From Snow to Flow: Predicting the Timing of Peak Streamflow Using SNOTEL Ablation Curves

Abstract:

It is well understood that there exists a relationship between mountainous snowpack and seasonal streamflow. For snow-dominated river basins within the western United States, ~ 80% of streamflow is a result of seasonal snowmelt. In Idaho, where a significant amount of winter precipitation is stored as snowpack, predictions for the timing and magnitude of peak streamflow are essential for many water users, including water resource and irrigation managers. Of particular interest is the relationship between the timing of snowmelt and the timing of peak streamflow. While agencies such as the Natural Resources Conservation Service (NRCS) are tasked with the responsibility of providing water supply forecasts, few methods have used a data-driven, empirical modeling approach to investigate relationships between ablation timing monitored by the NRCS SNOTEL program and streamflow monitored by the United States Geological Survey (USGS). Such analyses could provide predictive tools that allow water users and managers to easily access and interpret information about peak streamflow. Idaho hosts over 80 SNOTEL sites that report hydrometeorological data, and 213 streamflow gauging stations. The goal of this study is to establish relationships between the timing of ablation at SNOTEL stations and peak streamflow within select basins in Idaho, and to synthesize results into user-friendly visualization tools that will provide estimates of the probability that peak streamflow will occur within a certain number of days as ablation progresses from 0 to 100%. In our initial test basin, the Boise River basin, we evaluated melt-out levels in increments of 10% from three SNOTEL sites located in the basin and produced cumulative density function (CDF) curves to illustrate the probability of peak streamflow occurring within a given number of days from the date at which the SNOTEL site reached each melt-out level.