ИСТИНА

Developed and tested algorithms for online humus stocks estimation on the basis of information accumulated in the Information system Soil-geographic database of Russia (ISSGDB). Zernogradsky district of Rostov region was chosen as model region with high degree of agricultural development (90%), about 90% of which is arable land. Soil cover of the area is quite homogeneous with domination of loamy Vorony-Calcic Chernozems. We used the following information: vectorized legacy medium- and large-scale soil maps, analytical characteristics of reference profiles and attribute data of the regular monitoring surveys of agricultural lands. Soil profiles with carbon content, texture and bulk density data of all horizons were selected.Two obtained maps were compared: The first one was constructed according to traditional approach by using the characteristics of reference soils, humus stocks averaged data and large-scale soil maps. The second one was constructed on a regular grid (30'' step), as a result of the online humus stocks estimation on the base of reference soil parameters and agrochemical monitoring data. Various calculation algorithms were tested by averaging data for observation period and by space for vicinity with variable radius. Improving the accuracy of the online estimation strategy was identified on the base of mapping analysis.