Bathymetry forms the basis for studies in marine geology, biology and oceanography and is
essential for the Extended Continental Shelf Claim (ECSC), a legal framework established by
the United Nations (UN) to encourage a nation’s governance and management of its marine
resources. This research provides the first digital, integrated, Geographical Information System
(GIS) based bathymetric dataset for KwaZulu-Natal that combines near-shore and deep-water
datasets for use in marine sciences.
A total of 32 datasets acquired using a range of techniques and instruments between 1911 and
2006 were considered. Twenty nine of these were near-shore datasets with data densities
varying from 6 to 57 406 points per km2. Of these, 15 were acquired by the Council for
Geoscience (CGS), 9 by the South African Navy and 5 by the African Coelacanth Ecosystem
Programme (ACEP). Two of the remaining 3 deep-water datasets were grids acquired digitally
for this work, while the third was a digitised contour dataset. The 2003 General Bathymetric
Chart of the Oceans (GEBCO) grid is based on digitised point and contour data with a point
every 1 852 m, while the 1997 Smith and Sandwell grid is based on predicted satellite altimetry
data with a point every 3 704. The third deep-water dataset was digitised from a northern Natal
Valley bathymetric contour map developed in 1978 and has data densities varying from 0.02 to
1 point per km2.
Datasets were prioritised in the following descending order of quality defined by the available
metadata: multi-beam echo-sounder-derived datasets, followed by single-beam
echo-sounder-derived datasets and lastly lead line datasets. The digitised northern Natal
Valley bathymetric contour dataset after Dingle et al. (1978) was considered authoritative for
the deep-water areas, while the 2-minute interval Smith and Sandwell satellite derived
bathymetry dataset was integrated south of 31o S where no other dataset coverage existed.
Availability of good metadata describing bathymetric dataset positioning and depth measuring
instruments were essential. Where good metadata did exist, interrogation, integration and
quality control were straightforward. However, where the year of acquisition and depth
measuring instrument type were the only available metadata, information about positioning and
depth measuring instruments were inferred. The digitised northern Natal Valley bathymetric
contour dataset offered the best deep-water coverage and was derived from heterogeneous
point datasets about which no metadata was available. Metadata for the Smith and Sandwell
satellite derived bathymetric dataset suggested limited ship track data control for the study
area, while it was known to contain noise caused by an unquantified, rough sea state.
The integration process was successful but noticeable artefacts were recognised. Concentric
contour artefacts were present where the digitised northern Natal Valley bathymetric contour
dataset and the South African Navy Admiralty Fair Chart 34 dataset were integrated. Regional
conjoined arc-like contour artefacts north of 31o S as well as bumpy seafloor textures south of
31o S in the deep water areas were also found. In addition, artefacts were discovered in one of
the multi-beam datasets, normally associated with good high-resolution data coverage.
Intuitive, user-friendly, Geographical Information System (GIS) software and mapping software
were used to aid visual interrogation of the final contour dataset and the contour editing
capabilities in ESRI ® ArcGIS ® were used to edit concentric contour and conjoined arc-like
contour artefacts north of 31o S. GIS software was further used as a visual filter to remove the
regional bumpy seafloor texture south of 31o S, caused by noise in the satellite altimetry
dataset. An edited point dataset component south of 31o S was re-interpolated and the
resultant grid re-mosaiced with the original final grid north of 31o S, yielding an improved final
contour dataset.
The 1:3 000 000 scale final contour dataset resolved regional features such as the Thukela
Cone, the Thukela and 29o 25’ Canyons along with a broad un-named valley, termed here as
the Maputaland Valley, which drains the Maputaland Canyons. Near-shore areas of the
continental shelf were also resolved at higher scales of up to 1:45 000. Obvious data gaps
emerged with five areas prioritised for the acquisition of new digital data as part of a systematic
mapping programme to improve the dataset.
Powerful, cost-effective computer hardware and cost-effective, intuitive, user-friendly computer
software driven by ongoing technological advances made this work possible. These
technology advances continue to improve bathymetric data acquisition, positioning and
processing methods as well as improving data interpolation and map development.
The usefulness of this digital, integrated, marine GIS contour dataset has been demonstrated
by the interest of KwaZulu-Natal based organisations such as the University of KwaZulu-Natal
(UKZN), the Oceanographic Research Institute (ORI), Ezemvelo KwaZulu-Natal Wildlife
(EKZNW) and Umgeni Water along with the Cape Town based Marine and Coastal
Management (MCM) and the Pretoria based Council for Scientific and Industrial Research
(CSIR). Establishing this dataset as a base map for a KwaZulu-Natal 3D marine cadastre to
add other GIS data must be encouraged to improve collaboration, promote research and
improve ocean governance in KwaZulu-Natal, after which this type of 3D marine cadastre
should be extended to include the whole of South Africa. / Thesis (M.Sc.)-University of KwaZulu-Natal, Westville, 2009.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/8294 |
| Date | January 2009 |
| Creators | Young, Paul Michael. |
| Contributors | Uken, Ronald., Ramsay, Peter John., Whitmore, Greg P. |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0029 seconds