The complexity of temporal and spatial changes of soil characteristics under shifting cultivation in the tropics and the expense of comprehensive data collection motivates the development of a minimum data set (MDS) for characterizing soil productivity status and potential. We define a multi-criteria quantitative procedure for MDS selection: (i) the development of selectors based on objective selection criteria; (ii) the transformation of these selectors into combinable scores; and (iii) the combination of transformed selectors scores into a single rating for each soil variable. Selectors are (a) the norm of the vector representing a soil property in the space spanned by the standardized principal components that explain most of the variance; (b) the coefficient of determination of a one-way ANOVA of land use change on a property; (c) time of earliest response; (d) recovery time; (e) expense of sampling and measurement. These are justified heuristically. The method was applied to a set of MDS candidates: 13 soil variables collected within a chronosequence of shifting cultivation system in southern Cameroon. In this case the method selected five soil properties (pH in water, exchangeable calcium, available phosphorous, bulk density and organic carbon). These can be used individually or in combination to assess the effect of this practice on soil condition. The selected variables were easily interpretable in terms of their relation to land management practices and land use changes. The procedure was robust to soil orders and depths at which properties were measured. This method of choosing a MDS is expected to work well for studies of soil dynamics in other agro-ecosystems. © 2005 Elsevier B.V. All rights reserved.
Mendeley saves you time finding and organizing research
Choose a citation style from the tabs below