摘要
在对数值模式预报产品进行精细化释用处理中,为考虑模式地形与实际地形之间的差异性,本文提出了一种用于复杂地形下,综合考虑模式地形与实际地形的精细化三维插值方法,并将该方法应用于北京冬奥会重点区域的100 m高分辨率精细化温度产品释用中。算法先根据模式地形和实际地形的临界高度进行不同的三维插值,然后使用高斯模糊算法对插值后的结果进行处理用于模式产品的释用。利用2019年2月4-19日的地面自动站观测资料,对比分析原始数值预报温度产品和经过插值得到的精细化释用产品,定性分析结果表明:高分辨率释用产品有效考虑了实际复杂地形的影响,比原始数值预报产品更加精细化、美观化。客观检验结果表明:以自动站观测为实况,经过本文插值方法得到的高分辨率地面温度场比原始数值模式温度场的均方根误差、绝对误差和偏差均显著减小。因此,本文提出的基于高斯模糊的复杂地形下高分辨率三维插值方法可以保证释用产品的美观性和精细化,更重要的是也可以减小误差以提升产品的准确性。
In the release of numerical model forecasting products,there is a need to consider the difference between the model terrain and the actual terrain.This paper proposes a combined approach of three-dimensional interpolation and Gaussian fuzzy algorithm method to high-resolution temperature in areas with complex terrain.Specifically,the release of numerical model forecasting products is achieved by two processes:(1)A three-dimensional interpolation is first applied to obtain interpolated results of the model terrain and the actual terrain for a given critical height;(2)The interpolated results are then processed with the Gaussian fuzzy algorithm.The proposed method is applied to study 100 m high-resolution refined temperature product released in the key area of the Beijing Winter Olympics.By comparing the original numerical prediction outputs and the refined interpolated released products with observed temperature at the automatic weather station from 4 to 19 February 2019,it is found that the refined interpolated released products are better than the original numerical outputs as demonstrated by its refined aesthetics and a significant reduction in associated mean absolute error,root-mean-square error and BIAS between estimated and observed values in the high resolution temperature fields.Hence,the results demonstrate that the three-dimensional interpolation method based on Gaussian fuzzy proposed in this study can not only ensure the aesthetics and refinement of the released products,but more importantly,reduce the errors and improve the accuracy of the products.
作者
陈康凯
宋林烨
杨璐
陈明轩
陈敏
韩雷
曹伟华
CHEN Kangkai;SONG Linye;YANG Lu;CHEN Mingxuan;CHEN Min;HAN Lei;CAO Weihua(Ocean University of China,Qingdao 266100,Shandong,China;Beijing Institute of Urban Meteorology,Beijing 100089,China)
出处
《高原气象》
CSCD
北大核心
2020年第2期367-377,共11页
Plateau Meteorology
基金
国家重点研发计划项目(2018YFF0300102,2018YFC1507504)
国家自然科学基金项目(41875049,41605031)
北京市自然科学基金项目(8192016)。
关键词
高分辨率
三维插值
地形差异
高斯模糊算法
High-resolution
three-dimensional interpolation method
terrain difference
Gaussian fuzzy algorithm
