Wave Measurement Methodology and Validation from Wave Glider Unmanned Surface Vehicles

Abstract

The Wave Glider, an unmanned surface vehicle (USV), has been using an implementation of the Datawell MOSE sensor to measure and report wave height and spectral energy since 2011. In 2015, Liquid Robotics developed its own wave height sensor, including accompanying software in consultation with Dr. Jim Thomson, Principal Oceanographer at University of Washington, Applied Physics Lab (APL). The solution uses a GPS+AHRS sensor (LORD MicroStrain® GX3-35) to provide wave height and spectral reports. In 2017, the solution was updated to use the newer MicroStrain GX4-45 sensor. This paper will explain the implementation, use, and validation of the Liquid Robotics Wave Glider GPSwaves sensor for measurements of surface gravity waves. For waves, the challenge is that the Wave Glider surface float is not truly surface following, at least not in terms of the pitch-roll-heave signals used by traditional buoys and because the vehicle uses waves for propulsion. The solution to these challenges is the use of GPS for measurements of horizontal motion as the raw wave data signal, instead of pitch-roll-heave, methods (based on Herbers et al. 2012) termed ‘GPS Waves’. More specifically, the Wave Glider GPSwaves system is based on GPS Doppler speed measurements of wave orbital motion, which are processed onboard the vehicle to deliver the bulk wave parameters of significant wave height, peak wave period, dominant direction, as well as the spectral parameters of scalar energy and directional moments. The validation process builds upon a body of previous scientific work done with previous and current versions of the Wave Glider (SV2 and SV3). This validation process uses datasets over longer durations from multiple locations, including Astoria Canyon, Oregon (USA) and the Southern Ocean. Wave reports are validated against the Datawell implementation on the Wave Glider. In addition, these data products are evaluated relative to moored Datawell Waverider buoys and other third-party systems in a series of trials totaling 111 days and covering the full range of open ocean conditions including wave heights that range from 1 to 8 m and wind speeds from 1 to 17 m/s. This paper will discuss the approach to data collection including wave measurement processing, data collection, data pre-processing, spectral processing, spectral wave energy, wave direction and other factors such as quality control and overall evaluation of wave data sets. Where and how data is collected and processed on the Wave Glider will also be explained to highlight the flexibility and capabilities that an unmanned surface vehicle like the Wave Glider can bring to wave measurements. Results show that measurements of wave directional spectra, along with bulk wave parameters, can be accurately measured from a Wave Glider. Bulk wave parameters agree within 5% error and 5% bias. Spectral comparisons show that the bias is isolated to a mid-range of frequencies, 0.1-0.2 Hz, which is likely a result of vehicle propulsion. The bias and error values are similar or less than the discrepancies between other commercially available wave measurement systems, and thus the Wave Glider is demonstrated as a functional platform for ocean wave measurements.