Les mangroves sont des forêts côtières des régions tropicales et subtropicales, situées dans la zone de balancement des marées. Elles fournissent un grand nombre de services écosystémiques aux sociétés humaines et aux milieux adjacents (ressources, habitats, protection). Les mangroves sont continuellement affectées par des variations environnementales naturelles (salinité, température, cycle des marées) mais elles sont également le réceptacle d'effluents anthropiques (aquacultures, centres urbains). Il a été suggéré que ces forêts humides pourraient participer à la bio-épuration d'eaux usées domestiques (EUD). Les objectifs de la thèse étaient à la fois d'apporter des éléments pouvant ultérieurement servir à définir un cadre pour cet usage spécifique d'un écosystème naturel, et, sur le plan plus fondamental, d'évaluer les capacités de résistance et de résilience à court et long termes d'une mangrove soumise à une perturbation anthropique contrôlée. Le travail de thèse s'est appuyé sur un système expérimental fonctionnant depuis 2008 dans la plus grande mangrove de l'île de Mayotte. Dans ce système pilote, des EUD prétraitées sont quotidiennement déversées dans deux zones de mangroves dominées par les palétuviers Ceriops tagal ou Rhizophora mucronata et ayant des caractéristiques contrastées. Plusieurs campagnes d'échantillonnage ont permis de suivre in situ l'impact de l'excès de nutriments et d'eau douce sur la végétation, les crabes, la méiofaune et les communautés microbiennes, et de déterminer leurs capacités de résistance et de résilience. Dans ce travail, les EUD ont fortement stimulé la croissance de la végétation, modifié la structure des populations de crabes, de méiofaune, et des communautés microbiennes, et induit une hausse des densités et des activités des microorganismes impliqués dans le cycle de l'azote. Ceci n'a cependant pas entrainé de dysfonctionnement majeur de l'écosystème (perte massive de couvert végétal ou forte régression des populations de crabes). L'observation de l'écosystème sur deux ans après l'arrêt d'apports d'EUD a également mis en évidence une capacité notable de résilience de la mangrove. Enfin, les résultats de cette étude indiquent que la zone dominée par R. mucronata serait plus apte à recevoir les EUD prétraitées dans le cadre de leur bio-épuration par la mangrove car moins impactée. / Mangroves are coastal forests in (sub)tropical regions, located in the intertidal zone. They provide a many ecosystem services to human societies and neighboring ecosystems (resources, habitats, protection). Mangroves are continually affected by natural stresses (salinity, temperature, tidal cycle) but they are also subjected to anthropogenic effluents (aquaculture, urban centers). It was suggested that mangroves may be used for the bio-epuration of domestic wastewaters. This thesis aims at providing elements that will help defining the framework for this specific usage of a natural ecosystem and evaluating the long- and short-term resistance and resilience abilities of mangroves subjected to an anthropic disturbance. To reach these objectives, we used an experimental system set up in the largest mangrove of the island Mayotte. In this system, pretreated wastewaters (PW) are daily discharged in two mangrove zones dominated by Ceriops tagal or Rhizophora mucronata. Several sampling campaigns allowed to monitor in situ the impact of nutrient and freshwater excess on vegetation, crabs, meiofauna, and microbial communities and to determine their resistance and resilience capacities. In this work, PW strongly stimulated vegetation growth, altered the structure of crabs and meiofauna, modified the microbial structure and increased microbial density and activity of microorganisms involved in nitrogen cycle. However, no major dysfunction of the ecosystem (loss of vegetation, strong decrease in crab density) was observed. The monitoring of the ecosystem during two years after the PW discharge was stopped showed a marked ability of the mangrove to recover from the disturbance. Finally, the results indicate that R. mucronata mangrove zone, less disturbed, is more able to receive discharges for bio-epuration of domestic PW by mangrove ecosystem.

Mangrove ecosystems thrive in (sub)tropical, intertidal areas where adaptations<p>like vivipary and the hydrochorous dispersal of propagules become an absolute<p>necessity. As propagule dispersal and early growth allow for the replenishment of<p>existing stands and colonization of new habitats, many authors recognize the<p>importance of these stages in structuring mangrove populations and communities.<p>However, when it comes to the actual propagule dispersal and recruitment<p>mechanisms, there is an apparent lacuna in the current understanding of<p>mangrove ecology. The period between the mature propagule falling from the<p>parental mangrove tree and the early growth of the established seedling, under<p>various possible circumstances, remains in the dark. In this study we focus on this<p>particular period by investigating both the places where these propagules end up<p>as the pathways their dispersal units follow. And we go one step further.<p><p>Mangrove forests are being destroyed worldwide at a threatening pace despite<p>their tremendous asset to coastal human communities and associated biological<p>species. The effect of human-induced (cutting and mangrove conversion to<p>aquaculture ponds) as well as indirectly and/or ‘naturally’ evolving disturbances<p>(sea level rise) on propagule hydrochory occupies an important place in this study.<p><p>Dispersal of water-buoyant propagules of the family Rhizophoraceae and<p>Acanthaceae (now including the Avicenniaceae) was studied in Gazi Bay (Kenya),<p>Galle and the Pambala-Chilaw Lagoon Complex (Sri Lanka). The study sites<p>differ both in tidal regime and vegetation structure, covering an interesting variety<p>of ecological settings to examine propagule dispersal. Field data and experiments<p>ranging from micro/ mesotopographical measurements and successive propagule<p>counts to hydrodynamic and propagule dispersal experiments were collected or<p>executed in situ.<p><p>Two main methodological approaches were employed. Firstly, the question on<p>mechanisms of propagule recruitment was addressed by statistically investigating<p>the effect of microtopography, top soil texture and above-ground-root complexes on<p>the stranding and self-planting of propagules (Chapter 2&3). Afterwards,<p>suitability maps were created using Geographical Information Systems (GIS) to<p>assess whether a particular mangrove stand has the ability to succesfully<p>rejuvenate. Furthermore, the effect of degradation (tree cutting) (Chapter 2&3),<p>sea level rise (Chapter 2&4) and microtopography-altering burrowing activities of<p>the mangrove mud lobster Thalassina anomala (Chapter 3), was incoporated in the<p>GIS-analyses. Secondly, the combined set-up of hydrodynamic modelling and<p>ecological dispersal modelling was developed to simulate propagule dispersal<p>pathways influenced by dispersal vectors (tidal flow, fresh water discharge, wind),<p>trapping agents (retention by vegetation or aerial root complexes) and seed<p>characteristics (buoyancy, obligated dispersal period) (Chapter 5&6). This type of<p>approach provided the possibility to explore propagule dispersal within its<p>ecological context, but was also applied to an implication of shrimp pond area<p>restoration (Pambala-Chilaw Lagoon Complex, Sri Lanka) (Chapter 5) and to<p>evaluate changes in propagule dispersal when sea level rises (Gazi Bay, Kenya)<p>(Chapter 6).<p><p>The main findings regarding propagule recruitment indicate that propagules are<p>not distributed equally or randomly within a mangrove stand, yet species-specific<p>distribution for anchorage occurs. Characteristics of the environment<p>(microtopography, top soil texture and above-ground root complex) influence<p>propagule recruitment in a way that complex root systems (e.g. pencil roots and<p>prop roots) facilitate the entanglement of dispersal units and a more compact soil<p>texture (like clay and silt) and a predominant flat topography creates suitable<p>areas for stranding and self-planting of propagules. This combines effects of<p>existing vegetation and abiotic factors on mangrove propagule establishment.<p>Since propagule dispersal is not solely determined by species-specific propagule<p>characteristics (e.g. buoyancy, longevity, etc.), I emphasize that propagule sorting<p>by hydrochory has to be viewed within its ecological context. Propagule retention<p>by vegetation and wind as a dispersal vector, deserve a prominent role in studies<p>on propagule dispersal. The significance of dense vegetation obstructing long<p>distance dispersal (LDD in its definition of this work), mainly in inner mangrove<p>zones, supports our main finding that propagule dispersal is largely a short<p>distance phenomenon. ‘Largely’ is here understood as quantitatively, not<p>excluding epic colonization events of rare but important nature.<p>In accordance with the Tidal Sorting Hypothesis (TSH) of Rabinowitz (1978a),<p>smaller, oval-shaped propagules were found to disperse over larger distances than<p>bigger, torpedo-shaped propagules. We can however not fully support the TSH<p>because (1) these differences are no longer valid when comparing between torpedoshaped<p>propagules of different sizes and (2) propagule dispersal is not always<p>directed towards areas more inland, but can be strongly concentrated towards the<p>edges of lagoons and channels<p><p>Anthropogenic pressure on mangrove ecosystems, more specifically clear-felling or<p>mangrove conversion to aquaculture ponds, imposes limitations on propagule<p>recruitment due to reduced propagule availability and a decrease in suitable<p>stranding areas where the architecture of certain root complexes, like prop roots<p>and pencil roots, function as propagule traps. These types of pressure appear to<p>have more severe consequences on propagule dispersal than the effect of sea level<p>rise on mangroves. Mangrove forests, which are not situated in an obviously<p>vulnerable setting, can be resilient to a relative rise in sea level if a landward shift<p>of vegetation assemblages and successful early colonization is not obstructed by<p>human-induced pressures. Also, and this renders mangrove forests vulnerable in<p>spite of their intrinsic resilience, when the ‘capital’ of forest is severely reduced or<p>impoverished as happens extensively worldwide, the ‘interest’ on this capital,<p>understood as propagule availability, delivery and trapping, will not allow them to<p>efficiently cope with sea level rise, putting sustainability of mangrove ecosystem<p>services and goods at risk.<p><p>In a larger framework of mangrove vegetation dynamics, knowledge on propagule<p>dispersal will benefit management strategies for the conservation of mangroves<p>worldwide, besides its fundamental interest to fully fathom the ecology of this<p>particular marine-terrestrial ecotone formation. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished