摘要
青海地区海拔差异大、区域气候背景不同,具有显著的区域差异性.为在青海地区开展GPS(全球定位系统,Global Position System)气象应用研究,获取不同地区中尺度、高时间分辨率的水汽信息,本文以青海省三江源区(达日)、西宁市为研究区域,利用2010~2017年的探空数据回归得出两地本地化大气加权平均温度(Weighted Average Temperature,Tm)模型,对比分析发现两地本地化回归模型的精度要优于最早建立的Bevis模型和全国范围精度较高的龚绍琦模型;同时,对比分析三江源区(达日)、西宁两地利用GPT2(全球气温和气压经验模型,Global Pressure and Temperature)对流层模型与实测气象数据进行GPS大气可降水量(Precipitation Water Vapor,PWV)反演的差异,发现在研究区可利用GPT2对流层模型作为无实测气象数据PWV反演的补充,且精度相当.最后,利用研究区2016年GPS观测数据,使用本地化T_m模型,并以GPT2对流层模型为补充进行PWV反演,利用探空数据计算得到的PWV作为参考值,发现GPS反演PWV与参考值的相关系数达到0.9以上,且GPS-PWV略大于基于无线电探空观测(Radio Sounding,RS)获得的RS-PWV.本文研究为在青海地区开展GPS水汽遥感工作奠定了基础.
The accurate calculation of water-vapour content provides an important basic for the study of meteorology and atmospheric geophysics.In addition,numerical weather prediction requires water-vapour information at large-scale spatial and high temporal resolutions.At present,the main technical means(radio sounding,water-vapour radiometry) for observing precipitable water-vapour(PWV) are deficiencies in both these techniques,such as the high observation cost,sparse number of observation sites,and low-resolution of the observation times.As such,it is difficult to obtain observation information regarding the continuous variations of water-vapour or to effectively identify various short-term meteorological processes.We selected the Three River Headwaters Region(TRHR)(Dari County) and Xining city,which are two regions with significant regional differences in altitude and climate.Using the linear regression of sounding data from 2010 to 2017,we obtained the localised atmospheric weighted average temperature(Tm) model and verified the model accuracy.the analysis verified whether the GPT2 tropospheric model satisfies the accuracy requirements of GPS-inversion PWV in the absence of measured meteorological data.Lastly,using GPS observation data and the localised T_m model of the study area in 2016,we supplemented the PWV inversion by the GPT2 tropospheric model and compared it with the RS-PWV.The results indicate that the localised T_m regression model of the TRHR(Dari County) and Xining City performs better than the Bevis and Gong Shaoqi models.For the PWV inversion in the study area,the GPT2 tropospheric model can be used as a substitute for unmeasured meteorological data with comparable accuracy.The results of our inversion analysis show that the correlation coefficient between the GPS-PWV and RS-PWV is higher than 0.9,with the GPS-PWV being slightly larger than the RS-PWV.The results lay the foundation for GPS water-vapour remote sensing work and provide an important guarantee of obtaining high-precision,high-time-resoluti
作者
乔禛
魏加华
袁晓伟
李琼
张令振
QIAO Zhen;WEI Jiahua;YUAN Xiaowei;LI Qiong;ZHANG Linzheng(State Key Laboratory of Plateau Ecology and Agriculture,Qinghai University,Xining 810016,China;State Key Laboratory of Hydro-science&Engineering,Tsinghua University,Beijing 100084,China;School of Hydraulic and Electric Engineering,Qinghai University,Xining 810016,China;Xining Weather Bureau,Xining 810001,China)
出处
《应用基础与工程科学学报》
EI
CSCD
北大核心
2019年第5期957-970,共14页
Journal of Basic Science and Engineering
基金
国家重点研发计划项目(2017YFC0403600)
清华大学水沙科学与水利水电工程国家重点实验室开放基金(sklhse-2017-A-02)
国电电网科技项目(52283014000T)
上海航天科技创新基金(SAST2017-54)
