This study investigates the influences of soil moisture and vegetation on 30 h convective precipitation forecasts using the Weather Research and Forecasting model over the United States Great Plains with explicit treatment of convection. North American Regional Reanalysis (NARR) data were used as initial and boundary conditions. We also used an adjusted soil moisture (uniformly adding 0.10m~3/m~3 over all soil layers based on NARR biases) to determine whether using a simple observationally based adjustment of soil moisture forcing would provide more accurate simulations and how the soil moisture addition would impact meteorological parameters for different vegetation types. Current and extreme (forest and barren) land covers were examined. Compared to the current vegetation cover, the complete removal of vegetation produced substantially less precipitation, while conversion to forest led to small differences in precipitation. Adding 0.10m~3/m~3 to the soil moisture with the current vegetation cover lowered the near surface temperature and increased the humidity to a similar degree as using a fully forested domain with no soil moisture adjustment. However, these temperature and humidity effects on convective available potential energy and moist enthalpy nearly canceled each other out, resulting in a limited precipitation response. Although no substantial changes in precipitation forecasts were found using the adjusted soil moisture, the similarity found between temperature and humidity forecasts using the increased soil moisture and those with a forested domain highlights the sensitivity of the model to soil moisture changes, reinforcing the need for accurate soil moisture initialization in numerical weather forecasting models.
展开▼
