An application of satellite information to numerical weather predictions (NWPs) is one of the most expected achievements in satellite remote sensing. In some meteorological agencies, the data of the Special Sensor Microwave Imager (SSM/I) on the US Defense Meteorological Satellite Program (DSMP) satellites and the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) have been or will be used in their operational forecasts. Recently, National Space Development Agency of Japan (NASDA) launched the two microwave radiometers. One is the Advanced Microwave Scanning Radiometer (AMSR) on board the Advanced Earth Observing Satellite II "Midori II" (ADEOS-II), launched in December 2002, and the other is the AMSR-E (modified version of the AMSR) on board the NASA's EOS Aqua satellite, launched in May 2002. The AMSR makes measurements of Earth at approximately 10:30 am/pm in local time, while the AMSR-E at 1:30 pm/am. The utilization of the AMSR and AMSR-E data in addition to the previous microwave radiometers is highly expected to fill a critical gap of global observations. The Japan Meteorological Agency (JMA) introduced the three-dimensional variational data assimilation (3D-VAR) system for the operational Global Spectral Model in September 2001, and the four-dimensional variational data assimilation (4D-VAR) system for the operational Meso-Scale Model in March 2002. Currently, the real-time SSM/I and TMI data are available and tested for assimilation at JMA, while the AMSR and AMSR-E data will be available after initial check-out. We performed impact studies of the retrieved total-column-precipitable water and rainfall by SSM/I and TMI with the JMA NWP systems, and obtained considerable improvements in the predictions. These experiments will be extended to the AMSR and AMSR-E data.