Tropospheric ozone is a greenhouse gas and a reactive and toxic pollutant detrimental to human health and ecosystem productivity. Therefore, the study of tropospheric ozone column (TrOC) variability by means of measurements and modeling is an essential task.

The IKFS-2 spectrometer aboard the Meteor-M N2 satellite measured outgoing thermal radiation in the 5-15 µm spectral range with an un-apodised spectral resolution of 0.4 cm^-1 in 2015-2022. A retrieval technique based on the artificial neural network (ANN) algorithm and the method of principal components has been developed for interpreting the IKFS-2 spectral measurements. For the ANN training, we used TrOCs derived from ozonesonde measurements at different ground-based sites taken from the archive created by the TOAR-II HEGIFTOM working group, thus solving the problem of calibration of the IKFS-2 TrOC data product. The uncertainty estimated for IKFS-2 TrOC measurements equals ~3 DU (~12-15 %). The IKFS-2 TrOC data product has been validated by comparison with independent TrOC measurements at the NDACC IRWG and with IASI satellite measurements.

The spatiotemporal variability of TrOC over Russia was studied based on IKFS-2 and IASI satellite measurements, as well as the WRF-Chem numerical modeling and EAC4 reanalysis data. The seasonal variability in TrOCs was estimated, and the observed anomalous values in TrOC distribution were analyzed. It was revealed that over most of the territory of Russia, except for certain regions of Siberia and the Far East (Kamchatka region), a decrease in TrOC was observed over the 2016-2022 period.

This research was supported by SPbU grant No. 116234986.