Landfalling Tropical Cyclones: Investigating Rainfall Trends under Climate Change

Landfalling tropical cyclones (TCs) often lead to widespread societal impacts due to their associated wind and flood hazards. Among these, pluvial and fluvial flooding depend primarily on the intensity and total rainfall released during TC events. As global warming increases atmospheric humidity according to the Clausius-Clapeyron relationship, TC rainfall is expected to intensify, exacerbating flood risks. However, additional climatic drivers may also contribute to long-term changes in TC-induced rainfall. To understand and disentangle these drivers, robust modeling efforts and reliable observational datasets are essential.

In this study, we utilize a dataset from the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP), which includes rainfall estimates for historical TCs from 1950 to 2023. These estimates are derived using IBTrACS best-track data, two parametric wind models, and a physics-based Tropical Cyclone Rainfall (TCR) model. We validate the TCR model simulations by comparing them with TC rainfall estimates from ERA5 reanalysis data and the Integrated Multi-Satellite Retrievals for GPM (IMERG). This validation includes comparisons of lifetime accumulated rainfall for individual events and associated temporal trends across all events. Additionally, we use the TCR model to assess the role of climate change in driving long-term trends in TC rainfall. By generating counterfactual rainfall estimates, where the influence of increasing global mean temperature is removed through detrending of the temperature input data, we isolate a thermodynamic contribution of climate change to observed trends.

We find that the TCR model produces higher maxima and more extreme rainfall events compared to ERA5, consistent with the tendency of reanalysis data to underestimate extremes. However, the relative intensity distribution of TC rainfall is captured in ERA5 and aligns with the patterns produced by the TCR model. The relative temporal trends between the datasets also align. Therefore, the TCR model might be a valuable tool for overcoming the underrepresentation of extreme TC rainfall in reanalysis data. Furthermore, our counterfactual estimates reveal that while the Clausius-Clapeyron relationship explains a significant portion of the observed increases in lifetime accumulated rainfall, residual trends persist, suggesting the influence of additional climatic drivers. This research highlights the importance of robust modeling frameworks, such as TCR, for understanding and attributing changes in TC rainfall, providing critical insights into the evolving hazards posed by tropical cyclones in a warming world.