A Taxonomically Based, Ordinal Estimate of Soil Productivity for Landscape-Scale Analyses


Randall J. Schaetzl, Frank J. Krist, Jr. and Brad A. Miller


In this paper, we introduce, evaluate and apply a new, ordinally based, Soil Productivity Index (PI). The PI uses family-level Soil Taxonomy information, i.e., interpretations of features or properties, recognized in Soil Taxonomy, that tend to be associated with low or high soil productivity, to rank soils from 0 (least productive) to 19 (most productive). The index has wide application, generally at landscape scales. Unlike competing indexes, it does not require copious amounts of soil data, e.g., pH, organic matter, or CEC, in its derivation. GIS applications of the PI, in particular, have great potential. Results confirmed that, for 1000 sites in southern Michigan, the mean PI of cultivated sites is significantly higher (10.94) than for forested sites (7.77). We also compared the PI with published productivity values for Illinois soils. The positive statistical correlations that resulted confirmed that the PI is an effective measure of productivity for areas that do not have robust productivity data or a wealth of local soils knowledge, as does Illinois. Lastly, 2009 crop yield data for 11 Midwestern states were used to further evaluate the PI. In a GIS, we determined the soils and crops in particular fields, and thus were able to ascertain the mean PI value per soil, per crop, per county. Statewide summaries of these data produced statistical correlations among yields of specific crops and PI values that were all positive; many exceeded 0.60. For regionally extensive applications, the PI may be as useful and robust as other indexes that have much more exacting data requirements.