Characterization of soil fertility using the Rasch model

Abstract

A quantification of the overall soil fertililty potential should integrate the main soil physical and chemical properties, with different units. The formulation of the Rasch model is proposed as an instrument to measure soil fertility potential, integrating 11 soil properties (clay, silt and sand content, organic matter -OM-, pH, total nitrogen -TN-, available phosphorus -AP- and potassium -AK-, cation exchange capacity -CEC-, and deep -ECd-and shallow -ECs- soil apparent electrical conductivity, 0-90 and 0-30 cm depth respectively) measured at 70 locations in a field. In the case study, the considered soil variables fit the model reasonably, having an important influence on soil fertility, except pH, probably due to its homogeneity in the field. Moreover, a ranking of all soil samples according to their fertility potential and the influence of each variable on soil fertility are provided, being ECd, ECs, and the textural fractions of soil the most influential properties on soil fertility and, on the other hand, AP and AK the less influential properties. Results are in accordance with a previous work in the same field considering only five soil properties, both ECd and ECs and texture, denoting the importance of these variables to estimate soil fertility potential. Rasch model resulted to be useful to rationally determine locations in a field where high soil fertility potential exists and establishing those soil samples or properties with anomalies. This information can be necessary to conduct site-specific treatments, leading to a more cost-effective and sustainable field management.