Using AIRS retrievals in the WRF-LETKF system to improve regional numerical weather prediction

Abstract

In addition to conventional observations, atmospheric temperature and humidity profile data from the Atmospheric Infrared Sounder (AIRS) Version 5 retrieval products are assimilated into the Weather Research and Forecasting (WRF) model, using the local ensemble transform Kalman filter (LETKF). Although a naive assimilation of all available quality-controlled AIRS retrieval data yields an inferior analysis, the additional enhancements of adaptive inflation and horizontal data thinning result in a general improvement of numerical weather prediction skill due to AIRS data. In particular, the adaptive inflation method is enhanced so that it no longer assumes temporal homogeneity of the observing network and allows for a better treatment of the temporally inhomogeneous AIRS data. Results indicate that the improvements due to AIRS data are more significant in longer-lead forecasts. Forecasts of Typhoons Sinlaku and Jangmi in September 2008 show improvements due to AIRS data.

Keywords

• data assimilation
• numerical weather prediction
• ensemble Kalman filter
• satellite sounding data