Includes bibliographical references / The ~135 Ma Koegel Fontein complex, 350 km north of Cape Town, has rocks with abnormally low δ 18 O values, as low as - 5.2 ‰. The rocks with the lowest δ 18 O values belong to an intrusive breccia and have a fine - grained, black matrix. The silica content of these breccia rocks ranges between 44 - 68 wt %, their whole - rock δ 18 O values vary between - 5.2 ‰ and +1.8 ‰ , and their water content is between 0.2 - 1.1 wt %. The major and trace element composition of the breccia rocks is consistent with them containing variable proportions of bentonite, alkali basalt, gneiss, and epidote - and xenolith - rich material. Mapping indicates that the initial intrusion was an alkali basalt dyke and a bostonite dyke that were then intruded by two breccia plugs that extended along the planes of weakness created by the pre - existing dykes, forming two sub - parallel breccia dykes. The water content of the breccia rocks is low (average <1.0 wt %) and is consistent with interaction with fluids at a relatively high temperature (> 300°C). These fluids interacted with the rocks from both dykes and are responsible for their 18 O - depletion. On the basis of the O - and H - isotope composition of the breccia rocks, the δ 18 O value calculated for the fluids in equilibrium with the breccia rock s is approximately - 10.8 ± 0.2 ‰. This low δ 18 O value of the meteoric fluids is too low for the low latitude and warm climate at ~135 Ma and indicates that 18 O - depletion at Koegel Fontein predates the intrusion of the complex in the Cretaceous. It is possible that low - δ 18 O fluids circulated through the country rock during the Pan African when crustal reworking enabled deep circulation of surface water. This period of crustal reworking coincides with a 'Snowball Earth' event and would have been accompanied by 18 O - depleted meteoric water. In the Cretaceous, it is assumed that a mantle plume heated the crust, causing dehydration before partial melting of the crust. The initial low - δ 18 O fluids are effectively metamorphic fluids in equilibrium with the rock. It i s proposed that the initial loss of these fluids became explosive. These fluids migrated along the pre - existing alkali basalt and bentonite dykes, incorporating fragments of alkali basalt, bostonite, and country rock gneiss. This xenolith - rich fluidised material was then emplaced rapidly into the crust

No description available.

Includes abstract. / Includes bibliographical references (leaves 65-71). / The aims of this study are to: (1) characterize different natural gas leakage features present throughout the basin, and (2) understand the relationship of natural gas leakage with structural and stratigraphic elements, and (3) quantify liquid/gas hydrocarbon generation, migration and seepage dynamics through the post-rift history of the basin.

Bibliography: leaf 78-83. / Analysis and interpretation was carried out on trace metal samples (Cu,Zn,Cd) collected in the Southern Ocean south of South Africa. The results tentatively show that there is a significant copper flux from interstitial waters of the Weddell - Enderby basins. There is little evidence that these trace metals follow the systematics of nutrients throughout the water column; advection appears to play an important role in determining the vertical distribution of these chemical components.

Geological 3D static modelling has become an integral tool during the appraisal and developmental stages of a hydrocarbon field lifecycle. The 3D model becomes the basis upon which reservoir heterogeneity and characterisation are understood, hydrocarbon volumetrics are calculated and field development plans are designed. Reservoir compartmentalisation and fault-seal analysis is also an industry topic which has drawn much interest. Having a 3D model allows for fault-seal analyses to be carried out and evaluated using the statistically distributed reservoir properties. This study incorporates the building of a 3D geo-cellular reservoir model with a fault seal analysis of the E-S field, which is located on the north flank of the Bredasdorp Basin. The reservoir model was built using geostatistical methods to populate the several reservoir parameters into the model to calculate a hydrocarbon volume. In addition, a fault-seal analysis was carried out in order to investigate the phenomenon of having an oil accumulation separated from a gas accumulation either side of a fault. The facies modelling was carried out using the object modelling technique, in order to produce a model which is geologically plausible. Most of the remaining reservoir parameters were modelled using a variogram except in the case of water saturation, which was modelled using a J function equation. The volumetrics were assigned per fault block. Using a recovery factor of 75% for gas and 11% for oil, the calculated total recoverable hydrocarbons were 12.6 Bscf and 1.3 MMbbl respectively. The fault-seal analysis showed that the faults separating two of the fault blocks are not completely sealing. All the calculated fault properties supported this view, with the Shale Gouge Ratio (SGR) and threshold pressure relationship indicating a high likelihood for leakage across parts the faults. Pressure data from Repeat Formation Tests (RFT) however, indicates that the hydrocarbon accumulations in both blocks are isolated from each other. This contradiction has informed the recommendation to drill a highly deviated or short horizontal well which will cross the fault and intersect both blocks, and to complete the well using a sliding sleeve, thus providing the flexibility needed in order to manage multi-phase flow.

The Colenso Fault Zone marks the boundary between the Tygerberg and Swartland Terranes in the Pan-African Saldania Orogenic Belt. The fault zone comprises several discrete shear discontinuities that in places cross-cut relatively undeformed granites of the Cape Granite Suite. It is NW-SE striking, ~150 km long, and ~7km wide, but poorly exposed except in coastal exposures at the NW end. The deformation sequence can be divided into 2 sequences: The oldest deformation started with the emplacement of the G1, G2 and then the G3 of the Cape Granite Suite, followed by formation of aplite veins and strike-slip faulting generating cataclasites. The initial strike-slip sense of movement along the fault was sinistral, followed by dextral strike-slip shearing and finally late stage jointing. The Colenso Fault Zone is host to 3 large zones of cataclasis that are in the order of up to a few hundred metres in exposed down-dip and along-strike lengths. The cataclasites are composed of quartz and plagioclase clasts, in a phyllosilicate matrix. Both the wide cataclasites (several metres) and small cataclasite zones (tens of centimeters) show a decrease in shear intensity away from the core of the fault zone. The cataclasites have the same bulk chemical composition as the surrounding granite.

Bibliography: leaves 79-83. / Water, soil and rock samples were taken from the Harold Porter National Botanical Gardens and the Kogelberg Biosphere Reserve in order to investigate interactions between the various nutrient pools in a pristine fynbos ecosystem. Physical variables (pH, EC, temperature, DO, alkalinity) were measured in the field, and water samples were collected and analysed for chemical composition. Chemical composition and mineralogy of soil and rock samples from the Peninsula, Cedarberg and Goudini formations were determined to assess the influence of parent geology on soil chemistry. The surface water chemistry in the study area was remarkably homogeneous, with only major ions Na, CI, Mg, Ca, Si and S04 exceeding concentrations of 1 ppm at any of the sample sites. Concentrations of other major ions AI, K, NH4, N03 and Fe ranged between 0.1 and 0.7 ppm. The minor and trace elements B, Sb, Sr, Zn and Mn were present in concentrations greater than 1 ppb. Compared to the seawater chloride ratio, Na, Mg, K and S04 were depleted in the streams relative to CI, showing that these elements are preferentially accumulated and are conserved within the fynbos ecosystem. In contrast, CalCI ratio is elevated in streams, indicating an external source of Ca other than the seawater. This source may be windblown dust. The Fe content of soils appears to be lower than that of the parent bedrock, especially in the case of the iron-rich Goudini Formation. This indicates an as yet undetermined pathway for iron to leave the system, which may be through deeper anoxic groundwaters in the area. Only one site from the Oudebosch River headwaters showed significant deviation from other sites in several of the elemental trends, and also had no measurable DOC content, compared to the other sites which ranged from 10 to 16 mg/L DOC. These differences are probably the result of the different vegetation (afromontane forest as opposed to fynbos) and geological setting (clay rich, Cedarberg Formation soils). Soils contained higher concentrations of water-soluble ions than the streams, and organic carbon content ranges from 6 to 34 wt%. Organic 0 horizons and Melanic/Humic A horizons constituted the major soil types. Cation exchange capacities of the soil samples ranges between 31.3-93 mmolJkg, and about half of the soil samples were determined to be acid saturated and fairly leached of ions, probably due to the high rainfall. The soil composition and mineralogy correlate well with that of the underlying bedrock, and was dominated by quartz, with some clays, micas, feldspars and hematite. The study was similar to one in the Cedarberg, which has the same geological setting and equal fynbos species richness. This is the first integrated geochemical study to characterise the pristine fynbos biome of the Harold Porter National Botanical Gardens and the Leopards Gorge River Catchment, and will allow rapid assessment of any future changes.

Includes abstract. / Includes bibliographical references. / The False Bay dyke swarm is the southern-most set of Early Cretaceous dolerite dyke intrusions along the western margin of southern Africa associated with rifting of Gondwana and opening of the South Atlantic Ocean. Previous work highlighted the evolution of a single magma type from olivine-tholeiite basalt to ferro-tholeiite andesite. This thesis presents detailed field observations of each dyke in the False Bay - Cape Peninsula region and the focus of the study is on the finer details of the differentiation process to define combined assimilation and fractional crystallisation models.

Includes bibliography. / The Rietfontein kimberlite is an off-craton kimberlite pipe, located west of the Kaapvaal Craton at 26.75°, 20.04°E and hosts a range of xenocryst lithologies, including peridotite, eclogite and a suite of megacryst minerals. This study focuses on a suite of eclogite xenoliths, which were subject to a detailed petrographical and geochemical study, aimed at their characterisation and comparison to eclogites from on-craton and other off-craton localities. Garnet, clinopyroxene, accessory and secondary minerals were analysed for major element compositions using electron microprobe techniques and garnet and clinopyroxene trace element compositions were determined by Laser Ablation Inductively-Coupled-Plasma Mass Spectrometry (LA-ICP-MS) techniques. Oxygen isotopic compositions of five garnet samples were obtained using laser flourination techniques, followed by analysis by gas source mass spectrometry.

Bibliography: leaves 139-146. / The Swartruggens (156 Ma) and Star (128 Ma) kimberlites are two Group II, diamondiferous, hypabyssal kimerlite dyke swarms, situated in the Northern Province and the Free State respectively, South Africa. Representative samples from all dykes exposed in the mining operations, the Main and Changhouse Dykes, South Fissure and the barren Muil Dyke at Swartruggens, and the Wynandsfontein, East Star, Clewer, Byrnes and Barren dykes at Star, have been analysed for their major and trace element contents and Sr, Nd and Hf isotope compositions. Primary kimberlite magma chemistry is subjected to considerable modification due to the incorporation of both mantle and crustal material during ascent to the surface, crystal fractionation, and post-emplacement alteration by deuteric fluids. This study aims to constrain the effects of these processes, and thus to identify least-modified, close-to-primary, parental magma compositions, with the view to understanding the source region characteristics of, and the petrogenetic processes giving rise to, these kimberlites.