Quantification of topographic effects on predicted precipitation for typical Elbe-catchments in the Erzgebirge

在2002年的洪水事件中，降水预报相当准确。整个厄尔士山脉斜坡上的总体降雨量都得到了很好的代表。然而，在单个像素上，预测值和观测值之间存在较大的局部和时间偏差。主要的洪水灾害发生在小河流和小溪。这些易北河的许多支流都是由水坝调节的。这些集水区的面积从5平方公里到200平方公里不等。这意味着典型的集水区由气象预报模型的一个像素覆盖。由于这些集水区位于山脉的斜坡上，它们各自的地形坡度将偏离模型中的坡度。特别地，这些梯度是三维的。2002年8月事件的降水在很大程度上是地形引起的。为了找出集水区降雨的分区是地形引起的，将对12个大坝提供的集水区的地形影响进行量化，以确定所选的降水事件。这意味着使用所有可用的集水区相关数据，包括用雨量计数据校准的雷达数据、这些数据本身以及大坝当局测量的水资源预算中可用的信息。集水区降水量与参考站点降水量之间的偏差将被用来寻找地形影响的经验校正因子。将利用真实的汇水区几何形状与预报模型几何形状之间的差异来推导垂直运动的差异，即几何形状因子。经验因子和几何因子被应用于预测降雨，与观测到的降雨进行比较，最后用于制定预报模型的校正因子。

In the flood event of 2002 precipitation forecasts were rather accurate. The overall rains on the whole slope of the Erzgebirge were rather well represented. Looking at individual pixels, however, large local and temporal deviations between forecast and observed values occurred. The major flood damage occurred in the small rivers and streams. Many of these Elbe contributaries are regulated by dams. The sizes of these catchments ranges from about 5 km² to about 200 km². This means that the typical catchment area is covered by one single pixel of a meteorological forecast model. As these catchments are located on the slope of the mountain ranges their individual topography gradient will deviate from the one in the model. In particular these gradients are threedimensional. The precipitations of the August 2002 event were to a large extent topography induced. To find out which partition of catchment wide rains is topography induced the effects of the topography of 12 dam provided catchment areas will be quantified for selected precipitation events. This means to use all available catchment area related data comprising the area covering radar-data calibrated with rain gauge data, these data themselves, and furthermore, the information available from the water budget as measured by dam authorities. The deviations between the catchment precipitations and the ones of a references site will be used to find an empirical correction factor for topographic effects. The differences between the geometries of real catchment areas and forecast model geometries will be used to derive the differences in the vertical motion, a geometry factor. The empirical and the geometry factor are applied to forecast rains, compared to the observed rains and finally used to formulate a correction factor for forecast models.