论文部分内容阅读
温度插值方法的精度对于公路气候区划有重要的影响,以新疆天山地区为例,利用天山及其周边地区共52个气象站点在1971~2000年间的气温数据,采用数理统计方法中的多元线性回归法分析了天山地区累年年平均气温与经度、纬度和海拔的相关关系,在基于DEM(数字高程模型)一般插值方法的基础上,引入残差提出了基于DEM的综合插值法,与反距离权重法、普通Kriging法以及基于DEM的一般插值方法进行了插值效果对比和精度比较,在此基础上对4种插值方法得到的温度分布进行了分区,并对分区结果进行了分析。研究结果表明:插值精度越高,插值得到的气温分布与实际气温分布越接近,温度区划结果就越精确;基于DEM的2种插值方法较反距离权重法、普通Kriging法精度提高了1倍,更适用于地形复杂地区的公路气候区划需求;基于DEM的综合插值法相较于基于DEM的一般插值法在计算精度上提高了10%以上,在公路气侯区划数据有限时具有一定的优越性。
The accuracy of the temperature interpolation method has an important impact on the highway climate zoning. Taking the Tianshan area of Xinjiang as an example, the temperature data of 52 weather stations in Tianshan and its surrounding areas during 1971 ~ 2000 are used, and the multiple linear regression Method was used to analyze the relationship between the annual average temperature and the longitude, latitude and altitude in the Tianshan Mountains. Based on the general interpolation method of DEM (Digital Elevation Model), the integrated interpolation method based on DEM was proposed and the inverse distance Weight method, ordinary Kriging method and DEM-based interpolation method were used to compare the interpolation results and accuracy. Based on these, the temperature distributions obtained by the four interpolation methods were partitioned and the partitioning results were analyzed. The results show that the higher the interpolation accuracy, the closer the temperature distribution is to the actual temperature distribution and the more accurate the temperature partitioning results. The accuracy of the two methods based on DEM is double that of the inverse distance weighting method by ordinary Kriging method, Which is more suitable for the needs of highway climate zoning in complex terrain areas. Compared with the general interpolation method based on DEM, the DEM-based interpolation method improves the calculation accuracy by more than 10%, and has certain superiority when the highway weather zoning data is limited.