Spatial data mining for enhanced soil map modelling

TLDR

Induction rules using spatial environmental data have been applied to soil mapping, yet small classes and fine spatial patterns were poorly represented. The study evaluates strategies to improve soil map models generated by rule induction. The authors incorporated 250‑m resolution terrain attributes, a proportional sampling strategy, boosting of decision trees, and spatial contextual information into the rule‑induction process. The optimal model combined proportional sampling, boosted decision trees, and spatial contextual information, yielding the highest soil map quality.

Abstract

The principle of using induction rules based on spatial environmental data to model a soil map has previously been demonstrated. Whilst the general pattern of classes of large spatial extent and those with close association with geology were delineated, small classes and the detailed spatial pattern of the map were less well rendered. Here we examine several strategies to improve the quality of the soil map models generated by rule induction. Terrain attributes that are better-suited to landscape description at a resolution of 250 m are introduced as predictors of soil type. A map sampling strategy is developed. Classification error is reduced by using boosting rather than cross-validation to improve the model. Further, the benefit of incorporating the local spatial context for each environmental variable into the rule induction is examined. The best model was achieved by sampling in proportion to the spatial extent of the mapped classes, boosting the decision trees, and using spatial contextual information extracted from the environmental variables.