Predictive Soil Mapping - Mojave Desert, CA
Goal: To integrate GIS and Remote Sensing technology into standard soil survey in order to make large-scale soil mapping more cost effective, and to produce more robust data products.
Rationale: The need for soil maps of the Mojave Desert of California is widely recognized, however the costs associated with traditional survey methods make comprehensive field-based mapping impossible. Predictive soil modeling has the potential to help reduce mapping costs and make sound ecological management of the Mojave a reality.
Nature of Collaboration: San Diego State University, the University of California at Santa Barbara, and the USDA - Natural Resources Conservation Service have formed a consortium to investigate technologically advanced soil modeling techniques to support future soil survey efforts in the Mojave Desert Eco-region (funded by NASA's Office of Earth Science Research).
Theoretical basis of PSM:
The State Factor soil-forming model forms the theoretical basis of both traditional soil survey and this research. Jenny (1941), who formalized Dokuchaiev's ideas of soil formation, popularized the theory. It remains today a paradigm through which soil formation and distribution can be studied. The theory states that soil profile character is a function of climate, organisms, relief, parent material, and time, implying that if the spatial distribution of the soil forming factors is known, then soil character may be inferred. Digitally mapped environmental variables serve as surrogates for the factors.
Predictive soil mapping (PSM)- what is it?
PSM is an explicit and quantitative method of predicting soil character (including both soil "type"and soil property information) across a landscape from digitally mapped environmental variables (surrogates for soil forming factors). PSM begins with the development of a numerical or statistical model of the relationship between environmental variables andsoil data. The model is then applied to a geographic database to create a predictive map.
Fort Irwin case study objectives:
General Approach: Samples are extracted from the existing soil map and used as dependent variables in a binary decision tree modeling approach. A separate model is developed for each mapping unit thought to occur in the area where the model will be applied. The approach yields a separate probability grid (30m) for each unit. The predictions are then used to label the landform polygons.
Early Observation: A NRCS level 4 soil map drawn by surveyors during the mapping of Fort Irwin was compared to a landform map produced by the Mojave Desert Ecosystem Project. Visual inspection indicated that the boundaries separating the respective units occur in roughly the same location. This observation suggests that the landform linework might be of use in early stages of soil mapping in areas of the Mojave where the landforms have been mapped but the soils have not.
For more information on this project please contact Pete Scull: email@example.com