–64 mm) Sand (0.0625–2 mm) Silt ( ! 0.0625 mm) 11.0 28.3 26.6 42.1 2.0 2037.0 29.2 30.9 2.9 3031.4 27.5 36.6 4.5 47.8 20.8 23.2 40.7 7.5 52.4 45.5 19.7 30.5 1.9 6R 038.1 19.2 38.7 4.0 6L 12.0 32.0 24.6 30.6 0.8 7043.3 23.1 33.3 0.3 8035.4 25.4 35.1 4.1 9059.5 12.5 22.8 5.2 10 2.0 39.7 23.3 30.5 4.5 11 030.8 21.2 41.0 7.0 Guest River 5.5 33.4 32.4 26.7 2.0 Case Study: Discrimination of Factors Affecting Unionid Mussel Distribution 317 4284X—CHAPTER 12—17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) TABLE 12.2 Mean Water and Sediment ChemistryMeasurements at 13 Clinch River Sites, 1992–1994 Sampling Station Parameter Units 123456L 6R 78910 11 TGR Water Column (nZ 8) Temperature 8 C18.3 19.5 18.4 19.6 17.8 19.4 17.4 18.8 19.5 19.0 17.5 17.4 17.3 pH s.u. 8.4 8.5 8.3 8.4 8.3 8.3 8.4 8.3 8.5 8.4 8.3 8.2 8.2 Conductivity m mhos/cm 294 297 300 299 300 315 320 316 313 301 297 303 373 Alkalinity mg/L 141 146 148 174 153 142 150 132 147 137 130 123 116 Hardness mg/L 152 161 165 177 160 159 181 158 168 178 200 186 250 Turbidity NTU 5.3 5.5 6.0 5.3 5.0 6.7 4.8 5.2 5.5 5.9 3.4 7.34.7 NH 3 –N mg/L 0.08 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 TOC mg/L 5345244111 223 PO 4 –P mg/L 0.09 0.07 0.09 0.02 ! 0.01 0.01 ! 0.01 ! 0.01 ! 0.01 ! 0.01 ! 0.01 ! 0.01 0.12 NO 3 –NO 2 mg/L 0.99 0.78 0.66 0.56 0.74 0.66 0.70 0.75 0.77 0.74 0.68 0.50 0.41 Interstitial Water (nZ 8) pH s.u. 7.87.9 8.0 7.87.9 8.0 7.97.9 8.0 8.0 7. 97.9 7.9 Conductivity m mhos/cm 271 266 312 304 303 327 303 284 315 320 329 298 465 NH 3 –N mg/L 0.05 ! 0.05 ! 0.05 0.06 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 ! 0.05 TOC mg/L 6344456587 334 PO 4 –P mg/L 1.40 0.45 0.14 0.97 0.25 0.30 0.34 0.32 0.14 0.12 0.06 0.15 0.75 NO 3 –NO 2 mg/L 0.71 0.50 0.23 0.46 0.44 0.53 0.46 0.52 0.39 0.44 0.96 0.60 0.59 Al-diss. m g/L ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 ! 50 Total m g/L 4570 10,425 3500 22,250 12,725 16,550 5550 11,950 8050 6305 17,825 21,600 7780 As-diss. m g/L ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 Total m g/L ! 4 ! 4 ! 45! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 Cu-diss. m g/L 5111131211 1 ! 11 Total m g/L 15 14 725144823231515142229 Mn-diss. m g/L ! 10 25 ! 10 10 ! 10 ! 10 10 ! 10 10 ! 10 12 12 1240 Total m g/L 1280 452 310 1120 768 727 760 773 515 842 2530 1500 3205 Ni-diss. m g/L ! 3 ! 3 ! 3 ! 3 ! 3 ! 3 ! 3 ! 3 ! 3 ! 3 ! 3 ! 33 Total m g/L 11 10 42513171214810 14 25 57 Pb-diss. m g/L ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 ! 2 Freshwater BivalveEcotoxicology318 4284X—CHAPTER 12—17/10/2006—10:22—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) Total m g/L 43 31 850241619191222204955 Zn-diss. m g/L 666711 76 75 5667 Total m g/L 79 66 28 155 74 73 77 72 62 61 72 118 201 Sediment (nZ 2) Al %1.16 2.66 0.68 5.67 2.47 2.67 3.69 3.73 2.05 1.22 2.88 3.49 2.43 As mg/kg ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 ! 4 Cu mg/kg 17 15 726182337271619183633 Mn mg/kg 890 310 220 780 280 210 220 280 360 310 730 430 1160 Ni mg/kg 14 12 5311226191919921 40 77 Pb mg/kg 18 13 5219717 28 58 94481 Zn mg/kg 48 40 37 94 41 45 73 53 38 34 44 112 148 Case Study: Discrimination of Factors Affecting Unionid Mussel Distribution 319 4284X—CHAPTER 12—17/10/2006—10:22—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) contamination indicated that nickel concentrations at the Guest Riversite(meanZ 77 mg/kg) exceeded the probable effect level of 48.6 mg/kg (MacDonald et al. 2000). F AUNAL S URVEYS Acombinedtotal of 29 species of unionid mussels were collectedfrom 1992 to 1994 at the 13 sampling sites (Table 12.3). Results at individual sites varied dramatically, with no live specimens collected from Stations 3, 6L,and TGR, while 23 species were collected at Station 5. Species dominance among sites conformed to an upstream–downstream trend. Dominant species at Stations 1–4 (CRM 307.7–327.5) were the wavyrayed lampmussel ( Lampsilis fasciola), Cumberland mocca- sinshell ( Medi onidus conradicus ), Tennessee clubshell(Pleurobema oviforme ), kidneyshell ( Ptychobranchusfasciolaris), andrainbow(Villosa iris). Downstream,fromStations 5to11 (CRM213.1–269.5), the dominantspecies were the mucket ( Actinonaias ligamentina), pheasant- shell(Actinonaias pectorosa ), threeridge ( Amblemaplicata), spike(Elli ptio dilatata), and Cumberland moccasinshell. Four federally listed endangeredspecies were collected: tan riffleshell ( E. florentina walkeri), shinypigtoe ( Fusconaia cor), finerayed pigtoe ( Fusconaia cuneolus), and rough rabbitsfoot ( Quadrulacylindrica strigillata), all of which were locally common at one or more sites. Mussel densities in quadrat samples alsovaried greatly betweensites, with the highest densities at Stations 1and 5inthe upper half of the studyarea.Examining the occurrence of young mussels aged five yearsorless at the 13 sites provided another perspective on mussel populations alongthe river. The percentage of young specimens at three upstreamsites (Stations 1, 2, and 5) was much higherthan downstream percentages, while two sites with fairlygood mussel assemblagesinterms of species richnessordensity (Stations 4and 11) produced no young specimens. Asian clam(C. fluminea)density varied significantly ( P ! .01) betweensites, ranging from amean density of 78 clams/m 2 at Station9to 715 clams/m 2 at StationTGR. Fingernail clams (Sphaeriidae) also exhibited significant differencesindensity betweensites, ranging from zero at several sites to 10.4 clams/m 2 at Station4.The fingernail clams were much more prevalent at the five upstreamsites than farther downstream. Electrofishing resultsindicated that good to excellent fish assemblages existed at all sampling sites, with the exception of the Guest River location. IBI scores ranged from 48 to 58 (maximum possiblescore Z 60) at the Clinch River sites, while StationTGR scored a32, for apoorrating. Species composition among the Clinch River mainstem sites was very similar. T ISSUE M EASUREMENTS Tissue metal concentrations in Asian clams exhibited little spatialvariation in three samples from 1991 to 1992. Mean tissue concentration ranges ( m g/g dry weight) among the 13 sites were: As, 2–5; Al, 1236–3070; Cu, 19–41; Pb,2–8; Mn, 92–520; Ni, 2–9; and Zn, 234–380. The only sites significantlydifferent from one or moreothersites wereStations 2and 10 for aluminum (respective meansof3030 and 3070 m g/g comparedto1236–2295 m g/g at remaining sites) and StationTGR for manganese (meanZ 520 m g/g comparedto92–253 m g/g at remaining sites). Measurement of Asian clamcellulolytic activityasanindicator of sublethal stress resulted in extremely variableresults, both spatially between sampling sites and temporally among the six to nine samples collectedateach site from 1991 to 1995. Station9produced the best results consist- ently, with clam cellulolytic activities averaging 97.8% of the activity of control clams. Sites with resultsthatweresignificantlydifferent from Station9included Stations 1(meanZ 49.8%), 4(17.7%), 6L (45.2%), 8(42.8%), 10 (39.4%), 11 (27.2%), andTGR (45.6%). For thetwo samplesmeasuring cholinesteraseactivity, no significant difference was found for any sampling station compared to control activity. Freshwater BivalveEcotoxicology320 4284X—CHAPTER 12—17/10/2006—10:22—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) [...]... Mussel Distribution 4284X CHAPTER 12 17/10/2006—10:22—JEBA—XML MODEL C – pp 311–333 TABLE 12. 3 Density (#/m2) of Unionid Mussels at 13 Clinch River, Virginia, Sites, 1992–1994 322 Freshwater Bivalve Ecotoxicology TOXICITY TESTING In situ toxicity testing of the rainbow (V iris) on two occasions in 1995 resulted in no significant effects on either survival or growth over six-week exposures In both tests,... evidence of likely mining-related effects Elevated concentrations of manganese, nickel, and zinc, although individually known from a variety of sources, together constitute a signature of contamination from coal mine sources (see Chapter 13) The impact is further illustrated in Figure 12. 3, © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) 4284X CHAPTER 12 17/10/2006—10:22—JEBA—XML... effects Channel morphology, as measured by riffle length, upper pool length, and bend-to-bend ratio, was significantly correlated to all three measures of mussel community structure (Table 12. 4) These results reflected a tendency for stronger mussel assemblages in areas of shorter riffle and pool lengths and smaller bend-to-bend ratios These sites offered a wider range of habitat choices for both mussels... of these species In fact, in a follow-up survey, Ahlstedt and Tuberville (1997) noted a lack of recruitment at many sites The nature of a possible negative influence of Asian clams on juvenile unionids is unknown © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) 4284X CHAPTER 12 17/10/2006—10:22—JEBA—XML MODEL C – pp 311–333 328 Freshwater Bivalve Ecotoxicology Unionids ≤ 5 Years... site-specific mussel impact assessments, other than that of Dunn et al (Chapter 13) summarized above, were found Fleming et al (1995) used tissue enzyme analysis to attribute a mussel die-off in Swift Creek, North Carolina, to agricultural pesticide use Goudreau et al (1993) performed a field and laboratory impact © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) 4284X CHAPTER 12 17/10/2006—10:22—JEBA—XML... downstream of the plant for a distance of approximately 12 miles (19 km) This finding precipitated NPDES permit-mandated studies by APCo that determined that the cause of the impact was elevated copper concentrations in a plant effluent—a situation that had most likely existed before the spill to when the facility was constructed in the 1950s A state-of-the-art treatment facility completed in 1993 resulted... compliance with technology-based and water quality-based standards In summary, evaluation of biological and environmental data from 13 sampling sites, encompassing most of the 146 miles (235 km) of the Virginia portion of the Clinch River, has TABLE 12. 6 Species Richness at Five Clinch River Sites Sampled in Three Surveys, 1 912 1994 Sampling Station 2 3 5 8 11 a Ortmann, 1 912 1913 Ahlstedt, 1978–1983... Sampling Station 2 3 5 8 11 a Ortmann, 1 912 1913 Ahlstedt, 1978–1983 This Study, 1992–1994 17 16 24 25 32 10 NSa 12 10 14 10 0 23 6 14 Not sampled © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) 4284X CHAPTER 12 17/10/2006—10:22—JEBA—XML MODEL C – pp 311–333 330 Freshwater Bivalve Ecotoxicology demonstrated strong statistical relationships between mussel community structure and the... River Basin, Tennessee, North Carolina, Virginia, and Georgia 1994–1998, U.S Geol Surv., Water Resour Div., Circ 120 5, Reston, VA, 2000 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) 4284X CHAPTER 12 17/10/2006—10:22—JEBA—XML MODEL C – pp 311–333 332 Freshwater Bivalve Ecotoxicology Helfrich, L A., Weigmann, D L., Neves, R J., and Bromley, P T., The Clinch, Powell, and Holston... transportation-related and other types of pollutant spills These types of ephemeral 30 Density (#/m2) No Species % Young 25 20 % Young 15 # Species 10 Density 5 0 5 7 9 11 13 15 Substrate Penetrability (cm) 17 19 FIGURE 12. 4 Relationship of substrate penetrability with three mussel community measures at 13 Clinch River sites © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) 4284X CHAPTER 12 17/10/2006—10:22—JEBA—XML . 4 Cu-diss. m g/L 51111 3121 1 1 ! 11 Total m g/L 15 14 725144823231515142229 Mn-diss. m g/L ! 10 25 ! 10 10 ! 10 ! 10 10 ! 10 10 ! 10 12 12 124 0 Total m g/L 128 0 452 310 1120 768 727 760 773 515. length, width-to-depth ratio (average width divided by average thalweg depth of runs and pools), riffle-to-riffle ratio (average distance betweenriffles divided by average width), and bend- to-bend ratio. at Freshwater BivalveEcotoxicology 312 4284X CHAPTER 12 17/10/2006—10:21—JEBA—XML MODEL C–pp. 311–333 © 2007 by the Society of Environmental Toxicology and Chemistry (SETAC) representative riffle-run