Postlaunch calibration and bias characterization of AMSU-A upper air sounding channels using GPS RO data

Abstract

While the calibration accuracy of Advanced Microwave Sounding Unit-A (AMSU-A) was well characterized during its prelaunch period, its on-orbit performance remains important after the launch of each satellite. In this study, Global Positioning System (GPS) radio occultation (RO) data are used to carry out a postlaunch calibration to obtain an estimate of the accuracy of brightness temperatures measured by an AMSU-A instrument. At each scan angle, the mean difference and a linear regression relationship between AMSU-A observations and GPS RO brightness temperature simulations can first be derived using data in clear-sky conditions over ocean. The AMSU-A upper air sounding channels are then calibrated to GPS RO brightness temperature simulations using either the mean difference or the linear regression relationship. The effect of the above two calibration methods on the biases of AMSU-A data with respect to National Centers for Environmental Prediction global forecast system (NCEP GFS) 6 h forecast fields is finally examined. It is found that the magnitude and sign of biases vary with channels. After GPS RO calibration, biases are negative and of the same magnitude for all AMSU-A upper air sounding channels. The negative biases of AMSU-A brightness temperature are associated with a small warm bias in the physical temperature of NCEP GFS 6 h forecasts compared with GPS RO temperature retrieval above the low troposphere.