Predicting salinity generation in the Upper Colorado River Basin: Modeling, uncertainty, and monitoring issues

Document Type


Journal/Book Title/Conference

Journal of Hydrologic Engineering





Publication Date


First Page


Last Page



Salinity in the Upper Colorado River Basin (UCRB ) is due to both natural sources and processes, and anthropogenic activities. Given economic damage due to salinity of $295 million in 2010, understanding salinity sources and production together with transport are of great importance. SPA tially Referenced Regressions On Watershed (SPARROW ) is a nonlinear regression water quality model that simulates sources and transport of contaminants such as dissolved‐solids. However, SPARROW simulations of dissolved‐solids in the UCRB only represent conditions through 1998 due to limited data availability. More importantly, prior simulations focused on a single year calibration and its transferability to other years, and the validity of this approach is questionable, given the changing hydrologic and climatic conditions. This study presents different calibration approaches to assess the best approach for reducing model uncertainty. This study conducted simulations from 1999 to 2011, and the results showed good model accuracy. However, the number of monitoring stations decreased significantly in recent years resulting in higher model uncertainty. The uncertainty analysis was conducted using SPARROW results and bootstrapping. The results suggest that the watershed rankings based on salinity yields changed due to the uncertainty analysis and therefore, uncertainty consideration should be an important part of the management strategy.

This document is currently not available here.