ИСТИНА

Pattern of agricultural regions is a dynamic system changing with uneven speed and scale in different parts of the world. This dynamics is expressed by expansion of agriculture towards new areas, abandoning of cropped area and restructuring of land use due to changes in specialization of agriculture. For detect changes in agricultural land use, remote sensing data analysis is usually applied. Geographers have developed methodology for detect land cover properties, which became a wide spread technology easy to reproduce. Nevertheless, it is not sufficient for understand and explain spatial dynamics of land use as a basic feature of agricultural regions if not combined with another kind of approach. This one is based on classical tools of social and economic geography as spatial analysis of crop allocation and specialization of agriculture and of differentiation of modes of production. Several data sources and corresponding indicators can be used, e.g. agricultural statistics, demographic data (density and ethnic structure of rural population). The present state of agricultural statistics coverage and advance of methods of big data processing provide new opportunities in this field. The paper aims to demonstrate how the above mentioned approaches can be applied jointly for benefit understanding dynamics of agricultural regions. We focus on two cases referring to the countries, where the most drastic land use changes are under way. The first one is Brazil as the area of modern agricultural frontier. The second one is Russia, where political system changes since the 1990s deeply influenced geography of agriculture.