A 30-year convection-permitting regional climate simulation over the interior western United States. Part I: Validation

A 30-year high-resolution (4 km) regional climate simulation (October 1981 to September 2011) is conducted in the interior western United States (IWUS) using the weather research and forecasting (WRF) model. The high spatial resolution is motivated by the IWUS complex terrain environment and the high spatial variability of atmospheric and land surface variables. The simulation, driven by the NCEP climate forecast system reanalysis, is convection-permitting and uses the Noah multi-physics land surface model. Modelled surface temperature, precipitation, and snow water equivalent (SWE) are evaluated against snowpack telemetry (SNOTEL) data and against the parameter-elevation regressions on independent slopes model (PRISM) data set. The modelled daily minimum and maximum surface temperatures match the 4-km resolution PRISM data, with spatial correlation coefficients close to 1. The simulation accurately captures the observed distribution and amount of seasonal precipitation and the mountain snowpack in the IWUS, although discrepancies exist, especially over the high-elevation ridges. Simulated seasonal precipitation correlates well with observations, with correlation coefficients exceeding 0.85 for PRISM over the whole domain and 0.88 for SNOTEL over the mountain ranges. The simulation also replicates the spatial pattern of extreme precipitation events well, although it overestimates precipitation intensity and maximum duration of dry spells and underestimates the frequency of wet days. The simulated seasonal mountain snowpack and its spring melt-off timing show a negative bias at most SNOTEL sites. This validation justifies the use of the 30-year IWUS data set as a high-resolution data source, almost equivalent to a reanalysis.