Geospatial Analysis and Multivariate Classification of Soil Properties in Nicaraguan Sun and Shade Grown Coffee Systems
Anderson, N.P. 2006. M.Sc. Thesis. Program in Environmental Science and Regional Planning, Washington State University, Pullman, WA
Abstract
The global economic and ecological significance of coffee production has been increasing over the last several decades as coffee has become a major commodity in international agricultural trade. The trend of reducing shade cover to increase crop yield has raised important concerns regarding the long-term sustainability of sun grown coffee production. This study presents a geospatial approach, using geostatistical and multivariate analysis techniques to characterize soil properties in both sun and shade
grown coffee production systems. Two small, equal-sized plots near Masatepe, Nicaragua were chosen to intensively sample soil in both the wet and dry seasons. Collected data included soil pH, % water content,total C, N and S, and available nutrient concentrations (N, P, K, Ca, Mg and Al). Results indicate larger percentages of organic
C and N in the sun system, while pH was much more acidic as compared to the shade grown system. Univariate statistics showed a larger degree of variability,among most chemical predictors, between the wet and dry season in the sun grown system. Classification and regression tree (CART) analyses were easily able to distinguish the sun
grown system from the shade grown system while bivariate variogram models showed smaller correlation spaces in the shade system, representing more spatial heterogeneity in the upper 10 cm soil layer. Spatial patterns among soil properties appear to be buffered
by the presence of the canopy and root systems of shade trees as well as traditional organic management techniques. It is concluded that traditional shade grown management could provide a more spatially suitable environment for the long-term sustainability of coffee plantation soils.
