Аннотация:The study of fires and their consequences using remote sensing methods in the optical range is widespread. However, many works are devoted to forest fires, a little less to fires in arid areas. At the same time, even in arid regions, weather conditions do not always allow us to get a correct picture of the spread of fires in a sufficiently high spatial resolution (the first tens of meters). Our research uses the study area of the separating parts of the launch vehicles drop zone in Central Kazakhstan. To study the possibilities of remote sensing in the radio range, we used data from Sentinel-1 satellites to the same territory in Central Kazakhstan. In the course of experimental work, we found fresh burns in VH polarization seem to be the most contrasting. Therefore, we used Google Earth Engine to process the data. Since most fires in this area occur in June and July, we took the average image in VH polarization for April and August and subtracted one from the other. Due to the many factors affecting the radio range’s backscattering, we also added the difference between May and September average images to the previous one. Segmentation of the image by the mean shift method and classification of the obtained range of values into two classes using the natural breaks method gave a result comparable in accuracy with the use of indices in the optical range.