The Carson Basin underlies the Grand Banks, offshore Newfoundland, and is composed of several depocenters, the deepest of which holds over 7 km of Mesozoic and Cenozoic strata. To date only four exploration wells have been drilled. The basin lies to the southeast of the more intensely studied and developed Jeanne d'Arc Basin, which contains the Hibernia production platform and other developing oil-fields. A basement high separates the Carson Basin from the southern Jeanne d'Arc. The basins formed in response to the opening of the North Atlantic Ocean in a complex series of rifling events.

Submarine canyons and erosional scours have been recognized in the northern end of the Carson Basin at depths of 1100 m, and are buried and filled by the Banquereau Formation. The canyons were mapped using industry seismic reflection profiles. Two canyon complexes, informally named the Bonnition and St. George Canyons, incise an interpreted paleocontinental shelf-break. The upper reaches of the Bonnition Canyon carve a V-shaped erosional notch at least 6.5 km wide, with canyon walls dipping as much as 34.5 degrees. The Bonnition Canyon is over 39 km long, and trends roughly northwest-southeast. The St. George Canyon is over 30 km long and also trends northwest-southeast. Both canyons have deposited submarine fans basinward over a wide area, with a maximum thickness of approximately 900 m.

Synthetic seismograms created from well logs, along with biostratigraphic studies, indicate that the canyon incision correlates with a basinwide erosional unconformity that occurred in the Early Bocene. The Early Eocene Unconformity corresponds with a relative drop in sealevel on the shelf adjacent to the canyons, resulting in a change in marine environment from outer neritic to nearshore marine.

Early Eocene erosional channels and gullies of the Jeanne d'Arc Basin have previously been interpreted as submarine canyons. The Early Eocene erosion and deposition occur stratigraphically higher than the ridge separating the Jeanne d'Arc Basin from the Carson Basin, indicating possible interaction between the two basins. With the recognition of large submarine canyons and fans to the southeast in the Carson Basin, the Early Eocene erosional features of the Jeanne d'Arc Basin may be interpreted as subaerially exposed incised valleys. The northwest-southeast direction of transport for the East and West Cormorant Canyons is very similar to the Bonnition and St. George Canyon trend. A prograding clastic wedge at the outlet of the Cormorant Canyons was deposited on the high between the two basins. The prograding package may be the primary source of unstable material ultimately transported to the submarine fans of Carson Basin.

The Carson Basin is relatively underexplored when compared to other basins on the Grand Banks. The submarine fans of the Carson Basin are areally extensive, thick deposits which may act as both reservoir and trap for hydrocarbons present in the basin. Stratigraphic pinchouts and salt tectonism create favourable conditions for hydrocarbon plays, however the deep water environment and questionable source rock potential will continue to curtail exploration in the near filture.

Supervisors: J. Shimeld / M. Gibling

The Early Carboniferous (Tournaisian) Grantmire Formation belongs to the Horton Group and is 800 m thick, based on exposures and drill core in the northern part of the Sydney Basin onshore. The 503 m measured section of the Grantmire Formation in drillcore PE 83-1 is dominantly pebble conglomerate with interbeds of siltstone and minor beds of sandstone. The conglomera~e (facies 1) is light to medium red, polymictic, poorly sorted, and clast supported with subangular to subrounded clasts. Conglomerate beds reach 15 m thickness with a maximum recorded clast size of 22 cm. They are divided into three subfacies: interbedded pebble conglomerate/sandstone, pebble to cobble conglomerate, and small boulder conglomerate. The other facies are sandstone (facies 2), siltstone with multiple sandy layers (facies 3), coarse siltstone (facies 4), and fine siltstone (facies 5). Siltstones are medium reddish brown and in two facies have calcareous nodules with green reduction patches and/or envelopes suggesting paleosol or shallow groundwater origin. Macroscale patterns suggest coarsening upward sequences on the 10-50 m scale and a rare 100 m scale are the result of fan progradation as indicated by thickening upward trends and increasing clast size. Siltstone-rich intervals suggest distal fan or interfan conditions. Mesoscale (<Sm) coarsening upward sequences may represent small lobe or levee progradation whereas large-scale fining upward sequences (5-10 m) are channel fills.

The Grantmire Formation has been interpreted as the clastic fill of fault-bounded basins within the region of the Sydney Basin. Currently, the Grantmire Formation is the only mapped unit in the Sydney Basin Horton Group. The presence of black shales in the Horton Group is important for hydrocarbon potential regionally; they are not presently identified in the Sydney Basin. The main clast types are chert, sedimentary lithoclasts, quartzite, volcanic clasts, and granitic clasts. Chert is derived form an older sedimentary source than the siltstone and sandstone clasts. Volcanic clasts are dominantly rhyolite with minor basalt that could have young or reworked origins. Acidic plutons are the origin of granitic clasts and likely provide a significant proportion of sand-sized quartz, feldspar, and mica.

Grantmire paragenesis begins with deposition ofsand- and gravel-sized clasts with iron-rich clay. The clays were oxidized at the surface or in the shallow subsurface early mt he depositional history forming hematite grain rims. Calcite nodules with fine mosaic textures in siltstone, are linked to shallow groundwaters. A locally pervasive poikilotopic calcite cement was emplaced prior to significant burial. Calcite commonly partially replaces potassium feldspar grains, possibly around the same time interval or subsequently. Dissolution of some grains, clays and calcite cement postdates consolidation and has generated secondary porosity.

Porosity of sandstone and conglomerates averages 9.6% and ranges from 4.2 to 15.7% and permeability averages 2.26 md and ranges from 0.06 to 7.72 md. Reservoir quality ranging from poor to good is likely controlled by variable amount of detrital clay, authigenic minerals, carbonate cement, paleosol development, and irregular laminac of finer material.

Supervisor: M. Gibling

This thesis provides an analysis of the sedimento logical and morphological changes observed in Chezzetcook Inlet from 1766 to 1998 using air photograph composites and maps from 1766 and 1 854.This study explores for the environmental factors responsible for change, while providing semi-quantitative data on the scale of morphology and sedimentation occurring in Chezzetcook Inlet.Chezzetcook Inlet is an estuary located in Eastern Shore, Nova Scotia, Canada at 44034'N longitude and 63 050'W latitude and is approximately seven kilometers in length and four kilometers in width. Two major sedimentological and morphological changes appear to have taken place in Chezzetcook Inlet over the past 232 years: a dramatic episode between 1766 and 1854, and a well documented period between 1945 and 1997. Changes between 1766 and 1854 are associated with the growth of sand and gravel spits near the inlet entrance. Changes between 1945 and 1997 are associated with the extensions of spits and marsh colonization. The drastic growth of Red Island, lagoon closure and sediment infilling at Cape Entry, beach migration and lagoon formation at Story Head, and the extension of spits and marsh colonization at the Three Islands highlight the changes in this period. Construction in the area, storm activity and rising sea level probably accounts for the increased sedimentation and/or erosion that is causing morphological changes in Chezzetcook Inlet.

Supervisor: D. B. Scott

Elevated levels of 238Uranium (U) and its daughter products 226Radium (Ra) and 222Radon (Rn) in groundwater, surface water, and soils have the potential of posing health risks. In particular, high levels of radioactive elements in water or soil may correlate with high Rn in indoor air, which is considered by some to be carcinogenic. The purpose of this thesis was to determine whether U, Ra, and Rn data from 1978 to 1981 mining exploration activities in Nova Scotia could be used to predict the potential occurrence of elevated Rn in indoor air. To this effect, the writer compiled and entered almost 5000 data points onto maps at a scale of 1:50 000, and generated the accompanying database, from exploration assessment report maps filed at the Nova Scotia Department of Natural Resources. The data have been integrated into 12 basic GIS (U, Ra, and Rn) geochemistry layers. These data have been spatially analyzed against previously existing digital data, using ArcView sofiware. Preexisting layers used include: geographic features, geology, faults, and unconformities, all available at a scale of 1:500 000.

Data synthesized using GIS allowed examination of a variety of combinations of data layers. Analyses performed included the generation of graphs, tables, queries, "contained in" operations, and "distance to" equations, as well as the synthesis of map data. Results indicate that although the study area typically has relatively low U values, Rn and Ra values can be considered elevated in some locations. In particular, this study confirms that there are areas of Nova Scotia where the risk of elevated levels of Rn in indoor air could exist, which were not widely recognized until now. These areas include the Panuke Road area, underlain by Horton Group sedimentary rocks, and the Rawdon Hills area, underlain by the Halifax Group. The results also suggest that the scale of the available digital data is critical in determining the level to which the data can be analyzed, and a limiting factor in the analysis was the generalized scale (<1:50 000) of much of the preexisting layers. Despite this limitation, the advantages of the GIS approach in the exploitation and analysis, of otherwise relatively unmanageable mineral exploration data, for environmental purposes are obvious. The data from this thesis are being used to update existing maps of Rn and U potential for the Nova Scotia Department of Environment.

Supervisor: M. Zentilli

The comparison of data from two cores collected in Bedford Basin, adjacent to the Mill Cove Sewage Treatment Plant, in August 1998, to data from 1968, 1993 and 1996 allows the impact of pollution to be determined. Benthic foraminiferal distributions respond to changing environmental conditions allowing these changes to be recorded in the fossil record.

Species diversity and abundance decreased dramatically from 1968 to 1993, and continued to decrease up to 1996. Degradation of the benthic community is the result of increased organic matter pollution. Remediation of the environment is under way, as foraminiferal diversity increases with a decrease in organic matter. The abundance of Eggerella advena, species indicative of pollution, has decreased since 1993, confirming a reduction in organic matter. Near normal conditions, determined by the presence of calcareous species, were noted at the bottom of core 1 C, at 57-59cm depth. Increased pollution resulted in low oxygen conditions, determined by the abundance of organic linings in the top 12cm of sediment. Organic linings are remnants of calcareous tests, which have dissolved due to low oxygen and pH conditions in the sediments.

Lower organic matter concentrations in the 1998 samples and higher species diversity and abundance indicate that the benthic community is returning as the environmental conditions improve. These improvements follow a large outfall event in 1996 from the sewage treatment plant in Mill Cove.

Supervisor: D. B. Scott

Development and construction projects that require the disturbance and excavation of sulphide rich bedrock in Halifax and southern Nova Scotia produces large quantities of waste rock material. This material poses a high risk potential for acid rock drainage (ARD), a commonly known geoenvironmental issue in Nova Scotia. Acid drainage is associated primarily with the iron sulphides, pyrrhotite and pyrite, in addition to marcasite, sphalerite, galena, arsenopyrite, and chalcopyrite found in the Cambro/Ordivician Meguma Supergroup, which underlies 200000km2 of Nova Scotia's most populated and developed region. Variations of sulphide mineral type, mineral texture, the presence or absence of acidophillic bacteria (Thiobacillusferroxidans and Thio bacillus thiooxidans) and the availability of oxygen and water determine the reactivity and consequently the rate of generation of acidic solutions. The technique of subaqueous disposal is currently a potential method for waste management of excavated sulphidic waste rock of the Meguma Supergroup in Nova Scotia. This concept is based on the premise that acid generation is chemically suppressed in low oxygen conditions at depth in submerged sites. This method has yet been evaluated with regards to the sulphide bearing waste materials of the Meguma Supergroup and in particular, to disposal in marine environments.

This project investigates in a laboratory experiment, subaqueous disposal of sulphide bearing rock from the Meguma Supergroup in seawater compared to fresh water. Polished thin sections were prepared from a drill core sample near the Halifax International Airport. The main sulphide minerals of focus were pyrrhotite,marcasite, pyrite, and chalcopyrite. The experiment was conducted using distilled water, seawater and lakewater from local, natural environments. One thin section was submerged in SOOmL of each of the three water samples at a depth of 32 cm in graduated cylinders unexposed to atmospheric conditions at room temperature (~21 ^oC). One thin section was maintained in air as a control. Dissolved oxygen and pH measurements were obtained for the water samples at initial and final stages. Observations of surface features under reflected light were recorded and digitally imaged at 6, 15, and 40 days. Surface coatings, color, and topography of sulphide minerals were used as an indication of sulphide mineral reaction. Results indicate that in general, the rate of sulphide reactivity is higher in seawater compared to freshwater.

Supervisor: M. Zentilli

Diamonds and an extensive assemblage of associated minerals were discovered in podiform chromitites of the Luobusa ophiolite, southern Tibet, China. Thus far, 25 diamonds have been recovered from the heavy mineral separates of samples collected in 1996.

All the diamonds are colourless and transparent. Most of them are euhedral crystals, showing sharp-edged octahedral morphology. Others are broken fragments. Euhedral crystals are 150 x 150 µm to 400 x 400 µm in size, whereas broken fragments range from 200 x 250 µm to 900 x 1000 µm. Both Raman spectra and X-ray diffraction techniques have been used to confirm the identification of the diamonds. One diamond fragment contains 3 discrete silicate mineral inclusions, which have Raman spectra patterns most similar to serpentine. One of the inclusions was analysed with the electron microprobe, and has an unusual composition, high in MgO (30.7 wt%) and SiO₂ (64.1 wt%). Its formula best fits that of clinoenstatite. SEM images show that the inclusion possesses an octahedral morphology typical of that imposed on many syngenetic inclusions by contemporaneous growth of the surrounding diamond.

Minerals associated with the diamonds include chromite, forsterite, enstatite, Cr-diopside, PGE minerals, graphite, SiC (moissanite), gehlenite, Si-Fe and Cr-C alloys, zircon, sphene, rutile, apatite, corundum, sillimanite, plagioclase, K-feldspar, amphibole, biotite, phlogopite, chlorite, serpentine, sulphides, carbonites, celestite, uvarovite, almandine, wollastonite, quartz, and Fe-Ni and Au-Ag alloys. All the minerals were recovered from heavy mineral separates and confirmed by either electron microprobe or X-ray diffraction studies. Most of the chromites are rich in magnesian and chromium, with Cr#s [100 Cr/(Cr+Al)] ranging from 77 to 84 and Mg#s [100 Mg/(Mg + Fe)] between 62 and 76, suggesting crystallisation from a boninitic melt. Olivines have Fo contents ranging from 91 to 98%, with NiO contents varying from 0.3 to 1.35 wt%, correspondingly. Enstatites contain 88-95% En end-member, with Cr₂O₃ contents varying from 0.15 to 0.90 wt%. Chromium diopsides show uniform compositions (En 46-48%, Wo 48-50% and Cr₂O₃ 0.99-1.75 wt%). SiC crystals are 0.1 to 1.1 mm in size, and transparent (if not deeply coloured) with a strong brilliant adamantine luster. Colours range from colourless to grey-blue to pale green to yellow to yellow-blue to bluish-green to blue-black. Many SiC grains are colour-zoned with graduations between zones. The Cr-C alloy is steel gray and displays well-developed acicular form. It has a Cr/C ratio of 1:1. The Si-Fe grains large up to 1 mm in size are black, very shiny, and fractured. The Fe/Si ratio averages 2.76/7, very similar to ferrosilicite (Fe3Si7) inclusions in SiC from the Yakutia kimberlite. The hydrous minerals are Ti rich and distinct in composition from those of secondary origin in Oman chromitites.

The sampling and separation procedures were designed to minimize any possibility of natural or anthropogenic contamination of the samples. The well-developed crystal morphology of diamond, SiC (moissanite) and Cr-C alloys signifies that they crystallized from melts/fluids, and they are interpreted as original phases in the Luobusa chromitites. The well-preserved character of the diamonds indicates they have not undergone any resorption after formation. Based on the available data, 2 models are proposed for formation of the Luobusa diamonds: they may have formed metastably in a superasubduction zone environment where the chromitites formed, or they may be xenocrysts formed at greater depths and later incorporated into the chromitites.

Supervisor: P. Robinson

The Dacha Daban ophiolite, located in the middle part of the North Qilian Mountains, northwest China, is composed of a basal peridotite unit, a middle sequence of gabbro-diabase and an upper assemblage of pillow lavas. Two geochemical suites are recognized in the extrusive sequence. A boninitic suite is characterized by relatively high SiO₂ (49-56 wt.%), MgO (4.5-15.6 wt.%), Cr and Ni and low TiO₂ (0.23-0.48 wt.%), HREE and HFSE. The V/Zr ratio for these rocks averages 6.6 and the Ti/V ratio averages 63, whereas Mg#s [Mg/(Mg+Fe)] vary widely from 0.64-0.77. A tholeiitic suite has higher TiO₂ (0.92-1.90 wt.%) and HREE and lower SiO₂ (<51%) than the boninitic rocks. Both suites have similarly-shaped REE patterns with LREE depletion but the boninites have chondrite-normalized values less than seven whereas the tholeiites have values greater than 10.

Nd(t) values for the boninitic rocks range from +1.8 to +6.3 whereas those for the tholeiitic rocks range from +6.4 to +8.9 (calculated from t=500 m.y.). These values indicate that both suites were derived from a depleted mantle source. The variation in values of the boninitic rocks may be explained by the addition of a subduction component. The boninitic lavas are believed to have been derived from a strongly depleted mantle source possibly by 20-30% partial melting under hydrous conditions. Although enriched in LILE relative to MORB, these lavas show only slight enrichment in LREE suggesting that the subduction component was mainly MORB-derived fluid.

Based on the available geological and geochemical data, I suggest that the Dacha Daban ophiolite was formed in a supra-subduction zone environment.

This study is part of a larger project: "The study of the characteristics of the North Qilian ophiolites, evolution of orogenic belts, and deep geologic process" by Professor Qi Zhang of the Institute of Geology, Chinese Academy of Science, Beijing, China. The successful complete of this project has contributed to the study of the North Qilian belt, helped clarify the tectonic environment of the area and elucidated direction of ancient subduction in Northwest China.

Supervisor: P. Robinson

The physical disruption of sulphide-bearing rocks leads to the oxidation of the two most common iron-sulphide minerals, pyrite (FeS2) and pyrrhotite (Fe1-xS), and the generation of acid rock drainage (ARD). ARD, also called acid mine drainage (AMD), is typically associated with mining operations that create waste rock piles and tailings impoundments. However, it also occurs in any area that causes physical disruption of the bedrock, such as highway construction, quarry operations, and urban development or expansion. The resulting drainage from these areas generally has acidic pH values in the range of 2 to 4, and high contents of potentially harmful elements that are toxic to local ecosystems.

ARD chemistry, and the overall intensity and duration of the drainage, very much depends on local conditions and the mineralogical components of the bedrock. Acidic drainage from bedrock dominated by pyrite may be very different from bedrock dominated by pyrrhotite, since pyrrhotite reacts much more quickly than pyrite. In pyrrhotite-rich areas, this difference in reactivity rate could lead to toxic "pulses" of low pH waters released into surrounding waterways.

The area selected to test these hypotheses is southern Nova Scotia, which includes the site of one of the most serious cases of ARD in Nova Scotia, the Halifax International Airport. In this study, sulphide mineral textures, compositions, and associations were analysed in detail throughout a regional-scale area of several hundred square kilometres. Monoclinic pyrrhotite, with varying proportions of pyrite, are the predominant sulphide minerals. The location of pyrrhotite can be detected by magnetic susceptibility measurements made with a hand-held meter, field-scale magnetometer surveys, and regional-scale, airborne magnetic surveys.

Regional-scale stratigraphic, structural, and geophysical data that are presently available in digital form, were incorporated into a geographical information system (GIS), and used as evidence to predict areas that have a high potential of generating ARD. The potential or "favourability" maps generated through expert-driven boolean logic, index overlay, and fuzzy logic modelling, as well as data-driven, weights of evidence modelling proved very useful for outlining areas that may produce ARD in the future, if the bedrock is disrupted and exposed to surface oxidizing conditions. Due to the high cost of ARD treatment, and the limited success of presently available treatment technology, prediction and avoidance is the best option. In areas where avoidance is impossible, detailed mineralogical studies are necessary in order to plan for, and establish, the best approach to treatment and amelioration.

The conclusions of this study should be applicable in other areas of the world underlain by sulphidic-rich bedrock, including areas in the southern Appalachians (e.g., Great Smoky Group), Finland, and the Czech Republic, as well as other areas that produce "natural" ARD such as British Columbia.

Supervisor: M. Zentilli

The southern Andes plate boundary zone records a protracted history of bulk transpressional deformation during the Cenozoic. Transpression has been related to both oblique subduction and ridge collision. However, few structural and chronological studies of regional deformation are available to support one hypothesis or the other. The present thesis addresses along and across-strike variations in the nature and timing of plate-boundary deformation to better understand the Cenozoic tectonics of the southern Andes. A general objective was to gain insights into the nature of transpressional deformation at obliquely convergent plate margins. Five transects were mapped along the southern Andes, from 39^oS to 46^oS. The northernmost, Liquine transect (39^oS), documents ductile deformation of pre-Late Cretaceous age. Brittle deformation is represented by a regional, high angle, northeast-trending reverse fault that places greenschist facies mylonites against an undeformed Miocene granitoid. In contrast, Late Cenozoic brittle faulting of Cretaceous and Miocene plutons is well developed farther south at Reloncavi (41^oS), where contractional and strike-slip kinematics are documented. At Hornopiren (42^oS), Late Cenozoic ductile to brittle dextral strike-slip deformation along northeast striking shear zones was continuous from syntectonic pluton emplacement at 10 Ma, to low temperature, solid-state deformation at ca. 4.3 Ma. Brittle faults indicate that dextral strike-slip deformation remained active after 3 Ma. Puyuhuapi and Aysen transects (44-46^oS), document a remarkable increase in the contractional component of ductile and brittle deformation. At Puyuhuapi (44^oS), north-south trending, high-angle contractional ductile shear zones that developed from plutons, coexist with moderately dipping dextral-oblique shear zones in the wallrocks. In Aysen (45-46^o), top to the southeast, oblique thrusting predominates to the west of the Cenozoic magmatic arc, whereas dextral strike-slip shear zones develop within it. New ⁴⁰Ar-³⁹Ar data from mylonites and undeformed rocks from the five transects suggest that dextral strike-slip and contractional deformation occurred at nearly the same time but within different structural domains along and across the orogen. For instance, ⁴⁰Ar-³⁹Ar laser dating on highly strained synkinematic biotite from plutonic rocks with S-C fabrics at 42^oS documents dextral ductile shear at 4.3+0.3 Ma. Similar ages were obtained on both high strain pelitic schists with dextral strike-slip kinematics (4.4+0.3 Ma, laser on muscovite-biotite aggregates, Aysen transect, 45^oS) and on mylonitic plutonic rocks with contractional deformation (3.8+0.2 to 4.2+0.2 Ma, fine-grained, recrystallized biotite, Puyuhuapi transect). Oblique-slip, dextral reverse kinematics of uncertain age is documented at the Canal Costa shear zone (45^oS) and at the Queulat shear zone at 44^oS. Published dates for the undeformed protholiths suggest both shear zones are likely Late Miocene or Pliocene, coeval with contractional and strike-slip shear zones further north. Coeval strike-slip, oblique-slip and contractional deformation on ductile shear zones of the southern Andes suggests different degrees of along- and across-strike deformation partitioning of bulk transpressional deformation. The long-term dextral transpressional regime appears to be driven by oblique subduction. The short-term deformation is in turn controlled by ridge collision from 6 Ma to present day. This is indicated by most deformation ages and by a southward increase in the contractional component of deformation. Oblique-slip to contractional shear zones at both western and eastern margins of the Miocene belt of the Patagonian batholith define a large-scale flower structure by which deeper levels of the crust have been differentially exhumed since the Pliocene.

Supervisor: M. Zentilli

Fieldwork was carried out in the Makkovik Bay area of the Paleoproterozoic Aillik domain of the Makkovik Province, Labrador, to determine the field relationships of units in the study area, which aided collection of samples for geochemical analysis and argon thermochronology. The Aillik domain is the middle and least understood of the three domains of the Makkovik Province. The Kaipokok domain to the north of the Aillik domain is composed of mainly Archean crust and the Paleoproterozoic Cape Harrison domain to the southeast is composed mainly of juvenile crust. The Aillik domain in the Makkovik Bay area consists mainly of the ca. 1860 Ma Upper Aillik Group, Measles Point Granite, Kennedy Mountain Granite and a mafic dyke swarm.

The Upper Aillik Group consists of a felsic-dominated bimodal volcanic suite with minor mafic and sedimentary rocks. Sedimentary rocks are associated mainly with the mafic volcanic rocks and although their stratigraphy is not fully understood, the limited stratigraphic information indicates that the sediments were deposited in a rift basin during extrusion of mafic lavas and minor felsic lavas. The felsic volcanic rocks which occur above the rift sediments represent the largest volume of volcanic rocks. Closely associated with the felsic volcanic pile is the Measles Point Granite, a hypabyssal intrusion recognised during the present study. Intruding only the Upper Aillik Group and Measles Point Granite is a mafic dyke swarm. Following intrusion of the dykes the Upper Aillik group was thrust westwards onto the Kaipokok domain. The Kennedy Mountain Granite is not intruded by the mafic dykes and field relationships indicate it was emplaced after thrusting. A significant feature of all these rocks is that they were metamorphosed regionally to amphibolite facies, and have locally retrogressed to greenschist facies within and near the straightened zone, a zone of shearing which cuts through the study area and was formed syn- thrusting and was reactivated following the intrusion of the Kennedy Mountain Granite.

Major, trace and rare element geochemical analysis of samples of felsic and mafic volcanic rocks of the Upper Aillik Group, amphibolite dykes, Measles Point Granite and Kennedy Mountain Granite has yielded insight into the development of the Makkovik Province. The A-type geochemical characteristics of the felsic volcanic rocks and Measles Point Granite indicate that they formed in a within-plate tectonic setting. The chemistry of the mafic dykes supports this interpretation, but the mafic samples from the Upper Aillik Group display a mixed chemical signature ranging from MORB to volcanic arc. The geochemical signature of the Kennedy Mountain Granite is similar to that of the felsic and Measles Point Granite samples, although it is interpreted to be younger.

⁴⁰Ar/³⁹Ar thermochronology of amphibole and muscovite separates from samples collected across the study area has revealed two domain-wide thermal events following the intrusion of the Kennedy Mountain Granite. The amphibole separates dated by incremental heating record an age range of 1740-1720 Ma, correlative with province-wide ca. 1720 Ma granitic plutonism. The muscovite separates were dated by incremental heating and with a laserprobe, producing an age range of 1640-1660Ma which correlates with ca. 1650 Ma and ca. 1640 Ma intrusions in the Aillik and Cape Harrison domains. This age range lies within the error ranges of a ca. 1635 Ma amphibole separate from the straightened zone.

The combined field study and geochemical data are interpreted to indicate that the Aillik domain was formed by rifting of a continental margin. The mechanism for rifting is postulated to be slab-rollback of oceanic crust, which could explain the mixed chemical signature in the mafic volcanic samples. Following rifting, the Upper Aillik Group was thrust onto the Kaipokok domain (postulated to have occurred around 1820 Ma) and argon thermochronology indicates the Aillik domain was tectonically active for at least 150 Ma following thrusting. The nature of these later tectonic events is unknown.

Supervisor: N. Culshaw

Study of surface sediment samples collected in Bermuda subtropical environments reveals foraminifera assemblages from lagoons, reefs, caves, mangroves, and ponds that are sufficiently different to serve in paleoenvironmental reconstructions. These assemblages are based on species diversity, suborder percentages, characteristic species and associations, and sediment types. Some subenvironments can even be recognized, such as outer/lagoonal reefs, open/semi- protected/protected lagoons, mangrove swamps/fringing mangroves, non-tidal marine landlocked ponds/mangroves and other ponds. The main factors controlling foraminifera distribution in Bermuda are salinity, sediment type, pH, water energy, oxygen content, and light penetration. Presence of phytal substrate is also very important but was not investigated in this more geological study of sediment assemblages.

Bermuda recent sediment hosts a benthic foraminifera fauna as diverse as in other subtropical and tropical areas, and the general trends of foraminifera distribution and morphology are similar despite some differences.

In Bermuda mangrove swamps, foraminifera distribution shows a zonation relative to higher high water level, that can be used in accurate paleosea-level reconstructions. The present work includes the first detailed measured transects carried out in mangrove swamps for foraminiferal studies. Study of vibracore and Davis core samples permitted the construction of a late Holocene sea-level curve for Bermuda. This curve shows a lowstand between 2,000 y BP and 1,000 y BP, and a relative sea level (RSL) acceleration in rise since about 500 y BP that preceded the onset of modern global warming. The lowstand occurred also in Connecticut and South Florida, but not in other areas, and coincided with a cool period (Early Medieval Ice Advances) also detected elsewhere, and possibly also with a displacement of the Bermuda High over Bermuda. This curve illustrates the high variability of late Holocene sea level and climate.

Supervisor: D. B. Scott

An evaluation of flexural-slip folding in the Meguma Group was conducted in the hinge area of the Lawrencetown Anticline (Halifax area) and the Ovens Anticline (Ovens area). Flexural slip in these areas was accommodated on mainly bedding-parallel movement horizons consisting of simple slip surfaces and flexural-slip duplexes. Lateral and frontal ramp movement horizons are locally associated with bedding-parallel movement horizons, forming a linked system accommodating flexural-slip strain. Thrusting locally accompanied flexural-slip. Discordant shear fractures in the Ovens Anticline are kinematically related to flexural slip. Movement horizons are typically coated with slickenfibres and quartz veins commonly occupy flexural-slip structures in the Ovens Anticline, and fluid pressure was likely important during flexural slip. Movement lineations are roughly perpendicular to the fold hinge or systematically form an acute angle with the hinge in the direction of the fold plunge, suggesting flexural slip during non-cylindrical fold growth. Shear sense indicators include slickenfibres, flexural-slip duplex geometry, displaced conjugate veins and thrust geometry, and invariably show a reverse sense which changes systematically across the fold.

The average spacing of bedding-parallel movement horizons ranges from approximately 1-4 metres and spacing is inversely related to limb dip. The flexural-slip amount determined for a section of the Ovens Anticline from displaced discordant veins indicates that flexural-slip accounts for 4o-8o of limb dip. Slip amount is variable, with the largest slip occurring on flexural-slip duplexes and flexural-slip shear strain is uniform across the limb. Spacing and slip amount data suggest that increasing flexural-slip strain during incremental fold growth is accommodated by increased movement on early formed movement horizons and continuous formation of new movement horizons.

Bedding-cleavage relations, the distribution of minor folds and cleavage-parallel strain defined by pressure shadows on arsenopyrite suggest a synfolding origin for cleavage and indicate flexural slip was accompanied by significant layer-parallel shortening in fold hinges, including the flat tops of box folds. Flexural slip is a brittle, post metamorphic deformation which deforms the contact aureole of the ca. 370 Ma South Mountain Batholith. A post metamorphic age is supported by 40Ar/39Ar age data which constrain regional metamorphism to ca. 410-385 Ma but synfolding cleavage-parallel strain and the emplacement of (flexural-slip) quartz veins at ca. 375 Ma. A model of folding has been proposed which includes (i) initial box-fold development, (ii) synfolding cleavage development and layer-parallel shortening in the flat segments of box folds with flexural- flow on the limbs under metamorphic conditions and (iii) post metamorphic flexural-slip accompanied by additional cleavage-parallel strain, including layer-parallel shortening in fold hinges.

A synfolding, flexural-slip model is proposed for the auriferous quartz vein array at the Ovens, including synchronous emplacement of bedding-parallel and discordant veins. Most Meguma Gold Districts share general features with the Ovens which are consistent with a flexural- slip model, suggesting a flexural-slip model may be an appropriate and testable model for Meguma gold deposits.

Supervisor: N. Culshaw

A lengthy historical record of coastal evolution was combined with measurements made at weekly to bi-weekly intervals over approximately 550 days of coastal monitoring to investigate the evolution of the McNab’s Island area, a transgressive drumlin shoreline hosting gravel beaches.

Sub-annual to annual-scale records of bluff retreat, bluff erosion, foreshore erosion, till water content, well water level, sea level, waves, winds, precipitation, and air temperature demonstrate that storms are important in causing bluff and beach failure and retreat. Increased wind speed, wave height, water level, and precipitation accompanying storms interact with barrier beaches and clay- rich till bluffs and cause geographically variable coastal change over sub-annual time scales.

Historical charts, airphotos and records of sea level, winds, and waves indicate that rates of coastal change are spatially and temporally variable and that the interactions of rising sea level, storminess and sediment supply to barrier beaches have controlled the evolution of the study area over decadal time scales.

Rapid coastal change occurs only when sediment supply limitation, rapidly rising sea level, and increased storminess coincide, as between 1955 and 1964. Storminess and sea-level rise are both related to the North Atlantic Oscillation and affect sediment supply, giving rise to nonlinear and cyclic behaviour. Episodes of rapid beach and bluff retreat are preceded by long periods of stability and beach progradation during which offshore sediment reserves are depleted and the beach is morphodynamically conditioned to future failure due to accelerated sea-level rise and increased storminess.

Supervisor: D. J. W. Piper