Nitrogen (N) losses from agricultural tile drainage systems are environmental and economic losses for producers. This field study quantified N losses from three reps of shallow (SD), deep/conventional (DD), and controlled drainage (CD) on farmland in Nova Scotia. Drainage systems were under corn and alfalfa–oats–clover production. Outflow water and gas samples were obtained and analyzed for nitrate and nitrous oxide. Nitrate-N loads were 5.0, 11.1, and 6.4 kg ha^-1 in 2015; 1.8, 6.7, and 2.8 kg ha^-1 in 2016; and 0.74, 1.8, and 1.6 kg ha^-1 in 2017 for SD, DD, and CD, respectively. Controlled drainage reduced NO₃^--N loading by 42.3%–58.2% when compared with the conventional/DD in 2 of 3 yr of study, whereas SD was found to reduce NO₃^--N loading by 54.9%–73.1% compared with DD in all years studied. Total NO₃^--N losses in this study were measured during the growing season (1 Apr. to 31 Oct.); the magnitude of NO₃^--N losses and treatment effects may vary if studied year-round. Nitrous oxide fluxes were variable and low in magnitude throughout the study. Cumulated N₂O losses were <1% of the applied N for all drainage types. Controlled drainage increased yields compared with SD and DD. The use of CD in the region could aid in reducing climate stresses, as well as overall NO₃^--N loads exiting drainage systems and may enhance crop yields compared with conventional systems. Future studies on dissolved N₂O losses from drainage water may provide important insight into whether dissolved N₂O losses exceed surface emissions.