Digital Soil Mapping: The Future Need of Sustainable Soil Management

Digital soil mapping (DSM) involves in development of a statistical or mathematical model to estimate soil class or properties at unsampled locations using information on spatial variation of soil properties and different covariates affecting soil formation process. There are three main approaches followed in DSM, and these are geostatistical approach, state-factor (clorpt) approach, and pedotransfer function (PTF) approach. In the geostatistical approach, spatial variation parameters (nugget, sill, and range) are identified from a spatial soil database using semivariogram followed by making unbiased estimate of soil properties at unsampled location through kriging. In the state-factor (clorpt) approach, the soil formation theory is the backbone. In this approach soil is considered to be influenced by five major factors: climate (cl), organism (o), relief (r), parent material (p), and time (t). Therefore, abundantly available information on these factors in different digital platforms are exploited to develop model to estimate soil properties at unsampled location. The PTF approach is used to develop digital soil maps of complex soil properties and difficult to measure soil properties. In this approach digital soil map of basic soil properties is first developed using the first two approaches, which are then combined to map of complex soil properties through PTF model. All these three approaches of DSM are discussed in detail along with assessment of its accuracy and uncertainty. Through the DSM approaches, available legacy soil data may be converted to digital products for its better accessibility and utility, e.g., through development of soil information system.