This work is a continuation of previous studies on the relation between surface methane concentration with meteorological parameters and data on gray tone in 12 spectral bands of Sentinel-2 in the vicinity of Tiksi station. Almost similar analysis was carried out using a sample of daily observations and satellite imagery in summer months for the period from 2016 until 2022 year. It is shown that the regression models, calculated using sample of data for winter period are not applicable to predict surface methane concentration in summer months: determination coefficient for all models was less than 0,5, application of nonlinear models resulted in negative methane concentration. Regression analysis revealed a relation of Sentinel-2 spectral bands, or indices which inclusion together with wind speed in single statistical model caused the highest increase of determination coefficient. Two regression models were developed to evaluate surface methane concentration, which include relations between 2, 4, 7, 8, 9, 10 and 11 Sentinel-2 MSI bands and data on wind speed. After comparison of given MSI spectral bands with absorption lines of atmospheric gas components, it was found, that developed models for summer period are describing same effects as models for winter period: methane concentrations variability in atmosphere and sink of methane due to its reaction with hydroxyl radical OH.
Popov et al. (Fri,) studied this question.