Abstract

Phosphorus losses from agriculture may enhance eutrophication of fresh water bodies. This thesis focuses on preferential flow as a phosphorus transport pathway. Both lysimeter and field plot observations were conducted to evaluate the significance of preferential flow for P losses and to test management practices to reduce P losses. A decision support system was also developed to identify critical source areas, to diagnose probable causes of P losses and to prescribe appropriate site-specific best management practices. Preferential flow pathways are an important transport mechanism for P displacement from topsoil to drain tiles. Substantial drainage losses (on average 4.0 kg P ha-') of surface-applied P fertilizer, as measured in one of the studies included in this thesis, indicate a very effective vertical transport in structured clay soils. High sorption potential of the subsoil is bypassed because only a small part of the total pore volume is active in watedsolute displacement. Incorporation of P fertilizer and an increased contact between applied fertilizer and sorbing soil particles may significantly decrease P losses via preferential flow. Ponded flow conditions enhance P losses not only by enhancing preferential flow, but also by prolonging loading of P into water laying temporarily on the surface or subsurface. During four years of observations, nine high discharge episodes were responsible for, more than a half of the total loads of suspended soil material and P losses from clay plots, although the water volume discharged during these episodes reached, only one fourth of the total discharge. Different parts of the watershed do not contribute equally to total P loads. Identification of high-risk areas, site-specific diagnosis and abatement efforts should reduce the losses more efficiently. For instance, high-risk areas occupied no more than 5-1 0% of the total watershed area, according to decision support system developed in the study included in this thesis. Research at different scales is necessary for better understanding of the P problem, as well as for an easier practical application of research results. Shifting the scales tests the relevancy of the research and helps the identification of the gaps in our knowledge and understanding of the proble