Modeling and seasonal analysis of ozone content in field automatic calibration

In recent years, the form of site calibration has gradually evolved from manual measurement to automated mode. The National Satellite Meteorological Centre built up Dunhuang Site Automatic Observation Radiometric Calibration Operational System (DARCOS) in 2018, which achieved initial results. Ozone content is a parameter that has a relatively large influence on the visible light calibration of solar reflection channels. In automatic calibration systems, instruments for measuring ozone content are generally not involved, climate data is used instead. For the refined requirement of the algorithm of the DARCOS, in order to improve the forward accuracy near the Chappuis absorption band, by drawing on OMI ozone content products in 2013-2018, we performed seasonal analysis of Dunhuang ozone content data and built a yearly-collected data set model. On which it was found that in February, the monthly average of the concentration of the ozone column in the Dunhuang, fluctuated very obviously. Methods of correlation coefficient and machine learning random forest were applied respectively, to judge the correlation between meteorological parameters and the Ozone Content in February. The results: the ozone content in the Dunhuang region is negatively correlated with the temperature, average vapor pressure and minimum horizontal visibility.