ИСТИНА |
Войти в систему Регистрация |
|
ИПМех РАН |
||
Detailed knowledge on the spatial distribution of soils is crucial for environmental monitoring, management, and modelling. Authors present a method for the refinement of complex soil map units in which two or more soil types are aggregated. The aim is to draw new boundaries inside the map polygons to represent a single soil type and no longer a mixture of several soil types. The basic idea for the method is the functional relationship between soil types and topographic position as formulated in the concept of the catena. Authors use a comprehensive soil profile database and topographic attributes derived from a 10 m digital elevation model as input data for the classification of soil types with random forest models. For prediction of the soil types, authors stratified the soil map into groups and apply a specific random forest model only to the associated map units. Results show a significant spatial refinement of the original soil polygons. Validation of predictions was estimated on 1812 independent soil profiles which were collected subsequent to prediction in the field. Field validation gave an overall accuracy of 70%. Map units, in which shallow soils were grouped together with deep soils could be separated best. Also, Histosols could be predicted successful. Highest error rate was found in map units, in which Gleysoils were grouped together with deep soils or Anthrosols. To check for the validity of results authors open the black box random forest model by calculating the variable importance for each predictor variable and plotting response surfaces. Authors found good confirmations of hypotheses, that topography has a significant influence on the spatial arrangement of soil types and that these relationships can be used for disaggregation.