On the Decrease of the Oceanic Drag Coefficient in High Winds

Abstract

AbstractThe sheltering coefficient – prefixing Jeffreys' concept of the exponential wave growth rate at a gas‐liquid interface – is shown to be Reynolds number dependent from laboratory measurements of waves and Reynolds stresses. There are two turbulent flow regimes: wind speed range of 2.5 to 30 m/s where the drag coefficients increase with wind speed, and wind speed range of 30 to 50 m/s where sheltering/drag coefficients decrease/saturate with wind speed. By comparing model calculations of drag coefficients – using a fixed sheltering coefficient – with ocean observations over a wind speed range of 1 to 50 m/s a similar Reynolds number dependence of the oceanic sheltering coefficient is revealed. In consequence the drag coefficient is a function of Reynolds number and wave age, and not just wind speed as frequently assumed. The resulting decreasing drag coefficient above 30 m/s is shown to be critical in explaining the rapid intensification so prominent in the climatology of Atlantic hurricanes. The Reynolds number dependence of the sheltering coefficient, when employed in coupled models, should lead to significant improvements in the prediction of intensification and decay of tropical cyclones. A calculation of curvature at the wave crest suggests that at wind speeds above 56.15 m/s all waves––breaking or not––induce steady flow separation leading to a minimum in the drag coefficient. This is further evidence of the veracity of the observations of the oceanic drag coefficient at high winds.