Stratospheric Temperature Trends in the 11 Years of AIRS Spectral Radiance Observations

Abstract:

The AIRS (Atmospheric Infrared Sounder) level-1b radiances have been shown to be well calibrated (~0.3K or higher) and have little secular drift (~4mK/year) since its operation started in 2002. Given the rich information contained in the spectral radiances, such impressive instrument performances make AIRS radiances a valuable data set in the study of stratospheric climate. We compile 11 years (Sep 2002- Aug 2013) of AIRS radiances at channels in the CO₂ v₂ band with weighting functions peaked in the stratosphere. Using a state-of-the-art fast and accurate radiance simulator based on the PCRTM (Principle Component-based Radiative Transfer Model), we also simulate synthetic AIRS radiances at these channels based on two types of inputs: one is simulations by a free-running GFDL AM3 model and the other is ECMWF ERA-interim reanalysis. AIRS lower-stratospheric channels indicate a cooling trend of no more than 0.23 K/decade while its middle-stratospheric channels show a statistically significant cooling trend as large as 0.58 K/decade. Compared with AIRS observations, GFDL AM3 simulations underestimate the cooling trends in the middle-stratospheric channels while overestimate in the lower-stratospheric channels. Further simulations with separately varying CO₂ and SST suggest that the change of CO₂ alone is responsible for majority of the cooling trend in the middle-stratospheric channels, but the contributions of time-varying CO₂ and SST are comparable in the lower-stratospheric channels. In contrast, the synthetic radiances based on ERA-interim reanalysis show statistically significant positive trends in virtually all stratospheric channels. We also compare the zonal-mean trends estimated from observed and synthetic AIRS spectral radiances and climate data records based on multi-decade SSU (Stratospheric Sounding Unit) measurements. Though discrepancies exist in terms of magnitude and seasonality of the cooling, they all show that most cooling occurs in the tropics instead of the extra-tropics. This work also highlights the challenges to GCM modeling, reanalysis community and satellite measurements for stratospheric trend detection and attribution.