Extreme climate and human interventions driving flood dynamics in arid inland rivers: regional and global perspectives.

Flooding plays a vital role in sustaining hydrological cycles in arid regions. While recent advances in flood characterization and attribution have enhanced management and forecasting, limited understanding remains on how extreme climate change and human interventions influence flood generation and propagation from the perspectives of global atmosphere-ocean interactions and regional runoff process. Taking the Tarim River as a case study, this research integrates atmospheric numerical simulations and hydrological analysis to identify key circulation pattern changes and reveal mechanisms behind extreme climate-induced inland flooding. A three-dimensional Copula function model is applied to quantify the impacts of human activities on flood propagation. Results show increasing trends in annual maximum flood (AMF), flood duration (DT), and flood frequency (NT), with AMF (June-September) increasing by an average of 17.43 % after the abrupt-change year. Extreme climate indices such as Rx5 (maximum consecutive five-day rainfall) and TN90 (number of warm nights) significantly increase flood magnitude. Human water diversion reduced the flood magnitude coincidence probability between the headwaters and mainstream by 0.22, while reducing diversions improved it by 0.18. On a global scale, increased June sea surface temperatures(SST) over the North Sea-Baltic Sea, Mediterranean, and Black Sea regions are linked to intensified storm-type floods, while SST warming over the key region(15-65°N, 75-12°W) in North Atlantic contributes to snowmelt/ice-type floods in July-August. These findings identify key climate drivers and oceanic regions influencing flood types, quantify human impacts, and offer insights for adaptive flood risk management and policy-making in arid inland basins under climate change.