Fusion Application Of GNSS And MODIS Precipitable Water Vapor In Mainland China

Precipitable water vapor(PWV)is one of the important greenhouse gases,which is actively changing in the atmosphere and is closely related to the occurrence of strong convective weather.At present,using global navigation satellite system(GNSS)to retrieve PWV has gradually developed and matured,with high precision and temporal resolution.However,due to the long distance between GNSS stations,low density,and uneven distribution of stations,PWV monitoring with large scale and high spatial resolution cannot be carried out.The water vapor products obtained from Moderate-resolution imaging spectroradiometer(MODIS)are one of the commonly used data in large-scale water vapor monitoring,but their accuracy is low,and it is necessary to carry out the calibration of MODIS PWV using GNSS in order to improve its data reliability.This paper used the GNSS observation data provided by the crustal movement observation network of China(CMONOC)to verify and correct the accuracy of MODIS PWV and generated high-precision PWV products to provide data support for monitoring and forecasting disaster weather.The main research contents and conclusions of this paper are as follows.(1)The accuracy of MODIS water vapor is evaluated.In this paper,the empirical orthogonal function(EOF)and rotational empirical orthogonal function(REOF)analyses suggest that the PWV content and its variation in mainland China show obvious longitudinal and latitudinal distributions,and based on this,mainland China was divided into 13 regions.The results show that the root mean square error(RMSE)reaches a maximum in summer(9.14 ～ 32.74 mm)and a minimum in winter(0.88 ～ 15.95 mm),and the RMSE is generally higher in low-latitude regions than in mid-latitude regions.The accuracy of MODIS PWV is strongly dependent on the PWV content.(2)Construction of MODIS PWV calibration model.The best model type was selected experimentally for different regions and seasons,and the regional MODIS PWV calibration model(simple model)was constructed according to the least squares method.After calibration,the accuracy of MODIS PWV in each region has been improved to different degrees,and the best calibration effect is achieved in the subtropical monsoon and tropical monsoon regions,and the RMSE can be reduced by up to 79.1% compared with that before calibration.(3)Establishment of a calibration model for MODIS PWV products based on residual frequency domain analysis.To further improve the goodness-of-fit and accuracy of calibration models,the periodicity of the model residuals was analyzed.The results show that in different regions,the model residuals show obvious cyclic characteristics such as annual cycle and semi-annual cycle.In this paper,based on the simple calibration model,a regression model including the periodic term was established.The RMSE of the corrected MODIS PWV calibration model based on residual frequency domain analysis was 34.4 %,56.9 %,33.4 % and 22.0 % in spring,summer,autumn and winter,respectively.Compared with the simple model,the goodness of fit of the optimized model increased by 0.012 ～ 0.566,and the RMSE values can be further reduced by 0.4% ～ 43.8%,an average of 13.6%,so the calibration effect is better.(4)Generation and application of MODIS PWV calibration products for the Chinese mainland.Based on the high spatial resolution of MODIS,the overall MODIS PWV products are calibrated to generate MODIS PWV calibration products for mainland China.Taking Typhoon Lekima as an example,the corrected MODIS PWV products are used to analyze the PWV changes before and after the typhoon landfall,and provide reference for monitoring and warning of extreme weather.