In the past few decades, forest fires have increased in number and severity, especially in the Mediterranean regions of Türkiye and Greece, where significant fires caused damage to thousands of hectares of land as well as wildlife. The main objective of the present study is to develop an emission estimation method with satellite-based burned area data from significant forest fire events in the Eastern Mediterranean in July–August 2021. In the first stage of this study, pre-fire and post-fire images of the study area acquired by the Sentinel-2 satellite were processed to calculate the normalized burn rate difference index (dNBR). Then, CORINE Land Cover (CLC) data were used for detecting land cover classes in the burned areas. Atmospheric emissions of NOx, CO, SO2, total suspended particulate matter (TSP), particulate matter with diameters that are equal to or smaller than 2.5 μm (PM2.5), and black carbon (BC) were estimated using the EMEP/EEA Tier 2 - technology-specific approach method, in which burned area maps were retrieved using Sentinel-2 imageries and later combined with land cover type and burning efficiency to estimate the quantity of burning biomass emissions. Emission factors were then used to estimate the fires' trace gas and aerosol emissions. The results showed that the highest burned areas were found in the western Mediterranean region in Türkiye and Central Greece (⁓50,000 ha). The atmospheric emissions from these fires were calculated to be similar in both countries. Furthermore, emission amounts were compared with three different global fire emission inventories including GFAS, GFED, and FINN. The emissions obtained from the GFAS database were the highest emissions of the four emission estimation approaches and our estimated emissions were close to the GFAS. Emissions calculated from the other two databases (FINN and GFED) mostly provided underestimated emissions. The emission uncertainties in this study mainly originated from assumptions regarding the inclusion of burned area efficiency in emission calculations, the landcover dataset, and the emission factors used. Overall, this study is considered a new approach to emission calculations using Sentinel-2 data. This research provides further insight into the use of Sentinel-2 data in emission calculation applications at the local to regional scales.