Buoy observation of high frequency ocean wave energy: accuracy, consistency, and concerns for predictive applications

Abstract

Observational data from buoys are of primary importance during the development, calibration, and evaluation of ocean wave models, and these data are also used to make real-time corrections to operational models via data assimilation. By association, systematic inaccuracies in any buoy data are equally important, and thus when two buoy types provide systematically inconsistent information, this is a concern for anyone using an ocean wave model. This report is concerned with the accuracy of the high frequency portion of the ocean wave spectrum commonly observable by buoys, roughly 0.2 to 0.6 Hz. We evaluate four buoy types (two moored, two drifting) using two quantitative measures. The first involves comparing each type with a co-located ocean wave model. The second method involves evaluation of high frequency energy level as a function of wind speed. Both evaluation methods suggest that the Datawell Waverider (DWR) buoys have a strong tendency to report higher energy levels than the other three buoy types. We evaluate high frequency energy level using three different metrics (mean square slope, energy in a band of high frequencies, and spectral density at a single, specific band, 0.4 Hz), and the conclusions are found to be insensitive to the parameter used.