Productivity Estimation of Pampean Soils By Inductive Approaches.

Soil productivity indices (PI) represent ratings of potential plant biomass production of soils. Inductive approaches determine productivity based on inferred effects of soil properties on crop yield. A multiplicative model was developed by FAO (1970) that was information demanding but was used in some countries and for some crops with good performance. This model was adapted for Argentinean conditions. More recently Kiniry and Pierce (1983) independently developed a simpler index based on the work of Neill. This latter method assumes that crop productivity is proportional to root growth and influenced by five soil variables. Several modified versions exist. The objective of this work was to evaluate the performance of both inductive approaches related to wheat yield in the Pampas. Soil productivity was rated by the FAO and this information was available in published soil surveys. The PI-FAO was available at soil series level and was aggregated up to the geographical unit level and regressed against average wheat yield (1967-2006). As for the Kiniry-Pierce method, we estimated the sufficiency of soil available water holding capacity, pH, bulk density, and weather. Aggregation up to the geographical unit level was also performed. We adapted this latter index by adding a sufficiency factor for organic matter as an additive factor. The FAO index varied between 0.33 and 1 for Pampean soils, with a low correlation with observed yield (R² = 0.433, P = 0.05). The Kiniry-Pierce index ranged from 0.33 and 0.76 and its modification obtained slightly higher values. Correlations of original and adapted index versions with observed yield data were low (R² = 0.223; R² = 0.253, P = 0.05). Concluding, our results showed that both tested inductive PI´s were not suitable tools for regional Pampean soil ratings related to wheat yield.