一种依据不同变量建立加权平均温度模型  

A weighted mean temperature model according to different variables

作　　者：孔晓宇 何士伟 从建锋 KONG Xiaoyu;HE Shiwei;CONG Jianfeng(College of Geomatics, Shandong University of Science and Technology, Qingdao 266590, China)

机构地区：[1]山东科技大学测绘科学与工程学院,山东青岛266590

出　　处：《测绘工程》2020年第6期31-36,共6页Engineering of Surveying and Mapping

摘　　要：随着地基GPS气象学快速的发展,天顶湿延迟转换至大气可降水量的关键因子∏成为研究热点,选择大气加权平均温度(Tm)与地表温度(Ts)、地表露点温度(Td)、地表水汽分压(Es)、地表水汽分压对数形式(lnEs)建立香港地区多元变量线性回归B1、B2、B3、B4模型,其中B2与B4模型拟合RMSE值均为2.27 K,预测RMSE为2.28 K,两者模型均优于Bevis模型和李建国模型,其对大气可降水量的影响为0.608 mm。考虑Tm具有季节性周期变化,以及昼夜的差别,分别建立8种时间分辨率更高的季节和昼夜多元线性回归模型。季节性昼夜模型预测精度除秋季昼夜模型外,均优于本地化B2模型,其中模型预测差值最大为0.432 K,说明在香港地区建立春、夏、冬季节性昼夜模型有利于提升区域转换因子∏精度。Along with the development of ground-based GPS meteorology quickly,the key factor of the conversion of zenith wet delay to the precipitable water vapor has become a research hotspot.In order to improve precision of∏,the weighted mean temperature of the atmosphere(Tm)and land surface temperature(Ts),dew point temperature(Td),water vapor partial pressure(Es),the surface of the water vapor partial pressure logarithmic form(lnEs)established Hong Kong multivariate linear regression,B1,B2,B3,B4 model.B2 and B4 model fitting RMSE value was 2.27 K,prediction RMSE were 2.28 K.Both models were superior to Bevis model and Li Jianguo model,and their influence on atmospheric precipitable water vapor was 0.608mm.Considering the seasonal variation of Tm and the difference between day and night,eight seasonal and day-night multiple linear regression models with higher temporal resolution were established.Except for the autumn day-night model,the prediction accuracy of the seasonal day-night model was better than that of the localized B2 model.The maximum prediction difference of the model was 0.432k,which indicated that the establishment of the spring,summer and winter seasonal day-night model in Hong Kong was conducive to the improvement of the regional conversion factor accuracy.

关 键 词：天顶湿延迟 大气可降水量 大气加权平均温度 

分 类 号：P228[天文地球—大地测量学与测量工程]