Radar Measurements of Ocean Surface Waves using Proper
Abstract
Radar Measurement of Ocean Surface Waves using Proper Orthogonal Decomposition Erin Hackett Coastal Carolina University, ehackett@coastal.edu Current state-of-the-art techniques for measurement of ocean surface waves based on radar measurements use Fourier transforms (FFTs) to extract information about the ocean surface wave field from the measured signal. In addition, the methods often rely on filtering the FFT of radar backscatter or Doppler velocities using the linear dispersion relationship for surface gravity waves to remove artifacts associated with the measurement (e.g., “sea spikes”). Because natural wave fields are neither linear nor stationary, one might raise the question of whether FFTs are the optimal tool? Furthermore, when the end-goal of the wave retrieval is space- and time-resolved sea surface elevation maps (i.e., rather than a wave spectrum), the back-and-forth translations between Fourier and spatiotemporal domains can introduce artifacts and impose additional sensing requirements that would not exist if data were processed exclusively in the spatiotemporal domain. The proposed research examines the use of proper orthogonal decomposition (POD) to analyze and extract wave field information from radar backscatter and/or Doppler velocities. POD is generally used to make a low-order approximation of a high-dimensional process, and is an optimal linear tool to represent non-linear processes. The difficulty in implementing POD is that the basis functions do not inherently have a physical meaning, unlike the basis functions of FFTs. However, preliminary research performed under ONR ILIR funding has demonstrated encouraging results for making physical interpretations by linking the POD basis functions to wave field characteristics using simulated wave fields of varying complexity. The proposed research directly extends this work to the application of these methods to radar (sea clutter) data. The Environmental Sensing and Ship Motion Forecasting (ESMF) FNC program relies on measurement of phase-resolved wave fields using radar, and all of the currently implemented methods rely on FFTs. Thus, this research provides an alternative approach reducing programmatic risk. The research can also complement these conventional techniques because the method could provide a means to reduce noise prior to application of FFTs as well as decrease data storage requirements for the developed shipboard systems.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 08, 2016
- Source ID
- N000141512044
Entities
People
- Erin E Hackett
Organizations
- Coastal Carolina University
- Office of Naval Research
- United States Navy