Constructing a spatially-resolved database for modelling soil organic carbon stocks of croplands in European Russia

This paper describes the conception and construction of a geographic information system (GIS) for use in modelling changes in soil organic carbon stocks in European Russia. A GIS of croplands for European Russia was constructed to allow the RothC and CANDY models, and a statistical model of humus balance, to estimate how soil carbon stocks change in time. The soil map of Russia, the database of soil properties, the map of administrative division, the land use map, the climatic grid, the map of natural and agricultural zoning and an economic database serve as a basis for this system. A map and database of homogeneous units, for maximum accuracy and minimum uncertainty, was created. Homogeneous characteristics are the parameters required for modelling. In the course of this work, the sources of errors in the database and the possible ways of improving calculation accuracy were determined and are described. The methods used and decisions taken in constructing this database are applicable to other studies in which GIS databases need to be constructing from disparate sources.