Supplementing Sparse Observations of Temperature and Precipitation with a High-Resolution Atmospheric Model

The spatial scarcity of meteorological observations represents a significant challenge for distributed hydrological modelers. Mesoscale models can provide a physically-based approach to supplement surface observations over high-elevation terrain. The heavily instrumented North Fork American River basin in California offers an optimal study location to verify mesoscale model output. Gridded 6km surface temperature and precipitation were obtained from the Weather and Research Forecasting (WRF) model, which uses lateral boundary conditions from the North American Regional Reanalysis. Results indicate that during the wet season, the WRF model was shown to well represent the inter-annual variability of the large scale temperature lapse rate and the orographic gradient of precipitation. Errors in the WRF model's temperature and precipitation were of a similar magnitude to those errors using a low-elevations station coupled to a standard lapse rate and Precipitation Regression on Independent Slopes Method (PRISM). A distributed hydrological study to evaluate if the WRF based forcing improves simulated snowpack and streamflow remains as future work.