ABSTRACT

Shenandoah National Park in northern and central Virginia protects 777 square kilometers of mountain terrain in the Blue Ridge physiographic province and more than 90 streams containing diverse aquatic biota. Park managers and visitors are interested in the water quality of park streams and its ability to support healthy coldwater communities and species, such as the native brook trout (Salvelinus fontinalis), that are at risk in the eastern United States. Despite protection from local stressors, however, the water quality of streams in the park is at risk from many regional stressors, including atmospheric pollution, decline in the health of the surrounding forests because of invasive forest pests, and global climate change. In 2010, the U.S. Geological Survey, in cooperation with the National Park Service, undertook a study to compile, analyze, and synthesize available data on water quality, aquatic macroinvertebrates, and fish within Shenandoah National Park. Specifically, the effort focused on creating a comprehensive water-resources database for the park that can be used to evaluate temporal trends and spatial patterns in the available data, and characterizing those data to better understand interrelations among water quality, aquatic macroinvertebrates, fish, and the landscape.

Data from three primary sources, namely the Shenandoah Watershed Study, the Shenandoah National Park Aquatic Macroinvertebrate Monitoring Program, and the Springs and Headwater Streams Study, were compiled and loaded into the National Park Service’s NPSTORET database. This effort yielded a comprehensive database containing nearly 1.3 million measurements of habitat characteristics, approximately 442,000 measurements of water-quality characteristics, and over 438,000 measurements of biological taxa (fish and aquatic macroinvertebrates), collected across 673 sites over a period of more than 30 years.

Temporal trends in water quality indicate conflicting patterns in terms of acidity. Long-term (20- and 30-year) trends in acid-neutralizing capacity (ANC) and pH may indicate some improvement (decreasing acidity), but short-term (5-year) trends suggest increasing acidity. The long-term increase in pH occurred park-wide, although the increases were minimal in watersheds having siliciclastic bedrock. The long-term increases in ANC were mostly limited to watersheds with basaltic bedrock. Trends in concentrations of stream-water sulfate, another constituent of atmospheric deposition, indicated long-term improvements (declines in concentration) in watersheds having basaltic bedrock, long-term increases in concentration in watersheds having granitic bedrock, but no trend in watersheds with siliciclastic bedrock. Park-wide increases in mean, median, and maximum water temperatures were detected over the last 20 years. The average annual increase in mean water temperature park-wide was 0.04 °C, which equates to about 1.2 °C over the last 30 years. Short-term trends generally coincided with long-term trends but were more variable. Water temperatures generally tracked air temperatures, and additional analyses of longer-term (greater than 80 years) regional air-temperature data showed that the most recent increases in air temperature are not unprecedented.

Analysis of spatial patterns in water quality demonstrated that watersheds with higher mean elevations, lower land-surface gradients, and greater proportions of basaltic and carbonate geology are least affected by acidification and tend to be improving over time. Watersheds having greater proportions of siliciclastic and granitic geology, with smaller watershed areas and higher minimum watershed elevation tend to be affected by acidification and are experiencing continued degradation in water quality. There was no apparent spatial pattern in the water-temperature trends.

Benthic macroinvertebrate community metrics were found to be highly correlated with geology and, to a lesser extent, watershed area. Temporal trends in benthic macroinvertebrates showed evidence of change in community structure over time, which in most cases indicated declines in stream condition. Although the overall condition of park streams would be considered by most measures to be relatively healthy, streams in siliciclastic watersheds, in particular, have been and continue to be affected by acidic deposition. In addition, park streams have warmed significantly over the last 20 years and evidence indicates that benthic macroinvertebrate communities have responded to the warming trend.

Analysis of temporal trends in fish-community metrics revealed increasing species richness from 1996 to 2010, with the greatest increases observed in the larger streams. Furthermore, increases in fish richness were detected only in watersheds underlain by granitic and basaltic geology; no trend in richness was detected in siliciclastic watersheds, which also had the lowest richness values. The low richness values and lack of improvement over time in siliciclastic watersheds further suggests that recovery from acidification is not yet occurring in these areas. Increases in richness may partially be due to increased water temperatures.

Analysis of brook-trout population data indicated a response to acidification in siliciclastic watersheds, as mean abundances of young-of-year (YOY) and adult (age greater than 1 year; age 1+) fish were consistently lower in these watersheds than in others. Long-term (15-year) increases were detected, however, in adult abundances in the siliciclastic watersheds, although YOY abundances showed a declining trend during this same period. Despite observed demographic and environmental variation, brook trout population growth rates were stable throughout the study area between 1996 and 2009.

Although the current monitoring design in the park is spatially and temporally robust, specifically in terms of the ability to detect changes in water quality and aquatic fauna, and to understand processes related to stream acidification, multiple potential improvements to the monitoring programs were identified. Potential changes that would allow for more efficient accomplishment of monitoring objectives or more complete representation of water-resources and aquatic fauna conditions include reducing the frequency of water-quality trend monitoring to quarterly sampling, operating additional streamgages within the park, and co-locating monitoring stations used for water-quality, aquatic macroinvertebrate, and fish monitoring programs. Although these changes would potentially reduce monitoring costs or improve the datasets, the current dataset was found to be sufficient to satisfy multiple objectives, including objectives or data analysis for which the monitoring program was not originally designed, such as water-temperature trend analysis.

First posted September 3, 2013