Le Platreef est une formation noritique contenant des minéralisations en Éléments du Groupe du Platine. Elle se situe en bordure de la partie septentrionale du complexe du Bushveld. Afin de définir les spécificités de cette unité marginale, nous avons tout d'abord étudié des échantillons pouvant servir de point de comparaison : ils proviennent d'une zone centrale de l'intrusion, le forage de Bellevue. Les données en isotopie de l'oxygène indiquent des valeurs de δ18O proches de 6,5 à 7 ‰ et homogènes à travers la pile magmatique. Ces valeurs, plus lourdes de 1 à 1,5 ‰ que celles des autres produits issus du manteau du Kaapvaal, sont l'indice d'une contamination crustale à l'échelle de tout le complexe. L'isotopie de l'hydrogène et les teneurs en eau témoignent de cumulats peu ou pas altérés. <br /><br />Sur la base de critères pétrographiques et géochimiques, le Platreef apparaît beaucoup plus affecté par des phénomènes d'assimilation et de circulations de fluides. Il a été échantillonné sur 4 sites, en contact avec différentes roches encaissantes : granites Archéens, dolomies, quartzites et cornéennes. Les xénolithes calco-silicatés sont omniprésents dans ces portions du Platreef. Une contamination par des liquides de fusion partielle du granite est également mise en évidence. Les cumulats les plus affectés par l'altération se trouvent au contact des granites, qui constituent l'encaissant le plus fracturé. <br /><br />À l'échelle du complexe, le rapport fluide/roche des interactions est très faible : les circulations ont été confinées à des chenaux préférentiels. Les caractéristiques isotopiques et géochimiques des fluides ne correspondent pas à celles des eaux météoritiques. Ils étaient probablement originaires des formations sédimentaires proches et ont été expulsés lors de la perturbation thermique due à la mise en place des magmas du Bushveld. Ils contenaient du chlore et du CO2 leur permettant de jouer un rôle dans la distribution des PGE.

Graduation date: 2003 / Presentation date: 2002-06-03

Graduation date: 1996

This dissertation aims to improve our understanding of how dams and reservoirs impact freshwater systems worldwide. The following questions were addressed specifically: 1) what are the spatial patterns and magnitudes of flow regulation and channel fragmentation by dams globally; 2) how are dam impacts distributed biogeographically, and which are the biogeographic regions and taxa most threatened by dam impacts; and 3) how can climate change and dams be expected to interact in basins, and what management actions would mitigate adverse interactions? Results show that the majority of the world’s large river systems are fragmented and have their flow altered by dams. Exceptions to this tend to lie in regions inhospitable to hydropower development, such as northern tundra, or in the least economically active regions. The biogeographic distribution of dam impact is widespread, both at terrestrial and freshwater scales, representing significant threat to global biodiversity. Relatively species-poor tundra is the world’s only terrestrial ecoregion which remains predominantly unaffected by dams. Nearly half of the world’s freshwater ecoregions are internally fragmented by dams, and ecoregional distinctions may be artifically imposed by dams in many cases. Freshwater ecoregions with the highest counts of total and endemic species remain relatively unobstructed, representing significant conservation potential. Diadromy is one of the few fish traits indicative of vulnerability to dams for which data are sufficient for global scale analysis. Lampreys (Lampetra spp.), Eels (Anguilla spp.) and Shad (Alosa spp.) are examples of genera particularly vulnerable to dams because their distributions coincide with the most heavily fragmented freshwater ecoregions, and a large proportion of the coincident species for each genera are diadromous. Due to changes in discharge and water stress, the area of large river basins in need of management interventions to protect ecosystems or people will be much greater for basins impacted by dams than for basins with free-flowing rivers. Proactive measures that restore the natural capacity of rivers to buffer climate-change impacts are more desirable than reactive actions since they may also lead to environmental benefits such as higher water quality and restored fish populations – benefits which may later be unattainable.

Certain alteration minerals are used to identify the hydrothermally altered rocks. Potassic, phyllic (sericitic), propylitic, argillic and silicification are main alteration types observed in volcanic rocks. The role of remote sensing in alteration mapping is the differentiation of the minerals that are unique for In this study, Landsat TM 5 images are used. General alteration trend in the area is mapped by conventional methods of color composite, band rationing, principal component analysis and multi-linear regression analysis. Detailed mineral mapping carried on by using the United States Geological Survey (USGS) spectral library data. Spectral reflectances of selected minerals are analyzed according to the TM band intervals and appropriate band ratios are selected. TM bands 1,2,3,4,5 and 7 are used. Outputs of mineral maps are investigated for zonal distribution. Mineral maps that are obtained by the mineral separation method reveal that the youngest volcanic complex Hasandag is poorly altered. Instead, Ke&ccedil / iboyduran, Melendiz and Tepek&ouml / y volcanic complexes are found to be highly altered with the pattern of clay dominant in the center and increasing oxidation towards flanks. Also the alteration along the previously mapped buried faults is a proof that the method can provide information about the alteration source.

Tracking and quantifying hydrologic change in urbanizing watersheds is a complex problem which can vary spatially and temporally throughout the effective catchment area as change occurs. Hydromodification due to urbanization usually results in a larger peak event stream discharge, a change in typical event volume, a reduced lag time between rainfall and stream discharge events, and a more complex falling hydrograph. Recently extracted Environment Canada data have allowed the creation of a high resolution instantaneous stream flow dataset dating to the late 1960s for many Ontario gauge stations. Hydrometric data were obtained for fifteen urban and semi-urban catchments within Southern Ontario ranging in size from ~50km² to 300 km² with urbanized land use assemblages varying from <5% to 80%. Utilizing automated methods, each individual runoff event from the hydrographic record was identified and characterized. Temporal changes to urban land area, land use, and road length were quantified for each watershed from aerial photography spanning the period of record at approximately 8 year intervals allowing identified trends in event hydrograph parameters to be correlated quantitatively with the alteration of the catchment over time. <br> Increasing trends in event peak discharge were identified in all but one study catchment. Event volume was found to be consistently increasing in most of the urban watershed, while trends in event duration were observed but with no clear increasing or decreasing trend. The lack of consistent trends in the timing and distribution of flow during runoff events suggest that build-out, drainage network design, and stormwater management systems play differing roles in the neighbouring urban catchments. Changes to flood recurrence intervals through the period of urbanization were also investigated; peak magnitude of high frequency events is affected to a greater extent than low frequency or flood events. The relative change in return frequency distribution is not consistent between catchments, also the degree of alteration can differ between various recurrence intervals at a gauge. Peak discharge of some return periods appeared to decrease with urban development suggesting that the increased detention brought with urban stormwater management systems have effectively offset the increased runoff due to additional impervious area and improved drainage efficiency. A consistent relationship defining the change in geomorphically significant return periods (i.e. channel forming flow) with urbanization was identified in neighbouring urban catchments.

Anthropogenic disruptions in the form of hydrological alterations, such as dam construction and the associated water diversions are a cause of much upheaval to local and regional ecosystems. Lake 626 within the Experimental Lakes Area of north-west Ontario, along with its downstream wetlands, 626A and 626B are one such system. Construction of a dam at the L626 inflow has completely restricted water flow, reducing and reshaping the watershed, increasing water retention time, and decreasing outflow into the wetlands. This study investigates the state of each wetland through physical and geochemical characterization during the first year following the diversion. Previous studies have found that hydrological diversions in wetlands can lower water table levels, altering soil chemistry and producing a shift in floral and faunal communities. Ultimate consequences involve significant loss of wetland area through conversion to upland habitat. This provides a model for climatic warming scenarios, wherein sustained drought conditions can produce the same result. Boreal wetlands are surprising fragile ecosystems that store massive quantities of carbon and are at risk of releasing it in such situations. One study showed that an extended summer drought in an otherwise average year with above average precipitation produced losses of 90 g C/m2 over the course of the year. Maintenance of reduced-flow in wetlands 626A and 626B is expected to convert the system into a carbon source and reduce overall wetland area. Radiocarbon dating has revealed that following deglaciation, both 626A and 626B basins were open water wetlands, depositing limnic peat for about 3200 and 1300 years respectively. Each site then transitioned into open sedge dominated fen – 626B to the present and 626A until about 2.5 ka BP when Sphagnum began to develop. Wetland 626B is decidedly an open shrub/sedge fen, supporting Myrica gale, Chamaedaphne calyculata and Carex rostrata / lasiocarpa communities. Wetland 626A is a bog/fen complex, sharing similar communities in the fen areas, but housing a large, centrally located bog of shrub species overlying Sphagnum hummocks. Tritium values in 626A were similar to cosmic background levels, indicating that recharge of basal pore water has not occurred in at least 60 years. Tritium in 626B was much higher, suggesting a substantial difference in hydrology or peat hydraulic conductivity between the basins. Measurement of DOC profiles showed high concentrations in near-surface water, reaching over 80 mg/L, and dropping to about 20 mg/L at maximum depths. An opposite trend was seen for DIC and CH4 profiles which increased concentration with depth (25 – 70 mg/L DIC; 75 – 700 μmol/L CH4). Isotopically however, 13C signatures from basal DIC were more positive while signatures from CH4 were typically more negative (-6 ‰ to +4 ‰ DIC; -57 ‰ to -73 ‰ CH4). Breakdown of DOC by LC-OCD showed high concentrations of humic substances and low molecular weight neutrals. The origin of humic substances in surface water became more pedogenic with increasing distance from the L626 outflow, indicating the influence of decaying wetland vegetation on the DOC of adjacent water. A comparison between contemporary and future characterization of boreal peatlands under drought-like conditions will provide a better understanding of the impacts suffered by wetlands during hydrological alterations. The high sensitivity of wetlands to changing hydrology should also provide a measure for gauging the effects of long term climate warming. This will assist in the development of environmental policies to better govern both the establishment of water diversions and the multitude of other practices leading to climate change.

Chlorite and corrensite are very common secondary minerals in very low-grade metamorphic or hydrothermally altered mafic rocks. Corrensite, although defined as a 1:1 regularly mixed-layered chlorite/smectite or chlorite/vermiculite, has been considered as a unique mineral phase and might thus be a potential index mineral for evaluating very low metamorphic grade. Many lenticular basaltic rocks, which are intercalated with unmetamorphosed to low-grade metamorphosed Tertiary sedimentary rocks, occur sporadically in the Western foothills and the Central Range in Taiwan. The lenticular basaltic rocks in the Western foothills (diagenetic zone) and the western flank of the Central Range (anchizone to epizone) were subjected to different degrees of hydrothermal alteration and/or metamorphism. However, only few occurrences of corrensite in these basaltic rocks were reported. In the present study, the occurrences and mixed-layering features of corrensite and coexisting chloritic minerals in the basaltic rocks were characterized by utilizing optical microscopy, X-ray diffraction, scanning electron microscopy with energy dispersive spectrometry, and transmission electron microscopy (TEM). The results showed that most of these basaltic rocks contain abundant chloritic minerals occurring as replacements of mafic minerals or interstitial glass, or as vein- or vesicle-filling materials. The chloritic minerals include smectite, corrensite, chlorite, mixed-layered chlorite/smectite, or mixed-layered chlorite/corrensite. The compositions of chlorite, corrensite, and smectite are distinctive in terms of their Si/Al ratios, interlayer cations, and total cation numbers. Corrensite, chlorite, and mixed-layered chlorite/corrensite are common secondary mineral assemblages in the anchizone basaltic rocks, whereas only discrete chlorite occur in the epizone. The basaltic rocks in the diagenetic zone contain different assemblages, such as smectite only, chlorite + smectite + mixed-layered chlorite/smectite + mixed-layered chlorite/corrensite, or chlorite + corrensite + mixed-layered chlorite/corrensite. These different assemblages of chloritic minerals and other secondary mineral assemblages might reflect different stages of hydrothermal alteration. Corrensite was positively identified with TEM lattice-fringe imaging in the diagenetic rocks collected from Nangang, Poneikeng, Shanzijiao, Nanshihjiao, Hsiungkong, Shihsiouping, Fusing, and Northern Shihmen Reservoir areas, and in the studied anchizone rocks from Junghua, Kaopo and Baolai areas. The disappearance of corrensite clearly marks the boundary between the anchizone and epizone basaltic rocks. Corrensite, as a chemically and structurally unique mineral phase, is therefore a potential index mineral in very low-grade metamorphic rocks.

Electricity construction is required for a country¡¦s fundamental economic development. It provides the needed power and plays an important role in modern society. In order to develop sufficient electricity, Taiwan government has launched a construction project in the whole country starting from the year of 2004. The thermal power station is the major source of electricity in Taiwan and will take long time to construct. Based on the guidelines of Executive Yuan¡¥s Public Construction Commission, the construction shall focus on material management system; coordinate, monitor and examine the project; review by routine and check on site; integrate technology and experts from the production field to raise the execution effect and process quality. Alteration of thermal power station consists of generating set and supportive equipment. The construction methodology will determine if the construction can or not be completed on time. With reference by managerial and operational scopes of some enterprises, this thesis sets up Architecture-Oriented Thermal Power Station Alteration Affairs Management Model, abbreviated as AOTPSAAMM, to describe the whole picture for the thermal power station alteration project and further to achieve the aims on controlling schedule and raising knowledge management. AOTPSAAMM contains the architecture of thermal power station alteration construction. It not only describes the coalescence of organizational structure and organizational behavior in enterprise value chains, but also integrates with technology application systems. With duplicating and expanding capabilities, AOTPSAAMM will easily apply to other generating electricity system and create contemporary enterprise excellences.

Modern computer-aided design (CAD) systems have contributed significantly towards product development cycle time reduction and product quality improvement. To enhance the performance of CAD systems engineers must be able to create CAD models of conceptual designs quickly; at the same time CAD models must be easy to alter, so as to accommodate the rapid changes that the design undergoes through the lifecycle. However, there is no agreement in the way CAD models should be created to accomplish these goals. This work attempts to assess the effects of incentives on CAD model attributes during model creation and alteration; the effects of CAD model attributes on alteration are also investigated. Its aim is to derive prescriptions based on empirical evidence to improve CAD model creation and alteration efficiency. The CAD models under study are created by three sets of participants – students from a junior level CAD course, students from a senior level CAD course and experienced engineers involved in product development activities. The participants are incentivized to create and alter CAD models of designs they are provided with. The results indicate that upon removal of incentives, engineers (both students and professionals) tend to compromise on proper modeling procedures. Experts are quicker and adhere to commonly agreed correct modeling procedures during CAD model creation and alteration than students. The results also indicate that it is beneficial for alteration to construct a model with several simple features as opposed to a few complex features and that these features be fully constrained. Maintaining the traditional feature sequence improves the perception of the model. The retention and alteration of features (as opposed to deletion) is also shown to be positively correlated with model perception ratings.