Diapycnal mixing in the deep ocean is known to be much stronger in the vicinity of rough topography of mid-ocean ridges than over abyssal plains. In this study a microstructure probe attached to an autonomous underwater vehicle (AUV) was used to infer the spa- tial distribution of the dissipation rate of turbulent kinetic energy (ε) in the central valley of the Mid-Atlantic Ridge. This represents the first successful realization of a horizontal, AUV-based, deep-ocean microstructure survey. The study focused on a channel with unidi- rectional, partially supercritical sill overflow. The density was found to decrease along the channel following the mean flow of 3 to 8 cm/s. The magnitude of the dissipation rate was distributed asymmetrically relative to the position of the sill. Elevated dissipation rates were present in a segment 1 to 4 km downstream of the sill, reaching 1 · 10−7 W/kg. Flow speeds of more than 20 cm/s and elevated density finestructure were observed within this segment. The average along-channel flow was found to be strongly modulated by the semi-diurnal M2 tidal flow. Supercritical flow down the lee slope of the sill was observed during strong flow velocity conditions, and a hydraulic jump is expected to occur downstream of the sill dur- ing these phases. Consistently, upward displacement of isopycnals was observed in the area where the velocity distribution suggested the presence of a hydraulic jump. Indications for upstream propagating hydraulic jumps were found during phases of decreasing flow veloci- ties. Upstream propagating hydraulic jumps offer a possibility of inducing turbulent mixing closer to the sill or even upstream of it. The distributions of the flow, density and mix- ing rate provide a consistent picture of the fundamental physical mechanisms controlling the mixing in this deep ocean channel, i.e. tidally modulated, jet unidirectional sill over- flow with a hydraulic jump inducing turbulent mixing downstream. These results indicate deep-ocean mixing to depend heavily on the local bottom topography and flow conditions. Although one particular channel was studied, the fundamental physical mechanisms iden- tified in this study are expected to be applicable to other, similar channels. Furthermore, the results nicely illustrate that horizontally-profiling AUV-based observations may be an efficient tool to study deep-ocean turbulence over complex terrain where free-falling and lowered turbulence measurements are inefficient and time-consuming.