Impact of Calibration Drift Removal on Tropospheric Warming Trends Observed from Satellite Microwave Sounders

Long-term observations of global tropospheric temperatures from satellite microwave sounders play a vital role in climate change research. These observations involved those from the Microwave Sounding Unit (MSU), Advanced Microwave Sounding Unit-A (AMSU-A), and Advanced Technology Microwave Sounder (ATMS) onboard the current and historical NOAA polar orbiting satellites. Different research groups produced different tropospheric warming trends using similar satellite observations but with different bias correction algorithms. One of the key issues that were not well handled in the past were instrument calibration drifts. These included 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. Possible mechanisms causing these drifts may include time-varying side-lobe effect and degradation in detector or amplifier that could cause changes in calibration nonlinearity. This study removes these bias drifts in the level-1 instrument recalibration process using regressions of simultaneous nadir overpass (SNO) matchups between satellite pairs. The SNO regressions produced new time-varying level-1 calibration coefficients that remove calibration drifts but are independent of specific mechanisms used to explain the drifts. The calibration drift removal and adjustments of biases from other sources of error resulted in inter-consistent temperature records spanning MSU, AMSU-A, and ATMS for reliable detection of climate trend. The total tropospheric temperature (TTT) trends from our analysis were 0.142±0.045 K/decade during 1979-2022 after removal of the calibration drifts. This was 35% smaller than the warming trends when calibration drifts were not removed. The multi-model averages for the TTT trends from CMIP5 and CMIP6 climate model simulations were 0.28-0.29 K/decade during 1979 to 2019. This means that the satellite-observed total TTT trend in our study was only one-half of the climate model simulations during the same period. Essentially, the large reduction in TTT trends obtained here were mainly caused by the cascading removal of the spurious warming drifts in NOAA-11 through NOAA-14 during 1991-2003. This had caused a much lower TTT trend during the earlier half period (1979-2002) of the entire satellite observations. Such a trend reduction resulted in another noticeable phenomenon: trends during the latest half period (2003-2023) nearly doubled the first half period over the global ocean. This suggested an accelerated global warming signal from this study.

Disclaimer: The views, opinions, and findings contained in this report are those of the authors and should not be construed as an official NOAA or U.S. Government position, policy, or decision.