| 中国区域加权平均温度的时空分析及模型研究 |
| |
| 引用本文： | 聂檄晨,胡伍生,朱明晨,于先文,王庆.中国区域加权平均温度的时空分析及模型研究[J].测绘科学,2021,46(2):159-164. |
| |
| 作者姓名： | 聂檄晨 胡伍生 朱明晨 于先文 王庆 |
| |
| 作者单位： | 东南大学交通学院,南京211100;南京市测绘勘察研究院股份有限公司,南京210019 |
| |
| 基金项目： | 国家自然科学基金项目(41574022,41674035)。 |
| |
| 摘 要： | 加权平均温度(Tm)是全球卫星导航系统技术反演大气可降水量的关键参数,影响着水汽反演的精度。针对传统的Bevis模型运用在中国区域精度不高的问题,该文提出新的增加时空参数的Tm多元线性回归模型。根据2013—2015年中国86个探空站点的探空资料,分析了Tm的时空特征;然后根据2013年站点资料,利用线性回归建模方法建立了中国区域的Tm单因子回归模型和增加了时空参数的Tm多因子回归模型,并利用2014—2015年的探空数据进行验证。Tm单因子回归模型和Tm多因子回归模型的精度分别为3.1 K和2.6 K,比Bevis模型(精度3.3 K)分别提高了约6.0%和21.2%。考虑到季节对Tm的影响,将Tm多因子回归模型按季节分段,得到按季节分段的Tm多因子回归模型,其精度与Tm多因子回归模型大致相当,但能更细致表达出不同季节Tm的精度情况。结果表明增加了时空参数的Tm多因子回归模型更加适合中国区域的加权平均温度Tm的计算。
|
| 关 键 词： | 加权平均温度 时空分析 线性回归 |
Spatio-temporal analysis and model study of regional weighted average temperature in China |
| |
| NIE Xichen,HU Wusheng,ZHU Mingchen,YU Xianwen,WANG Qing.Spatio-temporal analysis and model study of regional weighted average temperature in China[J].Science of Surveying and Mapping,2021,46(2):159-164. |
| |
| Authors: | NIE Xichen HU Wusheng ZHU Mingchen YU Xianwen WANG Qing |
| |
| Institution: | (School of Transportation,Southeast University,Nanjing 211100,China;Nanjing Institute of Surveying,Mapping&Geotechnical Investigation Co.,Ltd.,Nanjing 210019,China) |
| |
| Abstract: | The weighted average temperature(Tm)is the key parameter of the global satellite navigation system technology to retrieve atmospheric precipitable water,which affects the accuracy of water vapor retrieval.In view of the low accuracy of traditional Bevis model in China,a new Tm multiple linear regression model with time-space parameters was proposed.According to the sounding data of 86 sounding stations in China from 2013 to 2015,the temporal and spatial characteristics of Tm were analyzed;then,according to the data of stations in 2013,the single factor regression model of Tm in China and the multi-factor regression model of Tm with additional temporal and spatial parameters were established by using the linear regression modeling method,and verified by the sounding data from 2014 to 2015.The accuracy of Tm single factor regression model and Tm multi-factor regression model were 3.1 K and 2.6 K,respectively,which were about 6.0% and 21.2% higher than Bevis model(the accuracy was 3.3 K).Considering the influence of season on Tm,the Tm multi-factor regression model was obtained by season segmentation.The accuracy of Tm multi-factor regression model was about the same as Tm multi-factor regression model,but it could express the accuracy of Tm in different seasons more carefully.The results showed that the multi factor regression model with time-space parameters was more suitable for the calculation of weighted average temperature Tm in China. |
| |
| Keywords: | weighted average temperature spatio-temporal analysis linear regression |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
