Mid‐Tropospheric Layer Temperature Record Derived From Satellite Microwave Sounder Observations With Backward Merging Approach.

We present a new version (v5.0) of the NOAA Center for Satellite Applications and Research (STAR) mid‐tropospheric temperature (TMT) time series. This data set uses a backward‐merging approach to intercalibrate 16 satellite‐based microwave sounding records. The instrument observations included those from the Microwave Sounding Unit (MSU) during 1979–2004, Advanced Microwave Sounding Unit‐A (AMSU‐A) during 1998–2017, and Advanced Technology Microwave Sounder (ATMS) from 2011 to present. A TMT time series during 2002–present based on satellite microwave observations in stable sun‐synchronous orbits was used as a reference in the backward merging process in which earlier satellites were adjusted and merged to the reference. Observations from earlier satellites were recalibrated to remove their calibration drifting errors relative to the reference using sequential overlapping observations. This included removal of spurious warming drifts in the MSU observations onboard NOAA‐11, NOAA‐12, and NOAA‐14 and a spurious cooling drift in the NOAA‐15 AMSU‐A observations. Temperature changes resulting from diurnal sampling drifts were corrected using an observation‐based semi‐physical model developed in this study. Other adjustments included channel frequency differences between MSU and AMSU‐A companion channels and instrument blackbody warm target effect on observed radiances. These adjustments resulted in inter‐consistent TMT records spanning MSU, AMSU‐A, and ATMS. The merged time series produced a global mean TMT trend of 0.092 ± 0.043 K/decade during 1979–2021 and a total tropospheric trend of 0.142 ± 0.045 K/decade after removal of a stratospheric cooling effect in TMT. Remarkably, the total tropospheric trends during the latest half period were nearly doubled the earlier half period over the global ocean. Plain Language Summary: Long‐term observations of global atmospheric temperatures from satellite microwave sounders play a vital role in climate change research. These observations involved multiple satellites spanning several decades. Careful intersatellite calibration and bias correction are needed to derive inter‐consistent records from multi‐satellite observations for reliable climate change detection. Here we develop a new version of the NOAA Center for Satellite Applications and Research (STAR) mid‐tropospheric temperature (TMT) time series for climate trend investigation. The time series includes instrument observations from three generations of satellite microwave sounders from 1979 to present. Intersatellite biases from several error sources were removed or minimized, including instrument degradation, changes in satellite local observation time, and spectral band differences between different generations of microwave sounders, etc. A unique feature in the time series was satellite merging started from the latest backward to the earlier ones. A TMT time series during 2002‐present was used as a reference in the backward merging, which was based on satellite microwave sounder observations with fixed local observation time. The reference TMT has a high accuracy in trend detection, allowing intercalibration and trend detection with better accuracy in time series of the entire period from 1979 to present. Key Points: A new version of the NOAA Center for Satellite Applications and Research (STAR) mid‐tropospheric layer temperature time series was developedInstrument recalibration has removed spurious warming drifts in observations from NOAA‐11 to NOAA‐14 and spurious cooling drifts in NOAA‐15The new record yields a trend of 0.14 K/decade during 1979–2021 with an even greater rate of warming after the year 2002 (0.22 K/decade) [ABSTRACT FROM AUTHOR]