Neural Network Estimation of Atmospheric Thermodynamic State for Weather Forecasting Applications

We present recent work using neural network estimation techniques to process satellite observation of the Earth's atmosphere to improve weather forecasting performance. A novel statistical method for the retrieval of atmospheric temperature and moisture (relative humidity) profiles has been developed and evaluated with sounding data from the Atmospheric InfraRed Sounder (AIRS) and the Advanced Microwave Sounding Unit (AMSU) on the NASA Aqua satellite and the Infrared Atmospheric Sounding Interferometer (IASI) and AMSU on the EUMETAT MetOp-A satellite. The present work focuses on the cloud impact on the AIRS and IASI radiances and explores the use of stochastic cloud clearing mechanisms together with neural network estimation. The algorithm outputs are ingested into a numerical model, and forecast information and decision support tools are then presented to a meteorologist. We discuss the underlying physical problem, the algorithmic framework, and the interaction with forecaster.