Diagnosing atmospheric heating rate changes using radiative kernels

Atmospheric radiative heating rate, which manifests radiative energy convergence in the atmosphere, is a fundamental factor shaping the Earth's climate and driving climate change.Compared to the radiative energy budget at the top of atmosphere (TOA) or surface, the atmospheric energy budget and heating rate are less studied and understood due to a lack of observational constraints and of diagnostic tools.Motivated by growing interest in atmospheric energy budget and particularly to facilitate the analysis of atmospheric heating rate, we innovate a set of radiative kernels, which quantitatively measure the sensitivity of atmospheric heating rate to different geophysical variables.When multiplied with the changes in these geophysical variables, these kernels can quantify the contributions of them to the heating rate change.A climate change experiment of Global Climate Models (GCMs) is used to test the application of the heating rate kernels.The results indicate that the radiative heating rate change simulated by the GCMs can be well reproduced by the kernels, which affirms the validity of the kernel method.The decomposition of the heating rate changes reveals rich information of the contributing mechanisms behind the changes.For example, in the tropical upper troposphere, the noticeably enhanced radiative cooling in a warmer climate is found to be dominated by atmospheric temperature and water vapor.Both of them increase the thermal radiation of the atmosphere, and are partially offset by a warming effect of the lifting high-cloud tops in this region.Moreover, we find that compared to the results corrected using the kernels, the cloud effect inferred from the radiative heating difference between clear-and all-skies (using the quantity termed "cloud radiative effect") has a non-negligible bias, which necessitates the use of kernels to quantify the cloud-induced heating rate changes.