Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 1 2018 NORTHEASTERN NATURALIST 25(1):1–20 Changes in the Status of Native Brook Trout on Laurel Hill, Southwestern Pennsylvania David G. Argent1.*, William G. Kimmel1, and Derek Gray2 Abstract - To evaluate the status of native Salvelinus fontinalis (Brook Trout) on Pennsylvania’s Laurel Hill, we sampled fish, assessed habitat, and documented water quality from 20 non-randomly selected headwater streams of northwest- and southeast-facing slopes. In late spring and early summer of 2011 and 2014–2016, we sampled fish communities and measured specific conductance (μS/cm), total alkalinity (mg/l as CaCO3), pH, and total dissolved aluminum (2011 and 2016). In addition, in 2015 we determined land-use patterns, riparian canopy, and substrate composition. Mean pH values among the streams recently assessed were significantly higher than historic values; however, all other water-quality parameters were similar. Native Brook Trout were present in all streams, and annual natural reproduction was evident in 90% of streams. Even though fish were present, we observed marked declines in total catch in both 0-age and adult trout; the overall reduction approached 60% when compared with those documented in 1983. We discuss possible causes for the observed declines, including acid deposition, introduction of nonnative/invasive species, water withdrawal, habitat fragmentation/alteration, predation, and climate change. Introduction Conservation of ecologically sensitive riverscapes and their attendant flora and fauna may require continued reassessments of established biolog ical baselines. Anthropogenic activities such as landscape disturbance (Kelly et al. 1980, Kocovsky and Carline 2006), nonnative fish introductions (Larson and Moore 1985), and climate change (Argent and Kimmel 2013) may combine to alter ecosystems at local and regional levels (Hudy et al. 2008). Some states have long-term historical profiles of their aquatic resources that are temporal baselines against which current environmental perturbations can be measured. The New York Department of Environmental Conservation maintains one such example of an extensive database documenting ichthyofaunal assemblages of lotic and lentic waters over time (Carlson et al. 2016). There is no such comprehensive monitoring and assessment program in Pennsylvania, and there are few historical accounts of the assemblage diversity and geographic distributions of its resident ichthyofauna. Data is particularly needed for the nearly 80,000 km of headwater streams in the state—less than half of which have ever been sampled (Argent et al. 2003)—and improved understanding of the fish assemblages in these waters is critical as emerging threats may irreversibly impact these fragile coldwater ecosystems where the native Salvelinus fontinalis (Mitchill) (Brook Trout) is a keystone species (Tzilkowski 2005). 1California University of Pennsylvania, 250 University Avenue, California, PA 15419. 2Wilfrid Laurier University, 75 University Avenue West, Waterloo, ON N2L 3C5, Canada. *Corresponding author - argent@calu.edu. Manuscript Editor: Stuart Welsh Northeastern Naturalist 2 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 Hudy et al. (2008) examined historical accounts of native Brook Trout distribution in the eastern US. They reported that native Brook Trout were not detected and were possibly extirpated from 1760 (33%) of 5279 sub-watersheds. In Pennsylvania, anthropogenic impacts of immediate concern are natural-gas extraction from the Marcellus shale layer (Wagner et al. 2014, Weltman-Fahs and Taylor 2013), water withdrawal (EBTJV 2011), and climate change (Argent and Kimmel 2013). These 3 stressors threaten the ecological integrity of the Commonwealth’s special protection waters, including those designated as High-Quality Coldwater Fishery and Exceptional Value (PADEP 2017). Historically, sulfur dioxide emitted largely from coal-fired power plants in the Ohio Valley caused wet and dry acid deposition within the Laurel Highlands, resulting in high sulfate loads in forest soils. Spring runoff from melting snow and storm events produced acidic pulses and elevated dissolved aluminum in poorly buffered headwater streams (Sharpe et al. 1984). Concern about water quality and fish assemblages in this region prompted a study of 61 Laurel Hill streams in 1983 by Sharpe et al. (1987) as part of the National Acid Precipitation Assessment Program (NAPAP 1998). Specifically, Sharpe et al. (1987) evaluated impacts of acid deposition and provided the only historic comprehensive assessment of ichthyofaunal assemblages of Laurel Hill streams, herein identified as the historical baseline. Since this study, passage of The Clean Air Act Amendments of 1990 has resulted in declines of sulfur dioxide emissions and improving water quality in the northeastern US (Stoddard et al. 2003). To assess the current status of historically self-sustaining populations of native Brook Trout on Laurel Hill, we compared results of our recent assessments (2011 and 2014–2016) with those of Sharpe et al. (1987). This region has been undersampled over the years and, therefore, represents a data gap in our understanding of Brook Trout distribution. Our objectives were to (1) compare the total catch of native Brook Trout from 1983 with contemporary collections, (2) document changes in fish-assemblage composition (e.g., shifts among resident species), and (3) discuss possible causations for the observed patterns. Methods Laurel Hill is a part of the Allegheny Mountain System of the Greater Appalachian Plateau Province of southwestern Pennsylvania. This 110-km anticlinal fold, located ~80 km southeast of Pittsburgh, is oriented along a northwest/southeast axis with an average elevation of 820 m (Shultz 1999). Laurel Hill lies in the mixed mesophytic forest region of Pennsylvania, and current canopy consists of 2nd and 3rd growth due to extensive logging in the late 1800s and early 1900s. At present, largely intact forests cover much of the area, which is protected from commercial development by substantial tracts of state parks, state forests, and state gamelands. Most recently, some state agencies have allocated leases for shale-gas development within public-land boundaries potentially resulting in habitat fragmentation and water withdrawal for hydraulic fracking (Weltman-Fahs and Taylor 2013). Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 3 We followed the methods of Sharpe et al. (1987) to classify streams based on the viability of their respective fish assemblages—fish present, fish absent, culturally impaired, and remnant fish population. We applied the fish present or culturally impaired designations to those streams harboring self-sustaining native Brook Trout populations that exhibited multiple year-class structures. Remnant populations consisted of a few adults only, with no evidence of stable year-class structure or reproduction. For this study, we selected a total of 20 streams: 19 classified as having fish present and 1 deemed to be culturally impaired. Ten were on the NW and 10 on the SE slopes of Laurel Hill. These streams spanned the length of the anticline in approximately paired positions on each facing slope (Fig. 1). During the 1983 study, all 20 streams harbored native Brook Trout populations consisting of at least 3 age-classes, including 0-age fish. Sampling locations and Pennsylvania special protection designations (PADEP 2017) of Laurel Hill streams are summarized in Table 1. Eighteen of the 20 headwater streams and their attendant watersheds are designated high-quality coldwater fishery or exceptional value status, and are afforded the highest levels of protection by the PADEP (2017; Table 1). The Laurel Hill streams are still bordered by largely intact riparian forest cover (PASDA 2013). Field notes taken in 1983 subjectively described benthic habitats as cobble, rubble, gravel, or sand, and provided species composition of canopy cover (W.E. Sharpe et al.,The Pennsylvania State University, University Park, PA, unpubl. field notes). To quantify substrate and riparian canopy-cover suitability for Figure 1. Locations of headwater streams surveyed on Laurel Hill. Delineated polygons demarcate the watershed boundary for each sampled stream. Northeastern Naturalist 4 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 native Brook Trout, we performed pebble counts using methods described by Bain and Stevenson (1999) and evaluated canopy cover using a densitometer (Johansson 1985) within each reach surveyed in 2015. During May and June of 2011 and 2014–2016, we employed the methodology of Sharpe et al. (1987) to document water quality and sample fish assemblages of each stream. We conducted all sampling within 100-m reaches located at or near reaches identified in field notes from the historical survey. We surveyed each site for ~30 minutes. We recorded specific conductance (μS/cm), pH, and temperature (o C) on-site and collected a sample for measurement of total alkalinity (mg/l as CaCO3) in the laboratory at California University of Pennsylvania. In 2011 and 2016, we took samples from each stream and sent them to H & H Water Controls Laboratory in Carmichaels, PA, for total dissolved aluminum (mg/l) analysis. We employed 1-pass backpack electrofishing (Model LR-24 Smith-Root Shocker, Smith-Root, Inc., Vancouver, WA; 300–400 volts, and 30–40 Hz) to sample the fish assemblage of each stream. We measured total length (TL) of each native Brook Trout to the nearest mm, and classified those less than 75 mm as young-of-the-year (YOY). We categorized all larger trout as adults. We used our best professional judgment to differentiate native trout from those of hatchery origin using size, shape, and color as primary determinants. We identified, enumerated, and released all other captured fish, which primarily comprised Cottus bairdii (Girard) (Mottled Sculpin). We used repeated-measures analysis of variance to test for differences in specific conductance (conductivity), alkalinity, pH, dissolved aluminum, and Table 1. Locations, Pennsylvania State designated uses, and canopy-cover proportions of sampled streams on Laurel Hill. * CWF = cold water fishery, EV = exceptional value, and HQ-CWF = high quality-cold water fishery (PADEP 2001). Stream name Slope Latitude (°N) Longitude (°W) Designated use % canopy cover Baldwin Run NW 40.33639 79.05039 EV 95 Bear Run South NW 39.89981 79.46442 EV 87 Lick Run NW 40.31831 79.08051 EV 88 M Fork Mill Creek NW 40.24911 79.14178 EV 78 Neals Run NW 40.03382 79.34998 HQ-CWF 76 N Fork Mill Creek NW 40.25198 79.14701 HQ-CWF 84 Powdermill Run N. NW 40.36241 79.02843 EV 89 Roaring Run South NW 40.06257 79.34493 EV 85 SF Sugar Run NW 40.38551 79.02408 CWF 74 Tubmill Run NW 40.31490 79.08909 EV 90 Allwine Creek SE 40.28449 79.03439 EV 90 Dalton Run SE 40.29488 79.01640 HQ-CWF 88 Little Glade Run SE 39.86293 79.36250 HQ-CWF 77 Little Mill Creek SE 40.31035 78.98829 EV 82 Mill Creek SE 40.30928 78.98798 EV 92 N Branch Bens Creek SE 40.23269 79.04709 EV 90 NF Bens Creek SE 40.26630 79.02016 EV 84 NF Jones Mill Run SE 40.02276 79.26656 EV 88 Shafer Run SE 40.08263 79.23398 CWF 62 SF Jones Mill Run SE 40.02212 79.26778 EV 77 Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 5 adult Brook Trout catch among sampling periods (1983, 2011, and 2014–2016). To meet the assumption of normality and homoscedasticity for ANOVAs, we reciprocal-transformed pH, and log-transformed alkalinity, conductivity, and adult Brook Trout catch. We confirmed normality by examining quantile–quantile plots and examined homoscedasticity with Levine tests (P > 0.05 for all tests). We performed Mauchly’s test to check the assumption of sphericity in our repeated-measures ANOVAs (all P-values > 0.05). If an ANOVA produced a significant result (P < 0.05), we ran Tukey’s honestly significant difference tests to determine which time periods were different from one another for the response variable of interest. We did not analyze Mottled Sculpin and YOY Brook Trout abundance with ANOVAs because the assumption for normal distribution of residuals was not met. Instead, we employed non-parametric Friedman tests with repeated-measures. If a result from the Friedman test was significant, we ran post-hoc tests suggested by Hollander and Wolfe (1999:295) to determine which time periods were different from one another for the response variable of interest. To examine if there was a relationship between Mottled Sculpin and Brook Trout catches, we fitted a linear mixed-effects model with Brook Trout catch as the response variable, Mottled Sculpin catch as a fixed factor, and stream identity as a random factor. A simple linear regression was not possible for this analysis because we collected the catch data via repeated sampling of the 20 streams through time (i.e., all data points were not independent). We loge-transformed Mottled Sculpin abundance and log10-transformed adult and YOY Brook Trout abundances. We performed a likelihood ratio test in the lmtest package (Zeileis and Hothorn 2002) to determine the significance of our mixed-effects model in comparison with an intercept-only null model without predictors. We determined R2 for the mixed-effects model according to Nakagawa and Schielzeth (2013), and calculated confidence intervals for predictions of the mixed-effects model using the bootMer function in the lme4 package (Bates et al. 2015). We conducted all analyses in the R programming language (Bliese 2016, R Core Team 2016). Results Our evaluation of water-quality parameters revealed no significant differences among sampling years for total alkalinity (Table 2, Fig. 2A). Specific conductance was consistent among years, with no clear trend over time (Table 2, Fig. 2B); pH values recorded after 1983 were significantly elevated between 2011 and 2015, but not in 2016 (Table 2, Fig. 2C). Although values of total dissolved aluminum for 2011 and 2016 were elevated in comparison with 1983 (Table 2, Fig. 2D), they were still far below the 200-μg/l toxicity threshold for Brook Trout and the historical threshold mean of 512 μg/l described for the fish absent category (Sharpe et al. 1987). Canopy cover varied from 62% to 95% (mean = 84%; Table 1) and a cumulative plot of substrate particle size among the 20 streams indicated that most stream substrates were 5–180 mm in diameter (Fig. 3). In summary, the physical habitat and water quality of sampled streams have remained relatively stable since 1983, with improving pH conditions. Northeastern Naturalist 6 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 Table 2. Results of repeated measures ANOVAs to compare water-quality parameters and adult Brook Trout catch among sampling periods in 20 streams. Degrees of Sum of Mean sum Variable Source freedom squares of squares F P Eta2 pH Year 4 0.0061 0.0015 29.7052 0.0000 0.4149 Error 76 0.0039 5.1938x10-5 Alkalinity Year 4 0.3172 0.0793 1.3555 0.2573 0.0300 Error 76 4.4462 0.0585 Conductivity Year 4 0.2877 0.0712 4.2748 0.0035 0.0888 Error 76 1.2787 0.0168 Dissolved Aluminum Year 2 2.8439 1.4219 13.7191 0.0000 0.3755 Error 30 3.1094 0.1036 Adult Brook Trout Year 4 5.7517 1.4379 24.2162 0.0000 0.4013 Error 76 4.5128 0.0593 Figure 2. Water-quality summary. (A) Total alkalinity (mg/l), (B) specific conductance (μS/ cm), (C) pH, and (D) total dissolved aluminum (mg/l). Lowercase letters above bars on the pH figure denote significant differences as determined by Tukey HSD tests. The dark line indicates the median, lower and upper hinges correspond to the 1st and 3rd quartiles, and the whiskers extend to the largest value no further than 1.5 times the interquartile range. The dots represent outlying data points. Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 7 In 1983, Sharpe et al. (1987) collected an average of about 35 native Brook Trout per stream. This total included 25 adult fish and 10 YOY per stream (Fig. 4). Although we found no data for surveys conducted between 1983 and 2011, comparisons with the data from 2011 and 2014–2016 reveal an overall decline in adult native Brook Trout of 64% from total catches reported in 1983 (Fig. 5, Appendix A). These declines were significant and relatively consistent between 2011 and 2014– 2016 (Tukey test; P < 0.001). We collected Brook Trout of hatchery origin from the North Fork of Jones Mill Run, North Branch of Bens Creek, and Shafer Run along with several Salmo trutta L. (Brown Trout). Overall, YOY native Brook Trout total catch showed highly variable recruitment rates among sampled streams, but we observed a general pattern of decline in comparison with 1983. Reductions of up to 50% were evident and significant between 1983 and 2011 and 1983 and 2016 (Friedman test with post-hoc analysis, P < 0.05; Fig. 4B, Appendix B), but not between 1983 and 2014 and 1983 and 2015 (Friedman test with post-hoc analysis, P >0.05, Figure 4B). The increase in 0-Age fishes during 2014 can be attributed to 2 streams: Powdermill Run North and Allwine Creek, which yielded 39 and 65 YOY, respectively. These 2 streams accounted for 47% of the total YOY total catch reported for 2014. Except for 2014 (as noted above), recruitment of YOY native Brook Trout dramatically declined when compared with levels documented in 1983 (Fig. 6). Similarly, every age class over our multi-year study revealed declines at time of sampling. Age 5+ fish (>250 mm TL) were present only in 1983. Of the 20 streams examined during our study, Sharpe et al. (1987) classified 19 streams as fish present and 1 as culturally impaired. We used the same historical criteria to reclassify the current status of these streams. We found that only 45% of those streams originally identified as fish present retained that classification during later surveys; 35% were re-classified as remnant fish (T able 3). Figure 3. Cumulative plot of mean substrate-particle size among 20 Laurel Hill streams. Error bars denote 1 standard deviation. Northeastern Naturalist 8 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 Mottled Sculpin, commonly found in association with native Brook Trout in headwater streams at Laurel Hill, were present in 13 of the 20 streams sampled. Unlike the native Brook Trout, they appear to have increased at a relatively steady pace since 2011. The Mottled Sculpin catch was greater than in 1983, but the difference between historic and current catches was not significant (P = 0.072; Fig. 4C, Appendix C). Our linear mixed-effects model demonstrated that there was a significant, but weak negative relationship between Brook Trout catch and Mottled Sculpin catch (P-value compared to null model = 0.003, R2 = 0.10, Fig. 7). Discussion Native Brook Trout populations in headwater streams are known to fluctuate widely in response to changing local abiotic characteristics of these inherently Figure 4. (A) Mean catch by stream of adult native Brook Trout, (B) YOY Brook Trout and (C) Mottled Sculpin among years in 20 Laurel Hill streams. See Figure 2 caption for information on formatting of the box plot. Lowercase letters above bars denote significant differences among years as determined by the Tukey HSD test (Adult Brook Trout) or Post-hoc tests following a Friedman test (YOY Brook Trout and Mottled Sculpin). Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 9 Figure 5. Percent change in catch of native Brook Trout (all age classes) from 1983 levels. Figure 6. Yearclass structure of native Brook Trout collected from 20 Laurel Hill streams comparing 1983 and 2011, and 2014–2016. Northeastern Naturalist 10 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 unstable environments (Hall and Knight 1981, Roghair et al. 2002). Anthropogenic factors may compound, mask, or exacerbate natural fluctuations. The EBTJV (2011) recognized the following as threats to native Brook Trout: climate change, acid deposition, rise in water temperature, urbanization, modification of hydrologic regime, dewatering events, invasive species, habitat degradation, fragmentation, and runoff from abandoned mine lands. Our comparisons of the results of surveys made over 24 years apart indicate a precipitous decline in Brook Trout numbers followed by a period of relative stability, suggesting a new baseline for this species on Laurel Hill. In the discussion below, we describe and speculate on possible impacts of historical and contemporary anthropogenic stressors on native Brook Trout populations in this geographic area, such as acid deposition, introduced/invasive species, competition, fishing pressure, water withdrawal, habitat fragmentation, and climate change. Acid deposition since the Sharpe et al. (1987) study appears to be an unlikely driver (Stoddard et al. 2003) of native Brook Trout declines on Laurel Hill. Although the area received high levels of sulfate deposition during the mid- to late 1980s and many streams showed fish declines due to episodic acidification, none of the 20 streams selected for our survey were identified as impacted by Sharpe et al. (1987). Buffering capacity of these streams prevented or reduced pH declines sufficient to mobilize soluble aluminum from forest soils. Total alkalinity and levels of total dissolved aluminum did not differ from historic values, which indicates retention of buffering capacity among the 20 streams over time. Further, wet sulfate deposition has been declining across the Northeast since the passage of the Clean Air Act Amendments of 1990, and surface waters have responded positively Table 3. An updated classification of Laurel Hill streams using criteria developed from the 1983 survey. Stream name 1983 Classification Contemporary classification Allwine Creek Culturally impacted Culturally impacted Baldwin Run Fish present Fish present Bear Run South Fish present Fish present Dalton Run Fish present Fish present Lick Run Fish present Remnant fish Little Glade Run Fish present Remnant fish Little Mill Creek Fish present Fish present M Ford Mill Creek Fish present Remnant fish Mill Creek Fish present Fish present N Branch Bens Creek Fish present Remnant fish N Fork Mill Creek Fish present Remnant fish Neals Run Fish present Fish present NF Bens Creek Fish present Fish present NF Jones Mill Run Fish present Fish absent Powdermill Run N. Fish present Fish present Roaring Run South Fish present Remnant fish SF Jones Mill Run Fish present Remnant fish SF Sugar Run Fish present Fish absent Shafer Run Fish present Fish absent Tubmill Run Fish present Fish present Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 11 (Stoddard et al. 2003). The significant increase in overall stream pH values may be a response to enhanced regulation of sulfur dioxide emissions and seems to have benefitted native Brook Trout populations on Laurel Hill. Interspecific and intraspecific competition for resources can influence growth rates, fecundity, and survival of native Brook Trout (Marchand and Boisclair 1997, Marschall and Crowder 1996). Several studies have documented the ability of non-native Brown Trout to (1) depress local densities of native Brook Trout, and Figure 7. (A) Relationship between Mottled Sculpin and native Brook Trout adult total catch among Laurel Hill streams. (B) Relationship between Mottled Sculpin and YOY native Brook Trout total catches. The shaded area represents the 95% confidence interval of the model predictions. Northeastern Naturalist 12 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 (2) exclude native Brook Trout from preferred habitat (DeWald and Wilzbach 1992, Fausch 1988, Fausch and White 1981). We found a few introduced (stocked) salmonids in several streams, but due to the relative isolation of the Laurel Hill collective, it seems unlikely that exotic species or fishing pressure (Marschall and Crowder 1996) played a major role in native Brook Trout population dynamics. The increase in Mottled Sculpin abundance may be associated with declines in native Brook Trout. Native Brook Trout and either Mottled Sculpin or Cottus cognatus Richardson (Slimy Sculpin) typically co-dominate the ichthyofaunal assemblages of many Pennsylvania headwater streams (Cooper 1983); hence, native Brook Trout and Mottled Sculpin have a long history of coexistence. Moreover, no studies have documented interspecific competition between native Brook Trout and sculpins at a level negatively impacting either species (Zimmerman and Vondracek 2006). However, a decline in native Brook Trout numbers may allow sculpins to increase, or perhaps some aspects of sculpin behavior may be a cause of or contributor to the observed decline of Brook Trout populations. For example, several studies have documented that some sculpin species in the genus Cottus are egg predators of salmonids (Biga et al. 2008, Fitzsimons et al. 2006, Marsden and Tobi 2014, Mirza and Chivers 2002). Additional studies are needed to determine if consumption of trout eggs by Mottled Sculpin may affect the native Brook Trout populations of Laurel Hill. Additionally, Mottled Sculpins may forage on other early life-stages of native Brook Trout, which could lead to a decline in recruitment of native Brook Trout. We did not detect Mottled Sculpin from Allwine Creek or Powdermill Run North; both streams experienced spikes in native Brook Trout recruitment. Recently, requests for water-withdrawal permits from surface and ground waters on Laurel Hill have been on the rise in support of a variety of development projects in the area. Water withdrawals effectively remove a portion of the streamflow with no return until late winter or spring when snowpack melts (Cunjak 1996). Competition for water resources has emerged as a contentious public issue on Laurel Hill among developers of recreational residences (American Rivers 2009), municipalities, and 2 large ski resorts. In addition, pending and realized gas extraction may further exacerbate conflicts among stakeholders. However, there is no documentation of historical or recent declines in stream discharge among the 20 Laurel Hill streams we surveyed. Several studies have focused on the effects of habitat fragmentation on fish populations including such factors as persistence, dispersal, growth, expression of life-history stages, and impediments to gene flow (Fahrig 2003, Letcher et al. 2007, Morita and Yamamoto 2002, Roberts et al. 2013). Fragmentation resulting in population isolation can occur because of a number of anthropogenic factors including pollution, road and dam construction, water diversion, climate change, and, most recently on Laurel Hill, shale-gas development (Hansbarger et al. 2010, Weltman- Fahs and Taylor 2013). Fragmentation can strongly influence population persistence and expression of life-history strategies in spatially structured populations. Letcher et al. (2007) Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 13 reported that, in naturally isolated tributaries, native Brook Trout persistence was associated with higher early juvenile survival (~45% greater), shorter generation time, and strong selection against large body size. Moreover, barriers to upstream migration caused rapid (2–6 generations) local extirpation. Although our quantitative measures of substrate and riparian canopy are not directly comparable to historical data, they do indicate the presence of suitable native Brook Trout spawning substrates and cover for YOY (Fig. 3). Raleigh (1982) described suitable spawning substrate for native Brook Trout as gravel 3–8 cm in diameter and ≤5% fines, criteria met among streams sampled on Laurel Hill. In addition, canopy cover varied from 62% to 95% (mean = 84%; Table 1), suggesting that streams were well shaded and stream banks were largely intact. Shafer Run, identified earlier in this paper as receiving Brook Trout of hatchery origin, maintains the lowest proportion of canopy cover. From these observations and comparative measures of water quality parameters, we concluded that habitat suitability for trout remained largely unchanged over the >24-year interval between our study and that of Sharpe et al. (1987). It would thus seem unlikely that the decline in Laurel Hill native Brook Trout populations can be attributed to a singular large-scale physical habitat change. However, climate change can also result in habitat fragmentation because elevated temperatures may restrict connectivity of watershed tributary networks (Hansbarger et al. 2010, Letcher et al. 2007, Meisner 1990), reduce fish survival (Xu et al. 2010a), and reduce fish growth (Xu et al. 2010b). Fragmentation of such habitats may ultimately lead to reductions in genetic variation among and within such isolated resident fish populations (Whiteley et al. 2013). Argent and Kimmel (2013) described the potential effects of climate change on Laurel Hill native Brook Trout populations, and documented varying patterns of air/ instream temperature relationships (thermal sensitivity [r]) in 6 (3 on each slope) of the 20 Laurel Hill streams described here. We documented largely intact riparian cover in the surveyed reaches; thus, it seems likely that the major factor controlling r would be variation in groundwater input. Canopy cover and groundwater input have both been documented as important factors in predicting r (Kelleher et al. 2012) and native Brook Trout occurrence (Kanno et al. 2015a, 2015b). In-stream/ air temperature profiles from the respective NW- and SE-slope receiving streams, suggest that avenues of tributary connectivity may be temporally constricted by elevated temperatures (Argent and Kimmel 2013). While speculative, the climatechange scenario is worthy of note and may impact streams exhibiting reductions in canopy cover. Further, in-stream temperature change may influence speciesassemblage dynamics. For example, Mottled Sculpins exhibit a greater maximum threshold-temperature tolerance (24.3 °C) than native Brook Trout (22.4 °C) (Eaton and Scheller 1996), and may experience an advantage if water temperatures increase in streams of Laurel Hill (Ar gent and Kimmel 2012). The similarity of water quality and habitat conditions documented between the historic and recent Laurel Hill surveys and the scope of the overall native Brook Trout population declines seem to rule out the aforementioned stressors acting Northeastern Naturalist 14 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 independently or in concert at the local level. Natural-gas extraction (Weltman- Fahs and Taylor 2013) may play a role in both water withdrawal and habitat fragmentation in the near future, but there is no available evidence of widespread historical impacts from such activities at this time. We recognize the large time gap between sampling periods (24+ years), but assert this comparison provides the only means for a much needed reassessment of the historical baseline. Moreover, we realize that native Brook Trout often experience unpredictable shifts in population structure (Hall and Knight 1981; Kanno et al. 2015b, 2016; Roghair et al. 2002), which explains why we extended our study to include 4 years of data. We believe that the longer sample period adds strength to our findings and provides a basis in which we establish a new contemporary baseline. In summary, it is not possible at this time to identify single or multiple causes for fish-assemblage changes in Laurel Hill streams. The literature provides some indication as to what might be happening, but a definitive reason for the nearly 60% decline since 1983 remains unknown. Companion studies in Maryland identify 5 reasons for native Brook Trout decline: high water temperature, agriculture, urbanization, non-native species invasions, and poor riparian habitat (Heft 2006). Based on our study, urbanization, agriculture, non-native species, and poor-quality riparian habitat seem unlikely causative agents for the observed decline in native Brook Trout populations among the Laurel Hill collective. Given the observed decline in resident stream-dwelling native Brook Trout populations on Laurel Hill, researchers and natural resource managers should consider further investigations on the reasons for decline, which could include systematic, temporal, and comprehensive surveys. The decline in resident native Brook Trout populations in Laurel Hill streams underscores the importance of biomonitoring and assessment of aquatic communities facing anthropogenic changes that may create new baselines of community diversity and structure. This study establishes a new baseline for native Brook Trout populations on Laurel Hill for the assessment of current and future anthropogenic stressors. Understanding of the limitations of adaptability and resilience (Adger and Kelly 2000) in these fish assemblages is crucial to their conservation, and a future program of dedicated monitoring may provide the necessary data to accomplish this goal. Acknowledgments We thank the Wild Resources Conservation Fund (Contract # WRCP – 10371) for providing financial support for this project. Jeff Ambrose, Chris Warden, Justin Peel, Benjamin Trask, Nathan Backenstose, and Austin Hess provided assistance with field collections, and Pat Suschak, Paul Knupp, Barry Zaffuto, and Lee Miller enabled access to various sampling sites. Lastly, we acknowledge the comments provided by 2 reviewers and the contributions of Manuscript Editor Dr. Stuart Welsh to improve this paper. Literature Cited Adger, P.M., and W.N. Kelly. 2000. Theory and practice in assessing vulnerability to climate change and facilitating adaptation. Climate Change 47:325– 352. Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 15 American Rivers. 2009. Americas most endangered rivers, 2009 Edition. Washington, DC. 4 pp. Argent, D.G., and W.G. Kimmel. 2012. Climate change in headwater streams: A baseline for monitoring and assessment. Final Report for Grant Agreement WRCP – 10371. Wild Resources Conservation Fund, Harrisburg, PA. 24 pp. Argent, D.G., and W.G. Kimmel. 2013. Potential impacts of climate change on Brook Trout, Salvelinus fontinalis, populations of streams draining the Laurel Hill in Pennsylvania. Journal of Freshwater Ecology 28:489–502. Argent, D.G., J.A. Bishop, J.R. Stauffer Jr, R.F. Carline, and W.L. Myers. 2003. Predicting fish habitat using landscape-level variables: A conservation approach. Fisheries Research 60:17–32. Bain, M.B., and N.J. Stevenson. 1999. Aquatic Habitat Assessment: Common Methods. American Fisheries Society, Bethesda, MD. 216 pp. Bates, D., M. Maechler, B. Bolker, and S. Walker. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67:1–48. Biga, H., J. Janssen, and J.E. Marsden. 1998. Effect of substrate size on Lake Trout egg predation by Mottled Sculpin. Journal of Great Lakes Research 2 4:464–473. Bliese, P. 2016. Multilevel modeling in R: A brief introduction to R, the multilevel package, and the nmle package. Available online at https://cran.r-project.org/doc/contrib/ Bliese_Multilevel.pdf. Accessed 5 September 2016. Carlson, D.M., R.A. Daniels, and J.J. Wright. 2017. Atlas of Inland Fishes of New York. New York State Museum Record 7. The New York State Education Department and the New York State Department of Environmental Conservation, Ithaca, NY. Available online at http://easternbrooktrout.org/ebtjv-reports/ebtjv-conservation-strategy/view/. Accessed 29 August 2017. Cooper, E.L. 1983. The Fishes of Pennsylvania and the Northeastern United States. The Pennsylvania State University Press, University Park, PA. 252 pp. Cunjak, R. 1996. Winter habitat of selected stream fishes and potential impacts from landuse activity. Canadian Journal of Fisheries and Aquatic Sciences 53 (Suppl. 1):267–282. DeWald, L., and M.A. Wilzbach. 1992. Interactions between native Brook Char and hatchery Brown Trout: Effects on habitat use, feeding, and growth. Transactions of the American Fisheries Society 121:287–296. Eastern Brook Trout Joint Venture (EBTJV). 2011. Conserving the eastern Brook Trout: Action strategies. Conservation Strategy/Habitat Work Group Eastern Brook Trout Joint Venture. Available online at http://easternbrooktrout.org/ebtjv-reports/ebtjv-conservation- strategy/view/. Accessed 3 May 2016. Eaton, J.G., and R.M. Scheller. 1996. Effects of climate warming on fish thermal habitat in streams of the United States. Limnology and Oceanography 41:1 109–1115. Fahrig, L. 2003. Effects of habitat fragmentation on biodiversity. Annual Review of Ecology, Evolution, and Systematics 34:487–515. Fausch, K.D. 1988. Tests of competition between native and introduced salmonids in streams: What have we learned? Canadian Journal of Fisheries and Aquatic Sciences 45:2238–2246. Fausch, K.D., and R.J. White. 1981. Competition between Brook Trout (Salvelinus fontinalis) and Brown Trout (Salmo trutta) for positions in a Michigan stream. Canadian Journal of Fisheries and Aquatic Sciences 38:1220–1227. Fitzsimons, J., B. Williston, G. Williston, G. Bravener, J.L. Jonas, R.M. Claramunt, J.E. Marsden, and B.J. Ellrott. 2006. Laboratory estimates of salmonine egg predation by Round Gobies (Neogobius melanostomus), sculpins (Cottus cognatus and C. bairdi), and crayfish (Orconectes propinquus). Journal of Great Lakes Research 32:227–241. Northeastern Naturalist 16 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 Hall, J.D., and N.J. Knight. 1981. Natural variation in abundance of salmonid populations in streams and its implications for design of impact studies, a review. EPA/600/3- 81/021. US Environmental Protection Agency, Corvallis, OR. 97 pp. Hansbarger, J.L., J.T. Petty, and P.M. Mazik. 2010. Brook Trout movement within a highelevation watershed: Consequences for watershed restoration. Pp. 74–84, In J.S. Rentch and T.M. Schuler (Eds.). Conference on the ecology and management of high-elevation forests in central and southern Appalachian Mountains General Technical Report NRSP- 64, US Department of Agriculture, Forest Service, Northern Research Station, Slaty - fork, WV. Heft, A.A. (Ed.). 2006. Maryland Brook Trout fisheries management plan. Maryland Department of Natural Resources Fisheries Service, Inland Fisheries Management Division, Annapolis, MD. 130 pp. Hollander, M., and D.A. Wolfe. 1999. Nonparametric Statistical Methods, 2nd Edition. John Wiley and Sons, New York, NY. 816 pp. Hudy, H., T.M. Thieling, N. Gillespie, and E.P. Smith. 2008. Distribution, status, and land-use characteristics of subwatersheds within the native range of Brook Trout in the eastern United States. North American Journal of Fisheries Management 28:1069–1085. Johansson, T. 1985. Estimating canopy density by the vertical-tube method. Forest Ecology Management 11:139–144. Kanno, Y., B.H. Letcher, A.L. Rosner, K.P. O’Neil, and K.H. Nislow. 2015a. Environmental factors affecting Brook Trout occurrence in headwater stream segments. Transactions of the American Fisheries Society 144:373–382. Kanno, Y., B.H. Letcher, N.P. Hitt, D.A. Boughton, J.E.B. Wofford, and E.F. Zipkin. 2015b. Seasonal weather patterns drive population vital rates and persistence in a stream fish. Global Change Biology 21:1856–1870. Kanno, Y., K.C. Pregler, N.P. Hitt, B.H. Letcher, D.J. Hocking, and J.E.B. Wofford. 2016. Seasonal temperature and precipitation regulate Brook Trout young-of-the-year abundance and population dynamics. Freshwater Biology 61:88–99. Kelleher, C., T. Wagener, M. Gooseff, B. McGlynn, K. McGuire, and L. Marshall. 2012. Investigating controls on the thermal sensitivity of Pennsylvania streams. Hydrological Processes 26:771–785. Kelly, G.A., J.S. Griffith, and R.D. Jones. 1980. Changes in distribution of trout in Great Smoky Mountains National Park, 1900–1970. US Fish and Wildlife Service Technical Paper 102. Washington, DC. 10 pp. Kocovsky, P.M., and R.F. Carline. 2006. Influence of landscape-scale factors in limiting Brook Trout populations in Pennsylvania streams. Transactions of the American Fisheries Society 135:76–88. Larson, G.L., and S.E. Moore. 1985. Encroachment of exotic Rainbow Trout into stream populations of native Brook Trout in the southern Appalachian Mountains. Transactions of the American Fisheries Society 114:195–203. Letcher, B.H., K.H. Nislow, J.A. Coombs, M.J. O’Donnell, and T.L. Dubreuil. 2007. Population response to habitat fragmentation in a stream-dwelling Brook Trout population. PLoS ONE 2(11):e1139. DOI:10.1371/journal.pone.0001139. Marchand, F., and D. Boisclair. 1997. Influence of fish density on the energy-allocation pattern of juvenile Brook Trout (Salvelinus fontinalis). Canadian Journal of Fisheries and Aquatic Sciences 55:796–805. Marschall, E.A., and L B. Crowder. 1996. Assessing population responses to multiple anthropogenic effects: A case study with Brook Trout. Ecological Applications 6:152–167. Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 17 Marsden, J.E, and H. Tobi. 2014. Sculpin predation on Lake Trout eggs in interstices: Skull compression as a novel foraging mechanism. Copeia 4:654–658. Meisner, J.D. 1990. Effect of climatic warming on the southern margins of the native range of Brook Trout, Salvelinus fontinalis. Canadian Journal of Fisheries and Aquatic Sciences 47:1065–1070. Mirza, R.S., and D.P. Chivers. 2002. Attraction of Slimy Sculpin to chemical cues of Brook Char eggs. Journal of Fish Biology 61:532–539. Morita, K., and S. Yamamoto. 2002. Effects of habitat fragmentation by damming on the persistence of stream-dwelling char populations. Conservation B iology 16:1318–1323. Nakagawa, S., and H. Schielzeth. 2013. A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods in Ecology and Evolution 4:133–142. National Acid Precipitation Assessment Program (NAPAP). 1998. NAPAP Biennial report to Congress: An integrated assessment. National Acid Precipitation Assessment Program. Silver Spring, MD. 152 pp. Pennsylvania Department of Environmental Protection (PADEP). 2017. Commonwealth of Pennsylvania, Pennsylvania Code, Title 25. Environmental Protection Agency. Harrisburg, PA. Available fromonline at http://www.pacode.com/secure/data/025/025toc.html. Accessed 29 August 2017. Pennsylvania Spatial Data Analysis (PASDA). 2013. PAMAP Program Land Cover for Pennsylvania, 2005. Available online at http://www.pasda.psu.edu/. Accessed 3 August 2015. R Core Team. 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available online at https://www.R-project. org/. Accessed 17 August. Raleigh, R.F. 1982. Habitat suitability-index models. Brook Trout. FWS/Obs-82/10.24. US Department of Interior, Fish and Wildlife Service, Washington, DC. 53 pp. Roberts, J.J., K.D. Fausch, D.P. Peterson, and M.B. Hooten. 2013. Fragmentation and thermal risks from climate change interact to affect persistence of native trout in the Colorado River basin. Global Change Biology 19:1383–1398. Roghair, C.N., C.A. Dolloff, and M.K. Underwood. 2002. Response of a Brook Trout population and instream habitat to a catastrophic flood and debris flow. Transactions of the American Fisheries Society 131:718–730. Sharpe, W.E., D.R. DeWalle, R.T. Leibfried, R.S. Dinicola, W.G. Kimmel, and L.S. Sherwin. 1984. Causes of acidification of four streams on Laurel Hill in Southwestern Pennsylvania. Journal of Environmental Quality 13:619–631. Sharpe, W.E., V.G. Leibfried,W.G. Kimmel, and D.R. DeWalle. 1987. The relationship of water quality and fish occurrence to soils and geology in an area of high hydrogen and sulfate ion deposition. Water Resources Bulletin 23:37–46. Shultz, C.H. 1999. The Geology of Pennsylvania. Pennsylvania Geological Survey and Pittsburgh Geological Society, Harrisburg, PA. 888 pp. Stoddard, J.L., J.S. Kahl, F.A. Deviney, D.R. DeWalle, C.T. Driscoll, A.T. Herlihy, J.H. Kellogg, P.S. Murdoch, J.R. Webb, and K.E. Webster. 2003. Response of surface-water chemistry to the Clean Air Act Amendments of 1990. EPA 620/R-03/00. National Health and Environmental Effects Research Laboratory Research Triangle Park, Raleigh, NC. 78 pp. Tzilkowski, C.J. 2005. Native Brook Trout and naturalized Brown Trout effects on two Pennsylvania headwater stream food chains. Ph.D. Dissertation. The Pennsylvania State University, University Park, PA. 216 pp. Northeastern Naturalist 18 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 Wagner, T., J.T. Deweber, J. Detar, D. Kristine, and J.A. Sweka. 2014. Spatial and temporal dynamics in Brook Trout density: Implications for population monitoring. North American Journal of Fisheries Management 34:258–269. Weltman-Fahs, M., and J.M. Taylor. 2013. Hydraulic fracturing and Brook Trout habitat in the Marcellus Shale region: Potential impact and research ne eds. Fisheries 38:4–15. Whiteley, A.R., J.A. Coombs, M. Hudy, R. Zachary, A.R. Colton, K.H. Nislow, and B.H. Letcher. 2013. Fragmentation and patch-size shape genetic structure of Brook Trout populations. Canadian Journal of Fisheries and Aquatic Sciences 70:678–688. Xu, C.L., B.H. Letcher, and K.H. Nislow. 2010a. Size-dependent survival of Brook Trout Salvelinus fontinalis in summer: Effects of water temperature and stream flow. Journal of Fish Biology 76:2342–2369. Xu, C.L., B.H. Letcher, and K.H. Nislow. 2010b. Context-specific influence of water temperature on Brook Trout growth rates in the field. Freshwater Biology 55:2253–2264. Zeileis, A., and T. Hothorn. 2002. Diagnostic checking in regression relationships. R News 2:7–10. Available online at https://CRAN.R-project.org/doc/Rnews/. Accessed 27 August 2017. Zimmerman, J.K.H., and B. Vondracek. 2006. Interactions of Slimy Sculpin (Cottus cognatus) with native and nonnative trout: Consequences for growth. Canadian Journal of Fisheries and Aquatic Sciences 63:1526–1535. Northeastern Naturalist Vol. 25, No. 1 D.G. Argent, W.G. Kimmel, and D. Gray 2018 19 Appendix A. Numbers of adult Brook Trout captured in 1983, 2011, and 2014–2016. Stream Name Slope 1983 2011 2014 2015 2016 Baldwin Run NW 19 15 4 17 7 Bear Run South NW 35 6 10 8 8 Lick Run NW 71 32 5 9 1 M Fork Mill Creek NW 11 6 5 5 3 N Fork Mill Creek NW 40 3 1 4 3 Neals Run NW 17 2 1 6 7 Powdermill Run N. NW 34 33 18 20 6 Roaring Run South NW 45 10 3 9 2 SF Sugar Run NW 31 4 0 2 0 Tubmill Run NW 36 9 0 8 7 Allwine Creek SE 17 2 10 22 3 Dalton Run SE 19 27 17 12 13 Little Glade Run SE 9 1 0 9 1 Little Mill Creek SE 31 7 10 11 15 Mill Creek SE 56 14 14 14 8 N Branch Bens Creek SE 21 5 14 9 1 NF Bens Creek SE 10 16 7 17 14 NF Jones Mill Run SE 24 2 2 7 5 SF Jones Mill Run SE 38 5 4 8 0 Shaffer Run SE 18 1 0 1 0 Total 582 200 125 198 104 Appendix B. Numbers of YOY Brook Trout captured in 1983, 2011, and 2014–2016. Stream Name Slope 1983 2011 2014 2015 2016 Baldwin Run NW 5 2 0 4 1 Bear Run South NW 9 1 3 4 7 Lick Run NW 10 22 12 2 13 SF Sugar Run NW 14 3 0 3 0 M Fork Mill Creek NW 3 3 3 0 1 N Fork Mill Creek NW 4 1 4 2 1 Neals Run NW 11 3 3 4 6 Powdermill Run N. NW 2 3 39 8 7 Roaring Run South NW 15 5 22 4 16 Tubmill Run NW 11 6 3 3 1 Allwine Creek SE 7 4 65 0 0 Dalton Run SE 1 6 3 15 17 Little Glade Run SE 29 0 0 0 1 Little Mill Creek SE 11 0 11 8 4 Mill Creek SE 10 0 13 13 13 N Branch Bens Creek SE 15 3 15 3 1 NF Bens Creek SE 0 5 13 8 7 NF Jones Mill Run SE 6 0 4 0 1 SF Jones Mill Run SE 3 0 4 0 0 Shaffer Run SE 15 0 0 0 0 Total 181 67 217 81 97 Northeastern Naturalist 20 D.G. Argent, W.G. Kimmel, and D. Gray 2018 Vol. 25, No. 1 Appendix C. Numbers of Mottled Sculpin captured in 1983, 201 1, and 2014–2016. Stream Name Slope 1983 2011 2014 2015 2016 Baldwin Run NW 0 0 0 0 0 Bear Run South NW 0 0 0 0 0 Lick Run NW 10 18 19 12 19 SF Sugar Run NW 0 27 22 45 0 M Fork Mill Creek NW 23 37 67 51 61 N Fork Mill Creek NW 22 27 73 49 30 Neals Run NW 26 20 31 54 76 Powdermill Run N. NW 0 0 0 0 0 Roaring Run South NW 58 76 27 46 38 Tubmill Run NW 2 30 19 53 32 Allwine Creek SE 0 0 0 0 0 Dalton Run SE 0 0 0 0 0 Little Glade Run SE 0 0 0 0 0 Little Mill Creek SE 0 0 4 21 36 Mill Creek SE 3 0 23 2 4 N Branch Bens Creek SE 29 33 46 65 0 NF Bens Creek SE 0 0 0 0 0 NF Jones Mill Run SE 3 36 44 50 47 SF Jones Mill Run SE 10 23 72 26 82 Shaffer Run SE 33 0 48 75 179 Total 219 327 495 549 604