Near-bottom turbulence measurements in a partially mixed estuary : Turbulent energy balance, velocity structure, and along-channel momentum balance

A set of moored, bottom-mounted and shipboard measurements, obtained in a straight section of the lower Hudson estuary during late summer and early fall of 1995, determine velocity, density, and along-channel pressure gradient throughout the 15-m water column, as well as providing direct eddy-correlation estimates of Reynolds stress and indirect inertial-range estimates of dissipation within 3 m of the bottom. The analysis focuses on testing 1) a simplified turbulent kinetic energy equation, in which production balances dissipation; 2) the Prandtl-Karman law of the wall, which is a relationship between bottom stress and near-bottom velocity gradient; and 3) a simplified depth-integrated along-channel momentum balance involving local acceleration, pressure gradient, and bottom stress. Estimates of production and dissipation agree well throughout the entire record. The relationship between bottom stress and velocity gradient is consistent with the law of the wall within approximately 1 m of the seafloor during flooding tides, but it departs from the law of the wall at greater heights during flooding tides and at all resolved heights during ebbing tides. The local stratification is too small to explain this effect, and the likely explanation is suppression of the turbulent length scale by the finite thickness of the relatively well-mixed layer beneath the pycnocline. Direct covariance estimates of bottom stress close the approximate momentum balance well during some periods, but are often smaller than the sum of the other terms in the balance by a factor of roughly up to 2. The agreement between stress estimates and the sum of the other terms is best during periods of strongest top-to-bottom stratification and worst during periods of weak stratification, for reasons that are not understood.