Three stations, at ∼80 m water depth on the Oregon shelf between 44.7°N and 43.9 N, were studied under hypoxic conditions in late spring and summer of 2009 to determine benthic oxygen consumption rates. Oxygen fluxes were derived from eddy correlation (EC) measurements made from an autonomous lander deployed for 11–15 h at a time. Average oxygen consumption rates ranged from 3.2 to 9.8 mmol m−2 d−1 and were highest at the southernmost station. Methods for separating eddy components and rotating coordinates were examined for effects on EC fluxes. It was found that oscillations at frequencies associated with surface and internal waves made significant contributions, but horizontal component biasing could be minimized by wave‐based rotation methods. Additional measurements included benthic boundary layer properties, and sediment permeability and profiles of sediment organic C, chlorophyll‐a, excess 210Pb and % fines. Comparative flux estimates were determined from benthic chamber measurements and microelectrode profiles at two of the stations. The chamber O2 consumption rates exceeded the EC fluxes by factors of 1.2–1.8, which may reflect enclosure effects, the different spatial and temporal scales of the measurements, and/or inhomogeneous benthic respiration rates. The magnitudes of the fluxes by either method, however, are low for shelf depths. Thus, for benthic O2 consumption to contribute to Oregon shelf hypoxia, bottom waters must be slowly renewed and minimally ventilated by along‐ or across‐shelf advection and turbulent mixing. Circulation studies indicate these conditions are favored by increased near‐bottom stratification during persistent summer upwelling‐ relaxation cycles. Key Points Novel eddy correlation measurements on the Oregon shelf Surface and internal waves make significant contributions to fluxes Low benthic oxygen consumption rates do not drive hypoxia