Adaptability Comparison of Climate Downscaling Methods and Future Climate Projections in the Pearl River Delta, China

Zhijun Yao,Guoru Huang
DOI: https://doi.org/10.1007/s11069-024-06993-4
IF: 3.158
2024-01-01
Natural Hazards
Abstract:Under the context of climate change, the projection of regional future climate can effectively deal with potential future meteorological and hydrological disasters. However, the accuracy of regional climate projection is often directly affected by downscaling methods. Therefore, this study selected dynamic downscaling, statistical downscaling, and dynamic-statistical downscaling to downscale global climate models in the Pearl River Delta (PRD), China, and compared the accuracy of the results on the annual, monthly, and daily scales to evaluate their adaptability. The downscaling method with the highest accuracy was used to project the future (2040–2100) climate of the PRD. The dynamic-statistical downscaling method exhibited the highest adaptability as it combines the advantages of the other two methods and performs better in both the trend and magnitude projections of climate variables. The future temperature, rainfall, and extreme climate indices in the PRD were generally projected to increase, especially in SSP585 scenario, but the increasing trends generally varied among the variables. The temperature was projected to gradually stabilize in the future low grade scenario, while under the future high-grade scenario it exhibited an increasing trend. Significant increases in rainfall were mainly observed in the highly urbanized central and coastal areas of the PRD during the flood season, and were projected to continue to increase under all future scenarios. The spatial characteristics of extreme climate indices are complex, yet they are all projected to increase in most future scenarios. Therefore, it is extremely important to carefully prepare in advance for disasters caused by extreme climate events.
What problem does this paper attempt to address?