Lucus Planum, extending for a radius of approximately 500km around 181 degrees E, 5 degrees S, is part of the Medusae Fossae Formation (MFF), a set of several discontinuous deposits of fine-grained, friable material straddling across the Martian highland-lowland boundary. The MFF has been variously hypothesized to consist of pyroclastic flows, pyroclastic airfall, paleopolar deposits, or atmospherically deposited icy dust driven by climate cycles. Mars Advanced Radar for Subsurface and Ionosphere Sounding (MARSIS), a low-frequency subsurface-sounding radar carried by European Space Agency's Mars Express, acquired 238 radar swaths across Lucus Planum, providing sufficient coverage for the study of its internal structure and dielectric properties. Subsurface reflections were found only in three areas, marked by a distinctive surface morphology, while the central part of Lucus Planum appears to be made of radar-attenuating material preventing the detection of basal echoes. The bulk dielectric properties of these areas were estimated and compared with those of volcanic rocks and ice-dust mixtures. Previous interpretations that east Lucus Planum and the deposits on the northwestern flanks of Apollinaris Patera consist of high-porosity pyroclastic material are strongly supported by the new results. The northwestern part of Lucus Planum is likely to be much less porous, although interpretations about the nature of the subsurface materials are not conclusive. The exact origin of the deposits cannot be constrained by radar data alone, but our results for east Lucus Planum are consistent with an overall pyroclastic origin, likely linked to Tharsis Hesperian and Amazonian activity. Plain Language Summary Lobe-shaped thick deposits, collectively known as the Medusae Fossae Formation, are found west of the Olympus Mons volcano on Mars. Visual observations of these smooth and relatively unremarkably looking materials have not definitively determined how they formed with hypotheses ranging from volcanic ash to remnants of a materials deposited at a previous location of the north pole, to accumulation of atmospheric dust. In this study we used the ground penetrating radar on board Mars Express to see through these deposits to derive information about Lucus Planum, the central lobe of the Medusae Fossae Formation. Through our analysis of the way the radar waves were reflected by subsurface layering, we concluded that the materials forming Lucus Planum are spatially variable: the east and west portions of the deposits are highly porous and probably composed of ashes and rocks from nearby volcanoes. In the north-west the deposits are much denser, but we could not unequivocally define their nature. Finally, our instrument could not detect signals from the central part of Lucus Planum, which suggests yet a different component in the deposits. This diversity points to a dynamic geological history in this unique region of Mars.

M.Ing. / As of 1991 new legislation demands that all existing and proposed mines have to submit and implement a rehabilitation plan. This plan consists of numerous phases and can cost a fair amount of money that almost none of the mines have budgeted for. During the rehabilitation process the improvement of the ground on, for example, discard dumps or slurry ponds, are not stable or adequate to enable the civil engineer to ensure stability for the new development. This can include a new sports field or low cost housing or just the peace of mind that the dump is stable and looks natural. This thesis will aim to introduce some of the numerous different ground improvement techniques that are available in South Africa and overseas.

Climate change stabilization requires an unprecedented effort to change our current approach to energy production and consumption. While rising energy prices are drawing increased attention to reducing energy demand, heightened concern about the environmental consequences of fuel choice requires that this demand be met at lower emission levels. In Canada, the realization of commitments to our GHG emission goals entails reducing residential energy use - a sector responsible for close to 20 percent of end-use energy consumption. This study focuses on the energy demand and emission levels of space and water heating, since these two components comprise 76 percent of residential energy demand. Ground source heat pumps (GSHPs) are a technology that provides heating at 25 to 30 percent of the energy consumed by even the most efficient conventional alternatives. GSHPs have been identified as the most energy-efficient, environmentally clean, and cost-effective space conditioning systems available. However, their drawbacks have been high capital costs, and uncertainty about whether the electric power used by heat pumps has higher system-wide emissions. This thesis delineates how adoption of GSHPs in the residential sector can help align Canada’s technology choices with commitments made to the Kyoto Protocol. The manuscripts delineate conditions under which GSHP systems achieve the largest net emission reductions relative natural gas, heating oil, and electric heat counterparts. Electricity generation methods and emissions embodied in inter-provincial and international electricity trade are shown to significantly affect the emission savings achievable through GSHP. The thesis quantifies how relative fuel prices influence annual operating savings that determine how rapidly the technology can achieve payback. This analysis reveals GSHPs to hold significant potential for substantial GHG reductions at a cost savings relative to conventional alternatives; the time horizons for payback are as short as nine years for average-sized homes, and significantly shorter for larger homes. / Science, Faculty of / Resources, Environment and Sustainability (IRES), Institute for / Graduate

This thesis presents ground control best practices in weak rock environments including an augmentation to the existing Span Design curve by adding 463 case histories of RMR76 values ranging from 25 to 60. A Neural Network analysis of this data has been added and compared to the existing Span Design data of 292 case histories. Ground support is almost always used in weak rock environments, though the type of support used can vary widely. The development of the weak rock augmented Span Design Curve has also been calibrated to four different support categories; Category A: Pattern Friction Sets, Category B: Pattern Friction Sets with Spot Bolting of Rebar, Category C: Pattern Friction Sets with Pattern Rebar Bolts and Category D: Cablebolting, Shotcrete, Spiling, Timber Sets or Underhand Cut and Fill. Category A is considered “Unsupported” with an average Factor of Safety less than 1.2. Categories B, C and D are considered “Supported” with average Factors of Safety greater than 1.2. All categories are compared the original Critical Span Design Curve presented by Lang (1994). However, only Category A can be accurately compared to the original Critical Span Design Curve as it is “Unsupported” as well. Category A yields good results, however, Categories B, C and D do not, but still demonstrate that spans can remain stable at lower RMR76 values. Design of underground man-entry type excavations in North America relies heavily upon empirical analysis. This design requires a higher Factor of Safety than other non-man entry type excavations. A comparison of the calculated ½ span failure Factor of Safety between all the categories is also presented. The contribution this research provides to the mining industry is the "Unsupported" Weak Rock Updated Span Design Curve and awareness pertaining to the potentially detrimental effects of using resin grounted rebar in weak rock masses and the false sense of security that the use of resin grouted rebar may instill. It is also shown that spans in the “Unstable” zone of the new “Unsupported” Weak Rock Updated Span Design Curve can possibly be stabilized if detailed engineering design is applied to obtain “Supported” status. / Applied Science, Faculty of / Mining Engineering, Keevil Institute of / Graduate

Ice formed by water vapour deposition has been identified in different terrestrial environments: 1) in the atmosphere; 2) at the ground’s surface; 3) in caves; 4) in seasonally frozen ground; and 5) in perennially frozen ground (permafrost). Thus far, ground ice formed by diffusion and deposition of vapour in soils (types 4 and 5) has rarely been studied in a natural setting and remains one of the most poorly described ice types on Earth. This thesis focuses on the dynamics of deposition and sublimation of atmospheric water vapour into permafrost and the isotopic signature (D/H and 18O/16O) of the emplaced ground ice under different experimental conditions. Ground ice was produced in sediments with different thermo-physical characteristics (glass beads, JSC Mars-1 simulant). After a two-month growth period, the higher porosity sediments (JSC) had more than 7x the gravimetric water content than the lower porosity soil. Ground ice profiles had a distinct concave downwards shape due to the decrease in saturation vapour pressure with depth. Results also indicate that vapour deposited ground ice has a distinct δD-δ18O composition that plots near regression slope value of 8. Pore water isotopes plot below the global meteoric water line (GMWL) when the source of moisture is directly on top of the sediments. If an air gap is introduced between the source of moisture and the sediments, the pore water isotopes shift above the GMWL due to re-sublimation at the ground surface. Overall, this thesis addressed some fundamental knowledge gaps required to better understand the growth and isotopic evolution of ground ice emplaced by vapour deposition.

The ongoing climate warming is expected to increase thermokarst activity and their impacts by inducing permafrost degradation and active layer deepening. A retrogressive thaw slump, which represents the most dynamic thermokarst landform, was investigated in the Richardson Mountains-Peel Plateau region. The exposed material at the thaw slump represents an opportunity to characterize the cryostratigraphy of the uppermost 5 m of permafrost. Analyses of the stratigraphy, sedimentology, isotope geochemistry and radiocarbon dating is presented. Six sites were also identified on an elevation-vegetation gradient to provide complementary data on thaw layer thickness. Summer air temperatures, vegetation cover type, mesoscale conditions modifying the snowpack, timing of the snow accumulation and winter air temperature inversions are identified as the main drivers of thaw layer thickness in the region. The physical and chemical parameters of the massive ground ice exposed at the thaw slump are characteristic of buried glacier ice that experienced water infiltration and partial refreezing. The layer between the massive ground ice units and the thaw layer in the thaw slump is identified as a relict thaw layer and represents the period of maximum active layer deepening. It dates to the Holocene thermal maximum, which represents a period of important thermokarst activity that resulted in widespread paleo-thaw unconformities across northwestern Canada. Association of the region’s thaw slump activity with paleoclimatic parameters provide indication that the combination of formerly glaciated continuous permafrost, hummocky rolling moraine terrain, stream-incised relief, and massive ground ice, coupled with major rainfall events, represents a set of condition that is favourable to thaw slump activity.

Garry Island, approximately 11 kilometres (7 miles) long and 0.8 to 3.2 kilometres (0.5 to 3.2 miles) wide, is located at about latitude 69° 28'N and longitude 135° 42'W in the southern part of the Beaufort Sea. The stratigraphy consists mainly of unconsolidated sands, silts, clays and stony clays which have been intensively deformed by the thrusting action of glacier-ice moving from the south. The deformed sediments are locally overlain by undisturbed sands and gravels containing marine fossils dated at >42,000 years. The absence of any evidence of glacial till on top of the sands suggests that Garry Island lay beyond the northwestern limits of the Laurentide ice sheet during the late-Wisconsin glaciation. Elevated strand-lines, which may be of great antiquity and occur at approximately 7.5 metre (25 feet) intervals to an altitude of almost 46 metres (150 feet), indicate the extent of Pleistocene fluctuations of sea level and the drowning of a pre-existing topography. The development of tundra polygons, in small flats behind sandspits or bars built across the drowned valleys in association with the former sea levels, has imparted a distinctive, stepped longitudinal profile to the stream courses. The tundra vegetation of Garry Island is classified into ten major habitats which are primarily related to drainage conditions and type of geomorphic activity. The island is underlain by permafrost and the thickness of the active layer is greatest, and ground temperatures in this layer are highest, beneath unvegetated surfaces and where the substrate is composed predominantly of mineral soil. Stratigraphic, geomorphic and historic evidence indicates considerable recession of the coastline in recent times. Current rates of retreat, reaching maxima of 10.5 metres (35 feet) per annum, are primarily related to the composition of the permafrost, being greatest in areas of fine-grained sediments, containing high ice contents, with a southerly exposure. Thermal erosion of the permafrost is the dominant process influencing cliff retreat and the primary role of wave action, on a short term basis, is in the removal of thawed debris from the base of the cliffs. Observations of three highly active mudslumps, created by the exposure of segregated ground ice, show that the rate of headwall recession is strongly correlated with ambient air temperatures. Maximum recession occurs where the ice content is high and the slumped debris is frequently removed from the base of the scarp. The cyclic development of a gully system on the ice face is described. The longevity of mudslump activity is prolonged where strong mudflows carry the thawed material away from the foot of the headwall, thus preventing the progressive burial of the scarp face. Mudflow velocities reveal a rhythmic pulsation related to periodic blocking of their channels. Mud levees, bordering the mudflows, result from the progressive bleeding of moisture from, and subsequent stagnation of, the mud rather than as residual features pushed aside by the advancing mudflow. Patterned ground on Garry Island is primarily restricted to non-sorted types. Angular intersections of thermal contraction cracks, representing the incipient stages of tundra polygons, exhibit a preferred tendency toward slightly-oriented, orthogonal systems. The initial micro-relief of earth hummocks is believed to originate through the accentuation of a miniature desiccation/frost crack pattern. Following the establishment of a vegetation cover, their subsequent growth involves further differential frost action and solifluction. Statistical tests show that the height, size and shape of earth hummocks are closely related to their position on the slope profile. / Arts, Faculty of / Geography, Department of / Graduate

Tai Chi Chuan, Tai Chi (TC) in short, is a popular form of Chinese martial arts which have been practised by millions of people and provides various health benefits, such as improving balance control, strengthening leg muscles, reducing fear of falling, enhancing flexibility and many more. The American Geriatrics Society and the British Geriatrics Society have recommended TC as a suitable exercise for preventing falls of older people. However, it is not clear which TC components are the most effective for balance improvement. In addition, it is also unknown what differences in biomechanics aspects are between non-TC participants and experienced TC practitioners. This thesis aims to provide new knowledge and understanding of the kinetics and kinematics characteristics of two most frequently presented TC foot movements: push-hand and Tai Chi gait (TCG), through a comparative experimental study on the ground reaction forces (GRFs) and plantar pressure distribution on both feet induced by TC and non-TC participants. Three hypotheses were proposed for facilitating this research. The characteristics of foot/leg movements in each of the simplified 24-form TC are investigated and each form is classified into one of four types. Two most frequently presented foot movements, push-hand and TCG foot movements, were identified, which are embedded in 18 of the 24 forms. This forms a basis for further biomechanics study. Ten male TC and ten male non-TC participants attended the experiments while performing push-hand, with and without an opponent, and TCG. The GRFs in the three perpendicular directions and plantar pressure distribution were measured simultaneously through using two force platforms and an insole system. A video recording system was also used to capture the movements of selected points on the subjects. The kinematics data were used to identify the critical positions that divide the leg stance phases of TCG.The characteristics of the foot forces during the two fundamental foot movements were examined and revealed. The differences and similarities on the GRFs and plantar pressure distribution generated by the two groups of participants were assessed and identified. The similarities show that the non-TC participants are able to achieve similar characteristics of foot forces as the TC participants do. The differences show that the TC participants have better control of their body movement, generating larger GRFs during push-hand and smaller GRFs during TCG in the two horizontal directions. These reflect the TC participants’ ability developed by practising TC for several years which cannot be achieved by the non-TC participants during the tests. It is also demonstrated that push-hand generates the lowest vertical foot forces among those generated by other possible human movements, such as TCG and walking.

The purpose of this study is to provide a qualitative, detailed description of individual and social behavior in a free-living population of thirteen-lined ground squirrels. Behavioral differences in relation to various periods of the annual cycle are also evaluated.

Steel Fibre Reinforced Concrete (SFRC) is defined as concrete containing randomly oriented discrete steel fibres. The main incentive of adding steel fibres to concrete is to control crack propagation and crack widening after the concrete matrix has cracked. Control of cracking automatically improves the mechanical properties of the composite material (SFRC). The most significant property of SFRC is its post-cracking strength that can impart the ability to absorb large amounts of energy before collapse. Ground slabs are structural applications that could benefit from these advantageous features of the SFRC. Many tests on SFRC ground slabs show that the material can offer distinct advantages compared to plain concrete. In concrete road pavements, SFRC is particularly suitable for increasing load-carrying capacity and fatigue resistance. Not surprisingly, recent years have witnessed acceleration in full-scale tests of SFRC and eventually acceptance of its use in concrete pavements. The use of SFRC in pavements has been slowed down by the absence of a reliable theoretical model to analyse and design these pavements. The analysis of ground slabs has traditionally been based on an elastic analysis assuming un-cracked concrete. Using such a method for SFRC would ignore the post-cracking contribution the SFRC can make to the flexural behaviour of the slab. Despite the growing trend of using methods of analysis based on yield-line theory, which can consider the post-cracking strength of SFRC, these methods were also found to underestimate the load-carrying capacity of SFRC ground slabs. To effectively account for the post-cracking strength of SFRC in the analysis of such slabs requires a method such as the finite element method. In the present work, non-linear methods are used to model the behaviour of SFRC ground slabs subjected to mechanical load. An analytical method is used to determine a tensile stress-strain response for SFRC. In this method, the post-cracking strength of SFRC is taken into account and hence the material model is sensitive to the element size used. The calculated stress-strain response is utilised in finite element analysis of SFRC beams and ground slabs. A smeared crack approach is used to simulate the behaviour of concrete cracking. The analytical method used to determine the tensile stress-strain response, as well as the finite element model, are evaluated using results from experiments on SFRC beams and ground slabs. The analytical results are found to compare well with the observations. The non-linear methods are further used to study the effect of the material model parameters as well as the support stiffness on load-displacement behaviour of SFRC ground slabs. The developed finite element model is shown to be more efficient compared to methods based on the yield-line theory. This is because it produces the load-displacement behaviour of the SFRC ground slab up to a reasonable limit and it provides the tensile stresses as well as the extent of cracking of the slab at every point on the load-displacement response. Using the developed finite element model will allow for considerable material saving since smaller slab thickness can be calculated compared to analytical models currently in use. / Thesis (PhD(Transportation Engineering))--University of Pretoria, 2008. / Civil Engineering / PhD / unrestricted