A computationally efficient method for obtaining model forecast winds in the vicinity of complex coastal orography

摘要

Advances in computers have provided the meansfor generating fine resolution mesoscale numericalweather predictions (NWPs). Each computeradvance brings demands for forecasts on eversmaller scales, especially by such disciplines as airpollution modeling and fire weather forecasting.Weather forecasts and observations on very smallscales are essential for driving the models used inthese important decision-making processes. Evenwith the improvements in mesoscale NWPs, thehorizontal scales desired by these communities arestill too small to be treated by current computertechnology in a timely and practical fashion. Evenif the computer resources were adequate,mesoscale model parameterizations are notnecessarily appropriate for these small scales,thereby potentially introducing significant modelerror in mesoscale NWPs.