Аннотация:The rapid growth of the archives of the publicly available remote sensing data (RSD) of the Earth turns them into big data. In fact, for every element of the RSD resolution, tens of thousands of scenes and hundreds of thousands of spectral values are accumulated. The processing of individual images and channels becomes difficult, if at all possible. On the other hand, the quality of traditional statistical information on the territory of Russia is decreasing. It is almost impossible to understand where, how much and what agricultural land is cultivated, and how much is abandoned. Specific methods for processing large arrays of multispectral RSD of different acquisition time make it possible to extract information on the intensity of land use. New methods of processing allow using new methods of measuring the state of land. It becomes possible to move from binary logic (processing-abandonment) to a continuous numerical scale, where the notion of land abandonment is only part of the values of the scale that are below a certain threshold. The scale of land exploitation intensity can be used to create intensity maps that are the result of extracting information from the entire RSD array. At present, the spatial accuracy of the analysis over the past 35 years is 30 m.