殷志远. 2017. WRF模式与TOPMODEL模型在洪水预报中的耦合预报试验研究[J]. 气象学报, ():-, doi:10.11676/qxxb2017.045
WRF模式与TOPMODEL模型在洪水预报中的耦合预报试验研究
The prediction experimental study on coupling WRF model and Topmodel model in flood forecasting
投稿时间:2016-09-06  最后修改时间:2017-02-07
DOI:10.11676/qxxb2017.045
中文关键词:  水文气象耦合预报,WRF,Topmodel,半分布式,漳河流域
英文关键词:Hydrological and meteorological coupling forecasting, WRF, Topmodel, ZhangHe basin
基金项目:
作者单位E-mail
殷志远 中国气象局武汉暴雨研究所 yzy87888073@126.com 
摘要点击次数: 456
全文下载次数: 299
中文摘要:
      基于空间分辨率90 m×90 m的湖北荆门漳河水库DEM地形数据,并从2012-2015年选取了20场洪水过程(其中16场为模拟,4场为检验),将华中区域数值预报业务模式WRF提供的三重嵌套空间分辨率3 km×3 km、9 km×9 km和27 km×27 km预报降雨与集总式新安江模型以及半分布式水文模型Topmodel耦合进行洪水预报试验。通过对比试验得到以下结论:当流域降雨的时空分布比较均匀时,集总式新安江模型可以较准确的预报出洪峰流量和峰现时间,而当降雨时空分布差异较大时,预报误差也会随之增大。基于DEM数据建立的Topmodel模型可以反映不同降雨时空分布下洪水预报结果的差异,试验结果表明,3 km×3 km和9 km×9k m洪水预报的输出结果比较接近,且在确定性系数和洪峰相对误差上要优于27 km×27 km的洪水预报结果,而在峰现时差的预报上,则是27km×27km的洪水预报结果与实测较吻合。通过研究还发现,虽然当流域降雨的时空分布存在一定差异时,三种空间分辨率的WRF预报降雨均无法预报出与实测相一致的降雨分布,但是在某些情况下,当降雨的时间分布误差和空间分布误差相抵消时,仍然可以得到较为准确的洪水预报结果。因此,高时空分辨率的模式预报降雨并不一定就能对洪水预报结果产生正贡献,需要通过反复尝试寻找水文模型和数值模式耦合的最佳时空分辨率。
英文摘要:
      Based on the spatial resolution 90m×90m DEM terrain data of Zhanghe reservoir in Jingmen city on Hubei province, twenty flood processes selected from 2012-2015(sixteen of which were simulated and four were tested),the study coupled the three nested spatial resolution 3km×3km、9km×9km and 27km×27km forecasting precipitation provided by the business model WRF of numerical forecast in Central China region with the lumped Xin''AnJiang model and the semi distributed hydrological model Topmodel respectively. The following conclusions were obtained by comparing the experiments: When the spatial and temporal distribution of rainfall in the basin was uniform, The lumped Xin''AnJiang model could predict the peak flow and peak time accurately. When the difference of rainfall time and space became larger, the prediction error was also increased. Because of basing on the DEM data, Topmodel could reflect the difference of flood forecasting results under different spatial and temporal distribution of rainfall. Test results show that the output result of 3km×3km and 9km×9km flood forecasting was close to each other and both of them were better than that of 27km×27km in the deterministic coefficient and relative error of peak. In the prediction of peak current time difference, the flood forecasting results of 27km×27km was in good agreement with the actual situation. Through the study also found that when there was a certain difference in the spatial and temporal distribution of the rainfall in basin, the WRF of the three spatial resolution couldn’t predict the rainfall distribution consistent with the fact, but in some cases, when the error of time distribution offset the error of space distribution, it still could get accurate flood forecasting results. Therefore, the model with high spatial and temporal resolution is not necessarily able to predict the flood forecasting results. The optimal spatial and temporal resolution for the coupling of hydrological model and numerical model was needed to find under the repeated attempts.
查看全文   查看/发表评论  下载PDF阅读器
分享按钮