RESUMEN

The transport of phosphorus (P) from agricultural fields to watercourses is a worldwide concern. Development of tools to evaluate the environmental P risk is required to assist farmers and field technicians to recommend best management practices. The objective of this work was to develop and test a new approach to assess P loss vulnerability for agricultural conservationist systems in southern Brazil using easily obtainable soil attributes: cover crop residue (CCR), soil organic carbon (OC), visual evaluation of soil structure (VESS), soil test P (STP) and clay content. For that, the principal component analysis was applied, and simulated rainfall was conducted in ten agricultural fields with three replicates under Oxisol, with slopes from 7% to 12%. The database was separated into main and test groups. Significant principal components were correlated with runoff coefficient (RC), final infiltration rate (FIR), total sediment (TS), total P (TP), particulate P (PP), soluble P (SP) and bioavailable P (BAP). Component 1, interpreted as 'increased risk of PP transport', explained 49% of the variation with a strong dependence on CCR, OC, STP and VESS, and correlated with decreased FIR, and increased RC, TS, PP and TP. Component 2, interpreted as 'increased risk of SP transport', explained 21% of the variation with a strong dependence on OC, STP and clay content, and correlated with decreased TS, and increased SP and BAP. The P-index scores from components 1 and 2 were divided into four classes of P loss vulnerability (low, medium, high and very high) considering the first, median and third quartile. The Euclidean distances demonstrated a high similarity of the P-index scores obtained for the main and test groups. Therefore, the developed model can estimate the risk of P loss in agricultural fields with conservative management and moderate slope in conditions similar to the database.