Characterizing Spatial Heterogeneity in Reservoir Evaporation within the Rio Grande Basin using a Coupled Version of the Weather, Research, and Forecasting Model

Increasing evaporative demand from storage reservoirs is aggravating water scarcity issues across the American West. In the Rio Grande Basin, open water evaporation estimates represent approximately one-fifth of all water losses from the Basin. However, most estimates of reservoir evaporation rely on outdated methods, point measurements, or simplistic models. Warming temperatures and increasing atmospheric evaporative demand are stressing over-allocated resources, increasing the need for improved evaporation estimates. In response to this need, we develop open water evaporation estimates at Elephant Butte Reservoir (EBR), New Mexico, using three evaporation models and field measurements. Few studies quantify spatial heterogeneity in evaporation rates across large reservoirs; we therefore focus our efforts on using the Weather, Research, and Forecasting model coupled to an energy budget lake model, WRF-Lake, to simulate evaporation across EBR over the course of two years. We compare results from WRF-Lake, which simulates lake heat storage, to results from the Complementary Relationship Lake Evaporation (CRLE) model and the Global Lake Evaporation Volume dataset (GLEV). Results indicate that monthly and annual evaporation totals from WRF-Lake and GLEV are similar, while CRLE overestimates annual evaporation totals, with monthly peak evaporation offset compared to WRF-Lake and GLEV. While WRF-Lake and GLEV appear to capture monthly and annual evaporation totals, only WRF-Lake simulates differences in evaporation totals across the reservoir surface. Average annual evaporation at EBR was approximately 1487 mm, yet annual totals differed by up to 545 mm depending on location. This study improves understanding of open water evaporation and elucidates limitations of extrapolating point in-situ or bulk evaporation estimates across large reservoirs.