中国天然径流量格点数据集CNRD v1.0（1961-2018）

水是人类赖以生存与发展的物质基础，也是我们感知和应对气候变化的重要媒介。受独特季风气候与阶梯状地形影响，中国水资源分布极不均匀，缺水问题突出，是全球水资源极度脆弱的地区之一。人类活动与气候变化的复合作用，进一步加剧了中国水循环过程研究的复杂性。因此，迫切需要一套质量可靠、时空连续，且剔除大规模人类活动影响下的天然径流数据，为水循环研究提供本底数据支持。然而，中国现有的天然径流资料缺失率较高，参考站点密度不足，在年际和季节变化尺度上存在较大偏差，难以客观揭示大尺度径流变化的自然规律。本研究建立了一套长时序、全覆盖、高质量、时空连续的天然河川径流资料，命名为CNRD v1.0（The China Natural Runoff Dataset version 1.0）。CNRD v1.0提供1961年1月1日至2018年12月31日中国0.25°×0.25°天然径流估算量日值、月值和年值。200个有资料水文站点率定结果显示，模型参数在大多数站点得到了充分校准，模型纳什效率系数（NSE）在率定期和验证期的平均值分别为0.83和0.80。无资料流域交叉验证结果显示，MPR方法提供了最佳的区域化方案，率定期 NSE中位数为0.76，验证期NSE中位数为0.72。结果总体显示水文模型参数率定和区域化表现良好，可用于长时序径流资料重建。另外，通过与两套全球径流格点数据集ISIMIP和GRUN比较，发现CNRD v1.0数据集的径流空间分布上过渡更加连续，且在表示中国复杂地形和气候理分划下的水资源空间分布方面优于全球径流数据集。

Water is the material foundation for human survival and development, and also an important medium for perceiving and addressing climate change. Affected by the unique monsoon climate and stepped terrain, water resources in China are extremely unevenly distributed, with prominent water shortage issues, making China one of the regions with extremely vulnerable water resources globally. The combined effects of human activities and climate change have further exacerbated the complexity of hydrological cycle research in China. Therefore, there is an urgent need for a set of natural runoff data that is reliable in quality, temporally and spatially continuous, and excludes the impacts of large-scale human activities, to provide baseline data support for hydrological cycle research. However, the existing natural runoff data in China has a high missing rate, insufficient density of reference stations, and large biases on inter-annual and seasonal change scales, making it difficult to objectively reveal the natural laws of large-scale runoff changes. This study developed a long-term, full-coverage, high-quality, temporally and spatially continuous natural river runoff dataset, named CNRD v1.0 (China Natural Runoff Dataset Version 1.0). CNRD v1.0 provides daily, monthly, and annual estimates of natural runoff at a 0.25° × 0.25° grid resolution across China, spanning from January 1, 1961 to December 31, 2018. Calibration results from 200 hydrological stations with available observational data show that model parameters were sufficiently calibrated at most stations, with the mean Nash-Sutcliffe Efficiency Coefficients (NSE) of the model reaching 0.83 and 0.80 during the calibration and validation periods, respectively. Cross-validation results for ungauged basins demonstrate that the MPR method yields the optimal regionalization scheme, with median NSE values of 0.76 and 0.72 during the calibration and validation periods, respectively. Overall, the results indicate that the hydrological model parameter calibration and regionalization perform excellently, and are suitable for the reconstruction of long-term runoff data. In addition, compared with two global runoff grid datasets, ISIMIP and GRUN, it is found that the spatial distribution of runoff in the CNRD v1.0 dataset exhibits more continuous transitions, and outperforms the global runoff datasets in representing the spatial distribution of water resources under China's complex terrain and climate zoning.

AI搜集汇总

背景与挑战

背景概述

该数据集是中国天然径流量格点数据集CNRD v1.0，覆盖1961年至2018年，提供日、月、年值的天然径流估算量，空间分辨率为0.01º - 0.05º，旨在为水循环研究提供高质量的本底数据。数据集基于水文模型率定和区域化方法构建，模型表现良好（纳什效率系数平均0.83），并优于全球径流数据集，但当前可开放获取的为月尺度数据，日尺度数据需联系作者。