Forecasting stream habitat and Brook Trout responses to climate change in Catoctin Mountain Park

Abstract

Anticipating and mitigating the effects of climate change is a fundamental challenge for natural resource conservation. In this report, we respond to research needs identified by Catoctin Mountain Park (CATO) for native Brook Trout (Salvelinus fontinalis) conservation and management as part of the US Geological Survey (USGS) Natural Resources Preservation Program in FY15-16. We addressed three overarching research questions: (1) How will anticipated changes in air temperature affect stream habitats? (2) How will changes to stream habitat affect the distribution of Brook Trout? (3) Which stream segments are most and least vulnerable to the effects of climate change? First, we surveyed Brook Trout abundance and fish community composition using electrofishing techniques within three watersheds: Owens Creek, upper Big Hunting Creek, and Blue Blazes Creek (a tributary to Big Hunting Creek). Second, we deployed a network of stream temperature gages to assess spatial variation in stream temperature and groundwater (GW) influence. Third, we used modeling techniques to forecast future stream temperatures that account for GW influences and air temperature scenarios. Fish sampling detected 13 species and 15,345 individual fish, the majority of which were Blacknose Dace (60%), Blue Ridge Sculpin (26%), and Brook Trout (6%). Brook Trout were not observed in Blue Blazes Creek and exhibited higher densities in Owens Creek than upper Big Hunting Creek (average densities = 19 fish/100 m and 4 fish/100 m, respectively). In contrast, Brown Trout were present in Blue Blazes Creek and exhibited greater density in Blue Blazes Creek than either Owens Creek or upper Big Hunting Creek (average densities = 3.0 fish/100 m, 0.3 fish/100 m, and 1.7 fish/100 m, respectively). Brown Trout occurred in sympatry with Brook Trout in Owens Creek and upper Big Hunting Creek, but appeared to have replaced Brook Trout in Blue Blazes Creek. Our fish surveys also revealed important locations for Brook Trout reproduction and young-of-year (YOY) dispersal within the Owens Creek watershed. Our study also revealed surprising differences in the distribution of Blue Ridge Sculpin among CATO streams. This species was abundant in Owens Creek (average density = 83 fish/100 m) but was less common in Blue Blazes Creek (average density = 12 fish/100 m) and was not detected in upper Big Hunting Creek. Histological examination of several specimens from Blue Blazes Creek by V. Blazer at the USGS Leetown Science Center revealed the presence of a novel parasite (Dermosystidium sp.) which has been linked to fish population declines elsewhere (Blazer et al. 2016). The parasite was not detected in Blue Ridge Sculpin samples from Owens Creek, and all trout appeared to be uninfected. Our survey results suggest that Blue Ridge Sculpin have been extirpated from upper Big Hunting Creek and have not recolonized from downstream source populations due to the fish passage barrier of Cunningham Falls. We recommend additional research to (1) evaluate the feasibility of reintroducing Blue Ridge Sculpin into upper Big Hunting Creek and (2) continue monitoring the distribution and potential spread of Dermocystidium in downstream waters. Stream temperatures ranged from 9.6 – 27.6 ºC during baseflow conditions in 2015 and 2016. Sites within upper Big Hunting Creek were consistently warmer than in Owens Creek or Blue Blazes Creek, suggesting an effect of headwater ponds outside CATO on upper Big Hunting Creek temperatures. For instance, in 2016 the maximum observed temperature in upper Big Hunting Creek was 27.6 ºC whereas Owens Creek reached a maximum of 23.7 ºC that year. Stream temperature data also revealed that 2016 was warmer than 2015 throughout the study area but did not exceed thermal tolerance limits for Brook Trout in either year. We estimated the influence of GW on stream temperatures using a statistical modeling approach based on the relationship between daily mean air temperature and stream temperature over time. Results indicated that effects of GW were generally stronger in the Owens Creek watershed than in Blue Blazes or upper Big Hunting Creek. However, we detected substantial spatial variation in GW influence among Owens Creek sites, with stream temperatures at some locations showing relatively little GW influence and others showing very strong influences (and correspondingly small influence of daily mean air temperatures). Although incoming lateral seeps were detected in upper Big Hunting Creek (D. Ferrier, Hood College, personal communication), the strongest effects of GW in the study area were due to GW upwelling within portions of the Owens Creek watershed (i.e., Tributary C in Figure 4) where we also observed high numbers of Brook Trout juveniles. Our results therefore identified potential high-priority areas for Brook Trout conservation in CATO. Finally, we modeled future stream temperatures based on scenarios characterizing GW sensitivity to air temperature and future air temperature increases. Stream temperature forecasts revealed important differences in habitat suitability for Brook Trout within and among watersheds. Big Hunting Creek sites were generally more sensitive to air temperature increases than sites in Owens Creek or Blue Blazes Creek. For instance, an increase in mean annual air temperature of 1.5 ºC (lowest level evaluated) exceeded thermal thresholds for Brook Trout in the majority of sites within that watershed, regardless of GW influence levels. In contrast, an air temperature increase of 1.5 ºC did not exceed thermal thresholds for Brook Trout in Owens Creek. However, modeled air temperature increases of 5 ºC resulted in a loss of Brook Trout thermal suitability throughout the study area. Model results revealed spatially patchy responses to air temperature increases that could provide an early-warning system for trout monitoring designs in CATO.