Large-Scale Influences on Tropical Cyclone Activities in a Superparameterized General Circulation Model Aquaplanet Experiment

Abstract Accurately predicting tropical cyclone (TC) activities is necessary for hazard mitigation and adaptation. However, general circulation model (GCM)-simulated TC behavior is sensitive to the horizontal resolution and to how convection is represented. A superparameterized (SP) GCM explicitly calculates convection and subgrid processes using slices of finer-resolution microscale models embedded in each GCM grid column. Hence, it better resolves convective processes while maintaining acceptable computational costs. Here, we compared TC activities from the Community Atmosphere Model v5.0 (CAM) (with parameterized physics, hereafter referred to as CPCAM) with SP-CAM v5.0 (SPCAM) by aquaplanet experiments with an off-equatorial, zonally symmetric sea surface temperature profile. It was found that SPCAM TC frequency was much higher than the CPCAM counterpart. In SPCAM, the TC wind–pressure relationship was better captured and storm intensity was higher. Additionally, more “TC seeds” were identified in SPCAM. Genesis potential index analysis suggested that the differences in the number of TCs can be attributed primarily to weaker vertical wind shear (VWS) and secondarily to stronger 850-hPa absolute vorticity in SPCAM. Further inspection revealed that the VWS difference was associated with a slightly weaker Hadley circulation, whereas the intertropical convergence zone was wider and extended further northward in the summer hemisphere, leading to stronger low-level absolute vorticity in SPCAM. Stronger convectively coupled equatorial wave activities were also found in SPCAM. Apparently, the well-captured equatorial Rossby and mixed Rossby–gravity waves promoted TC genesis. These results highlight the value of applying multiscale modeling methods in climate modeling of extreme events with enhanced computational efficiency. Significance Statement Accurately predicting tropical cyclone (TC) genesis through computer simulations is important for hazard mitigation. However, the number of TCs simulated is sensitive to the model’s horizontal resolution and to how convection is represented. Alternatively, “superparameterization” (SP) explicitly calculates convection and unresolved processes using a simplified high-resolution model embedded in a low-resolution model, balancing accuracy and computational cost. Here, we investigate TC frequency using a convective-parameterized (CP) general circulation model (GCM) and an SP-GCM. There was much more TC genesis in the SP-GCM than in the CP-GCM. This was attributed to a weaker tropical overturning circulation, a wider and further northward intertropical convergence zone, and improved simulation of tropical waves. More advanced SP techniques can improve extreme weather modeling.