A framework for quantifying climate-informed heavy rainfall change: Implications for adaptation strategies.

Area of Science:

• Climate Science
• Hydrology
• Environmental Science

Background:

• Climate change poses significant risks due to increased heavy rainfall.
• Understanding spatiotemporal rainfall patterns is crucial for effective flood risk management.
• Cambodia is a flood-prone nation requiring tailored climate adaptation strategies.

Purpose of the Study:

• To develop a climate-informed framework for assessing heavy rainfall and associated risks.
• To quantify future spatiotemporal changes in heavy rainfall characteristics for Cambodia.
• To create province-level flood risk change maps to guide adaptation strategies.

Main Methods:

• Quantified spatiotemporal rainfall characteristics using climate model datasets and ERA5 reanalysis data.
• Employed a conditional artificial neural network (CANN) for temporal disaggregation of rainfall data.
• Applied a zero-inflated generalized extreme value (ZIGEV) function for extreme value analysis (EVA).

Main Results:

• Rainfall intensity is projected to increase, particularly at higher return periods and shorter durations towards late-century.
• Significant increases in heavy rainfall frequency are anticipated in Phnom Penh by mid-century (~10-fold).
• Flood risk change maps highlight Southeastern and Northwestern regions as priorities for adaptation.

Conclusions:

• The developed framework, combining CANN and ZIGEV, improves EVA by reducing uncertainty and bias.
• Findings provide crucial data for policymakers to map flood susceptibility and vulnerability.
• The study supports targeted adaptation strategies to mitigate multidimensional risks from heavy rainfall.