Date of Award

Spring 1-1-2013

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Aerospace Engineering Sciences

First Advisor

William Emery

Second Advisor

Baylor Fox-Kemper

Third Advisor

Scott Palo

Fourth Advisor

Roland Romeiser

Fifth Advisor

Jeffrey P. Thayer

Abstract

Ocean surface currents play an important role in ocean-atmosphere interactions and global ocean circulation, and are also significant for fishing, ocean navigation, and search & rescue. Existing in-situ and remote sensing techniques for measuring ocean surface currents are limited by spatial and temporal data coverage, and thermal IR feature tracking methods are limited by clouds and weak thermal gradients. High-resolution spaceborne Synthetic Aperture Radar (SAR) offers repeatable cloud-penetrating measurements of the ocean surface. This research explores methods for ocean surface current measurement through satellite-based SAR.

The major part of this research is concerned with the development and application of a semi-automated algorithm to generate ocean surface currents at ≈1.9 km resolution from sequential spaceborne C-band SAR intensity images using the Maximum Cross-Correlation (MCC) method. The primary geographical area of study is the coastal California Current System (CCS), and nearly two years (2008-2009) of 30-min lag data from the Envisat ASAR and ERS-2 AMI SAR sensors is analyzed. The velocity wavenumber spectrum of the derived MCC SAR currents agrees with the k-2 power law as predicted by submesoscale resolution models, and also shows seasonal mesoscale variability. The derived MCC SAR currents are validated against High frequency (HF) radar currents, and the two show some agreement in vector direction, with MCC SAR vectors oriented slightly anti-clockwise relative to HF radar vectors. The unimodal mean-symmetric residual histograms indicate that errors between the two datasets are random, except for a mean positive bias of ≈11 cm/s in MCC SAR currents relative to HF radar currents. This magnitude difference occurs primarily in the along-shore component (≈6 cm/s) and is negligible in the cross-shore component. Doppler Centroid Cross-Track (XT) radial currents from Envisat Wide Swath Mode (WSM) scenes are compared with HF radar radial currents, and are seen to have much larger extreme values, which is attributed to the Doppler wind correction process. Ignoring the extreme values, errors between the two datasets appear to be random, with a near-zero mean bias, and are also linked with the Doppler radial estimation errors attributed to model wind corrections. Comparison of Doppler radials with MCC SAR radials for two ≈12-hour lag cases also shows promising results. Finally, experiments conducted with TerraSAR-X experimental Dual Receive Aperture (DRA) mode Along-Track Interferometry (ATI) datasets suggest possible solutions for the absolute phase calibration problem using interferometric phase over ocean only.

Share

COinS