Estimates of the tropical cloud feedback of global climate models (GCMs) show a large inter-model spread due to the small-scale nature of convective cloud processes. Estimates from large-eddy simulations (LES) are more consistent among themselves, but difficult to scale to the climate system. Here we consider a compromise between GCMs and LES, and study how tropical clouds over the Atlantic respond to a climate perturbation in a regional kilometer-resolution model. We perform two 4-year-long simulations at 3.3 km horizontal grid spacing with the limited-area model COSMO on a 9,000 × 7,000 km2 domain covering the tropical Atlantic: a control simulation and a climate change simulation using the pseudo-global warming approach. In a previous publication we have demonstrated that this approach yields a credible representation of the tropical climate without the double-ITCZ bias commonly seen in GCMs. Here we address the cloud feedback and find a reduction of ITCZ high-cloud cover resulting in a negative longwave cloud feedback similar to the CMIP6 ensemble mean. We find a reduction of stratocumulus clouds, which we argue is primarily a thermodynamic response. More surprisingly, the shallow cumulus cloud cover over the West Atlantic increases, which we argue is due to increased stability resulting in weaker entrainment and a more humid boundary layer. The mean shortwave cloud feedback over the tropical Atlantic is positive, comparable to the CMIP6 ensemble mean. The emerging estimate of the total cloud-radiative feedback over the Atlantic is slightly negative, consistent with CMIP6 GCMs, but it is substantially stronger than that of the driving GCM.
CITATION STYLE
Heim, C., & Schär, C. (2024). Climate Change Response of Tropical Atlantic Clouds in a Kilometer-Resolution Model. Journal of Geophysical Research: Atmospheres, 129(1). https://doi.org/10.1029/2023JD038947
Mendeley helps you to discover research relevant for your work.