Estuaries at the global scale are significant but highly uncertain sources of atmospheric nitrous oxide (N 2 O), which is an intense greenhouse gas and ozone depletion agent. As the largest estuary in the United States, the Chesapeake Bay is suggested to be a spatially and temporally variable source and sink of N 2 O. However, limited observations of N 2 O cycling preclude us from estimating and predicting its net N 2 O flux. To improve our mechanistic understanding of the processes that control the N 2 O flux at the point of production, we applied multiple 15 N tracers ( 15 NH þ 4 , 15 N-urea, 15 NO À 2 , and 15 NO À 3 ) to separately track N 2 O production from nitrification and denitrification under in situ and manipulated O 2 concentrations in the Chesapeake Bay. Nitrification was the major N 2 O production pathway in oxic waters (up to 7.5 nmol N 2 O L À1 d À1 ). In contrast, denitrification dominated N 2 O production from hypoxic/anoxic waters (up to 20 nmol N 2 O L À1 d À1 ). N 2 O production from urea was observed for the first time in estuarine waters. The contribution from urea was small, but interestingly, showed a depth pattern distinct from other N 2 O precursors. Experimentally lowering the O 2 concentration substantially enhanced N 2 O production. Therefore, the expansion of hypoxic and anoxic zones in the Chesapeake Bay under climate change as suggested by some climate models may favor the production of N 2 O, potentially providing positive feedback on warming. Overall, our study provides mechanistic constraints on N 2 O dynamics that could benefit modeling studies to better estimate the N 2 O flux in the Chesapeake Bay and other coastal environments.