A copula-based multivariate flood frequency analysis under climate change effects

Area of Science:

• Hydrology and Climate Science
• Environmental Risk Assessment

Background:

• Floods are severe natural hazards with significant societal impacts.
• Climate change is projected to alter flood frequency and intensity.
• Traditional bivariate analyses are insufficient for future flood risk assessment.

Purpose of the Study:

• To assess projected changes in flood characteristics (duration, volume, peak).
• To analyze flood events using a trivariate copula-based framework under climate change scenarios.
• To evaluate flood risk in the Kan River basin, Iran.

Main Methods:

• Utilized eight downscaled and bias-corrected General Circulation Models (GCMs) from CMIP6.
• Employed a machine learning model to simulate daily streamflow using GCM outputs.
• Applied a trivariate copula-based framework (hierarchical Archimedean copula) to analyze flood events.

Main Results:

• Heterogeneous asymmetric copulas provided more accurate flood frequency estimations than symmetric or homogeneous asymmetric copulas.
• Climate change is projected to significantly influence trivariate joint return periods, especially in the far future.
• Flood frequency may increase by approximately 50% in the far future compared to historical and mid-term periods.

Conclusions:

• Flood characteristics are expected to exhibit nonstationary behavior due to climate change.
• Projected changes necessitate adaptive flood risk management strategies.
• The findings offer crucial insights for managing flood risk in a changing climate.