Les berges des grands fleuves soumis à des crues saisonnières et aux effets de la marée sont l’objet de glissements et d’éboulements causant des dégâts matériels et parfois des victimes humaines. Le présent travail consiste à étudier la stabilité des berges de rivière soumises à la marée en considérant l’influence des écoulements externe et interne de l’eau sur le massif du sol. On considère en particulier l’effet de la fluctuation du niveau d’eau à l’extérieur et à l’intérieur du massif sur la pression interstitielle dans le sol. Le modèle d’écoulement de Dupuit est adopté pour trouver la variation de la surface libre dans le massif. Une méthode des tranches est programmée pour calculer les coefficients de sécurité au glissement des berges soumises à la marée. Des études paramétriques sur la géométrique de la berge, sur les caractéristiques hydromécaniques du sol et sur celles de la marée sont ensuite présentées. L’érosion superficielle de la berge est simulée par différentes méthodes et couplée avec le calcul de stabilité. Enfin, des études des cas sur la baie du Mont-Saint-Michel et sur les rives du Mékong inférieur sont présentées.La comparaison entre les simulations et les mesures de surface libre pendant les marées nous permet de conclure quant à la pertinence du modèle d’écoulement de Dupuit dans le cas de milieux relativement homogènes et isotropes. Mais celui-ci s’avère moins précis pour des milieux plus complexes, tels que les milieux multicouches. La marée a un effet important sur les pressions interstitielles dans le sol, lesquelles subissent des phénomènes d’amortissement et de retard.Les facteurs de sécurité au glissement varient en fonction de la marée et ils sont minimaux pendant le reflux avant la basse mer à cause du retard de la surface libre dans le sol et de la décélération de la vitesse descendante de la marée avant la basse mer. Des abaques sont proposés donnant les domaines de sécurité en fonction de différents paramètres du sol, de la géométrie de la berge et de la marée. Ces abaques permettent également le calcul du coefficient de sécurité par interpolation linéaire.Quand la berge est submergée pendant un temps suffisamment long, la zone de rupture du massif se localise dans la partie supérieure de la berge et se produit à la pleine mer. Ce mode de rupture est observé aussi sur les sites étudiés (Mont-Saint-Michel). Au reflux, on met évidence un deuxième mode de rupture correspondant à des surfaces de glissement plus profondes liées à des pressions interstitielles résiduelles élevées.Dans la période de décrue (site de Kaoh Chorram), le coefficient de sécurité diminue avec la baisse du niveau de l’eau sous le sommet de la berge. A même niveau d’eau dans la rivière, le coefficient de sécurité à la décrue est inférieur à celui résultant d’un état d’équilibre. Des différences importantes s’observent aussi sur les surfaces de glissement entre les deux régimes. / The riverbanks, subjected to seasonal flooding and tidal variation, are submitted to landslides causing property damage and sometimes human victims. The presented work is conducted to analyze the riverbank stability subjected to the tide variation by considering the influence of external and internal water flows on the bank soil. More particularly, we consider the effect of ground water variation on the soil pore pressure. The Dupuit’s flow model is adopted to find the variation of the ground water table. A method of slices is developed to calculate the safety factor of riverbank slides as a function of tide. Parametric studies of the effect of bank geometry, hydromechanical soil properties and the characteristics of tide variation are then presented. Then the bank surface erosion is simulated by different methods and coupled with the calculation of landslide. Finally, case studies at Mont-Saint-Michel bay and on the lower Mekong riverbank are conducted.The comparison between simulations and field measurements of ground water table during the tide variation let us conclude that the Dupuit’s flow model can be adopted for homogeneous and isotropic soils. However, the mode is less accurate for more complex media, such as multilayered soils. The tide has a significant effect on the pore pressure in the soil, which creates damping and delayed phase phenomena with respect to the tide.Slide safety factors vary as a function of tide and they are minimal during the falling of water level before low tide due to the delay of the ground water table and the slowing down of the speed of the tide. Charts of safety factor have been proposed providing the security domains based on various soil parameters, bank geometry and tide variation. These charts also allow calculation of the safety factor by linear interpolation.When the bank is submerged for a time period long enough, the failure surface is located in the upper part of the bank and occurs at high tide. This failure mode is also observed in the field study (Mont-Saint-Michel). During the falling of tide from the bank surface, the second mode of failure is observed corresponding to deeper slip surfaces associated with high residual pore pressures.In the period of decline after flooding of the bank (Kaoh Chorram site), the bank safety factor decreases with the falling of water level below the top of the bank. At the same water level in the river, the safety factor during decreasing water level is smaller than that during the steady state. Significant different failure surfaces are observed between the two regimes.

Arctic soils store close to 14% of the global soil carbon. Most of arctic carbon is stored below ground in the permafrost. With climate warming the decomposition of the soil carbon could represent a significant positive feedback to global greenhouse warming. Recent evidence has shown that the temperature of the Arctic is already increasing, and this change is associated mostly with anthropogenic activities. Warmer soils will contribute to permafrost degradation and accelerate organic matter decay and thus increase the flux of carbon dioxide and methane into the atmosphere. Temperature and water availability are also important drivers of ecosystem performance, but effects can be complex and in opposition. Temperature and moisture changes can affect ecosystem respiration (ER) and gross primary productivity (GPP) independently; an increase in the net ecosystem exchange can be a result of either a decrease in ER or an increase in GPP. Therefore, understanding the effects of changes in ecosystem water and temperature on the carbon flux components becomes key to predicting the responses of the Arctic to climate change. The overall goal of this work was to determine the response of arctic systems to simulated climate change scenarios with simultaneous changes in temperature and moisture. A temperature and hydrological manipulation in a naturally-drained lakebed was used to assess the short-term effect of changes in water and temperature on the carbon cycle. Also, as part of International Tundra Experiment Network (ITEX), I determined the long-term effect of warming on the carbon cycle in a natural hydrological gradient established in the mid 90’s. I found that the carbon balance is highly sensitive to short-term changes in water table and warming. However, over longer time periods, hydrological and temperature changed soil biophysical properties, nutrient cycles, and other ecosystem structural and functional components that down regulated GPP and ER, especially in wet areas.

No description available.

>Magister Scientiae - MSc / The research was driven by a need to increase the knowledge base concerning wetland ecological responses, as well as to identify and evaluate the factors driving the functioning of the Franschhoek Trust Wetland. An ecohydrological study was undertaken in which vegetation cover, depth to groundwater, water and soil chemistry were monitored at 14 sites along three transects for a 12 month period. The parameters used include temperature, pH, electrical conductivity (EC), sodium, potassium, magnesium, calcium, iron, chloride, bicarbonate, sulphate, total nitrogen, ammonia, nitrate, nitrite and phosphorus. T-tests and Principal Component Analysis (PCA) were used to analyze trends and to express the relationship between abiotic factors and vegetation. Results reflect the strong influence of hydrology, microtopography and nutrient availability in structuring vegetation composition in the wetland. The wetland has been classified as a palustrine valley bottom with channel wetland, which is predominantly groundwater-fed (phreatrotropic), but receives surface water inputs as well. Small scale gradients of microtopography allow for differences in flooding frequency and duration resulting in hydrologically distinct sites which differ chemically. Three zones were distinguished in the wetland. Hollows or low sites were characterized by intermittent flooding and drying and higher nutrient concentrations in soil and groundwater. High sites which were rarely or never flooded exhibited higher groundwater temperature and ammonia as well as iron in soils and groundwater. The inundated sites remained flooded throughout the year and were characterized by high nitrate and nitrite in soil as well as high EC, magnesium, bicarbonate, sulphate and phosphorus in groundwater. The limited availability of nitrogen in the wetland favoured plant types Typha capensis, Paspalum urvillei and Juncus .kraussii which are able to either fix nitrogen or store nitrogen during more favorable conditions. The main chemical concentration changes take place between summer and winter. The Principal Component Analyses suggest that sodium, chloride, potassium, ammonia and phosphorus are the dominant ions determining the chemistry of groundwater. Increased abstraction from the table mountain aquifer to supplement human demand may put the wetland at risk of degradation. Intensified agriculture and other land use in the area are likely to increase pollution loads into the wetland causing shifts in nutrient availability and vegetation composition. Continued and long term monitoring is essential to ensure effective management of the wetland and is highly recommended. Closer partnerships between wetland managers and scientists as well as community awareness and involvement through a volunteer monitoring programme should be encouraged

Arsenic (As) concentrations in groundwater that exceed the Maximum Contaminant Level (MCL) (10 μg/L) for drinking water have been reported throughout the United States, with higher occurrences in the Southwestern basin-fill aquifers. Levels of As above the MCL were measured in wells throughout the Cache Valley Basin, Utah. The As is naturally occurring in geologic material from the soil surface to depths of groundwater. This study reports on the mechanisms of retention and solubilization of As through these zones using geochemical modeling and microcosm studies. Two cores (NP 9 and NP 13) were collected from the soil surface to the depth of groundwater and sectioned based on observed redoximorphic features. Pore water was analyzed for As and iron( Fe) redox species, general water quality parameters and solid phase As, Fe and Mn using sequential extractions. These data were used in PHREEQC and MINTEQ geochemical models to predict mechanisms of As retention. Microcosm studies were performed using sediments from the water table zone. The sediments were exposed to oxidized, reduced, and poisoned conditions over time to evaluate the effect of the seasonal fluctuating water table on As release. Modeling results indicated As(V) was dominantly sorbed to hydrous ferric oxides (HFO) throughout both profiles. Although much less As(V) was sorbed to CaCO3, the percentage associated with calcite was 1.7 to 3.3% and 6 to 59% in the surface and water table zones for NP 9 and NP 13, respectively. As(III) solubility was controlled by the formation of an As-S mineral, orpiment. Microcosm findings, over 113-day incubation, concluded that regardless of treatment condition, As is released. For reduced samples As in solution was primarily As(III), while oxidized and poisoned samples only released As(V). The release of As under every condition, and the lack of reduced As and Fe in the poisoned samples, indicates that As release is abiotically controlled, while reduction is microbially driven. Carbonate minerals were the source of As(V) under treatment conditions as determined using an acetate extraction. Desorption of As(V) from carbonate minerals and the reduction of As(V) to As(III) played a significant role in explaining solution phase As(III) concentrations.

No description available.

No description available.

No description available.

No description available.

No description available.