Exploring Western North Pacific Tropical Cyclone Activity in the High-Resolution Community Atmosphere Model

High-resolution climate models (similar to 28 km grid spacing) can permit realistic simulations of tropical cyclones (TCs), thus enabling their investigation in relation to the climate system. On the global scale, previous works have demonstrated that the Community Atmosphere Model (CAM) version 5 presents a reasonable TC climatology under prescribed present-day (1980-2005) forcing. However, for the Western North Pacific (WNP) region, known biases in simulated TC genesis frequency and location under-represent the basin's dominant share in observations. This study addresses these model biases in WNP by evaluating WNP TCs in a decadal simulation, and exploring potential improvements through nudging experiments. Among the major environmental controls of TC genesis, the lack of mid-level moisture is identified as the leading cause of the deficit in simulated WNP TC genesis over the Pacific Warm Pool. Subsequent seasonal experiments explore the effect of constraining the large-scale environment on TC development by nudging WNP temperature field toward reanalysis at various strengths. Temperature nudging elicits a significant response in TC genesis and intensity development, as well as in moisture and convection over the Warm Pool. These responses are sensitive to the choice of nudging timescale. Overall, the nudging experiments demonstrate that improvements in the large-scale environment can lead to improvements in simulated TCs, suggesting future model developments in relation to model physics. In this way, the potential improvements in model fidelity will contribute to the understanding of how the mean state of current or future climates may give rise to extremes such as TCs. Plain Language Summary Climate models at high resolution (similar to 28 km pixel size) can realistically represent tropical cyclones (TCs). Previous work has shown that on the global scale, the spatial pattern and annual frequency of TC formation are well simulated in a high-resolution climate model. However, for the typhoon-prone Western North Pacific (WNP) region, the number of TCs as simulated by the model is much lower than observation, and the location of their formation is too much toward the north. Our study finds out that these differences between simulated and observed TCs are due to the simulated environment: over the region in the WNP where the ocean temperatures are the warmest and TCs typically form, the simulated environment is too dry and cold. To address these issues, we conducted experiments that nudge the temperature over the WNP region toward observation at various strengths. The environment, with lessened cold bias, generally becomes more moistened and gives rise to more realistic TC formation and intensity evolution. Our results suggest that by improving the simulation of the background mean state, climate models can better represent extreme events such as TCs.

Automated summary

High-resolution climate models can realistically simulate tropical cyclones. However, the simulated number and location of TCs in the Western North Pacific region are different from observations. The lack of mid-level moisture is identified as the leading cause of the simulated WNP TC genesis deficit, and improving the large-scale environment can lead to better simulation of TCs.