Abstract

This study proposes a novel dynamical model-based method to estimate Probable Maximum Precipitation (PMP) under climate change. The proposed method fundamentally builds on the Atmospheric Initial and Boundary condition shifting method to realize storm transposition and the pseudo-global warming (PGW) method in the Weather Research and Forecasting (WRF) model. This study presents a case study of the developed methodology, focusing on two Atmospheric River-induced heavy rainfall events across three watersheds in central Chile. As a result, maximum precipitation with shorter durations generally increased with future warming in the target basins. The three basins' average increments in the 12-hr maximum precipitation from the current climate run were 12.7 % (SD: 5.7 %) in the 2050 s PGW run and 22.6 % (SD: 2.8 %) in the 2090 s PGW run. Warming effects on maximum precipitation with longer durations vary in the three basins studied. The Maipo River Basin and Rapel River Basin showed increases in the 48-hr maximum precipitation from the current climate run to the 2090 s PGW run of 8.4 % and 6.0 %, respectively, while the Aconcagua River Basin showed a significant decrease in the 48-hr maximum precipitation. Such inconsistent precipitation responses to warming were well explained by changes in the Integrated water Vapor Transport (IVT) and Vertically Integrated Moisture Flux Convergence (VIMFC), demonstrating the importance of storm dynamics changes due to warming. Maximum precipitation with shorter durations showed larger warming scaling rates, even exceeding double Clausius-Clapeyron scaling (14 %/K). Future studies should apply the proposed method to more storm events, which will enable a robust PMP estimation with its uncertainty under climate change.