David L. Burton ^1* Holly Wiltz ² John MacLeod ³

• ¹ Dalhousie University, Halifax, Canada
• ² University of Prince Edward Island, Charlottetown, Prince Edward Island, Canada
• ³ Charlottetown Research and Development Centre, Agriculture and Agri-Food Canada (AAFC), Charlottetown, Prince Edward Island, Canada

Nitrous oxide (N2O) is highly water soluble and can be readily transported in waters draining from agricultural fields. Relatively few studies have quantified N2O losses through agricultural tile drainage systems and none have compared the effect of different sources of applied nitrogen or their timing of application. While IPCC guidelines provide estimates of emissions from agricultural drainage water, the uncertainty in these estimates is relatively high. This research quantifies N2O loss in tile drainage water, as influenced by nitrogen source and timing. The study site was located at Agriculture Canada research station, Harrington, PE, Canada and consisted of 12 plots with subsurface drainage systems installed at approximately 80 cm, separated by buffer drains. Three swine manure treatments were considered with inorganic fertilizer (ammonium nitrate) as a control, each replicated three times. Manure treatments included fall and spring application of solid swine manure and spring application of liquid swine manure, all applied to supply 120 kg N ha -1 . The magnitude of N2O loss, as measured from samples collected at the tile outlets, demonstrated significant episodic emissions. Annual cumulative dissolved N2O emissions ranged from 0.1 to 5.69 kg N ha -1 (mean 0.83 kg N ha -1 ), while emissions from the soil surface were 0.09 to 1.16 kg N ha -1 (mean 0.33 kg N ha -1 ). N2O emissions in tile water were not significantly affected by the form of N applied, however tile drain length significantly impacted tile water N2O concentration. IPCC coefficients for N2O emissions from agricultural drainage water would underestimate actual N2O emissions at this site.