Publication Date:
2016-12-13
Description:
Stream temperature is a critical habitat parameter for cold-water fish, many species of which now exist in geographically fragmented populations within the western United States. To assist managers in identifying thermally suitable fish habitat, we used data from 31 pools on streams of the White River National Forest in Colorado, USA to create multiple regression models to predict summer pool temperature metrics related to lethal and sublethal thermal tolerances of fish. We modeled the 7-day mean of daily maximum pool temperature for the warmest 7 days and the mean temperature of the warmest month, using air temperature and several geomorphic parameters. The strongest predictor variables of these temperature metrics were drainage area, discharge, and residual pool volume. Most previous studies found air temperature to be the strongest predictor variable for pool temperature, but for the mountain streams in this study, variables related to stream flow volume and stream morphology had better predictive power. The models, created from and tested against field data, were able to explain 66% and 51% of the variability in monthly mean and 7-day mean pool temperatures, respectively, and had prediction errors of less than 2 °C. The reach-scale approach developed here, which includes geomorphically relevant predictors of pool temperature, should be applicable to other mountainous river networks. This article is protected by copyright. All rights reserved.
Print ISSN:
0197-9337
Electronic ISSN:
1096-9837
Topics:
Geography
,
Geosciences
