| 基于DEM修订的新疆区域气温空间插值精度优化研究 |
| |
| 引用本文： | 武鹏飞,张钧泳,谭娇.基于DEM修订的新疆区域气温空间插值精度优化研究[J].新疆气象,2019,13(2):108-114. |
| |
| 作者姓名： | 武鹏飞 张钧泳 谭娇 |
| |
| 作者单位： | 新疆乌鲁木齐气象卫星地面站,新疆大学 资源与环境科学学院,新疆大学 资源与环境科学学院 |
| |
| 基金项目： | 国家自然科学基金(41771470,U1303381,41261090);自治区科技支疆项目(201591101);自治区重点实验室专项基金(2016D03001, 2014KL005);教育部促进与美大地区科研合作与高层次人才培养项目;新疆大学博士生科技创新项目(XJUBSCX-2016015) |
| |
| 摘 要： | 为研究新疆地区气温的空间变异性,以新疆66个国家气象台站1981—2010年月平均气温和30 m空间分辨率DEM数据为基础,采用传统插值法、基于DEM多元线性回归插值和基于DEM修正的空间插值方法对新疆区域气温数据进行栅格化,并分析年平均气温与海拔的相关关系。通过采用反距离权重法(IDW),普通克里格法(Kriging),样条函数法(Spline)和趋势面分析法(Trend)4种空间插值方法对气象要素进行直接插值、气温多元回归模型残差结果插值、基于DEM修正插值对比分析。通过针对插值方法进行基于MAE和RMSIE的交叉验证,结果表明传统插值方法、基于多元线性回归和基于DEM修正4种空间插值精度均为IDWKrigingSplineTrend。反距离权重(IDW)空间插值方法最优,基于DEM修正IDW插值、基于多元线性回归IDW插值与传统IDW插值精度分别是0.039、0.477、1.038,插值结果客观的表达了新疆区域气温随空间梯度的变化趋势。
|
| 关 键 词： | 数字高程模型(DEM) 气温 空间插值 新疆 |
| 收稿时间： | 2018/1/17 0:00:00 |
| 修稿时间： | 2018/4/12 0:00:00 |
Research on Spatial Interpolation Accuracy Optimization of Air Temperature in Xinjiang Based on DEM |
| |
| WU Pengfei,ZHANG Junyong and TAN Jiao.Research on Spatial Interpolation Accuracy Optimization of Air Temperature in Xinjiang Based on DEM[J].Bimonthly of Xinjiang Meteorology,2019,13(2):108-114. |
| |
| Authors: | WU Pengfei ZHANG Junyong and TAN Jiao |
| |
| Institution: | Urumqi Meteorological Satellite Ground Station,College of Resources and Environment Science,Xinjiang University,College of Resources and Environment Science,Xinjiang University |
| |
| Abstract: | In order to study the spatial variability of temperature in Xinjiang,based on the DEM data of the average monthly temperature and spatial resolution of 30m from 66 national meteorological observatories in Xinjiang during 1981-2010.Direct space interpolation was widely used in meteorological rasterization of air temperature data such as the spatial interpolation technique of IDW,Kriging,Spline,Trend,Lagrangian,etc,without considerding the effects of topography on the air temperature,in recent years with the expansion of GIS technology applications,the method of regression model by geographic factors about longitude,latitude,elevation,etc combined with spatial interpolation was used in grid-based regional climate factors and achieve some result of With Surfer,ArcGIS as a platform.Through cross-validation of MAE and RMSIE for the interpolation method,the results show that the annual average temperature multiple temperature regression model for Xinjiang residual interpolation and DEM correction interpolation accuracy are higher than the meteorological element direct interpolation,and the DEM correction interpolation accuracy is the highest. The results showed that among the four interpolation method combined with GIS,the accuracy of the interpolation method of IDW is higher than that of other interpolation methods,however,the results of multivariate linear regression is better than direct interpolation, The results show that the accuracy of the four spatial interpolation methods based on multiple linear regression and DEM correction is higher than the traditional interpolation method of meteorological elements. The inverse distance weighting (IDW) spatial interpolation method is optimal. The IDW interpolation based on DEM, the IDW interpolation based on multiple linear regression, and the traditional IDW interpolation accuracy are 0.039, 0.477, and 1.038.The accuracy of IDW interpolation based on DEM correction is generally higher than other interpolation methods,and the interpolation result objectively expresses the variation trend of Xinjiang air temperature with spatial gradient. |
| |
| Keywords: | Digital elevation model(DEM) Temperature Spatial interpolation Xinjiang |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《新疆气象》浏览原始摘要信息 |
| 点击此处可从《新疆气象》下载免费的PDF全文 |
|
