GPT2w模型在南极地区精度分析  

The Accuracy Analysis of GPT2w at the Antarctic Area

作　　者：孔建[1,2] 姚宜斌 单路路[3] 王泽民 KONG Jian;YAO Yibin;SHAN Lulu;WANG Zemin(Chinese Antarctic Center of Surveying and Mapping,Wuhan University,Wuhan 430079,China;School of Geodesy and Geomatics,Wuhan University,Wuhan 430079,China;Key Laboratory of Geospace Environment and Geodesy,Ministry of Education,Wuhan University,Wuhan 430079,China)

机构地区：[1]武汉大学中国南极测绘研究中心,湖北武汉430079 [2]武汉大学地球空间环境与大地测量教育部重点实验室,湖北武汉430079 [3]武汉大学测绘学院,湖北武汉430079

出　　处：《测绘学报》2018年第10期1316-1325,共10页Acta Geodaetica et Cartographica Sinica

基　　金：国家重点研发计划(2016YFB0501803);国家自然科学基金(41574028;41604002);中央高校基本科研业务费专项资金(2042016kf0037);武汉大学地球空间环境与大地测量教育部重点实验室测绘基础研究基金(16-02-04)~~

摘　　要：GPT2w(global pressure and temperature 2wet)是目前应用较为广泛的对流层延迟经验模型之一,可提供气压、温度、水汽压等气象参数。为验证和分析GPT2w模型在南极地区的精度,本文利用分布在南极区域的探空站数据和中国第33次南极科考期间的实测探空气球数据对模型气压、温度、水汽压参数进行分层精度检验。与探空站数据比较发现,在南极地区地面高度上,GPT2w模型精度较高,与全球其他区域精度较为一致;进一步通过对比1月和7月统计结果,发现Bias和RMS呈现出季节特性;同时发现模型在垂直方向存在较大误差,表现为随着高度的增加,精度随之下降并逐步趋于稳定。实测数据对比方面,首先利用ECMWF(European Centre for Medium-range Weather Forecasts)气压分层数据对实测数据的可靠性进行验证,结果显示,实测数据与ECMWF分层数据符合得较好;同时通过比对发现,GPT2w天内精度在地面高度上仍与月平均精度相当,但垂直方向随着高度的增加精度相比于暖季精度会有所下滑,说明未考虑日周期项变化对模型精度存在一定影响。用探空数据计算的对流层延迟(zenith tropospheric delay,ZTD)来分析GPT2w的计算精度,结果表明GPT2w在南极区域ZTD计算精度在厘米级,与全球其他位置计算精度相当。GPT2w(global pressure and temperature 2 wet)is one of the widely used empirical troposphere models,which is usually used to compute the tropospheric parameters,such as pressure,temperature,water vapor pressure etc.In order to analyze the model accuracy around the Antarctic Pole area,the actual data from the radiosonde stations and the measured data during the Chinese 33th Antarctica Expedition are used to test the model parameter’s accuracy on different height level.Based on the comparison with the data from the radiosonde stations and GPT2w,it is found that on ground level,the model accuracy at Antarctic Pole area is high,which is consistent with the accuracy at other areas;then by comparing the statistics in January and July,it is found that Bias and RMS showed seasonal characteristics;at the same time,it is found that there is a large error in the vertical direction of the model,which shows that as the height increases,the accuracy decreases and gradually stabilizes.Before the comparison with the measured data and GPT2w,the ECMWF(European Centre for Medium-range Weather Forecasts)data are used to confirm the reliability of the measured data,while the results showed that the two dataset consistent with each other.Meanwhile,by comparison,it is found that the GPT2w accuracy among a day is equal to monthly average accuracy on ground level,but the vertical accuracy decreases with the increase of height compared with the accuracy of the warm season,which shows that there is a certain impact on the accuracy of the model without considering the variation of the daily period.zenith tropospheric delay(ZTD)calculated by sounding data is used to analyze the accuracy of GPT2w.The result shows that the accuracy of GPT2w ZTD in the Antarctic is centimeter accuracy,which is equivalent to the accuracy of other areas in the world.

关 键 词：GPT2w 南极地区 探空站 探空气球 ECMWF 

分 类 号：P228[天文地球—大地测量学与测量工程]