Denitrification, the microbially-mediated process by which nitrate is converted to nitrogenous gases, may be an important sink of anthropogenic nitrogen (N) in eastern US watersheds. Actual rates of denitrification, however, have been difficult to quantify and remain one of the critical unresolved N processes at the landscape scale. This project focuses on improving our understanding of the spatiotemporal distribution of denitrification in an agricultural landscape, in hopes of improving our strategies for controlling N loading to streams and rivers. During the period from 2008 to 2010, we quantified denitrification using the push-pull technique (Addy et al. 2002) at Cornell's Harford Teaching & Research farm, at a range of locations from upslope to riparian. We also used the detailed information available on N inputs and outputs at the farm to construct an N balance. Preliminary results are supporting our hypothesis that most denitrification occurs over a relatively small portion of the landscape and correlates with those areas prone to saturating. This has important management implications since agricultural lands are often intentionally drained, removing these biogeochemical hotspots.