A System to Evaluate Prime Farmland Reclamation Success Based on Spatial Soil Properties

Abstract. Scholars, governmental agencies, and concerned citizens are interested in developing empirical predictive models to quantitatively assess the vegetative productivity potentials of reconstructed soils (neo- sols). This research presents equations for a northern Michigan mining region in the Upper Peninsula of Michigan, based on data derived from the National Resources Conservation Service. We employed principal component analysis to develop models to predict the vegetative productivity of corn, corn silage, oats, alfalfa/hay, Irish potatoes, red maple (Acer rubrum L.), white spruce (Picea glauca [Moench] Voss), red pine (Pinus resinosa Aniton), eastern white pine (Pinus strobus L.), jack pine (Pinus banksiana Lamb.), and lilac (Syringa vulgaris L.). Soil attributes that were examined in this research include: available water holding capacity, moist bulk density, % clay, % rock fragments, hydraulic conductivity, % organic matter, soil reactivity, % slope, and topographic position. Four predictive equations based on landscape topography have been developed and are described as an all-mesic woody plant and crop equation, a xeric equation, an equation specific to jack pine, and a wet environment equation. The models were highly significant (p<0.0001) and explained 87.93%, 74.52%, 65.33%, and 87.68% of the variation in site productivity of the respective landscape setting. These equations are intended to assist in efforts to assess the vegetative productivity potentials of reconstructed soils on post-mined landscapes and other disturbed landscapes.