Assessing compound flood hazards in the Pearl river Delta: A Scenario-Based Integration of trivariate fluvial conditions and extreme storm events

Abstract

In coastal delta regions, typhoons not only generate storm surges but also threaten local communities by causing extreme fluvial flooding due to intense rainfall. Traditionally, upstream river discharges are neglected and not linked to these compound events, leading to an underestimation of flood extent and impacts. Quantifying the joint occurrences of extreme fluvial and coastal conditions is essential for accurately predicting future compound flood hazards. This study proposes an integrated statistical-numerical modeling approach to assess compound fluvial-coastal flood hazards. The approach combines a trivariate statistical analysis to characterize extreme fluvial conditions with a numerical model to simulate coastal storm surges and their compound effects. The methodology is applied to the Pearl River Delta in southern China, a region highly vulnerable to compound flooding. Trivariate joint statistical relationships are developed using historical river discharge records during typhoon events from 1957 to 2022 to characterize extreme fluvial conditions. While the correlation among the three connected rivers is weak under low flow conditions, a high dependence exists during extreme events, increasing the likelihood of concurrent flooding. The trivariate fluvial conditions are then integrated with univariate coastal storm surge conditions to project the compound flood hazard scenario. The results show that high-risk scenarios (100-year river discharge combined with 100-year storm tide) could inundate over 24% of the Pearl River Delta’s land area. Neglecting river discharges underestimates the floodplain extent by up to 32%. Transition zones influenced by both river flow and storm surges are identified along midstream river networks and upstream floodplains. These regions experience significant expansion with rising hazard levels, suggesting larger compound flood areas under future extreme conditions. This scenario-based approach provides valuable insights into characterizing and mapping compound flooding risks in vulnerable coastal regions.