Intersatellite calibration of AMSU-A observations for weather and climate applications



[1] Long-term observations from the Advanced Microwave Sounding Unit-A (AMSU-A) onboard polar-orbiting satellites NOAA 15, 16, 17, and 18 and European Meteorological Operational satellite program-A (MetOp-A) were intercalibrated using their overlap observations. Simultaneous nadir overpasses (SNOs) and global ocean mean differences between these satellites were used to characterize calibration errors and to obtain calibration coefficients. Calibration errors were found manifesting themselves as certain scatter or temporal patterns of intersatellite biases, such as well-defined seasonal cycles in the Arctic and Antarctic SNO difference time series or a unique pattern closely correlated to the instrument temperature variability induced by Solar Beta Angle (SBA) variations in global ocean mean difference time series. Analyses of these patterns revealed five different types of biases that need to be removed from existing prelaunch-calibrated AMSU-A observations, which include relatively stable intersatellite biases between most satellite pairs, bias drifts on NOAA 16 and channel 7 of MetOp-A, sun-heating-induced instrument temperature variability in radiances, scene temperature dependency in biases due to inaccurate calibration nonlinearity, and biases due to channel frequency shift from its prelaunch measurement in certain satellite channels. Level-1c time-dependent calibration offsets and nonlinear coefficients were introduced and determined from SNO and global ocean mean temperature regressions to remove or minimize the first four types of biases. Channel frequency shift in NOAA 15 channel 6 was obtained from the radiative transfer model simulation experiments. The new calibration coefficients and channel frequency values have significantly reduced the five different types of biases and resulted in more consistent multisatellite radiance observations for intercalibrated satellite channels. The intercalibrated AMSU-A observations have been merged with its precursor, the intercalibrated microwave sounding unit (MSU), to generate the NOAA/Center for Satellite Applications and Research (STAR) version 2.0 upper-air temperature climate data record (CDR) for climate trend and variability monitoring from 1979 to the present. The intercalibrated AMSU-A radiance data are expected to further improve accuracies of numerical weather prediction and consistencies in climate reanalysis and CDR developments.