Tropical cyclone strength, precipitation, and environment in variable resolution CAM-MPAS simulations over Western North Pacific

The multi-year simulation of tropical cyclones (TCs) over the Western North Pacific (WNP) in the variable resolution (VR) CAM-MPAS model is studied. Experiments with the global quasi-uniform low resolution of 120 km (MPAS-UR) and the variable resolution mesh of 30–120 km refined over East Asia (MPAS-VR) are integrated from 1980 to 2005 following the Atmospheric Model Intercomparison Project protocol. By utilizing an objective detection method, TCs in ERA5 reanalysis and model simulations are tracked and compared against observations. MPAS-VR shows significant advantages over MPAS-UR as indicated by more realistic TC counts, intensities, lifetime distribution, and seasonal variation. The large-scale circulation and precipitation patterns associated with TCs are also improved in MPAS-VR relative to MPAS-UR. Based on the theory of Dynamic Genesis Potential Index, the multi-year TC records are further used to quantify the dependence of TC genesis on various dynamical environmental factors from the perspective of seasonal variation. We find that in ERA5, the relative contribution of the 500 hPa vertical pressure velocity term to TC genesis exceeds that of the 200–850 hPa vertical wind shear term, which is responsible for the August peak and strong seasonal variation of TC genesis. MPAS-UR fails to capture such relationship while MPAS-VR performs much better in this regard, suggesting that the higher skills in simulating the relative contributions from different dynamical environmental factors to the simulated seasonal cycle of TC genesis may explain the improvements from MPAS-UR to MPAS-VR.