Abstract The evolution of spectrally resolved outgoing longwave radiation measured at the top of the atmosphere (TOA) reflects the fingerprints of key geophysical variables, serving as a powerful tool for studying climate change. In this work, trends in TOA brightness temperature (BT) in the mid-infrared spectral range observed by the Infrared Atmospheric Sounding Interferometer (IASI) are compared with trends in synthetic BTs generated from a set of atmosphere-only simulations with the EC-Earth3 climate model (v3.3.3), over the period 2008–2019. Despite the presence of spectral biases, the model simulations effectively reproduce the IASI trends in the thermal infrared. A spectral kernel analysis is then applied to the synthetic radiances to quantify the contributions of temperature, surface temperature, water vapor, clouds, and greenhouse gases to these trends. The negative trend found in the core of the CO 2 band is attributed to the stratospheric cooling, which is overestimated in the climate model simulations. In the wing of the CO 2 band, the negative trend in radiance results from the combined effect of a positive contribution from the increasing tropospheric temperature and a negative contribution driven by rising atmospheric CO 2 concentration. In the atmospheric windows, clouds have a negative impact on the radiance trend and also significantly affect the inter-annual variability of the model’s radiance. Lastly, the near-zero trend in the water vapor band reflects a balance between the positive trend driven by temperature increases and the negative trend associated with water vapor changes. This work highlights the utility of spectrally resolved radiances to disentangle forcing and feedback processes, improving climate model evaluation.
Fera et al. (Thu,) studied this question.