Dynamical measurements in shallow seas from ocean colour and infrared imagery
Dynamical measurements in shallow seas from ocean colour and infrared imagery
This work applies remotely-sensed visible wavelength and infrared radiance data to measure sea surface velocities and to infer the degree of dispersal processes in the tidally-dominated regime of the shallow waters around the British Isles. To achieve this, an integrated scheme which produces sea surface velocities from atmospherically and geometrically corrected satellite digital data has been developed. Because the validity of the extracted velocities depends strongly upon the geometrically accuracy of the sequential images, an improved scheme for correcting geometrical distortions is also developed. Sea surface velocities are extracted from cross-correlation analysis between the sea surface reflectance (or radiance) signal from time sequential imagery series. The method, known as maximum cross-correlation (MCC), is applied to sequential images of visible wavelength reflectance from channel 3 (540-560 nm) of the Coastal Zone Colour Scanner (CZCS) and infrared radiance from the Advanced Very High Resolution Radiometer (AVHRR). In this way, coherent velocity distributions have been obtained for the United Kingdom shelf waters. Various deficiencies of the MCC method are identified and the corresponding strategies to improve the method are discussed. Because the elapsed time interval between successive CZCS images is approximately 2 M2 tidal cycles, the satellite-extracted velocity field can be interpreted as residual in areas of weak tidal currents. In the eastern English Channel, the remaining tidal signal, predicted by a tidally driven model, is removed from the computed velocities. The resulting snapshot of the residual velocities in the eastern English C annel agreed with previous observations. For the AVHRR sequence, the satellite-extracted velocities in the eastern English Channel are compard with depth-averaged currents derived from the tidally induced model, and the degree of agreement between both techniques is assessed. This provides one possible way of testing hydrodynamical models in shallow seas where vertical homogeneity is assumed. The potential use of colour imagery for extracting diffusion coefficients in the marine environment is also discussed. The mean effective longitudinal diffusivity coefficient in the eastern English Channel has been estimated through autocorrelation analysis of subtle sea surface colour patterns. Diffusion coefficients of the order of 104 to 105 cm2 s-1 are determined from the CZCS reflectance data, and the accuracy of the values is discussed. Visible wavelength radiance data from the high spatial resolution Airborne Thematic Mapper (ATM) data has also been utilized for dynamical measurements. In Southampton Water, strong blooms of Mesodinium rubrum during the summer provide a distinctive red colouration to the sea water which reveals streakiness and patchiness depending on the tidal stage. An algorithm relating cell concentration and spectral radiance has been produced with an accuracy of approximately 400 cell ml-1. The cell-calibrated ATM images have been examined and dynamical measurements are made by tracking near-surface features. Estimates of the longitudinal diffusion coefficient are also made from the cell-calibrated images. It is concluded that fruitful and reliable information can be gained about dynamical and dispersal processes in shallow seas and estuaries from imaging remote sensing devices.
University of Southampton
Garcia, Carlos Alberto Eiras
cadd1c0c-3909-423f-8af1-6c3ef702cfd5
1989
Garcia, Carlos Alberto Eiras
cadd1c0c-3909-423f-8af1-6c3ef702cfd5
Garcia, Carlos Alberto Eiras
(1989)
Dynamical measurements in shallow seas from ocean colour and infrared imagery.
University of Southampton, Doctoral Thesis.
Record type:
Thesis
(Doctoral)
Abstract
This work applies remotely-sensed visible wavelength and infrared radiance data to measure sea surface velocities and to infer the degree of dispersal processes in the tidally-dominated regime of the shallow waters around the British Isles. To achieve this, an integrated scheme which produces sea surface velocities from atmospherically and geometrically corrected satellite digital data has been developed. Because the validity of the extracted velocities depends strongly upon the geometrically accuracy of the sequential images, an improved scheme for correcting geometrical distortions is also developed. Sea surface velocities are extracted from cross-correlation analysis between the sea surface reflectance (or radiance) signal from time sequential imagery series. The method, known as maximum cross-correlation (MCC), is applied to sequential images of visible wavelength reflectance from channel 3 (540-560 nm) of the Coastal Zone Colour Scanner (CZCS) and infrared radiance from the Advanced Very High Resolution Radiometer (AVHRR). In this way, coherent velocity distributions have been obtained for the United Kingdom shelf waters. Various deficiencies of the MCC method are identified and the corresponding strategies to improve the method are discussed. Because the elapsed time interval between successive CZCS images is approximately 2 M2 tidal cycles, the satellite-extracted velocity field can be interpreted as residual in areas of weak tidal currents. In the eastern English Channel, the remaining tidal signal, predicted by a tidally driven model, is removed from the computed velocities. The resulting snapshot of the residual velocities in the eastern English C annel agreed with previous observations. For the AVHRR sequence, the satellite-extracted velocities in the eastern English Channel are compard with depth-averaged currents derived from the tidally induced model, and the degree of agreement between both techniques is assessed. This provides one possible way of testing hydrodynamical models in shallow seas where vertical homogeneity is assumed. The potential use of colour imagery for extracting diffusion coefficients in the marine environment is also discussed. The mean effective longitudinal diffusivity coefficient in the eastern English Channel has been estimated through autocorrelation analysis of subtle sea surface colour patterns. Diffusion coefficients of the order of 104 to 105 cm2 s-1 are determined from the CZCS reflectance data, and the accuracy of the values is discussed. Visible wavelength radiance data from the high spatial resolution Airborne Thematic Mapper (ATM) data has also been utilized for dynamical measurements. In Southampton Water, strong blooms of Mesodinium rubrum during the summer provide a distinctive red colouration to the sea water which reveals streakiness and patchiness depending on the tidal stage. An algorithm relating cell concentration and spectral radiance has been produced with an accuracy of approximately 400 cell ml-1. The cell-calibrated ATM images have been examined and dynamical measurements are made by tracking near-surface features. Estimates of the longitudinal diffusion coefficient are also made from the cell-calibrated images. It is concluded that fruitful and reliable information can be gained about dynamical and dispersal processes in shallow seas and estuaries from imaging remote sensing devices.
This record has no associated files available for download.
More information
Published date: 1989
Identifiers
Local EPrints ID: 461502
URI: http://eprints.soton.ac.uk/id/eprint/461502
PURE UUID: 75227fcd-22be-4b28-b342-2b9d0863b599
Catalogue record
Date deposited: 04 Jul 2022 18:48
Last modified: 23 Jul 2022 01:08
Export record
Contributors
Author:
Carlos Alberto Eiras Garcia
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics