Methods for Field Scale Digital Mapping of Soil Functional Properties.

Traditional soil survey delineated soil units based on their physical, morphological differences, not their functionality in terms of crop production or other ecosystem functions. The goal of this study was to delineate and classify soils based on their functional similarities rather than morphological differences. In an effort to provide meaningful, relevant soils information at field scales, a digital soil mapping (DSM) technique, known as terrain attribute mapping (TASM), was implemented, under the soil similarity model, to generate a soil class map related to the yield functionality of the soils. From these “functional classes”, a continuous property map was created, predicting the depth to a water limiting paleosol layer, using the Soil-land Inference Model (SoLIM). This was done under the assumption that water movement and redistribution across the landscape is the driving factor in soil functional differentiation and that topography controls water movement. Terrain attributes were derived from a 5 meter Digital Elevation Model (DEM) for the Southeastern Purdue Agricultural Center (SEPAC), an aproximatly 4.1 Ha farm field. Field sampling of texture, redoxomorphic features, depth to fragipan and depth to paleosol were performed at 20 points selected using a Latin hyper cube sampling design, for the purpose of model validation. Rules were created based on terrain attributes to create a fuzzy membership map of five classes, each representing zones thought to differ in soil function. Validation found all five classes to be significantly different in terms of corn and soybean yield, determined by yield monitor data. These types of functional class and property maps may be used, in combination with field trial data, to inform management decisions and feed various systems models.