Phosphorus and sediment loading to surface waters from liquid swine manure application under different drainage and tillage practices

Phosphorus (P) and sediment can move from agricultural land to surface waters, deteriorating its quality. This study was undertaken to improve understanding of partitioning of P and sediment to surface water via overland runoff and underground drainage pathways, and identify control measures. Over two full years, and including important winter events, P and sediment load overland and through tile were quantified from micro-catchments with relevant drainage and management practices imposed. Crop nutrients were supplied by liquid swine manure, either injected under minimum tillage management or surface-applied and incorporated under conventional till. Winters were temporally important for loadings from both runoff and drainage tile, particularly during rain on snow. A single event of 50 mm rain on snow over 2 days contributed more than 80% of the Pdop (dissolved organic + particulate P) and sediment that moved overland, and contributed 28% of Pdop and 20% of the sediment that moved through drainage tile during that season. Loads of P and sediment in both overland runoff and tile drainage were greater in non-growing seasons (NGSs) than growing seasons (GSs). For example, loading overland averaged 0.14 kg ha−1 dissolved reactive phosphate (DRP) and 1551 kg ha−1 sediment in NGSs, and 0.04 kg ha−1 DRP and 42 kg ha−1 sediment in GSs (four catchments, two seasons, runoff DRP first GS only). Through drainage tile, DRP load averaged 0.08 kg ha−1 in NGSs and 0.01 kg ha−1 in GSs from one field, A, and 0.02 kg ha−1 in NGSs and 0.003 kg ha−1 in GSs from another field, B; Pdop load was 0.07 kg ha−1 in NGSs and 0.02 kg ha−1 in GSs, similar from both Fields A and B; and sediment load was 23 kg ha−1 in NGSs from Field A, 8 kg ha−1 in NGSs from Field B, and 2 kg ha−1 in GSs from both fields. It is therefore important to manage movement during NGSs, particularly when runoff occurs over frozen soil. Movement through drainage tile comprised 31, 24 and 16% of the overland + subsurface DRP, total P (Pt) and sediment loads, respectively. Presence or type (blind inlet or hickenbottom) of surface inlet had little impact on P and sediment loading. Artificial drainage reduced overland + subsurface load to surface water to one-third for Pt and one-tenth for sediment, and is therefore a suitable strategy for controlling both P and turbidity in surface water. Overland + subsurface DRP load was unchanged by artificial drainage. Preferential flow of liquid swine manure to drainage tile only occurred with injection, in the year the drains were installed, in one of two fields. Along with being infrequent, the incidental DRP load through tile drains comprised only 2% of the annual Pt load from the catchment. The associated minimum tillage system reduced overland Pt and sediment runoff load 3- and 6-fold, respectively, relative to conventional till with broadcast incorporated manure.