Retrieval of atmospheric water vapor from AMSR-E and the application for NWP at JMA

Atmospheric water vapor is a key geophysical parameter for Numerical Weather Prediction (NWP). For operational forecasts of heavy precipitation and typhoon, accurate moisture analysis fields as the initial conditions of NWP are required. Atmospheric water vapor is a source of clouds, rains and snows, and the distribution and amount play important roles for a representation of latent heat releases and absorptions in NWP model. Numerous moisture observations are required to produce accurate and realistic analyses and forecasts in the data assimilation. In this paper, a retrieval algorithm of atmospheric water vapor for AMSR-E and results of the product verification and comparison are presented. In addition, some NWP experiment results in JMA Meso-Scale Model data assimilation system are shown as an application example of this algorithm.