Time-Lapse Acoustic Imaging of the Oceanic Energy Cascade

Active-source marine seismic reflection records have been used to construct acoustic images of oceanic thermohaline structure. In the water column, acoustic boundaries are produced by variations in temperature and, to a lesser extent, salinity. Acoustic waves generated by an array of airguns suspend...

Full description

Bibliographic Details
Main Author: Dickinson, Nicholas Alexander
Format: Doctoral or Postdoctoral Thesis
Language:English
Published: University of Cambridge 2019
Subjects:
Online Access:https://www.repository.cam.ac.uk/handle/1810/293721
https://doi.org/10.17863/CAM.40834
Description
Summary:Active-source marine seismic reflection records have been used to construct acoustic images of oceanic thermohaline structure. In the water column, acoustic boundaries are produced by variations in temperature and, to a lesser extent, salinity. Acoustic waves generated by an array of airguns suspended behind a vessel are reflected from these boundaries and recorded by long cables of hydrophones which are towed beneath the sea surface. Careful signal processing of these records yields acoustic images of thermohaline structure with both horizontal and vertical resolutions of $O(10)$~m. In this dissertation, a three-dimensional seismic survey from the Faroe-Shetland Channel and a two-dimensional seismic survey from the Gulf of Mexico are presented and analysed. In the Faroe-Shetland Channel, 54 seismic transects were acquired over 25 days in July-August 1997. These transects were recorded sequentially, with adjacent transects separated by $\sim 0.3$~km and $\sim 1$~day. The survey thus provides time-lapse sampling of local thermohaline structure within an area of $\sim 150$~km$^2$. Processed acoustic images are dominated by a prominent band of bright reflections at depth. Calibration using nearby hydrographic measurements shows that this band corresponds to a pycnocline which separates warm near-surface waters from cold deep waters. The depth of the pycnocline varies by $\sim 350$~m over $\sim 10$~days. Acoustic reflections within the pycnocline are tilted by up to 1\textdegree \ with respect to the horizontal, suggesting currents in geostrophic balance. Temporal variations in pycnocline depth and in reflection tilt are consistent with northeastward advection of a mesoscale anticyclonic vortex. Comparison with contemporaneous satellite measurements of sea-surface temperature and sea-surface height suggests that this vortex is caused by meandering of the Continental Slope Current. These results have significant implications for our understanding of how mesoscale activity in the Faroe-Shetland Channel affects flux of heat and salt to the Nordic Seas. At finer scales, the seismic images show clearly defined internal waves. The geometry of these waves is compared with theoretical descriptions of linear and nonlinear waves in a two-layer fluid. The form of the internal wave field and of localised turbulence is investigated by spectrally analysing the vertical displacements of automatically tracked reflections. Internal wave spectra differ markedly from the open-ocean Garrett-Munk model spectrum. Diapycnal diffusivities are estimated from clear turbulent spectral subranges. Temporal variations in the form of spectra are related to changes in shear and stratification associated with passage of the mesoscale vortex. In the Gulf of Mexico, a single seismic image shows clear reflections in the uppermost $\sim 1000$~m adjacent to and overlying the continental slope. Spectral analysis of these reflections shows that the local internal wave field is accurately described by the Garrett-Munk spectrum. Diapycnal diffusivities are estimated using a semi-empirical parametrisation of internal wave energy. Comparison with diffusivities estimated from nearby hydrographic records and with previous measurements indicates that this method is robust. The analyses presented in this dissertation suggest that automated signal processing and interpretation of existing seismic reflection records has the potential to strengthen our understanding of the dynamical links between oceanic mesoscale activity, internal waves and turbulence. This project was sponsored by the Natural Environment Research Council (NERC).