Using Sporadic Streamflow Measurements to Improve and Evaluate a Streamflow Model in Ungauged Basins in Wisconsin

Abstract

Streamflows derived from hydrological models are widely used in decision-making processes in a broad array of natural resources applications. There remain substantial challenges in quantifying error and uncertainty in hydrological models, but understanding the sources and magnitudes of error and uncertainty are essential to support robust decision making. In this study, the accuracy of a mixed-effects model for streamflow (flow-duration curves) across the state of Wisconsin, the Natural Community Model (NCM), was evaluated. The NCM is used as the basis for scientific studies and management decisions in Wisconsin, but uncertainty in the NCM has not yet been quantified, and performance has not been assessed formally except at continuously monitored streamflow stations. Although there are a couple hundred long-term monitoring stations, there are thousands of short-term and sporadic monitoring stations in Wisconsin. To take advantage of the vast number of sparsely monitored and short-term stations, an index gauge approach was used to estimate long-term streamflow percentiles and flow-duration curves (with uncertainty). These flow-duration targets formed the basis for an assessment of NCM accuracy in ungauged streams. A random forest model for NCM error was developed that provides a qualitative understanding of sources of error in the NCM as well as a quantitative way to correct the NCM using information from the sporadic/short-term streamflow stations that could not be included in the original NCM training set. By combining the original NCM and the random forest model, an updated NCM was produced with reduced error (75th percentile of errors dropped from 0.23 to 0.07 m3/s), and uncertainty estimates were defined for use with the updated NCM in decision making and research applications.