2012 NORTHEASTERN NATURALIST 19(3):431–444 Physiographic and Anthropogenic Factors Influencing fish Community Composition in Tributaries to the Youghiogheny River in Pennsylvania David G. Argent1,* and William G. Kimmel1 Abstract - We evaluated the status of fi sh communities from 30 fi rst- to fourth-order wadeable Youghiogheny River (YR) tributaries located in the Allegheny Mountain Section (AMS) and Pittsburgh Low Plateau (PLP) physiographic provinces in Pennsylvania. The AMS is underlain primarily by geologic formations dominated by the Allegheny Group, while the Casselman Formation underlies much of the PLP. Streams draining the AMS were primarily forested, low-order (1–2), high-gradient, and poorly-buffered with minimal cultural influences. By contrast, the majority of the PLP was drained by low-gradient, well-buffered, valley streams of order >2, with varying anthropogenic influences. In order to characterize connectivity, all streams were sampled from their mainstream junctions to an endpoint 200 m upstream utilizing back-pack electrofi shing. A total of 366 fi shes representing seven families, 17 genera, and 22 species were collected from AMS streams, while the PLP yielded 2738 fi shes representing 8 families, 25 genera, and 41 species. five AMS streams and one from the PLP were fi shless. The principal factors influencing fi sh distribution, species richness, and density on AMS streams were order, total alkalinity, and barriers to mainstem connectivity; in addition to stream order and total alkalinity, fi sh communities of the PLP were affected to varying degrees by cultural stressors including surface mining and urbanization. This approach could be applied throughout the United States at regional appropriate scales to document zoogeographic patterns of fi sh distribution relative to contemporary and future anthropogenic influences. Introduction The Youghiogheny River (YR) flows north from its headwaters in West Virginia 214 km to its confluence with the Monongahela River (MR) near Mckeesport, Pennsylvania (Bryant and McPhilliamy 1973). Two mainstem reaches of the YR have been impounded, creating Deep Creek Lake in Maryland and the Youghiogheny Reservoir in Pennsylvania. The 4585-km2 YR Basin (USEPA 1971) encompasses two distinct Physiographic Provinces: the Allegheny Mountain Section (AMS) and the Pittsburgh Low Plateau (PLP) within Pennsylvania (Sevon 2000), extending from West Virginia to Connellsville and from Connellsville to its MR confluence, respectively (fig. 1). The entire watershed lies within the Appalachian Ecoregion level III 70 – Western Allegheny Plateau, encompassing parts of Pennsylvania, Ohio, West Virginia, and Virginia (Omernik 1987). The AMS is characterized by ridges, steep-sided gorges, and rocky floors. The underlying geology is largely dominated by the Allegheny Group consisting of shale and sandstone with some coal, fi reclay, and limestone (Hickok and Moyer 1940). Small pockets of the Pottsville Group, which contain thin shales and coal, are also present (Dicken et al. 2008). Along its unimpounded 120-km reach below !California University of Pennsylvania, Department of Biological and Environmental Sciences, California, PA 15419. *Corresponding author - argent@calu.edu. 432 Northeastern Naturalist Vol. 19, No. 3 the Youghiogheny Reservoir, the YR is joined by a number of tributaries which have been designated as High Quality Cold Water fisheries (HQ-CWF), Cold Water fisheries (CW), or Exceptional Value (EV) by the Pennsylvania Department of Environmental Protection (PA DEP) ( (PA Code 2001; WRAS 2001, 2003). The PLP by contrast, consists of relatively shallow valleys with smooth, rolling upland surfaces. The uplands are developed on rocks underlain by signifi cant bituminous coal reserves contained within the Cassleman Formation and Monongahela Group (Dicken et al. 2008). Tributaries draining the PLP include those designated as Trout Stocking Waters (TSF) and Warm Water fisheries (WWF) (PA Code 2001). Deposits of bituminous coal within both the AMS and PLP were extensively mined and subsequently abandoned, leaving a landscape of open pits and refuse piles. Long after the cessation of active operations, a number of these sites continue to pollute surface streams with acid (AMD) and net alkaline (NAMD) mine drainage from point and non-point sources. Acid mine drainage has resulted in widespread degradation of water quality and impairment of aquatic communities throughout coal mining regions of the northeastern United States. Hendricks (1980) stated that acidic coal mine drainage and siltation are the most severe and limiting pollutants to aquatic organisms in the YR. In addition, acid deposition (AD) throughout the Basin has negatively impacted aquatic life in the poorly buffered HQ-CWF and EV streams that join the YR from Chestnut Ridge and Laurel Hill (Kimmel 1999; Kimmel et al. 1996; LeFevre and Sharpe 2002; Sharpe et al. 1985, 1987). By the early 1950s, AMD had resulted in several major fi sh kills and a seemingly lifeless river (Palmer 1984). Like the MR (Argent et al. 2007), the YR figure 1. The YR Basin in Pennsylvania and locations of sampling stations. 2012 D.G. Argent and W.G. Kimmel 433 benefi tted from passage of environmental legislation and improving water quality; however, the threats from its mining legacy remain. Periodic acid discharges from tributaries like the Casselman River (Ziemkiewicz and Brant 1997) and Cherry Creek (MDE 2003) serve as a reminder of the fragility of its recovery. Over the years, the severity of acidifi cation throughout the YR Basin prompted a number of local watershed groups and Chestnut Ridge Trout Unlimited to acquire fi nancial support for a variety of mitigation projects (e.g., Laurel Hill Creek Water Resources Plan, Glade Run Alkaline Sand Project, and Morgan Run Acid Mine Remediation Project). While signifi cant funds have been expended to ameliorate the effects of acidity within the Basin, long-term monitoring and bioassessment have largely been absent. The earliest published accounts of YR fi shes are those of Cooper (1983) from collections made in the mid-1960s, and, more recently, those of Hendricks (1980) and Hendricks et al. (1983), who documented 57 species from 172 stations across Pennsylvania and Maryland. The Pennsylvania fish and Boat Commission also maintains an unpublished database as part of its statewide stream-assessment protocols. However, of the named Pennsylvania tributaries north of Confluence, fewer than 10% have been sampled since the late 1970s, and no concerted effort has been made to sample tributary/mainstem junctions (Argent et al. 1997). Moreover, two fi sh species of conservation concern—Catostomus catostomus (Forster) (Longnose Sucker), a YR endemic, and Pararhinichthys bowersi (Goldsborough and Clark) (Cheat Minnow)—may be extant. Both are considered critically imperiled by the state’s Natural Heritage Inventory and are listed as species of “immediate concern” in Section 15.7 of Pennsylvania’s State Wildlife Action Plan (PWAP 2005). In 2008, the Pennsylvania Department of Conservation and Natural Resources (DCNR) recognized the YR as its “River of the Year” (DiBerardinis 2008). Several conservation plans have been developed for the YR Basin to ensure its continued protection: “River Conservation Plan for the Middle Youghiogheny River Corridor” and the “Youghiogheny Rivers Conservation Plan” (PA Bulletin 2001). Each plan identifi es a need for the establishment of a monitoring network for the Basin and an evaluation of aquatic fauna. Of particular interest are the tributaries, which may serve as temporal refuges, breeding grounds, and nursery areas for fi sh, and as sources of recolonizers for the YR concomitant with improving water quality. The principal objectives of this study were to characterize fi sh communities in wadeable tributary/mainstream junctions of the YR Basin and to identify physiographic and anthropogenic factors which influence community structure. The area of study encompassed named wadeable tributaries that join the YR over its 120-km free-flowing reach downstream of the Youghiogheny Reservoir dam from Confluence to McKeesport, PA (fig. 1). Methods We sampled 30 named wadeable YR tributaries, 13 and 17 on the AMS and PLP, respectively, during May and June 2009. Major tributaries such as the Casselman River and Laurel Hill Creek were not included in this survey because their size and depth precluded comparable sampling effort with our selected 434 Northeastern Naturalist Vol. 19, No. 3 protocol (Kimmel and Argent 2006). A standard sampling distance beginning just above the tributary/river confluence and extending 200 m upstream or to the nearest blockage to fi sh passage comprised a sampling station. If no fi sh were captured during the fi rst 100 m, electrofi shing was terminated. The midpoint of each sampling station was identifi ed by GPS coordinates, and the total length (m) and mean width (m) were determined. Streams were classifi ed by size using Strahler (1952). Two-passes were employed to collect fi sh from each sampling station, utilizing DC-pulsed back-pack electrofi shing gear (Kimmel and Argent 2006). In order to optimize and to standardize gear effi ciency, output was adjusted according to local specifi c conductance to produce approximately 100–300 volts and draw 2–3 amps. Game fi shes and large specimens were identifi ed in the fi eld and released. All others were fi xed in 10% formalin on-site, taken to the laboratory at California University of Pennsylvania, and refrigerated for approximately 24 hr. All specimens were then preserved in 70% isopropyl alcohol and identifi ed to the lowest practicable taxonomic level, typically species. Measurements of temperature (oC), pH, and specifi c conductance (μS/cm) were taken in the fi eld using a standard thermometer, pH Testr3+ (Oakton Instruments), and a conductivity meter (YSI Model 33 S-C-T, Yellow Springs Instruments), respectively. A 500-ml water sample was collected at each site and brought to the laboratory for analysis of total alkalinity (mg/l as CaCO3) by titration. Observable incidences of pollution (e.g., AMD, NAMD, and point sources of raw or treated sewage effluent) were noted. Because the YR Basin encompasses two distinct physiographic provinces and associated geologic formations, we elected to evaluate our data by summarizing fi sh community and water quality from each province separately. Employing ArcGIS ver. 9.3 (ESRI 2008), we identifi ed the proportional contribution of the respective geologic formations (PBTGS 1995) underlying each watershed. We computed species richness and density (fi sh/m2) for fi sh collected from each site. fish community metrics were compared with water quality parameters and geologic data to defi ne patterns of fi sh distribution. Because collected fi sh and water quality data did not meet the assumptions of normality, the nonparametric Mann-Whitney U-test was utilized to determine if differences existed between stations within the AMS and PLP. To evaluate the relationships between fi sh community metrics (e.g., richness and density) and water quality/geology variables, we used Spearman’s rank correlation. All statistical tests were evaluated at a signifi cance level of P < 0.05. Results/Discussion Sampling-reach lengths ranged from 80 to 200 m, while mean widths ranged from 2.1 to 23.8 m and stream orders were from 1–4 (Table 1). Although the PA DEP designated use classifi cation (PA Code 2001) identifi es most all AMS and PLP streams as CW or WWF, respectively (Table 2), stream temperatures at time of sampling did not differ signifi cantly (Mann-Whitney U-test: P-value > 0.05; Table 3). Likewise, pH values were not signifi cantly different between the two physiographic regions (Mann-Whitney U-test: P-value > 0.05; Table 3). Total alkalinity and specifi c conductance, both of which were signifi cantly higher in 2012 D.G. Argent and W.G. Kimmel 435 the PLP (Mann Whitney U-test: P-value < 0.05; Table 3), differentiate water quality between the two physiographic regions. A total of 3104 fi shes were collected from 24 of the 30 streams sampled. No signifi cant relationship existed between area sampled and species richness or density (Spearman’s rank correlation: P-value > 0.05). Neither Longnose Sucker nor Cheat Minnow were captured during this survey. Allegheny Mountain Section The majority of AMS streams were fi rst or second order (Table 1) and drain forested landscapes underlain largely by the Allegheny Group with local influences from the Pottsville Group (Table 4). These exhibited depressed alkalinity and specifi c conductance values accompanied by anomalous elevated pH readings (Table 3). The pH and alkalinities documented here (Table 3) were not predictable from the local geology (DeWalle et al. 1987), and no evidence of Table 1. Stream orders and dimensions of sampling stations on YR tributaries draining the AMS and PLP physiographic provinces in Pennsylvania. Streams are arranged in geographic order from south to north within each physiographic province. Stream Stream order Reach length (m) Mean reach width (m) Allegheny Mountain streams Hen Run 1 200 4.7 Ramcat Run 1 200 6.1 Camp Run 1 100 2.6 Lick Run 1 200 5.0 Cucumber Run 3 80 7.0 Meadow Run 4 200 12.5 Bear Run 2 100 5.5 Jonathan Run 2 100 7.7 Sugar Run 1 100 1.8 Bruner Run 1 200 5.9 Johnson Run 1 144 4.3 Morgan Run 2 200 10.0 Indian Creek 4 200 23.8 Pittsburgh Low Plateau streams Dunbar Creek 3 200 11.6 Opossum Run 2 200 6.7 Mounts Creek 3 200 7.2 Galley Run 2 200 2.9 Hickman Run 2 100 2.1 Dickerson Run 2 200 5.4 Smiley Run 2 200 7.5 Washington Run 2 200 6.8 Browneller Run 2 200 3.4 Jacobs Creek 3 200 25.7 Furnace Run 2 200 2.2 Virgin Run 2 200 5.2 Cedar Creek 2 200 6.7 Sewickley Creek 4 200 21.5 Pollock Run 2 200 8.1 Douglas Run 2 200 9.0 Long Run 3 200 7.9 436 Northeastern Naturalist Vol. 19, No. 3 confounding cultural influences was observed. The local geology, mostly lacking carbonate-bearing strata, would predict sub-neutral pH values associated with the lack of buffering capacity. Alkalinity in surface streams is imparted to groundwater flowing through deposits of carbonate minerals, typically limestone. In well-buffered streams, aquatic life is protected from rapid pH change due to acid inputs and benefi ts from nutrients associated with alkalinity. Values of baseflow alkalinity identifi ed as less than 10 mg/l (Omernik 1983) and less than 20 mg/l as CaCO3, except where natural conditions are less (PA Code 2001), are indicative of systems sensitive to acidifi cation. Weiner (2000) further states that for protection of aquatic life, the buffering capacity should be at least 20 mg/l. Therefore, these low-order tributaries may be considered as extremely sensitive to acidifi cation. Three of the AMS streams, Hen and Meadow runs and Indian Creek, are separated from the rest on the basis on alkalinity and conductivity and, in the cases of Meadow Run and Indian Creek, by size (3rd and 4th order, respectively). A total of 366 fi shes representing 7 families, 17 genera, and 22 species were collected here (Table 5). first- and second-order streams here were largely depauperate (fig. 2) five fi rst-order streams were fi shless (Table 1, fig. 2), and of these, Johnson and Cucumber runs contained impediments to fi sh passage that prevented sampling over the prescribed 200-m reach (Table 1). Camp and Lick runs, which harbored 0 and 3 fi shes, respectively, are underlain by 70% Pottsville Group (Table 4), an acidic stratum (Sharpe et al. 1987). Overall, streams devoid of fi sh and lacking impediments to passage exhibited alkalinity values ranging from 4 to 10 mg/l, while those which yielded fi sh ranged from 2 to 10 mg/l total alkalinity (Table 3, fig. 2). Table 2. Designated uses of AMS and PLP tributaries to the YR in Pennsylvania. Streams are arranged in geographic order from south to north. Allegheny Mountain Pittsburgh Low Plateau streams Designated use* streams Designated use* Hen Run CWF Dunbar Creek TSF Ramcat Run CWF Opossum Run WWF Camp Run HQ-CWF Mounts Creek WWF Lick Run HQ-CWF Galley Run WWF Cucumber Run CWF Hickman Run WWF Meadow Run HQ-CWF Dickerson Run WWF Bear Run EV Smiley Run WWF Jonathan Run EV Washington Run WWF Sugar Run EV Browneller Run WWF Bruner Run EV Jacobs Creek WWF Johnson Run HQ-CWF Furnace Run WWF Morgan Run HQ-CWF Virgin Run TSF Indian Creek HQ-CWF Cedar Creek TSF Sewickley Creek WWF Pollock Run WWF Douglas Run WWF Long Run TSF *PA DEP Designated Use: CWF = Coldwater fishery, EV = Exceptional Value, HQ-CWF = High Quality Coldwater fishery, TSF = Trout Stocking fishery, and WWF = Warmwater fishery. 2012 D.G. Argent and W.G. Kimmel 437 Twelve of the 13 AMS streams are located within the confi nes of Ohiopyle State Park (fig. 1) and classifi ed as either EV or HQ-CWF. The dominant geological formation is the Allegheny Group, ranging from 0–99% coverage (Table 4), with the proportion of Glenshaw Formation (Table 4) a modest predictor of species richness (non-linear regression: P-value < 0.05, R2 = 0.50). fish associations, depauperate in species richness and density where present, were typically located within the fi rst 20 m of a sampling reach. Such “remnant” or “transient” communities have been described from poorly buffered streams on nearby Laurel Hill as a result of AD (Sharpe et al. 1987). Studies by Haines (1981), Sharpe et al. (1985), and others reported episodic acidifi cation associated with high flows during snowmelt, which may depress pH values and elevate dissolved aluminum concentrations to levels toxic to fi sh. Among streams within the AMS, Hen Run, Meadow Run, and Indian Creek exhibited the highest values of total alkalinity, specifi c conductance (Table 3), Table 3. Selected water quality parameters of AMS and PLP streams. Streams are arranged in geographic order from south to north within each physiographic province. Conductivity Alkalinity Stream Temperature (oC) pH (μS/cm) (mg/l) Allegheny Mountain streams Hen Run 9.0 8.70 80 16 Ramcat Run 9.0 8.50 25 4 Camp Run 9.2 8.80 60 8 Lick Run 10.3 9.03 38 8 Cucumber Run 10.0 8.20 55 10 Meadow Run 17.0 7.60 150 24 Bear Run 11.9 8.79 40 6 Jonathan Run 9.6 9.15 73 4 Sugar Run 9.2 8.75 44 10 Bruner Run 10.0 8.50 50 4 Johnson Run 10.5 8.54 54 14 Morgan Run 15.0 7.70 185 2 Indian Creek 19.5 7.70 240 16 Pittsburgh Low Plateau Streams Dunbar Creek 17.5 8.20 160 34 Opossum Run 14.5 8.70 405 150 Mounts Creek 15.7 8.56 264 54 Galley Run 12.0 7.00 700 56 Hickman Run 9.0 6.10 600 4 Dickerson Run 10.0 8.60 470 150 Smiley Run 12.5 8.10 175 34 Washington Run 11.5 8.80 420 138 Browneller Run 12.0 8.90 435 220 Jacobs Creek 13.0 8.40 340 52 Furnace Run 17.0 7.90 223 68 Virgin Run 20.0 8.50 160 48 Cedar Creek 10.0 8.90 500 224 Sewickley Creek 19.0 8.50 800 136 Pollock Run 8.0 8.80 500 234 Douglas Run 7.0 7.90 700 236 Long Run 11.2 8.30 700 130 438 Northeastern Naturalist Vol. 19, No. 3 Table 4. Geologic strata expressed as percent contribution within the AMS. Streams are arranged in geographic order from south to north. Allegheny Glenshaw Pottsville Burgoon Stream Group Formation Group Sandstone Hen Run 93 7 0 0 Ramcat Run 91 7 2 0 Camp Run 0 0 70 10 Lick Run 0 0 70 10 Cucumber Run 99 1 0 0 Meadow Run 98 1 1 0 Bear Run 97 0 3 0 Jonathan Run 98 2 0 0 Sugar Run 99 0 1 0 Bruner Run 97 2 1 0 Johnson Run 99 0 1 0 Morgan Run 97 1 1 0 Indian Creek 81 17 1 0 Table 5. Ichthyofauna collected from one or more YR tributaries draining the AMS and PLP in Pennsylvania. Family/common name Scientifi c name AMS PLP Salominidae Brook Trout Salvelinus fontinalis Mitchill 6 2 Rainbow Trout Oncorhynchus mykiss (Waldbaum) 3 1 Brown Trout Salmo trutta L. 2 Cyprinidae Central Stoneroller Campostoma anomalum (Rafi nesque) 1 174 Striped Shiner Luxilus chrysocephalus Rafi nesque 1 Common Shiner Luxilus cornutus (Mitchill) 5 Spotfi n Shiner Cyprinella spiloptera (Cope) 31 River Chub Nocomis micropogon (Cope) 57 1 Golden Shiner Notemigonus crysoleucus (Mitchill) 4 Emerald Shiner Notropis atherinoides Rafi nesque 66 Sand Shiner Notropis stramineus (Cope) 4 Rosyface Shiner Notropis rubellus (Agassiz) 3 40 Mimic Shiner Notropis volucellus (Cope) 606 Channel Shiner Notropis wickliffi Trautman 5 Shiner sp. Notropis sp. 7 Bluntnose Minnow Pimephales notatus (Rafi nesque) 2 285 Fathead Minnow Pimephales promelas Rafi nesque 1 21 Blacknose Dace Rhinichthys atratulus (Hermann) 58 112 Longnose Dace Rhinichthys cataractae (Valenciennes in Cuvier 52 11 and Valenciennes) Creek Chub Semotilus atromaculatus (Mitchill) 15 204 Catostomidae White Sucker Catostomus commersoni (Lacépède) 3 120 Northern Hogsucker Hypentelium nigricans (Lesueur) 7 104 Black Redhorse Moxostoma duquesnei (Lesueur) 2 Golden Redhorse Moxostoma erythrurum (Rafi nesque) 8 Ictaluridae Brown Bullhead Ameiurus nebulosus (Lesueur) 10 Yellow Bullhead Ameiurus natalis (Lesueur) 2 Stonecat Noturus flavus Rafi nesque 10 2012 D.G. Argent and W.G. Kimmel 439 species richness, and density (fig. 3). These streams are underlain by 81–99% Allegheny Group (Table 4). Only Hen Run harbored native Salvelinus fontinalis (Brook Trout), while Ameiurus nebulosus (Brown Bullhead) was collected only from Meadow Run. Indian Creek yielded 57 of the 58 Nocomis micropogon (River Chub) captured. Together, these three streams accounted for 76% of the total number of fi shes and 77% of overall species richness (fig. 3). The paucity of fi shes within the low-order tributaries draining the AMS (Table 1, fig. 3) may be indicative of historical and/or current episodic acidifi cation resulting from low buffering capacity, stream order, thermal preference, and/ or or obstacles to riverine connectivity. The near total absence of Brook Trout is noteworthy, as these streams fall within the expected geographic range of this species in the forested cold-water environments designated for its special protection by the PA DEP (e.g., EV and HQ-CW; PA Code 2001). Along with vulnerability to acidifi cation, coldwater species here may also be vulnerable to modest increases in temperature as predicted from several climate change models (Eaton and Sheller 1996). Further, issues of watershed connectivity may exist as a result of natural barriers to passage such as waterfalls and log jams as well as culverts under the YR Trail. Stations within the AMS that yielded no fi sh in 2009 were resampled in 2010 along their designated reach and an additional 100-m reach upstream of their impediments to fi sh passage. Low numbers of Rhinichthys atratulus (Blacknose Dace), Rhinichthys cataractae (Longnose Dace), and Cottus bairdi (Mottled Table 5, continued. Family/common name Scientifi c name AMS PLP Centrarchidae Rock Bass Ambloplites rupestris (Rafi nesque) 23 Green Sunfi sh Lepomis cyanellus Rafi nesque 1 48 Pumpkinseed Lepomis gibbosus (L.) 10 Bluegill Lepomis macrochirus Rafi nesque 4 60 Sunfi sh Hybrid Lepomis hybrid 1 Smallmouth Bass Micropterus dolomieui Lacépède 14 33 Largemouth Bass Micropterus salmoides (Lacépède) 1 Spotted Bass Micropterus punctulatus (Rafi nesque) 1 Black Crappie Pomoxis nigromaculatus (Lesueur) 1 Percidae Greenside Darter Etheostoma blennioides Rafi nesque 2 180 Rainbow Darter Etheostoma caeruleum Storer 339 Fantail Darter Etheostoma flabellare Rafi nesque 15 55 Johnny Darter Etheostoma nigrum Rafi nesque 2 Banded Darter Etheostoma zonale (Cope) 50 Yellow Perch Perca flavescens (Mitchill) 2 3 Logperch Percina caprodes (Rafi nesque) 1 Darter Hybrid Percina sp. 1 Scianidae Freshwater Drum Aplodinotus grunniens Rafi nesque 2 Cottidae Mottled Sculpin Cottus bairdi Girard 106 70 Slimy Sculpin Cottus cognatus (Richardson) 33 Total 366 2738 440 Northeastern Naturalist Vol. 19, No. 3 Sculpin) were collected near each mainstem confluence, suggesting the presence of transient associations derived from the mainstem. Pittsburgh Low Plateau section Valley streams here, second to fourth order (Table 1) and influenced by a variety of anthropogenic factors, are designated as WWF or TSF by the PA DEP (PA Code 2001; Table 2). All but Hickman Run exhibited concomitant elevated values of total alkalinity, pH, and specific conductance (Table 3). These values are consistent with the presence of the Monongahela Group and Glenshaw Formations, which frequently contains deposits of limestone, within the underlying geology (Table 6). A total of 2738 fishes representing 8 families, 25 genera, and 41 species were collected from PLP streams (Table 5). Although cultural influences such as AMD, NAMD, and untreated/treated sewage discharges impact a number of streams here, fish community species richness and density greatly exceeded those of the AMS (Mann-Whitney U-test: P-value < 0.05). Hickman Run was the only stream that yielded no fish (fig. 2), a likely consequence of persistent raw sewage observed in the field. Historically, both Galley Run and Sewickley Creek were impacted by AMD. However, a comprehensive remediation strategy implemented on Sewickley Creek (Capacasa 2009) has initiated a recovery that is evidenced by its high species richness (fig. 2); Galley Run, currently impacted by NAMD, exhibits higher species richness and density than most AMS streams (fig. 2). figure 2. A comparison of fi sh species richness and density among YR tributaries draining the PLP in Pennsylvania. Bars denote species richness, and black squares denote density. 2012 D.G. Argent and W.G. Kimmel 441 figure 3. A comparison of fi sh species richness and density among YR tributaries draining the AMS in Pennsylvania. Bars denote species richness, and black squares denote density. Table 6. Geologic strata expressed as percent composition within the PLP. Streams are arranged in geographic order from south to north. Allegheny Casselman Glenshaw Monongahela Stream Group Formation Formation Group Dunbar Creek 47 9 43 0 Opossum Run 0 0 0 75 Mounts Creek 44 12 43 1 Galley Run 0 92 0 8 Hickman Run 0 92 0 8 Dickerson Run 0 89 0 7 Smiley Run 0 100 0 0 Washington Run 0 20 61 18 Browneller Run 0 53 0 47 Jacobs Creek 0 23 70 6 Furnace Run 9 48 43 0 Virgin Run 0 81 13 5 Cedar Creek 0 0 0 99 Sewickley Creek 0 22 56 22 Pollock Run 0 53 0 47 Douglas Run 0 53 0 47 Long Run 0 62 0 37 442 Northeastern Naturalist Vol. 19, No. 3 The combination of alkalinity, stream size (Karr 1981), and riverine connectivity likely account for the increased species richness and density over those of AMS streams (fig. 2). Along with buffering capacity, alkalinity is associated with increased secondary biological productivity (Death and Winterbourn 1995, Rahel 1986). However, the presence of the Casselman Formation in excess of 30% negatively affected species richness for most streams here (Spearman’s Rank Correlation: P-value < 0.05, r = -0.52). The role of connectivity among PLP streams may also be a contributor to diversity and density, as no stream contained a barrier to fi sh passage within 200 m of its YR confluence (Table 1). Summary This study represents the first comprehensive survey of YR tributary fish associations at their mainstream/tributary junctions. Historically, the YR was severely impacted by AMD, but now supports a significant recreational fishery as a result of a comprehensive mitigation strategy. The two physiographic provinces traversed by the River are drained by streams very different in their underlying geology, water quality, size, cultural influences, and ichthyofauna. The low-order forested AMS streams are characterized by poor buffering capacity, low conductivity, and anomalous elevated values of pH. The ichthyofauna here is depauperate in species richness and density, with 5 of the 13 tributaries being fishless, although all are afforded special protection as cold-water fisheries. Further, only one stream supported Brook Trout. Major threats to these fragile streams and their remnant fish populations include acid deposition and impediments to mainstream connectivity. By contrast, the larger valley streams of the PLP are well-buffered, and in spite of a variety of pollutants, overall support a healthy and diverse ichthyofauna. The AMS and PLP tributaries represent geologically distinct components of the YR river basin in spite of their geographic proximity, and highlight the need for comprehensive watershed surveys. Conservation of biological resources depends upon focused management strategies in light of threats that, in turn, require accurate and timely inventories of faunal associations along with their spatial distributions. Such is the case with expanding natural gas development from the Marcellus and Utica formations which underlie large portions of Pennsylvania (including the YR Basin), West Virginia, New York, Kentucky, and Ohio. Tributary/mainstream junction surveys on large watersheds such as this provide baseline information on water quality and ichthyofaunal communities against which future degradations can be assessed. Acknowledgments We wish to thank the PA Wild Resource Conservation Fund for providing fi nancial support for this survey (WRCP Grant #08289) and Michael Brady, Michael McCoy, Carlyle Meekins, and Mark Novak, fi eld technicians. Andrew Saint John and Barbara Wallace from the PA Department of Conservation and Natural Resources assisted with stream access and permitting. 2012 D.G. Argent and W.G. Kimmel 443 Literature Cited Argent, D.G., R.F. Carline, and J.R. Stauffer, Jr. 1997. Historical and contemporary distribution of fi shes in Pennsylvania Technical Report. RWO - 47. USGS Biological Resources Division, University Park, PA. 278 pp. Argent, D.G., W.G. Kimmel, R. Lorson, and E. Emery. 2007. Ichthyofauna of the Monongahela River Basin in Pennsylvania: A contemporary evaluation. Journal of Freshwater Ecology 22:617–628. Bryant, G., and S. McPhilliamy. 1973. Water quality assessment report: Monongahela River Basin. US EPA, Region II, Wheeling Offi ce, WV. 18 pp. Capacasa, J.M. 2009. Decision rationale total maximum daily loads for acid mine drainage affected segments Sewickly Creek Watershed TMDL, Westmoreland County, Pennsylvania. US Environmental Protection Agency, Philadelphia, PA. 10 pp. Cooper, E.L. 1983. The fishes of Pennsylvania and the Northeastern United States. The Pennsylvania State University Press, University Park, PA. 243 pp. Death, R.G., and M.J. Winterbourn 1995. Diversity patterns in stream benthic invertebrate communities: The influence of habitat stability. Ecology 76:1446–1460. DeWalle, D.R., R.S. Dinicola, and W.E. Sharpe. 1987. Predicting baseflow alkalinity as an index to episodic stream acidifi cation and fi sh presence. Journal of the American Water Resources Association 23:29–35. DiBerardinis, M. 2008. Pennsylvania Department of Conservation and Natural Resources press conference. Harrisburg, PA. 18 March 2008. Dicken, C.L., S.W. Nicholson, J.D. Horton, S.A. Kinney, G. Gunther, M.P. Foose, and J.A.L. Mueller. 2008. Preliminary integrated geologic map databases for the United States: Delaware, Maryland, New York, Pennsylvania, and Virginia. Available online at http://pubs.usgs.gov/of/2005/1325/. Accessed 1 March 2011. Eaton, J.G., and R.M. Scheller. 1996. Effects of climate warming on fi sh thermal habitat in streams of the United States. Limnology and Oceanography 41:1109–1115. ESRI 2008. ArcGIS Desktop: Release 9.3. Environmental Systems Research Institute. Redlands, CA. 600 pp. Haines, T.A. 1981. Acidic precipitation and its consequences for aquatic ecosystems: A review. Transactions of the American fisheries Society 110: 669–707. Hendricks, M.L. 1980. The distribution of fi shes of the Youghiogheny River. M.Sc. Thesis. Frostburg State College, Frostburg, MD. 216 pp. Hendricks, M.L., J.R. Stauffer, Jr., and C.H. Hocutt. 1983. The zoogeography of fi shes of the Youghiogheny River system, Pennsylvania, Maryland, and West Virginia. American Midland Naturalist 110(1):145–164. Hickok, W.O. IV, and F.T. Moyer. 1940. Geology and mineral resources of Fayette County, Pennsylvania. Pennsylvania Geological Survey County Report 26. Harrisburg, PA. 530 pp. Karr, J. 1981. Assessment of biotic integrity using fi sh communities. fisheries 6:21–27. Kimmel, W.G. 1999. Macroinvertebrate community responses to episodic stream acidifi cation on the Laurel Hill in Southwestern Pennsylvania. Pp. 17–22, In W.E. Sharpe and J.R. Drohan (Eds.). The Effects of Acidic Deposition on Aquatic Ecosystems in Pennsylvnaia. Proceedings of the 1998 PA Acidic Deposition Conference, Vol. II. Environmental Resources Research Institute, University Park, PA. 61 pp. Kimmel, W.G., and D.G. Argent. 2006. Effi cacy of two-pass electrofi shing employing multiple units to assess stream fi sh species richness. fisheries Research 82:14–18. Kimmel, W.G., E.L. Cooper, and C.C. Wagner. 1996. Macroinvertebrate and fi sh populations of four streams receiving high rates of hydrogen and sulfate ion deposition. Journal of Freshwater Ecology 11:493–511. LeFevre, S.R., and W.E. Sharpe. 2002. Acid stream water remediation using limestone sand on Bear Run in southwestern Pennsylvania. Restoration Ecology 10:223–236. 444 Northeastern Naturalist Vol. 19, No. 3 Maryland Department of the Environment (MDE). 2003. Total maximum daily loads to address low pH in Cherry Creek in the Deep Creek Watershed, Garrett County, Maryland. Montgomery Business Park Center, Baltimore, MD. 27 pp. Omernik, J.M. 1983. Total alkalinity of surface waters: A national map. Annals of the Association of American Geographers 73:133–136. Omernik, J.M. 1987. Ecoregions of the conterminous United States. Annals of the Association of American Geographers 77:118–125. Palmer, T. 1984. Youghiogheny: Appalachian River. University of Pittsburgh Press, Pittsburgh, PA. 352 pp. Pennsylvania (PA) Bulletin. 2001. Department of Conservation and Natural Resources notices: Middle Youghiogheny River Corridor Conservation Plan. 31:6523–6646. Pennsylvania Bureau of Topographic and Geologic Survey (PBTGS). 1995. Physiographic provinces 1:100,000. Department of Conservation and Natural Resources. Harrisburg, PA. Pennsylvania (PA) Code. 2001. Title 25. Environmental Protection – Chapter §93. Department of Environmental Protection, Bureau of Water Supply, Harrisburg, PA. 280 pp. Pennsylvania’s Wildlife Action Plan (PWAP). 2005. Pennsylvania’s Wildlife Action Plan: Version 1.0a. Compiled and edited by the Pennsylvania Game Commission and Pennsylvania fish and Boat Commission, Harrisburg, PA. 762 pp. Rahel, F. 1986. Biogeographic influences on fi sh species composition of northern Wisconsin lakes with applications for lake acidifi cation studies. Canadian Journal of fisheries and Aquatic Sciences 43:124–134. Sevon, W.D. 2000. Physiographic provinces of Pennsylvania. Pennsylvania Department of Conservation and Natural Resources, Bureau of Topographic and Geologic Survey, Harrisburg, PA. 4th Series, Map 13, scale 1:2,000,000. Sharpe, W.E., D.R. DeWalle, R.T. Leibfried, R.S. Dinicola, W.G. Kimmel, and L.S. Sherwin. 1985. Causes of acidifi cation of four streams on Laurel Hill in southwestern Pennsylvania. Journal of Environmental Quality 13:3619–631. Sharpe, W.E., V.G. Leibfried, W.G. Kimmel, and D.R. DeWalle. 1987. The relationship of water quality and fi sh occurrence to soils and geology in an area of high hydrogen and sulfate ion deposition. Journal of American Water Resources Association 23:37–46. Strahler, A.N. 1952. Dynamic basis of geomorphology. Geological Society of America Bulletin 63:923–938. US Enviornmental Protection Agency (USEPA). 1971. A report on pollution of the Ohio River and its tributaries in Pittsburgh, Pa., area. USEPA Region III, Cincinnati, OH. 93 pp. Watershed Restoration Action Strategy (WRAS). 2001. State Water Plan Subbasin 19D. Lower Youghiogheny River Watershed, Westmoreland and Fayette counties. DEP Bureau of Watershed Management, Harrisburg, PA. 8 pp. WRAS. 2003. State Water Plan Subbasin 19E. Upper Youghiogheny River Watershed, (Laurel Hill Creek and Indian Creek), Fayette, Somerset, and Westmoreland counties. DEP Bureau of Watershed Management, Harrisburg, PA. 13 pp. Weiner, E.R. 2000. Applications of Environmental Chemistry: A Practical Guide for Environmental Professionals. CRC Press LLC, Boca Raton, FL. 456 pp. Ziemkiewicz, P.F., and D.L. Brant. 1997. The Casselman River restoration project. West Virginia Surface Mine Drainage Task Force Symposium, Morgantown, WV, April 15–16, 1997. National Mine Land Reclamation Center, Morgantown, WV.