| At present,China has accumulated meteorological satellite data for more than 20 years,but the observation data from different series of satellites and different sensors lacks a uniform radiation calibration standard,and the data quality is uneven.It is difficult to use it for environmental and climate change monitoring with maximum effectiveness.This requires solving the problems of insufficient accuracy and mutual consistency of long-term series of meteorological satellite data,that is,recalibrating the historical satellite data according to the same radiation standard.This study is based on a variety of satellite remote sensing data,according to the needs of the calibration site for the temporal stability,spatial uniformity of the radiation characteristics and atmospheric dryness,referring to similar international studies to set the index parameters and select peseudo-invariant calibration sites(PICS)in the northwest of China to serve as a unified calibration benchmark for long-term serial domestic satellite data.Northwest China has arid climate and high altitude.It has a vast desert Gobi surface and low aerosol and water vapor content.It has the conditions and potential for selecting PICS.According to the spatio-temporal variation coefficient of satellite observation data,the temporal stability and spatial uniformity of ROI are evaluated.This paper mainly uses MODIS land surface reflectance and BRDF products,combined with MODIS aerosol and water vapor and other atmospheric remote sensing products,to carry out many years of satellite remote sensing data analysis in Northwest China,and selects PICS sites with relatively excellent radiation performance in Northwest China.The sites are mainly distributed in the Taklimakan Desert in Xinjiang,the Chaidamu Basin in Qinghai,the Hexi Corridor,and the Tengger Desert in Inner Mongolia.By evaluating and comparing the characteristics of atmospheric parameters(aerosol,water vapor,ozone)and cloudiness of 4 typical sites,the atmospheric characteristics of PICS in different spatial regions are distinguished,and Landsat 8 OLI remote sensing data with a spatial resolution of 30 m is used in 250 m and 1000 m evaluation scale,the spatial uniformity of the site was evaluated.The preliminary selection of PICS was evaluated in terms of atmospheric characteristics and spatial scale characteristics,so as to achieve a more comprehensive andreasonable selection of PICS sites in northwestern China.Comparing the selected site with similar international sites horizontally,China has better site characteristics under the same geographical conditions and has a significant advantage of high altitude.Based on the above preliminary site selection results,the key features(spectral reflectance and BRDF)measurements of some sites were carried out in the summer of 2019,and the FY-3D MERSI site vcarious calibration test was also considered.Based on the measured surface reflectance spectral data of XCDH?W and SHDBAN sites around the Qaidam Basin,the effectiveness of the selected PICS calibration application was verified.The shapes of all the surface reflectance spectral conform to the characteristics of the spectral curve of the sandy land.Comparing the spectral observed by the UAV at a height of 100 m and the MODIS Terra satellite platform with ground observations,the results are highly consistent.It shows that the local measured spectral of the surface can characterize the radiation characteristics of the site at a larger scale,which verifies the uniformity of the site.Based on the 4-parameter spectral model in VNIR,the measured spectral of the site is simulated,the RMSE is less than 0.5%,and the determination coefficient reaches 0.995.Based on the measured and simulated surface reflectance spectral,the TOA reflectance was compared with the satellite observations of FY-3D MERSI.The relative percentage deviations were all less than 5%.The results show that the XCDH?W and SHDBAN sites can be effectively used in the calibration of FY-3D MERSI and have the ability to provide radiation benchmarks for domestic meteorological satellites.This paper also demonstrates the application of the automatic radiation calibration method in the typical PICS DHANG to the satellite instruments' radiation calibration performance tracking ability.The accuracy of the automated radiation calibration method was verified using MODIS observation data with higher accuracy.The results show that the method has a calibration accuracy of 3% in VNIR,5% in SWIR,and a standard deviation ranging from1.95% To 3.67%.Based on this method,the radiation performance of domestic meteorological satellite optical loads FY-3B VIRR and FY-3C VIRR has been tracked for more than 3 years,and FY-3D MERSI has been tracked for more than 2 years.The results of FY-3C VIRR of multi-site calibration methods and the automated radiation calibration method are compared.The calibration slope of each channel of the three optical loads has seasonal changes,which reflects the seasonal changes of the DHANG field in the calibration,but the overall stability is stable.Compared with the multi-site calibration method,the average value over the entire3-year time scale has good consistency.The automatic radiation calibration method has reliable accuracy and the ability to track radiation performance.It can be considered to be extended to other sites and applied to more optical load automated site calibration. |
PDF Full Text Request