J. R. Alder

Thermal controls of Yellowstone cutthroat trout and invasive fishes under climate change.

We combine large observed data sets and dynamically downscaled climate data to explore historic and future (2050-2069) stream temperature changes over the topographically diverse Greater Yellowstone Ecosystem (elevation range = 824-4017 m). We link future stream temperatures with fish growth models to investigate how changing thermal regimes could influence the future distribution and persistence of native Yellowstone cutthroat trout (YCT) and competing invasive species. We find that stream temperatures during the recent decade (2000-2009) surpass the anomalously warm period of the 1930s. Climate simulations indicate air temperatures will warm by 1 °C to >3 °C over the Greater Yellowstone by mid-21st century, resulting in concomitant increases in 2050-2069 peak stream temperatures and protracted periods of warming from May to September (MJJAS). Projected changes in thermal regimes during the MJJAS growing season modify the trajectories of daily growth rates at all elevations with pronounced growth during early and late summer. For high-elevation populations, we find considerable increases in fish body mass attributable both to warming of cold-water temperatures and to extended growing seasons. During peak July to August warming, mid-21st century temperatures will cause periods of increased thermal stress, rendering some low-elevation streams less suitable for YCT. The majority (80%) of sites currently inhabited by YCT, however, display minimal loss (<10%) or positive changes in total body mass by midcentury; we attribute this response to the fact that many low-elevation populations of YCT have already been extirpated by historical changes in land use and invasions of non-native species. Our results further suggest that benefits to YCT populations due to warmer stream temperatures at currently cold sites could be offset by the interspecific effects of corresponding growth of sympatric, non-native species, underscoring the importance of developing climate adaptation strategies that reduce limiting factors such as non-native species and habitat degradation.

Implementation and evaluation of a monthly water balance model over the US on an 800 m grid

We simulate the 1950-2010 water balance for the conterminous U.S. (CONUS) with a monthly water balance model (MWBM) using the 800-m Parameter-elevation Regression on Independent Slopes Model (PRISM) data set as model input. We employed observed snow and streamflow data sets to guide modification of the snow and potential evapotranspiration components in the default model and to evaluate model performance. Based on various metrics and sensitivity tests, the modified model yields reasonably good simulations of seasonal snowpack in the West (range of bias of ±50 mm at 68% of 713 SNOTEL sites), the gradients and magnitudes of actual evapotranspiration, and runoff (median correlation of 0.83 and median Nash-Sutcliff efficiency of 0.6 between simulated and observed annual time series at 1427 USGS gage sites). The model generally performs well along the Pacific Coast, the high elevations of the Basin and Range and over the Midwest and East, but not as well over the dry areas of the Southwest and upper Plains regions due, in part, to the apportioning of direct versus delayed runoff. Sensitivity testing and application of the MWBM to simulate the future water balance at four National Parks when driven by 30 climate models from the Climate Model Intercomparison Program Phase 5 (CMIP5) demonstrate that the model is useful for evaluating first-order, climate driven hydrologic change on monthly and annual time scales. This article is protected by copyright. All rights reserved.