| The measurements from Microwave Sounding Unit (MSU) and Advanced Microwave Sounding Unit-A (AMSU-A) have been extensively used for atmospheric temperature trend study during last several decades. The existences of inter-sensor biases and orbital drift, orbital differences among different satellites are two major challenges for climate study using long-term time series of satellite measurements. In this study, the impacts of orbital drift and orbital differences of satellites on AMSU-A derived temperature trends over Amazon rainforest are investigated. The AMSU-A near-nadir observations from NOAA-15, NOAA-18, NOAA-19, and MetOp-A during 1998 - 2014 are employed. The double difference method is firstly applied to obtain the estimates of inter-sensor biases for each paired AMSU-A instruments, in which NOAA-18 is used as a reference satellite. The inter-calibrated observations from the four satellites mentioned above are used to calculate monthly mean diurnal cycles of brightness temperature for each of the 15 AMSU-A channels. The diurnal correction method is then applied to all AMSU-A data using the estimated diurnal-cycle variations in order to obtain corrected data valid at the same local time. Finally, it is shown that the inter-sensor bias correction and diurnal correction have significant impacts on the AMSU-A derived long-term atmospheric temperature trends. |
