Abstract
dc:description.abstractHigh Frequency radar systems provide a unique opportunity to measure evolving littoral oceanic dynamics at high temporal and spatial resolution. Backscattered electromagnetic signals from ocean waves are modulated by Bragg resonant scattering. A perturbation analysis yields an expression for the spectral content of radar signals which can be exploited to provide estimates of oceanographic parameters: the radial component of surface current can be extracted from the frequency locations of the first order peaks; the ocean wave directional spectrum is related to the second order continuum via a non-linear integral equation. The periodogram, based on a Fourier decomposition of radax data, is the standard method used to derive frequency spectra. Limitations in this approach, caused by inhomogeneities in the underlying ocean field, are investigated. An instantaneous frequency technique is proposed in this thesis which mitigates the spectral distortion by demodulating the backscattered radax signals -a filtering procedure is developed which measures the temporally varying Bragg components. Alternative spectral techniques are analysed in order to validate the filter: an autoregressive paxa, metric modelling approach and an eigendecomposition method. The filter is evaluated, using radar and in situ data, which establishes its potential for ocean remote sensing. Significant improvements in the quantity and accuracy of wave measurements are demonstrated. Properties and constraints of the filter are derived using simulated data. Finally, the generic structure of the extracted instantaneous frequency signals is investigated and related to oceanographic processes.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Sheffield
- Year dc:date.issued
- 2006
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Middleditch, Andrew
Rights
- Language dc:language.dcterms_iso639_2
- eng
Identifiers
dc:identifier.*- Identifier
- uk.bl.ethos.434473