A comparison of soil quality indexing methods for vegetable production systems in Northern California

Consultants, farm advisors, resource conservationists, and other land managers may benefit from decision tools that help identify the most sustainable management practices. Indices of soil quality (SQIs) can provide this service. Various methods were tested for choosing a minimum data set (MDS), transforming the indicators, and calculating indices using data from alternative vegetable production systems being evaluated near Davis, California. The MDS components were chosen using expert opinion (EO) or principal components analysis (PCA) as a data reduction technique. Multiple regressions of the MDS indicators (as independent variables) against indicators representing management goals (as iterative dependent variables) showed no significant differences between the EO and PCA selection techniques in their abilities to explain variability within each sustainable management goal. Linear and non-linear scoring techniques were also compared for MDS indicators. The non-linear scoring method was determined to be more representative of system function than the linear method. Finally, indicator scores were combined using either an additive index, a weighted additive index, or a decision support system. For almost all indexing combinations, the organic system received significantly higher SQI values than the low input or conventional treatments. The efficacy of the indices was tested by comparisons with individual indicators, variables representative of management goals, and another multivariate technique for decision making that used all available data rather than a subset (MDS). Comparison with the comprehensive multivariate technique showed results similar to all of the indexing combinations except the additive and weighted indices using the linearly scored, EO-selected MDS. This suggests that a small number of carefully chosen soil quality indicators, when used in a simple, non-linearly scored index, can adequately provide information needed for selection of best management practices. © 2002 Elsevier Science B.V. All rights reserved.