Unveiling the dynamics of flood characteristics and inundation patterns in the Poyang Lake Basin under a changing climate

Study region The Poyang Lake Basin, China. Study focus The Poyang Lake Basin is frequently affected by floods, posing serious threats to lives and property. Assessing potential changes in flood characteristics and inundation patterns across its five major tributaries under climate change is important for flood risk assessment and adaptation planning. This study developed a hydrological-hydrodynamic model (DTVGM-CaMa-Flood) to simulate the flood processes during 2001–2021, and assessed future changes in flood characteristics and inundation patterns during 2030–2100 under two emission scenarios (SSP1–2.6 and SSP5–8.5). New hydrological insights for the region The model reasonably reproduces monthly discharge and water levels at six stations, with Nash-Sutcliffe efficiency coefficient (NSE) over 0.88, and also captures typical daily flood event hydrographs, with NSE above 0.79. The simulated inundation extent generally agrees with MODIS observations, with approximately 84.3%–88.9% inundation occurrence frequency within the model capability. Future projections show that flood characteristics across the five tributaries exhibit spatial heterogeneity, with peak discharges increasing mainly in the eastern and decreasing in the western tributaries, whereas changes in 7-day flood volumes are weaker. Peak discharges at Waizhou, Lijiadu and Meigang may exceed historical maxima, especially under SSP5–8.5, indicating a greater potential for extreme floods. Projected inundated areas generally decrease across depth classes in the Ganjiang River Basin, whereas deep inundation expands in the Fuhe, Xinjiang and Raohe River Basins. In the Xiushui River Basin, moderate-depth inundated areas increase while deep inundated areas decrease.