Ren Zhihui, Sang Yan-Fang, Cui Peng, Chen Fei, Chen Deliang
Key Laboratory of Water Cycle & Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, 100101, China.
University of Chinese Academy of Sciences, Beijing, 100049, China.
Sci Data. 2025 Jan 2;12(1):3. doi: 10.1038/s41597-024-04362-1.
The Qinghai-Tibet Plateau (QTP), a high mountain area prone to destructive rainstorm hazards and inducing natural disasters, underscores the importance of developing precipitation intensity-duration-frequency (IDF) curves for estimating extreme precipitation characteristics. Here we introduce the Qinghai-Tibet Plateau Precipitation Intensity-Duration-Frequency Curves (QTPPIDFC) dataset, the first gridded dataset tailored for estimating extreme precipitation characteristics in QTP. The generalized extreme value distribution is chosen to fit the annual maximum precipitation samples at 203 weather stations, based on which the at-site IDF curves are estimated; then, principal component analysis is done to identify the southeast-northwest spatial pattern of at-site IDF curves, and its first principal component gives a 96% explained variance; finally, spatial interpolation is done to estimate gridded IDF curves by using the random forest model with geographical and climatic variables as predictors. The dataset provides precipitation information within 1, 2, 3, 6, 12, 24 hours and 5, 10, 20, 50,100 return years, with a 1/30° spatial resolution. The QTPPIDFC dataset can solidly serve for hydrometeorological-related risk management and hydraulic/hydrologic engineering design in QTP.
青藏高原是一个容易遭受破坏性暴雨灾害并引发自然灾害的高山地区，这凸显了开发降水强度-历时-频率（IDF）曲线以估算极端降水特征的重要性。在此，我们介绍青藏高原降水强度-历时-频率曲线（QTPPIDFC）数据集，这是首个专门用于估算青藏高原极端降水特征的网格化数据集。选择广义极值分布来拟合203个气象站的年最大降水样本，并据此估算站点IDF曲线；然后，进行主成分分析以识别站点IDF曲线的东南-西北空间模式，其第一主成分的解释方差为96%；最后，以地理和气候变量作为预测因子，利用随机森林模型进行空间插值以估算网格化IDF曲线。该数据集提供了1、2、3、6、12、24小时以及5、10、20、50、100重现期的降水信息，空间分辨率为1/30°。QTPPIDFC数据集可为青藏高原与水文气象相关的风险管理以及水利/水文工程设计提供有力支持。
