Diabatic heating of mesoscale convective cloud systems from synergistic satellite data

This study investigates the relationship between latent and radiative heating of tropical mesoscale convective systems (MCSs). Therefore, we expanded heating rates derived from active sensors, which have a sparse sampling, by applying artificial neural networks on cloud properties from IR sounder data and atmospheric and surface properties from meteorological reanalyses. With the help of a cloud system reconstruction, we determined the MCSs and their size. The radiative enhancement of mature MCSs rises with their released latent heat. Furthermore, larger, more organized MCSs generally heat the atmosphere more than smaller MCSs for a similar rain intensity. The projection of the MCS properties onto a plane of vertically integrated latent heating (LP) and radiative enhancement (ACRE) reveals that the MCSs tend to converge towards a mean LP-ACRE value during their life cycle, with larger initial LP-ACRE values requiring longer lifetimes to reach this settling point.