Global climate change is one of the most important scientific, societal and economic contemporary challenges. Fundamental understanding of the major processes driving climate change is the key problem which is to be solved not only on a global but also on regional scales. The accuracy of regional climate modelling depends on a number of factors. One of these factors is the adequate and comprehensive information on the anthropogenic impact which is highest in industrial regions and areas with dense population-modern megacities. Megacities are not only "heat islands", but also significant sources of emissions of various substances into the atmosphere, including greenhouse and reactive gases. In 2019, the mobile experiment EMME (Emission Monitoring Mobile Experiment) was conducted within the St. Petersburg agglomeration (Russia) aiming to estimate the emission intensity of greenhouse (CO2, CH4) and reactive (CO, NOx) gases for St. Petersburg which is the largest Northern megacity. St. Petersburg State University (Russia), Karlsruhe Institute of Technology (Germany) and the University of Bremen (Germany) jointly ran this experiment. The core instruments of the campaign were two portable FTIR spectrometers Bruker EM27/SUN which were used for ground-based remote sensing measurements of the total column amount of CO2, CH4 and CO at upwind and downwind locations on the opposite sides of the city. The NO2 tropospheric column amount was observed along a circular highway around the city by continuous mobile measurements of scattered solar visible radiation with OceanOptics HR4000 spectrometer using the DOAS technique. Simultaneously, air 1 5 10 15 20 25 https://doi.org/10.5194/amt-2020-87 Preprint. Discussion started: 22 April 2020 c Author(s) 2020. CC BY 4.0 License. samples were collected in air bags for subsequent laboratory analysis. The air samples were taken at the locations of FTIR observations at the ground level and also at altitudes of about hundred meters when airbags were lifted by a kite (in case of suitable landscape and favourable wind conditions). The entire campaign consisted of 11 mostly cloudless days of measurements in March-April 2019. Planning of measurements for each day included the determination of optimal location for FTIR spectrometers based on weather forecasts combined with the numerical modelling of the pollution transport in the megacity area. The real-time corrections of the FTIR operation sites were performed depending on the actual evolution of the megacity NOx plume as detected by the mobile DOAS observations. The data processing activities included the following steps: (1) the generation of calibrated spectra from raw interferograms; (2) the retrievals of the CO2, CH4, and CO column averaged abundances using the software tools provided by the COCCON (Collaborative Carbon Column Observing Network); (3) the retrieval of tropospheric NO2 amount from DOAS measurements; (4) the laboratory analysis of air samples; (5) the numerical modelling of the plume movement based on the actual meteorological information. The estimates of the St. Petersburg emission intensities for the considered greenhouse and reactive gases were obtained by coupling a box model and the results of the EMME observational campaign using the mass balance approach. The CO2 emission flux for St. Petersburg as an area source was estimated as 89±28 kt km-2 yr-1 which is two times higher than the corresponding value in the EDGAR database. The experiment revealed the CH4 emission flux of 135± 68 t km-2 yr-1 which is about one order of magnitude greater than the value reported by the official inventories of St. Petersburg emissions (~17 t km-2 yr-1 for 2017). At the same time, for the urban territory of St. Petersburg, both the EMME experiment and the official inventories for 2017 give similar results for the CO anthropogenic flux (251±104 t km-2 yr-1 vs. 280 t km-2 yr-1) and for the NOx anthropogenic flux (66±28 t km-2 yr-1 vs. 47 t km-2 yr-1).