Land resources management, hydrological modeling and ecological
environmental decision-making require knowledge about the
spatial distribution of
soil properties to improve their accuracy and reliability. Traditional soil survey technology can not meet this demand. Based on the case study of Longyou County, Zhejiang Province, a GIS database including the soil attributes data which originated from the Second Chinese Soil Survey and some easily observable environmental information were designed and built.
Decision tree modeling was introduced to deducing and expressing spatial distribution of 4 soil properties, i.e. pH, organic matter, available P and available K. The trees associated content class of each soil property with envirnmental attributes, such as landform, geology, landuse and remote sensing images, was translated into a clear, quantitative, environmental factor-associated regular system. This system can be used to predict continuous spatial distribution of soil properties. Results of the
prediction showed that approximately 75～81% spatial variance of soil properties could be explained by the
decision tree models.
More abstracts about the Spatial Prediction of Soil Properties Based on Decision Tree Modeling