Based on the ATMS(Advanced Technology Microwave Sounder) satellite microwave data during 1—30 August 2012,this paper analyzes the characteristics of data and does the bias correction tests in WRFDA,and verifies the impact of assimilation ATMS data on track forecasts of typhoons Kai-tak(the 13rd),Tembin(the 14th) and Bolaven(the 15th) in August 2012.Results show that,before the bias correction,the deviations for most ATMS sounding channels are less than those for corresponding AMSUA/MHS channels,except for some window channels and humidity channels.The bias variations across scan position for ATMS channels are less than those for corresponding AMSUA/MHS channels,so the data from all 96 scan positions for ATMS channels can be used.The biases are maximum for ATMS window channels and minimum for ATMS channels 6—10.After using the bias correction coefficients derived from ATMS microwave data,it can reduce the biases for temperature channels and humidity channels and improve the prediction of track and minmum pressure of typhoon compared to directly use the bias correction coefficients from NOAA-18.Case studies show that assimilation of ATMS satellite microwave data can improve the prediction of typhoon tracks.