Dynamic Assimilation of MODIS-Retrieved Humidity Profiles within a Regional Model for High-Latitude Forecast Applications

Abstract

A ?hot start? technique is applied to the fifth-generation Pennsylvania State University?National Center for Atmospheric Research (PSU?NCAR) Mesoscale Model (MM5) to dynamically assimilate cloud properties and humidity profiles retrieved from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument on board the NASA Earth Observing System polar-orbiting satellites. The assimilation approach has been studied through extensive numerical experimentation for high-latitude rain events to demonstrate the feasibility and the benefit of the approach on the model cloud and precipitation simulation/forecast. The ingestion of MODIS-retrieved cloud and clear-air humidity information impacts MM5 cloud fields on both a microphysical and macrophysical level. From short-term (6?12 h) forecast experiments conducted for a preliminary test case and 16 extensive summer and winter experiments, the following primary conclusions have been reached. 1) It is feasible to introduce MODIS-retrieved cloud-top properties and humidity profiles into the MM5 model in a hot start mode without disrupting model stability and evolutionary continuity. 2) The introduction of high-resolution MODIS information produced more accurate humidity fields and resulted in increased mesoscale structure in the cloud and precipitation fields. 3) The opportunistic ingestion of MODIS data at its observation time into the model leads to improved 6?12-h model precipitation forecasts with respect to not only the frequency of occurrences, but also the magnitude of precipitation amounts. 4) Verification with three-dimensional analyses indicates some improvement in model forecasts of temperature, wind, pressure perturbation, and sea level pressure as well. 5) Verification with surface station observations indicates that model forecasts of 2-m temperature, 2-m relative humidity, 10-m winds, and sea level pressure are also improved, most notably for the summer cases. The largest improvement in forecast skill is for 2-m relative humidity (12%).