Wavenumber Dependence of Surface Roughness Over a Variety of Wind Conditions

The observation of small-scale ocean surface features is of great importance to a number of air-sea interaction processes. These include (but are not limited to) gas transfer, remote sensing, and upper-ocean turbulence. It has been stated that the discrepancy between leading model spectra underlines the need for a greater body of small-scale wave measurements. One approach to bridging this gap is the quantification of energy present (and specific contribution to surface roughness) in specific wavenumber bands. For such quantification, it is necessary to be able to resolve surface features well into the capillary regime. Presented here is an application of shipboard polarimetry to the measurement of short-scale ocean waves. The polarimetric slope sensing (PSS) method, first proposed by Dr. Chris Zappa in 2008, provides a unique mode of observation for field wave measurements that is both sensitive enough to resolve capillary waves and rugged enough to survive the motions of the observation platform. The camera apparatus has no in-water component; as such, platform effects and other distorting influences may be minimized by orienting the device sufficiently far away from the ship. Field data used in this paper were collected aboard the R/V F.G. Walton Smith during the Grand LAgrangian Deployment (GLAD), an upper-ocean drifter experiment that took place during the Summer of 2012. To derive wavenumber spectra from the polarimetry requires image rectification and normalization considering boat motions. The PSS method allows for time-dependent evaluation of spatial slope spectra, enabling analysis of sea surface waves in nonstationary conditions. Selections of these spectra are first applied to the study of wavenumber-specific contributions to mean square slope; the latter portion of the work focuses on the response of the sea surface to changes in wind forcing. In all forcing conditions, waves in the gravity-capillary regime were found to provide the greatest share of overall mean square slope and rate of change of mean square slope. Waves in the pure capillary regime were found to contribute very little to the overall mean square slope; pure capillary waves were also observed to be the least sensitive to changes in mean square slope. Whereas surface roughness measurements in stationary conditions were found to follow the slick linear fit of Cox & Munk 1954, measurements made in non-stationary conditions were scattered between the clean and slick lines, indicating a possible connection between steadily increasing wind speed and changes in the spatial distribution of surfactant.