THE PROSPER MODEL: LEVERAGING BIG DATA TO PREDICT FLOW PERMANENCE IN SNOW-DOMINATED RIVER NETWORKS

Despite the importance of accurate streamflow permanence classifications, our understanding and available measurements of streamflow permanence at a landscape extent is surprisingly incomplete. The U.S. Geological Survey (USGS) has developed the PRObability Of Streamflow PERmanence (PROSPER) model, a GIS-based empirical model that provides predictions of the annual probability of a stream channel having year-round flow for any unregulated and minimally-impaired stream channel in the Pacific Northwest. The model provides annual predictions for 2004-2016 at a 30-m spatial resolution based on monthly updated values of climatic conditions and static physiographic variables associated with the upstream basin. Predictions correspond to the channel network consistent with the National Hydrography Dataset stream grid and are publicly available through the USGS StreamStats platform (https://water.usgs.gov/osw/streamstats/). A total of 10,740 streamflow observations were included in analysis, compiled and filtered from 11 independent data sets across the study area. The PROSPER model performed well in snow-dominated regions that have minimal groundwater influences. In particular, the most informative predictor variable was April Snow Water Equivalent, which highlights the influence of late spring snow cover for supporting streamflow in mountain river networks. Streamflow permanence probabilities (SPP) varied across the study area by geography and from year-to-year. Notably lower SPP corresponded to the climatically drier subregions of the study area. Higher SPP were concentrated in coastal and higher elevation mountain regions. In addition, SPP appeared to trend with average hydroclimatic conditions, which were also geographically coherent. The year-to-year variability lends support for the growing recognition of the spatiotemporal dynamism of flow permanence. Results suggest that PROSPER model can be a useful tool to evaluate regions of the landscape that may be resilient or sensitive to drought conditions, allowing for targeted management efforts to protect critical reaches. Additionally, PROSPER’s predictive ability underscores the importance of including the simple metric of streamflow observations into any field survey and is especially well suited for citizen science efforts.