En Guinée, la disponibilité des ressources en eau est soumise à de graves problèmes de gestion avec des conséquences importantes pour les populations, notamment à Kankan où la plupart des puits tarissent pendant la saison sèche, et où la fourniture régulière de l’eau à partir de la rivière Milo n'est pas assurée dans tous les quartiers de la ville. Le Milo et son bassin versant représente donc un enjeu important pour les habitants de cette ville. Le Milo (480 km de long) qui draine un bassin versant 13 810 km2 est le plus important des affluents rive droite du Niger. Pour évaluer l’impact du bassin versant du Milo et de la ville de Kankan sur la qualité des eaux et des sédiments de ce cours d’eau, 30 échantillons de sédiments de fond ont été prélevés au cours d’une première campagne (juin-juillet 2013) sur l’ensemble des cours d’eau du bassin versant, tandis que durant une deuxième campagne (avril 2014-mai 2015), un suivi régulier a été mis en place sur le Milo en amont et en aval de Kankan. Au cours de cette dernière, 232 échantillons d’eau du Milo (dont 116 utilisés pour l’isotopie), et 26 échantillons de sédiments de fond ont été prélevés sur les stations de Bordo (amont de Kankan) et de Karifamoriah (aval de Kankan). Les concentrations en éléments majeurs et traces, terres rares, COP, COD, pH, alcalinité, l’azote organique, les isotopes de l’oxygène et de l’hydrogène, ainsi que la micro-granulométrie des sédiments ont été mesurés. Les résultats obtenus durant le cycle hydrologique 2014-2015 ont permis d’estimer le flux de matières exportées en solution par le Milo à Kankan à 47 863 t.an-1 et le flux de matières particulaires à 76 759 t.an-1. 92% de ce tonnage est exporté durant la période des hautes eaux. Ces flux permettent d’estimer des vitesses moyennes d’érosion physique (8 t.km-2.an-1) et d’altération chimique (5 t.km-2.an-1) du bassin relativement faibles. Le flux spécifique de CO2 consommé par l’altération chimique est lui aussi relativement faible (76.103 mole.km-2.an-1), mais il reste dans la moyenne des flux de CO2 mesurés sur les bassins couverts de sols latéritiques. Le degré de contamination des sédiments et sa variation spatio-temporelle ont été évalués grâce au calcul du facteur d’enrichissement (FE), en normalisant les concentrations en éléments traces par rapport à Al, Sc et Ti et en prenant le PAAS et l’UCC comme matériaux de référence. La quasi-totalité des éléments traces proviennent essentiellement des processus d’altération et sont d’origine naturelle. Toutefois, un enrichissement modéré à significatif (5 < FE < 20) a été mis en évidence pour Zr et Hf (enrichissements naturels) et As et Sb (impacts anthropiques). Finalement, l’apport anthropique en éléments traces, quand il existe, reste modéré. Les profils de concentrations en terres rares ont permis de montrer que la signature géochimique des sédiments de fond du Milo est homogène sur l’ensemble du bassin et caractéristique des sols latéritiques, non perturbée par la ville de Kankan et elle est proche des signatures du PAAS et de l’UCC. Pour l’ensemble des éléments traces et des terres rares la contribution anthropique reste faible et les fractions disponibles (phases labiles) sur les sédiments (extraction à l’EDTA) sont également faibles, confirmant ainsi que l’essentiel de ces éléments rentrent principalement dans la composition des phases résiduelles. / In Guinea, the availability of water resources is subjected to serious problems of management with important consequences for the populations, particularly in Kankan city where most of the wells dry up during the dry season and where the water supply from the Milo river is not assured for all the districts. Then the Milo river and its watershed represent an important issue for the inhabitants of Kankan. The Milo river (480 km in length) which drains an area of 13 810 km2 is the most important right-bank tributary of the Niger river. To evaluate the impact of the Milo watershed and of the Kankan city on the water and sediments of this river, 30 samples of bottom sediments have been collected during a first campaign (June-July 2013) on the main course from upstream to downstream and on the main tributaries, whereas during a second campaign (April 2014-May 2015) a regular monitoring was set up on the Milo river, upstream and downstream Kankan city. During this campaign, 232 river water samples (of which 116 for isotopic analyses) and 26 bottom sediment samples have been collected at Bordo (upstream) and Karifamoriah (downstream) stations. The concentrations of major and trace elements, rare earth elements, POC and PON, DOC, pH, alkalinity, the isotopic composition of O, H and the micro-granulometry of sediments have been measured. The results obtained during the hydrological cycle 2014-2015 allowed to estimate the fluxes of dissolved (47 863 t.y-1) and particulate (76 759 t.y-1) matters exported by the Milo river at Kankan. 92% of these fluxes are exported during the high flow period. These fluxes allowed us to estimate very low average rates of physical erosion (8 t.km-2.y-1) and chemical weathering (5 t.km-2.y-1) of the Milo catchment. The specific flux of CO2 consumed by chemical weathering (76.103 mole.km-2.y-1) is relatively low but within the range of CO2 fluxes measured in catchments draining lateritic soils. The degree of sediment contamination and its spatio-temporal variation have been assessed using the enrichment factor (EF), by normalizing the trace element concentrations with Al, Sc and Ti and by taking PAAS and UCC as the reference materials. Almost all the trace elements mainly originate from weathering processes and are of natural origin. Nevertheless, moderate to significant enrichments (5 < EF < 20) have been calculated for Zr and Hf (natural enrichments) and As and Sb (anthropogenic impacts). Finally, the anthropogenic contribution of trace elements, when it exists, remains moderate. The rare earth concentration patterns allowed to show the geochemical signature of river bottom sediments is homogeneous within the Milo catchment and characteristic of lateritic soils, non- perturbed by the city of Kankan and close to the PAAS and UCC patterns. For most of the trace and rare earth elements, the anthropogenic contribution remains low and the available fractions (labile phases) in the sediment (EDTA extraction) are also low, confirming that the main part of these elements are mainly in the residual phases.

O presente estudo teve por objetivo avaliar a influência de esgotos domésticos e petroquímicos, dispostos por via de emissário submarino, nas composições físico-químicas e microbiológicas da coluna dágua e na geoquímica dos sedimentos, bem como investigar uma possível relação entre o meio e a composição química das carapaças do foraminífero Ammonia tepida. Para atingir tal objetivo foram coletadas 10 amostras de água e de sedimento superficial, no entorno dos emissários do Araçá e do Terminal Petrolífero Almirante Barroso (TEBAR), além de um ponto controle, no canal de São Sebastião, SP, em março e abril de 2007. Em cada amostra de água foram realizadas análises físico-químicas e microbiológicas. Nos sedimentos, foram realizadas análises granulométricas e geoquímicas (macronutrientes, elementos maiores e traços). Do sedimento, carapaças coradas (indivíduos vivos) e não coradas (mortos) de A. tepida foram triadas e analisadas em espectrômetros de energia dispersiva (MEV-EDS) e de emissão atômica (ICP-OES) para análise química dos elementostraços. Os resultados das análises físico-químicas e microbiológicas da água, indicaram que a coluna dágua não estava contaminada, no momento da amostragem, pois todos os valores obtidos estavam dentro dos limites exigidos pela resolução do CONAMA, nº 357/2005. Os resultados granulométricos mostraram que o sedimento é litoclástico e mais psamítico no Araçá, e mais pelítico, no TEBAR. Já os dados geoquímicos de carbono orgânico, nitrogênio, enxofre e fósforo revelaram que ambas as regiões são propícias ao enriquecimento desses elementos. A concentração de elementos-traço, na maioria dos pontos estudados, nas duas áreas, quando comparada aos valores guia de causaefeito, não foi considerada tóxica para a biota marinha. Contudo, foi observado enriquecimento significativo de cádmio, em certas áreas no Araçá, e de bário, no TEBAR. A concentração deste último elemento, provavelmente, está relacionada à água de produção do terminal. Os elementos-traço das carapaças de A. tepida coradas, em ambos os emissários, apresentaram maior número de correlações significativas com o sedimento, sugerindo que carapaças de foraminíferos mortos estão mais susceptíveis ao viés dos processos diagenéticos. No TEBAR, foram observadas correlações significativas positivas entre os teores de Co, Cr, Mn e Pb nos sedimentos e os elementos-traço das carapaças coradas. Já no Araçá, somente os teores de Cr correlacionaram-se positiva e significativamente aos teores encontrados nas carapaças coradas. Isto corrobora com a influência da composição do sedimento nas reações intracelulares dos foraminíferos, refletidas na construção do exoesqueleto. Diante do exposto, os resultados deste estudo trouxeram luz a novas questões que permitiram inferir que a auto-depuração da água do mar, em ambas as regiões, e a estação de tratamento de efluentes, no TEBAR, não se revelaram eficientes na diluição e/ou remoção do conteúdo de nutrientes, matéria orgânica e elementos maiores e traço. / The goal of this study was to evaluate the influence of domestic and petrochemical sewage, discharged through a submarine sewage outfall, on the physicochemical and microbiological composition of the water column and the geochemistry of sediments. As well as to investigate a possible relationship between the environment and the chemical composition of Ammonia tepida foraminifer\'s tests. The area where this study was focused on two sewage emissaries: Araçá and Almirante Barroso Petroliferous Terminal (TEBAR), both located in São Sebastião Channel, São Paulo, Brazil. To achieve this goal, 10 water and bottom sediments samples in in March and April of 2007 were collected form each outfall. A background sample also collected as a control. For each water sample physicochemical and microbiological analyses were carried out and, in the sediments, grainsize and geochemical (macronutrients, major and trace elements) were analyzed. Trace elements found in the tests of stained (live) and unstained (dead) Ammonia tepida were identified by scanning electron microscopy using energy dispersive spectroscopy (SEM-EDS) and by inductively coupled plasma optical emission spectrometry (ICP-OES). The analysis of the water samples showed that the water column was not contaminated. All the concentrations obtained were in accordance with the National Environmental Council (CONAMA) limits. Grain size analysis, demonstrated that the sediments are of terriginous origin, being mainly psamitic in Araçá and mainly pelitic in TEBAR. The geochemical data of organic Carbon, Nitrogen, Sulphur and Phosphorus revealed that both regions are susceptible to enrichment in these elements. In both areas, the concentration of trace elements, when compared to quality guideline values, was not considered toxic for marine life. However, significative enrichment in Cadmium in certain areas of Araçá and Barium in TEBAR. The latter is probably related to the terminal\'s water production. In both outfalls, it was observed a significant correlation between the trace elements in the live A. tepida\'s tests and the sediment, which suggests that the dead foraminifera are more susceptible to diagenical bias. In TEBAR the correlations observed were significative and positive among the elements Co, Cr, Mn and Pb present in the sediments and the trace elements of the stained tests. In the Araçá outfall, only the Cr levels correlate significative and positively with the levels found in the tinged tests. This shows that sediment composition influences the intracelular reactions in the forminifera, which reflected in the exoskeleton\'s building. Given the above, the results of this study brought up new questions that allowed to infer that sea water depuration at both outfalls, as well as at sewage treatment plant, in TEBAR was not sufficient to dilute and/or remove the nutrient content, the organic matter and major and trace elements of the sewage.

Philosophiae Doctor - PhD / The management and disposal of huge volumes of coal combustion by products such as fly ash has constituted a major challenge to the environment. In most cases due to the inadequate alternative use of coal fly ash, the discarded waste is stored in holding ponds, slag heaps, or stock piled in ash dumps. This practice has raised concerns on the prospect of inorganic metals release to the surface and groundwater in the vicinity of the ash dump. Acceptable scientific studies are lacking to determine the best ash disposal practices. Moreover, knowledge about the mobility patterns of inorganic species as a function of mineralogical association or pH susceptibility of the dry disposed ash dump under natural weathering conditions are scarce in the literature. Fundamental understanding of chemical interactions of dry disposed ash with ingressed CO2 from atmosphere, percolating rain water and brine irrigation within ash disposal sites were seen as key areas requiring investigation. The mineralogical association of inorganic species in the dry disposed ash cores can be identified and quantified. This would provide a basis for understanding of chemical weathering, mineralogical transformations or mobility patterns of these inorganic species in the dry ash disposal scenario. The current study therefore aims to provide a comprehensive characterisation of weathered dry disposed ash cores, to reveal mobility patterns of chemical species as a function of depth and age of ash, with a view to assessing the potential environmental impacts. Fifty-nine samples were taken from 3 drilled cores obtained respectively from the 1 year, 8 year and 20-year-old sections of sequentially dumped, weathered, dry disposed ash in an ash dump site at Tutuka - a South African coal burning power station. The core samples were characterized using standard analytical procedures viz: X-ray fluorescence (XRF), X-ray diffraction (XRD), Fourier transforms infrared (FTIR) techniques, Scanning electron microscopy/energy dispersive spectroscopy (SEM/EDS) and Acid neutralisation capacity (ANC) test. A modified sequential extraction (SE) method was used in this study. The chemical partitioning, mobility and weathering patterns in 1 year, 8 year and 20-year-old sections of the ash dump were respectively investigated using this modified sequential extraction scheme. The sequence of the extractions was as follows: (1) water soluble, (2) exchangeable, (3) carbonate, (4) iron and manganese and (5) residual. The results obtained from the 5 steps sequential extraction scheme were validated with the total metal content of the original sample using mass balance method. The distribution of major and trace elements in the different liquid fractions obtained after each step of sequential extraction of the 59 drilled core samples was determined by inductively coupled plasma mass spectrometry (ICPMS). The data generated for various ash core samples were explored for the systematic analysis of mineralogical transformation and change in ash chemistry with ageing of the ash. Furthermore, the data was analyzed to reveal the impact of ingressed CO2 from atmosphere, infiltrating rain water and brine irrigation on the chemistry of ash core samples. Major mineral phases in original ash core samples prior to extraction are quartz (SiO2) and mullite (Al2O3·2SiO2). Other minor mineral phases identified were hematite (Fe2O3), calcite (CaCO3), lime (CaO), anorthite (CaAl2Si2O8), mica (Ca (Mg, Al)3 (Al3Si) O10 (OH)2), and enstatite (Mg2Si2O6). X-ray diffraction results show significant loss of crystallinity in the older ash cores. The presence of minor phases of calcite and mica in dry disposed ash cores are attributed to reduction in the pore water pH due to hydration, carbonation and pozzolanic reactions. The X-ray diffraction technique was unable to detect Fe-oxyhydroxide phase and morealuminosilicate phases in ash core samples due to their low abundance and amorphous character. X-ray fluorescence results of the original ash core samples showed the presence of major oxides, such as SiO2, Al2O3, Fe2O3, while CaO, K2O, TiO2, Na2O, MnO, MgO, P2O5, and SO3 occur in minor concentrations. The ratio of SiO2/Al2O3 classified the original core samples prior to extraction as a silico-aluminate class F fly ash. The ternary plot of major elements in 1-year-old ash core samples was both sialic and ferrocalsialic but 8 year and 20-year-old ash core samples were sialic in chemical composition. It is noteworthy that the mass % of SiO2 varies through the depth of the core with an increase of nearly 3 %, to 58 mass % of SiO2 at a depth of 6 m in the 1-year-old core whereas in the case of the 8-year-old core a 2 % increase of SiO2 to a level of 57.5 mass % can be observed at levels between 4-8 m, showing dissolution of major components in the matrix of older ash cores.. The Na2O content of the Tutuka ash cores was low and varied between 0.6-1.1 mass % for 1-year-old ash cores to around 0.6-0.8 mass % for 8-year-old ash cores. Sodium levels were higher in 1-year-old ash cores compared to 8 year and 20-year-old ashcores. Observed trends indicate that quick weathering of the ash (within a year) leached out Na+ from the ash dump. No evidence of Na+ encapsulation even though the ash dump was brine irrigated. Thus the dry disposal ash placement method does not result in a sustainable salt sink for Na-containing species over time. The total content of each of the elements in 1 year and 20-year-old ash cores was normalised with their total content in fresh ash from same power station to show enrichment and depletion factor. Major elements such as K+, Mn showed enrichment in 1-year-old ash cores whereas Al, Si, Na+, Ti, Ca, Mg, S and Fe showed depletion due to over time erosion. Trace elements such as Cr, Sr, P, Ba, Pb, V and Zn showed enrichment but Ni, Y, Zr showed depletion attributed to over time erosion. In 20-year-old ash cores, major elements such as Al, Na+ and Mn showed enrichment while Si, K+, Fe, Mg and Ca showed depletion highlighting their mobility. Trends indicated intensive flushing of major soluble components such as buffering constituents (CaO) by percolating rain water. The 1-year-old and 20-year-old coal ash cores showed a lower pH and greater loss/depletion of the soluble buffering constituents than the 2-week-old placed ash, indicating significant chemical weathering within a year. Based on ANC results the leaching behaviours of Ca, Mg, Na+, K+, Se, Cr, and Sr were found to be controlled by the pH of the leachant indicating high mobility of major soluble species in the ash cores when in contact with slightly acid rain water. Other investigated toxic metals such as As, Mo and Pb showed amphoteric behaviour with respect to the pH of the leachant. Chemical alterations and formation of transient minor secondary mineral phases was found to have a significant effect on the acid susceptibility and depletion pattern of chemical species in the core ash samples when compared to fresh ash. These ANC results correlated well with the data generated from the sequential extraction scheme. Based on sequential extraction results elements, showed noticeable mobility in the water soluble, exchangeable and carbonate fractions due to adsorption and desorption caused by variations in the pore water pH. In contrast, slight mobility of elements in the Fe and Mn, and residual fractions of dry disposed fly ashes are attributed to the co-precipitation and dissolution of minor amount of less soluble secondary phase overtime. The 1-year-old dry disposed ash cores were the least weathered among the 3 drilled ash cores. Therefore low concentration of toxic metals in older ash cores were ascribed to extensive weathering with slower release from residual mineral phases over time. Elements were found to associate with different mineral phases depending on the age or depth of the core samples showing greater heterogeneity in dispersion. For instance the average amount of total calcium in different mineral associations of 1-year-old ash cores is as follows; water soluble (10.2 %), exchangeable (37.04 %), carbonate (37.9 %), Fe and Mn (7.1 %) and residual (2.97 %). The amount of total Na+ in different mineral phases of 1-year-old ash cores followed this trend: water soluble (21 %), exchangeable (11.26 %), carbonate (2.6 %), Fe and Mn (4.7 %) and residual (53.9 %). The non-leachable portion of the total Na+ content (namely that contained in the residual fraction) in the 1-year-old ash core samples under conditions found in nature ranged between 5-91 %. This non-leachable portion of the Na+ showed the metastability of the mineral phases with which residual Na+ associates. Results showed older ash cores are enriched in toxic elements. Toxic elements such as As, B, Cr, Mo and Pb are enriched in the residual fraction of older ash cores. For instance As concentration in the residual fraction varied between 0.0003- 0.00043 mg kg-1 for 1-year-old ash cores to around 0.0003-0.0015 mg kg-1 for 20-year-old ash cores. This suggests that the older ash is enriched in toxic elements hence dust from the ash dump would be toxic to human health. The knowledge of mobility and ecotoxicological significance of coal fly ash is needed when considering its disposal or reuse in the environment. The mobility and ecotoxicology of inorganic metals in coal fly ash are determined by (i) mineralogical associations of inorganic species (ii) in-homogeneity in the ash dumps (iii) long and short term exposure to ingress CO2 and percolating rain water. Management issues such as inconsistent placement of ash in the dumps, poor choice of ash dump site, in-homogeneity in brine irrigation, no record of salt load put on the ash dumps and lack of proper monitoring requires improvement. The thesis provides justification for the use of the modified sequential extraction scheme as a predictive tool and could be employed in a similar research work. This thesis also proved that the dry ash disposal method was not environmental friendly in terms of overall leaching potential after significant chemical weathering. Moreover the study proved that the practice of brine co-disposal or irrigation on ash dumps is not sustainable as the ash dump did not act as a salt sink. / South Africa