Evaluation of a climate simulation in Europe based on the WRF–NOAH model system: precipitation in Germany
Evaluation of a climate simulation in Europe based on the WRF–NOAH model system: precipitation in Germany
复制标题
DOI：
10.1007/s00382-013-1727-7
复制
发表时间：
2013-03
期刊：
影响因子：
4.6
通讯作者：
K. Warrach‐Sagi;T. Schwitalla;V. Wulfmeyer;H. Bauer
中科院分区：
文献类型：
--
作者：
K. Warrach‐Sagi;T. Schwitalla;V. Wulfmeyer;H. Bauer
建立了WRF模式及其陆面模式NOAH，并将其作为欧洲区域气候模式。它是根据1989年至2008年最新的ERA中期再分析数据强制进行的，并以0.33°和0.11°分辨率进行操作。本研究的重点是验证德国的月平均和季节平均降水量，在德国气象局的高质量降水数据集。特别是，降水的研究在地形的西南部德国和干燥的低地东北部德国。在这两个地区，降水数据对于水文学家和农民等最终用户来说非常重要。两个WRF模拟表明，系统的正降水偏置不明显的ERA中期和高估的湿天频率。降尺度试验改善了ERA-interim低估的降水强度年周期。归一化泰勒图，即，通过对数量进行标准化而消除系统偏差的结果表明，在德国西南部和整个德国大部分地区，WRF降尺度比ERA中期降水分析提供了更好的空间分布，但降低了德国东北部的结果。在应用模式分辨率为0.11°时，WRF显示了地形上典型的RCM系统误差，如迎风-背风效应。对流允许的情况下，2007年夏季成立的研究，提高了降水模拟的降水最大值在山区和降水的空间相关性的位置。这一结果表明，在对流允许的尺度上，区域气候模拟的价值很高。
The Weather Research and Forecast (WRF) model with its land surface model NOAH was set up and applied as regional climate model over Europe. It was forced with the latest ERA-interim reanalysis data from 1989 to 2008 and operated with 0.33° and 0.11° resolution. This study focuses on the verification of monthly and seasonal mean precipitation over Germany, where a high quality precipitation dataset of the German Weather Service is available. In particular, the precipitation is studied in the orographic terrain of southwestern Germany and the dry lowlands of northeastern Germany. In both regions precipitation data is very important for end users such as hydrologists and farmers. Both WRF simulations show a systematic positive precipitation bias not apparent in ERA-interim and an overestimation of wet day frequency. The downscaling experiment improved the annual cycle of the precipitation intensity, which is underestimated by ERA-interim. Normalized Taylor diagrams, i.e., those discarding the systematic bias by normalizing the quantities, demonstrate that downscaling with WRF provides a better spatial distribution than the ERA interim precipitation analyses in southwestern Germany and most of the whole of Germany but degrades the results for northeastern Germany. At the applied model resolution of 0.11°, WRF shows typical systematic errors of RCMs in orographic terrain such as the windward–lee effect. A convection permitting case study set up for summer 2007 improved the precipitation simulations with respect to the location of precipitation maxima in the mountainous regions and the spatial correlation of precipitation. This result indicates the high value of regional climate simulations on the convection-permitting scale.