Merging Multiple-Partial-Depth Data Time Series Using Objective Empirical Orthogonal Function Fitting

Abstract

In this paper, a method for merging partial overlapping time series of ocean profiles into a single time series of profiles using empirical orthogonal function (EOF) decomposition with the objective analysis is presented. The method is used to handle internal waves passing two or more mooring locations from multiple directions, a situation where patterns of variability cannot be accounted for with a simple time lag. Data from one mooring are decomposed into linear combination of EOFs. Objective analysis using data from another mooring and these patterns is then used to build the necessary profile for merging the data, which is a linear combination of the EOFs. This method is applied to temperature data collected at a two vertical moorings in the 2006 New Jersey Shelf Shallow Water Experiment (SW06). Resulting profiles specify conditions for 35 days from sea surface to seafloor at a primary site and allow for reliable acoustic propagation modeling mode decomposition, and beamforming.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2010
Accession Number
ADA533647

Entities

People

  • Arthur E. Newhall
  • Patrick J. Haley
  • Pierre F. J. Lermusiaux
  • Timothy F Duda
  • Ying-Tsong Lin

Organizations

  • Woods Hole Oceanographic Institution

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Acoustic Navigation
  • Acoustic Propagation
  • Acoustics
  • Birds
  • Data Sets
  • Differential Equations
  • Eigenvalues
  • Equations
  • Internal Waves
  • Measurement
  • Navigation
  • New Jersey
  • Order Statistics
  • Ray Tracing
  • Shallow Water
  • Statistical Analysis
  • Statistics

Readers

  • Approximation Theory.
  • Computational Modeling and Simulation
  • Oceanography.