The impact of the resolution of meteorological data sets on catchment‐scale precipitation and drought studies
The impact of the resolution of meteorological data sets on catchment‐scale precipitation and drought studies
复制标题
气象数据集分辨率对流域尺度降水和干旱研究的影响
DOI：
10.1002/joc.5483
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
A. Becker
中科院分区：
文献类型：
--
作者：
J. Hellwig;K. Stahl;M. Ziese;A. Becker
气象数据为在一系列尺度上研究水文学提供了基础，包括流域尺度的干旱传播研究，以预警水文干旱影响。网格化的气象数据集很容易获得，并可用于此目的。由于这些数据集在空间/时间覆盖范围和空间/时间分辨率上不同，对于大多数研究来说，这两个方面之间存在权衡。虽然这种权衡主要是针对降雨量描述的，但其他特征也会产生影响。因此，我们在分布在德国各地的小流域(小于200 km2)的尺度上调查了从低分辨率输入数据得出的气象指数的偏差。比较了REGNIE(覆盖德国，1 × 1 km网格)、E-OBS(欧洲，0.2 5°网格)和GPCC(全球，1°网格)数据集之间的差异。一般来说，对于小流域，降水偏差随着分辨率的降低而增加，因为低分辨率的数据集无法解决相关的空间变异性随海拔的变化。此外，不同的内插方法导致干旱天数差异较大。相关系数等指标表明，干、湿期具有较好的一致性。通常用来指示干旱的标准化降水指数(SPI)值在平均水平上吻合较好，但也有一些变化。总体而言，结果表明，低分辨率数据集的绝对值可能不适合用于评估小型源头集水区的水文条件，而确定干旱时期的相对措施则更可信。对于大流域，数据集的分辨率相关性较小，但不同的内插方法仍然会对不同的产品产生不同的结果。因此，直接利用气象数据进行集水规模应用的研究应选择适当规模的数据，并可能需要考虑根据海拔高度调整降雨量和干旱天数的定义。
Meteorological data provide the basis to study hydrology at a range of scales, including catchment‐scale drought propagation studies for early warning of hydrological drought impacts. Gridded meteorological data sets are readily available and used for this purpose. As these data sets differ in spatial/temporal coverage and spatial/temporal resolution, for most studies there is a trade‐off between these two aspects. While this trade‐off has mostly been described for precipitation sums, other characteristics will also matter. We therefore investigated biases in meteorological indices derived from low‐resolution input data at the scale of small catchments (smaller than 200 km2) distributed over Germany. A comparison among the data sets REGNIE (covering Germany, 1 × 1 km grid), E‐OBS (Europe, 0.25° grid) and GPCC (whole world, 1° grid) was carried out. Generally, for small catchments biases in precipitation increase with decreasing resolution because low‐resolution data sets are not able to resolve the relevant spatial variability with elevation. In addition, different interpolation methods lead to high differences in the number of dry days. Relative measures such as the correlation coefficient reveal good consistencies of dry and wet periods. Standardized precipitation index (SPI) values, which are often used to indicate drought, match well on average but show some variations. Overall, the results suggest that absolute values of low‐resolution data sets may not be suitable to use for an assessment of the hydrological conditions at the scale of small headwater catchments, whereas relative measures for determining periods of drought are more trustworthy. For large river basins the resolution of the data set is less relevant, but different interpolation methods still lead to different results for different products. Therefore, studies that directly use meteorological data for catchment‐scale applications should choose data at the appropriate scale and may need to consider adjustments of precipitation amounts with elevation and of the definition of dry days.
影响因子：
8.2
作者：
Bachmair, Sophie;Stahl, Kerstin;Overton, Ian C.
通讯作者：
Overton, Ian C.
影响因子：
4.6
作者：
Bachmair, S.;Kohn, I.;Stahl, K.
通讯作者：
Stahl, K.