Estimation of total plant available phosphorus in representative soils from Mediterranean areas

The sustainable use of P in agriculture, which is a non-renewable and strategic resource, relies on soil P tests. Traditional soil P tests, however, may not accurately assess plant-available Pin soils ranging widely in properties. The main objectives of this work were: (i) the study of the accuracy of the widely used Olsen Pin assessing available P to plants, and (ii) testing other extraction methods to assess phytoavailable P, in a representative group of soils from Mediterranean areas. To this end, a P depletion experiment was conducted using 18 soils to estimate total available P to plants in soil. In addition to Olsen P, other P extractions were performed to check their accuracy in predicting total available P to plants, including anion exchange resins in Cl- and HCO3- forms. Poor predictions of total available P with Olsen P (R-2 = 0.53, P < 0.001) were improved when clay and carbonates were taken into account in a predictive model (R-2 = 0.8, P < 0.001). The best estimations of total available P were obtained with resins, that in HCO3- form (R-2 = 0.87, P < 0.001) performing better than that in Cl form (R-2 = 0.77, P < 0.001). Estimations with resins improved when potential organic P inputs to available P were taken into account. This reveals the role of organic P to the availability of this nutrient to plants. It can be concluded that single models based on Olsen P plus routinely determined soil properties, and single extractions with resins predict total available Pin studied soils with sufficient accuracy. (C) 2017 Elsevier B.V. All rights reserved.