Many studies have used data from climate models, such as global climate models (GCMs) and regional climate models (RCMs), to predict the future impact of climate change. However, a bias exists between simulated climate model data and observed data. Therefore, various bias correction methods have been developed to reduce this imbalance. The quantile mapping (QM) method is one of the most widely used approaches worldwide. However, the QM method does not account for the relative change in the raw data since bias correction is only performed on observed data. To solve this problem, the detrended quantile mapping (DQM) and quantile delta mapping (QDM) methods were developed to consider the relative changes in the raw data. Generally, the QM method is performed assuming that the statistical characteristics are the same when using cumulative density functions (CDFs) obtained from climate models and observations. However, the general QM (or QDM) method is performed using daily data, and thus the outcome may be slightly different from the quantiles of observed data in the statistical analysis of extreme data. This difference can lead to distortions when estimating relative changes in rainfall quantiles. Herein, the regional quantile delta mapping (RQDM) method for bias correction, which can solve the problems of the QDM method, was proposed. Additionally, evaluations were performed for the RQDM and existing QDM methods using several statistical approaches for a historical period (S0; 1979–2005). The results revealed that the RQDM method was similarly corrected to the observed results than the QDM method and produced more appropriate outcomes based on statistical evaluations. Moreover, the RQDM method showed a well-preserved relative changes in rainfall quantiles of the raw data, unlike the QDM method. Thus, it was found that the proposed RQDM method showed more robust results than the conventional QDM method.
Bibliographical noteFunding Information:
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. 2019R1A2C2010854 ). This work was also supported by the Yonsei University Research Fund of 2018 (2018-22-0070).
© 2020 Elsevier B.V.
All Science Journal Classification (ASJC) codes
- Water Science and Technology