Linear trends and closures of 10-yr observations of AIRS stratospheric channels

The Atmospheric Infrared Sounder (AIRS) level-1b radiances have been shown to be well calibrated (~0.3 K or higher) and have little secular drift (~4 mK yr⁻¹) since operation started in September 2002. This paper investigates the linear trends of 10 years (2003–12) of AIRS global-mean radiances in the CO₂ v₂ band that are sensitive to emissions from the stratosphere (stratospheric channels). AIRS lower-stratospheric channels have a cooling trend of no more than 0.23 K decade⁻¹ whereas the midstratospheric channels consistently show a statistically significant cooling trend as large as 0.58 K decade⁻¹. The 95% confidence interval for the trend is ~±0.20 K decade⁻¹. Two sets of synthetic AIRS radiances are computed using the principal component–based radiative transfer model (PCRTM), one based on a free-running GFDL Atmospheric Model, version 3 (AM3), over the same period and one based on ERA-Interim. The GFDL AM3 simulations overestimate the cooling trends in the mid- to upper-stratospheric channels but slightly underestimate them in the lower-stratospheric channels. The synthetic radiances based on ERA-Interim, however, have statistically significant positive trends at virtually all stratospheric channels. This confirms the challenge to the GCM modeling and reanalysis community to create a better simulation or assimilation of the stratospheric climate. It is shown that the linear trends in AIRS radiances can be reproduced to a large extent by the spectral radiative kernel technique and the trends from the AIRS L2 temperature retrievals and from the change of CO₂. This suggests a closure between AIRS L1 radiances and L2 retrievals and the potential merit of AIRS data in studies of stratosphere changes.