The interactions of clouds with radiation influence climate. Many of these impacts appear to be related to the radiative heating and cooling from high-level clouds in the upper troposphere, but few studies have explicitly tested this. Here, we use simulations with the ICON-ESM global atmosphere model to understand how high-level clouds through their radiative heating and cooling of the atmosphere, influence the large-scale atmospheric circulation and precipitation in the present-day climate. We introduce a new method to diagnose the radiative heating of high-level clouds: we use a temperature threshold of -35°C to define high-level clouds and also include the lower parts of these clouds at warmer temperatures. The inclusion of the lower cloud parts circumvents the creation of artificial cloud boundaries and strong artificial radiative heating at the temperature threshold. To isolate the impact of high-level clouds, we analyze simulations with active cloud-radiative heating, with the radiative heating from high-level clouds set to zero, and with the radiative heating from all clouds set to zero. We show that the radiative interactions of high-level clouds warm the troposphere and strengthen the eddy-driven jet streams, but have no impact on the strength of the Hadley circulation and the latitude of the Intertropical Convergence Zone. Consistent with their positive radiative heating and energetic arguments, high-level clouds reduce precipitation throughout the tropics and lower midlatitudes. Overall, our results confirm that the radiative interactions of high-level clouds have important impacts on climate and highlight the need for better representing their radiative interactions in models.

We propose the lag 1 autocorrelation of daily precipitation as a simple diagnostic of tropical precipitation in climate models. This metric generally has a relatively uniform distribution of positive values over the tropics. However, selected land regions are characterized by exceptionally low autocorrelation values. Low values correspond to the dominance of high-frequency variance in precipitation. Consistent with previous work, we show that CMIP6 climate models overestimate the autocorrelation. Global kilometer-scale models capture the observed autocorrelation pattern when deep convection is explicitly simulated. When a deep convection parameterization is used, the autocorrelation increases across the tropics, suggesting that land surface-atmosphere interactions are not responsible for the changes in precipitation variability. Furthermore, an accurate simulation of convectively coupled equatorial waves does not necessarily lead to a correct representation of the autocorrelation, and vice versa. This suggests other driving processes for the autocorrelation pattern.

The TRACMIP ensemble includes slab-ocean aquaplanet control simulations and experiments with a highly idealized narrow tropical continent (0-45ºW; 30ºS - 30ºN). We compare the two setups to contrast the characteristics of oceanic and continental rain bands and investigate monsoon development in GCMs with CMIP5-class dynamics and physics. Over land, the rainy season occurs close to the time of maximum insolation. Other than in its timing, the continental rain band remains in an ITCZ-like regime akin deep-tropical monsoons, with a smooth latitudinal transition, a poleward reach only slightly farther than the oceanic ITCZ’s (about 10º), and a constant width throughout the year. This confinement of the monsoon to the deep tropics is the result of a tight coupling between regional rainfall and circulation anomalies: ventilation of the lower troposphere by the anomalous meridional circulation is the main limiting mechanism, while ventilation by the mean westerly jet aloft is secondary. Comparison of two sub-sets of TRACMIP simulations indicates that a low heat capacity determines, to a first degree, both the timing and the strength of the regional solsticial circulation; this lends support to the choice of idealizing land as a thin slab ocean in much theoretical literature on monsoon dynamics. Yet, the timing and strength of the monsoon are modulated by the treatment of evaporation over land, especially when moisture and radiation can interact. This points to the need for a fuller exploration of land characteristics in the hierarchical modeling of the tropical rain bands.