An improved statistical bias correction method for Global Climate Model (GCM) precipitation projection: A case study on the CMCC-CM2-SR5 model projection in China’s Huaihe River Basin
Study region: We selected the China’s Huaihe River Basin as the study region.Study focus: Bias correction is crucial for improving the accuracy of Global Climate Model (GCM) predictions. We proposed an improved statistical bias correction method called the EquiDistant Cumulative Distribution Functio...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-02-01
|
| Series: | Journal of Hydrology: Regional Studies |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2214581824004956 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Study region: We selected the China’s Huaihe River Basin as the study region.Study focus: Bias correction is crucial for improving the accuracy of Global Climate Model (GCM) predictions. We proposed an improved statistical bias correction method called the EquiDistant Cumulative Distribution Function (CDF) matching method with least square fitting of CDF’s deviation (LS-EDCDF) to correct biases in GCM precipitation. The LS-EDCDF method incorporates recurrence periods and least squares with the determination of transfer functions relating observation to model simulation. We applied this and three other methods—EquiDistant Cumulative Distribution Function matching method (EDCDF), combined Linear Scaling and Cumulative Distribution Function matching method (LS-CDF) and Quantile-quantile mapping method (QUANT)—to correct biases in the CMCC-CM2-SR5 model. Performance was evaluated in four aspects. Taylor diagram statistical metrics and correlation coefficients are used to measure agreements between the basin-average observations and bias-corrected model precipitation and their CDFs during the validation period (2015–2020), respectively. The Expert Team on Climate Change Detection and Indices (ETCCDI) extreme precipitation indices and spatial patterns of extreme precipitation in July 2020 in the observations and bias-corrected data were then compared to evaluate the effectiveness of these methods on capturing precipitation extremes.New hydrological insights for the region: Results show that the LS-EDCDF method outperforms the other three methods in correcting monthly CDFs, reducing biases, and preserving extreme precipitation. All results of these methods indicate that the CMCC-CM2-SR5 model tends to overestimate precipitation but performs well during the main flood seasons, especially for extreme precipitation. Future projections suggest decreasing interannual variability of precipitation, with a lower change rate during the flood season and a higher rate during the non-flood season. The bias-corrected data from this method can serve as a valuable input for projecting future floods in this important area of China. |
|---|---|
| ISSN: | 2214-5818 |