The microbial life in the soil is essential for providing a functioning habitat for plants to grow. A literature study was conducted to investigate the knowledge and science behind soil biology. The purpose of this study was to define what is soil health and how it is influenced by the soil microbial communities. The literature study concluded that the ability of soil biology to benefit plants includes a variety of aspects. Nutrient availability, soil structure and pest resistance are all greatly influenced by soil microbes.  To practically examine these theories, an experiment was conducted where Brandywine tomatoes where grown in three different scenarios. A commercial potting soil, Hasselfors ekojord, was used as substrate in all groups. In the control group (C) the plants were grown only in the substrate. In the second group (R), the seeds where treated with a microbial inoculum and then planted in the substrate. In the third group (RE), the same treatment as in R was done to the seeds and here, compost extract were also added to the RE group. The plants were grown in separate pots in a greenhouse and the growth rate was observed and documented as well as the total harvest. In the end of the growing season a chemical and biological analysis was done to the soil as well as a sap analysis on the leaves. The plant growth where similar among the groups although R and RE showed slightly higher growth rates in the later stages of the growing season. The harvested fruit was highest in C but not significantly. The microbial contents were high in all soils though more fungi communities in the RE and bacterial communities in C. The chemical analysis showed high nitrate concentrations in the leaves in C. In R and especially RE the nitrate conversion into amino acids and proteins where higher wish indicates that these groups are more resilient to pests like aphids. / Det mikrobiella livet i jorden är avgörande för att skapa en fungerande livsmiljö för växter. En litteraturstudie genomfördes för att undersöka nuvarande kunskap och vetenskap bakom markbiologi. Syftet med denna studie var att definiera markhälsa och hur den påverkas av det mikrobiella livet i jorden. Slutsatsen från denna litteraturstudie var att jordbiologins förmåga att gynna växter innefattar en rad olika aspekter. Näringstillgänglighet, markstruktur och skadedjursbeständighet påverkas starkt av jordmikrober.  För att praktiskt granska dessa teorier genomfördes ett experiment där Brandywine-tomater odlades i tre olika scenarier. En kommersiell plantjord, Hasselfors ekojord, användes som huvudsubstrat i alla grupper. I kontrollgruppen (C) odlades växterna endast i substratet. I den andra gruppen (R) behandlades frön med en mikrobiell ympning innan de såddes i substratet.I den tredje gruppen (RE) utfördes samma fröbehandling som i R och kompostextrakt tillsattes också till RE-gruppen. Växterna odlades i separata krukor i ett växthus och tillväxthastigheten observerades och dokumenterades liksom den totala skörden. I slutet av växtsäsongen gjordes en kemisk och biologisk analys av jorden samt en savanalys på bladen. Tillväxten var likartad bland grupperna även om R och RE visade något högre tillväxttakt i de senare stadierna av växtsäsongen. Skördad frukt per planta var högst i C, dock intesignifikant. Den mikrobiella koncentrationen var hög i alla jordar men mer svamporienterat i RE och bakterieorienterat i C. Den kemiska analysen visade högt nitratinnehåll i bladen i C. I R och särskilt i RE var nitratomvandlingen till aminosyror och proteiner högre vilket indikerar att dessa grupper är mer motståndskraftiga mot skadedjur så som bladlöss.

No description available.

Radionuclides are widely used in energy production and medical, military and industrial applications. Thus, understanding the behaviour of radionuclides which have been or may be released into ecosystems is important for human and environmental risk assessment. Modelling of radionuclides or their stable element analogues is the only tool that can predict the consequences of accidental release. In this thesis, two dynamic stochastic compartment models for radionuclide/element transfer in a marine coastal ecosystem and a freshwater lake were developed and implemented (Paper I and III), in order to model a hypothetical future release of multiple radionuclides from a nuclear waste disposal site. Element-specific mechanisms such as element uptake via diet and adsorption of elements to organic surfaces were connected to ecosystem carbon models. Element transport in two specific coastal and lake ecosystems were simulated for 26 and 13 elements, respectively (Papers I and III). Using the models, the concentration ratios (CR: the ratio of the element or radionuclide concentration in an organism to the concentration in water) were estimated for different groups of aquatic organisms. The coastal model was also compared with a 3D hydrodynamic spatial model (Paper II) for Cs, Ni and Th, and estimated confidence limits for their modelled CRs. In the absence of site-specific CR data, being able to estimate a range of CR values with such models is an alternative to relying on literature CR values that are not always relevant to the site of interest. Water chemistry was also found to influence uptake of contaminants by aquatic organisms. Empirical inverse relationships were derived between CRs of fish for stable Sr (CRSr) and Cs (CRCs) and water concentrations of their biochemical analogues Ca and K, respectively (Paper IV), illustrating how such relationships could be used in the prediction of more site-specific CRCs and CRSr in fish simply from water chemistry measurements. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Manuscript. Paper 4: Manuscript.</p>

La vasière intertidale de Brouage, écosystème très étudié localement de longue date, a fait l’objet depuis 5 ans d’analyses plus approfondies notamment sur le devenir du biofilm microbien se formant à la surface des sédiments à marée basse. Ce travail de thèse s’inscrit dans cette optique. Il propose, par le couplage d’outils mathématiques et écologiques, d’analyser le réseau trophique de la vasière de Brouage sous différentes conditions abiotiques à partir de la synthèse des données les plus récentes. La modélisation inverse a permis d’estimer les flux manquants du réseau trophique. Les calculs d’indices d’analyse des réseaux écologiques ou ENA (Ecological Network Analysis) ont caractérisé la structure et le fonctionnement du réseau trophique.Un premier travail méthodologique, basé sur la dégradation-reconstruction d’un jeu de données très complet, a défini la meilleure fonction, la moyenne, pour le choix d’une seule solution pour chaque flux à la sortie de la modélisation inverse. La moyenne est la meilleure fonction puisque son utilisation aboutit à une bonne estimation des flux manquants quel que soit le niveau d’information intégré au modèle. Cette fonction permet aussi une bonne préservation de la structure et du fonctionnement du réseau trophique ainsi reconstruit. Le réseau trophique de la vasière de Brouage à basse mer a été étudié sous un aspect saisonnier par l’opposition de l’été et de l’hiver, l’hiver étant caractérisé par la présence d’oiseaux migrateurs sur la vasière. L’activité limicole des oiseaux est soutenue en hiver par une forte production du microphytobenthos et un fort recyclage de la matière. L’organisation interne et la spécialisation des flux restent cependant semblables pour les deux saisons considérées. Dans un deuxième temps, le réseau trophique à pleine mer en été a été modélisé sous différentes conditions hydrodynamiques permettant ou pas la remise en suspension du biofilm microbien. La comparaison entre ces deux situations met en évidence l’effet de la remise en suspension sur le couplage benthos-pelagos. Le système soumis à la remise en suspension présente un très faible indice de recyclage lié à la faible intégration au réseau trophique pélagique du carbone particulaire benthique remis en suspension. De plus, la remise en suspension provoque une augmentation de l’activité du système couplée à une faible spécialisation des flux, ce qui est supposé déstabiliser le système. / The Brouage mudflat is an intertidal ecosystem which has been studied locally for a long time. During the last 5 years, in depth studies have especially focused on the fate of the microbial biofilm which develops at the surface of the sediments at low tide. The present Ph.D. work was accomplished in this framework. It proposes, through the synthesis of the most recent data, to combine mathematical and ecological tools to analyze the Brouage food web under different abiotic conditions. The inverse modeling allowed estimating the unknown flows of the food web. The calculation of the ENA (Ecological Network Analysis) indices characterized the structure and the functioning of the food web. A methodological work, based on the degradation/reconstruction of a complete data set, determined the best function, the mean, to choose a single solution for each flow among the set of values estimated by the inverse modeling. The mean is the best function because it leads to a good estimation of the unknown flows irrespective of the quantity of information integrated in the model. Moreover, this function allows a good conservation of the structure and functioning of the reconstructed food web.The Brouage food web at low tide was investigated with a seasonal approach by the opposition of the summer and the winter food webs, the latest being characterized by the presence of migratory shorebirds. The shorebird activity is sustained by a strong primary production and a strong recycling. Nevertheless, the internal organization and the specialization remain similar for the two seasons. In a second step, the food web at high tide was modeled under different hydrodynamical conditions sufficient or not to induce the resuspension of the microbial biofilm. The comparison between the two food webs highlights the effect of the resuspension on the benthos-pelagos coupling. The recycling activity is low due to the low integration of the resuspended benthic particulate carbon in the pelagic food web during resuspension. Additionally, the resuspension leads to an increase of the activity of the whole system coupled with a drop of the specialization of the trophic pathways that is supposed to destabilize the system.

Os ecossistemas aquáticos têm sido afetados pelo uso intensivo de agrotóxicos e pela introdução de espécies exóticas. Para melhor avaliar os efeitos diretos e indiretos dessas atividades sobre as comunidades aquáticas, é essencial a utilização de experimentos em escala mais próxima a real. O objetivo desta pesquisa foi analisar os efeitos do agrotóxico Vertimec® 18 CE, cujo princípio ativo é a abamectina, sobre a comunidade planctônica através de um experimento em mesocosmos com duração de 20 dias. Para reproduzir relações biológicas mais complexas e analisar o impacto da introdução de uma espécie exótica sobre o plâncton, inseriram-se girinos de Lithobates catesbeianus em alguns tratamentos. Foram consideradas duas concentrações do agrotóxico (2 &#956;L L-1 em dose múltipla e 8 &#956;L L-1 em dose única) na ausência e presença de girinos, além de dois tratamentos não contaminados (controle com e sem girinos). Variáveis físicas, químicas e biológicas (clorofila a, aspectos qualitativos e quantitativos do fitoplâncton e do zooplâncton) da água e do sedimento foram monitoradas. A partir dos resultados obtidos, verificou-se que os girinos causaram um aumento na concentração de nutrientes na água e uma diminuição do oxigênio dissolvido e da densidade planctônica. Nos tratamentos contaminados, como um efeito indireto do incremento da concentração de nutrientes e da mortalidade das espécies de Cladocera e Copepoda, a densidade do fitoplâncton e de algumas espécies tolerantes de Rotifera aumentou. No tratamento com a concentração mais elevada do agrotóxico, registrou-se a morte de todos os girinos, proporcionando um aumento na concentração de nutrientes na água e na densidade de algumas espécies fitoplanctônicas, que alcançaram uma densidade total de 96.919 ind mL-1. Estes resultados indicam que a inserção dos girinos e a adição do agrotóxico causaram efeitos diretos e indiretos sobre a comunidade planctônica e a qualidade da água. Conclui-se que os mesocosmos podem contribuir para a compreensão de tais efeitos e oferecer subsídios ao manejo de ecossistemas de água doce impactados por atividades antrópicas. / Aquatic ecosystems have been impacted due to the intensive use of pesticides and the introduction of exotic species. It is essential to use more realistic experiments for assessing direct and indirect effects of these activities on aquatic communities. This thesis aimed to analyze the effects of the pesticide Vertimec® 18 EC, whose active ingredient is abamectin, on plankton community through a mesocosm experiment during 20 days. Tadpoles of the species Lithobates catesbeianus were introduced in some treatments to simulate complex biological relationships and to evaluate the impact of the introduction of such exotic species on plankton. Different treatments were analyzed, including one case of no contamination (control) and two cases of contamination: 2 &#956;L L-1 (multiple dose) and 8 &#956;L L-1 (single dose). Such conditions were considered both with absence and presence of tadpoles. Physical, chemical and biological (chlorophyll a, qualitative and quantitative aspects of phytoplankton and zooplankton) variables of water and sediment were also monitored. The tadpoles caused an increase in the concentration of nutrients and a decrease of dissolved oxygen concentration and plankton density. In the contaminated treatments, as an indirect effect of the nutrient increase and the mortality of species of Cladocera and Copepoda, the density of phytoplankton and some tolerant species of rotifers increased. In the treatment with the highest concentration of the pesticide, all the tadpoles died. As a result of the death of the tadpoles, there was nutrient enrichment in the water, with an associated increase in the density of some phytoplankton species, reaching a total density of 96,919 ind mL-1. These results suggested that direct and indirect effects occurred on plankton community and water quality through the addition of the contaminant and the tadpoles. We conclude that the mesocosms might contribute to the understanding of these effects and provide subsidies for management of freshwater ecosystems impacted by human activities.

A demanda da população humana por alimentos, fibras e biocombustíveis impõe crescimento pronunciado na agricultura intensiva e, com ela, no consumo de fertilizantes. Este manejo de fertilizantes leva a alterações dramáticas nas quantidades e proporções de nutrientes em agroecossistemas que, por sua vez, podem influenciar comunidades biológicas por meio de alterações na biomassa, qualidade nutricional e composição de produtores. Uma tendência no cenário produtivo brasileiro recente é a expansão das culturas de biocombustíveis, notadamente da cana-de-açúcar (etanol) e da soja (biodiesel), sobre ambientes de cerrado. Estas culturas, que juntas cobrem nada menos que 30 milhões de hectares do território brasileiro, demandam padrões contrastantes de fertilização. A cana-de-açúcar exige altas proporções de nitrogênio relativo a fósforo, enquanto que a soja exige altas proporções de fósforo relativo a nitrogênio. Esta dissertação teve como objetivo testar os efeitos diretos e indiretos que as alterações nas quantidades e proporções de N e P que se seguem à conversão de ambientes nativos para o cultivo de cana-de-açúcar e soja têm sobre comunidades biológicas. Estes objetivos foram testados por meio de dois experimentos em mesocosmos aquáticos manipulando nutrientes, algas, girinos e ninfas de libélulas. O primeiro experimento seguiu um delineamento fatorial completo cruzando três concentrações de N (Nreferência, Nsoja, Ncana) e três concentrações de P (Preferência, Psoja, Pcana), e dois níveis tróficos (algas e girinos de Physalaemus cuvieri). Níveis de nutrientes manipulados simularam condições medidas em corpos d´água de cerrado, ou padrões de fertilização conhecidos para culturas de cana-de-açúcar e de soja. No segundo experimento pareamos essas concentrações (Nreferência Preferência, Nsoja, Psoja, Ncana Pcana) de forma a reduzir a dimensionalidade do experimento e assim permitir a inclusão de um terceiro nível trófico (algas, girinos de Hypsiboas faber e ninfas de libélulas). Incluímos ainda um tratamento no qual as ninfas de libélulas estiveram engaioladas para separar os efeitos da predação de girinos, dos efeitos da alteração nos seus atributos. As principais variáveis de resposta em ambos experimentos foram concentração de clorofila a como indicadora de biomassa de fitoplâncton, e mortalidade, massa final e estágio de desenvolvimento final de girinos como indicadores de rendimento de consumidores. Uma comparação na concentração de clorofila antes e depois da introdução de girinos procurou testar o efeito da adição de consumidores na biomassa de produtores. Conforme esperado, NT e PT medidos responderam expressivamente à manipulação de N e P. Nas semanas que antecederam a introdução de girinos, a biomassa algal cresceu em resposta ao aumento na concentração do nitrogênio, mas não do fósforo; esta tendência se manteve ao longo dos experimentos. Isso provavelmente ocorreu porque concentrações medidas de NT variaram da oligotrofia à hipertrofia; por sua vez, concentrações medidas de PT estiveram consistentemente em valores considerados hipertróficos. A introdução de girinos não exerceu efeitos evidentes sobre a concentração de clorofila a, possivelmente indicando que o alimento era abundante o suficiente para não ocorrer competição entre consumidores. O rendimento de consumidores foi significativamente influenciado pela manipulação de nutrientes. Tratamentos ricos em P levaram a uma redução no crescimento de Physalaemus cuvieri, e uma tendência à redução no seu desenvolvimento e sobrevivência. De forma similar, o cenário de fertilização de soja - portanto rico em P - levou a uma redução significativa na sobrevivência de H. faber. Por sua vez, o cenário de fertilização de cana-de-açúcar levou a um ganho de massa de H. faber 30% superior ao controle. Formulamos a hipótese que esta diminuição no crescimento e sobrevivência de girinos em cenários de fertilização ricos em P teria sido causada pela proliferação de cianobactérias, frequentemente impalatáveis ou mesmo tóxicas, e que tendem a ser favorecidas em ambientes de baixa proporção N:P. Ao contrário do que esperávamos, ninfas de libélulas não foram capazes de promover redução direta ou indireta no crescimento ou sobrevivência de girinos, ou influenciar a biomassa de fitoplâncton. Concluímos que as alterações nos níveis de nutrientes que acompanham a conversão de ambientes nativos para paisagens agrícolas tem efeitos importantes sobre cadeias alimentares aquáticas (controle \'bottom-up\'), e que estes efeitos são muito mais fortes do que aqueles mediados pela presença de predadores (controle \'top-down\'). No entanto, estes efeitos não são necessariamente intuitivos, uma vez que determinados padrões de fertilização influenciam negativamente o rendimento de elos intermediários da cadeia e, portanto, a transmissão da biomassa para níveis tróficos superiores. / The human demand for food, fibers and biofuels imposes a pronounced growth in intensive agriculture and, with it, in the consumption of fertilizers. Fertilizer management promotes dramatic changes in the quantities and proportions of nutrients in agroecosystems which, in turn, influence biological communities through changes in the biomass, nutritional quality, and composition of producers. A recent trend in the Brazilian productive scenario is the expansion of biofuel crops - notably sugarcane (ethanol) and soybean (biodiesel) - over the cerrados. These crops, covering no less than 30 million hectares in the country, demand contrasting fertilization patterns: sugarcane requires high proportions of nitrogen relative to phosphorus, whereas soybean requires high proportions of phosphorus relative to nitrogen. The objective of this thesis was to test the direct and indirect effects that changes in the quantities and proportions of N and P that follow conversion of native habitats to sugarcane and soybean fields have on biological communities. These objectives were achieved by means of two mesocosm experiments manipulating nutrients, algae, tadpoles and dragonfly naiads. The first experiment followed a full factorial design crossing three concentrations of N (Nreference, Nsoy, Ncane) and three concentrations of P (Preference, Psoy, Pcane), as well as two trophic levels (algae and Physalaemus cuvieri larvae). Nutrient levels simulated conditions measured in water bodies of the cerrado, or patterns of fertilization known for sugarcane or soybean crops. In the second experiment we paired these concentrations (Nreference Preference, Nsoy Psoy, Ncane Pcane) so as to reduce dimensionality and as such to permit inclusion of a third trophic level (algae, Hypsiboas faber larvae, dragonfly naiads). We included an additional treament with caged naiads to separate the effects of actual tadpole predation from those of changes in tadpole traits. The main response variables in both experiments were cholophyll a concentrations as na indicator of phytoplankton biomass, and tadpole mortality, final mass and final developmental stage as indicators of consumer performance. A comparison in the concentration of chlorophyll a before and after the introduction of tadpoles tested the effect of consumers on producer biomass. As expected, measured TN and TP responded strongly to N and P manipulations. In the weeks preceding the introduction of tadpoles, algal biomass increased in response to the concentration of N, but not to the concentration of P; this trend was maintained throughout the experiment. This probably happened because measured TN varied from oligotrophic to hypertrophic conditions; in contrast, measured TP consistenly corresponded to hypertrophic conditions. Tadpole introduction had no evident effects on chlorophyll concentrations, possibly indicating that food was abundant and competition among tadpoles, if present, was weak. Consumer performance was significantly influenced by nutrient manipulations. P-rich treatments led to a reduction in growth, and to a reduction trend in development and survivorship, of Physalaemus cuvieri. Similarly, the scenario of soybean fertilization - therefore P-rich - led to a significant reduction in Hypsiboas faber survivorship. The scenario of sugarcane fertilization led to a 30% increase in mass gain in Hypsiboas faber. We hypothesized that this reduction in tadpole growth and survivorship in P-rich environments was caused by a proliferation of unpalatable or even toxic cyanobacteria, which are favored in low N:P ratio environments. Contrary to our expectations, dragonfly naiads did not cause a direct or indirect reduction in tadpole growth or survivorship, nor influence phytoplankton biomass. We conclude that changes in nutrient levels that accompany the conversion of native habitats to agricultural fields can have strong effects on freshwater food chains (\'bottom-up\' control), and that these effects are much stronger than those mediated by predator presence (\'top-down\' control). However, bottom-up effects are not necessarily intuitive as certain fertilization patterns negatively influence the performance of intermediate consumers and, as such, the transmission of biomass to higher trophic levels.

Dans les zones euphotiques des écosystèmes côtiers, les producteurs primaires se développent à la fois dans l’espace pélagique (colonne d’eau) et benthique (sur ou près du fond). Dans ces écosystèmes, le couplage bentho-pélagique prend une place non négligeable qu’il convient d’étudier. Les consommateurs primaires benthiques en sont des acteurs à travers leur activité alimentaire. La relation trophique entre les différents producteurs primaires et les consommateurs primaires benthiques reste toutefois mal élucidée. Pour apporter des éléments d’informations, cette thèse s’est focalisée dans un premier temps sur la dynamique spatio-temporelle des producteurs primaires au sein de la matière organique particulaire en suspension (MOPS), notamment au niveau de l’interface eau-sédiment. Cette étude a permis de montrer une forte variation saisonnière de ce peuplement algal à toutes les profondeurs, mais également une importante contribution relative des diatomées benthiques dans la MOPS près du fond en hiver. Le second axe de recherche s’est concentré sur la relation trophique entre des consommateurs primaires benthiques (coquille Saint-Jacques, pétoncle noir, moule et ormeau) et les producteurs primaires présents dans la MOPS. L’analyse des isotopes stables dans les tissus mous des pétoncles noirs et des moules a indiqué une prédominance du phytoplancton assimilé et une potentielle contribution du microphytobenthos dans leur régime alimentaire (plus importante pour la moule), notamment en période hivernale et près du fond. Cependant, ces études ont mis en lumière l’effet du métabolisme sur les valeurs isotopiques dans les tissus mous. Un aspect novateur de cette thèse a été d’utiliser les exosquelettes des coquilles Saint-Jacques et des ormeaux pour caractériser leur régime alimentaire dans le milieu naturel, afin de s’affranchir des problèmes liés à l’utilisation des isotopes stables dans les tissus mous. Des études en conditions contrôlées ont validé l’intégration des sources d’alimentation à travers le carbone métabolique dans les carbonates de la coquille Saint-Jacques et la couleur de la coquille pour l’ormeau. Des études complémentaires doivent toutefois être menées. Enfin, une dernière étude s’est focalisée sur la capacité de la coquille Saint-Jacques à mettre en suspension les particules benthiques par leurs mouvements de valves. / In coastal euphotic areas, primary producers grow both in the pelagic (water column) and in the benthic zone (in or near the sediment). In these ecosystems, benthic-pelagic coupling is an important process which must be well described. Benthic primary consumers modify this coupling through their diet. Trophic relationships between primary producers and benthic primary consumers are not thoroughly understood. To gain additional information, this thesis was focused on the spatial and temporal dynamic of primary producers in suspended particulate organic matter (SPOM), especially at water-sediment interface. This study highlighted a seasonal variation of microalgal populations at all depths, and a relative dominance of benthic diatoms in SPOM near sediment and in winter. The second objective of the thesis was focused on the trophic relationships between benthic primary consumers (great scallop, black scallop, mussel and ormer) and primary producers in SPOM. Stable isotope analysis in soft tissues of black scallops and mussels showed a predominance of assimilated phytoplankton and the potential contribution of microphytobenthos in their diet (more important for mussel), especially in winter and near the sediment. However, this study highlighted the impact of metabolism on stable isotopic values of soft tissues. An innovative aspect of this thesis was the use of exoskeleton of great scallops and ormers to characterize their diet in natural environment and thus overcome problems related to the utilization of stable isotopes in soft tissues. Experiments in controlled conditions validated food sources integration through metabolic carbon in carbonates of great scallops and shell color for ormer but complementary studies are required on mollusk shell utilization. The last part of this thesis was focused on the ability of great scallops to achieve resuspension of benthic particles by valve movements.

Os ecossistemas aquáticos têm sido afetados pelo uso intensivo de agrotóxicos e pela introdução de espécies exóticas. Para melhor avaliar os efeitos diretos e indiretos dessas atividades sobre as comunidades aquáticas, é essencial a utilização de experimentos em escala mais próxima a real. O objetivo desta pesquisa foi analisar os efeitos do agrotóxico Vertimec® 18 CE, cujo princípio ativo é a abamectina, sobre a comunidade planctônica através de um experimento em mesocosmos com duração de 20 dias. Para reproduzir relações biológicas mais complexas e analisar o impacto da introdução de uma espécie exótica sobre o plâncton, inseriram-se girinos de Lithobates catesbeianus em alguns tratamentos. Foram consideradas duas concentrações do agrotóxico (2 &#956;L L-1 em dose múltipla e 8 &#956;L L-1 em dose única) na ausência e presença de girinos, além de dois tratamentos não contaminados (controle com e sem girinos). Variáveis físicas, químicas e biológicas (clorofila a, aspectos qualitativos e quantitativos do fitoplâncton e do zooplâncton) da água e do sedimento foram monitoradas. A partir dos resultados obtidos, verificou-se que os girinos causaram um aumento na concentração de nutrientes na água e uma diminuição do oxigênio dissolvido e da densidade planctônica. Nos tratamentos contaminados, como um efeito indireto do incremento da concentração de nutrientes e da mortalidade das espécies de Cladocera e Copepoda, a densidade do fitoplâncton e de algumas espécies tolerantes de Rotifera aumentou. No tratamento com a concentração mais elevada do agrotóxico, registrou-se a morte de todos os girinos, proporcionando um aumento na concentração de nutrientes na água e na densidade de algumas espécies fitoplanctônicas, que alcançaram uma densidade total de 96.919 ind mL-1. Estes resultados indicam que a inserção dos girinos e a adição do agrotóxico causaram efeitos diretos e indiretos sobre a comunidade planctônica e a qualidade da água. Conclui-se que os mesocosmos podem contribuir para a compreensão de tais efeitos e oferecer subsídios ao manejo de ecossistemas de água doce impactados por atividades antrópicas. / Aquatic ecosystems have been impacted due to the intensive use of pesticides and the introduction of exotic species. It is essential to use more realistic experiments for assessing direct and indirect effects of these activities on aquatic communities. This thesis aimed to analyze the effects of the pesticide Vertimec® 18 EC, whose active ingredient is abamectin, on plankton community through a mesocosm experiment during 20 days. Tadpoles of the species Lithobates catesbeianus were introduced in some treatments to simulate complex biological relationships and to evaluate the impact of the introduction of such exotic species on plankton. Different treatments were analyzed, including one case of no contamination (control) and two cases of contamination: 2 &#956;L L-1 (multiple dose) and 8 &#956;L L-1 (single dose). Such conditions were considered both with absence and presence of tadpoles. Physical, chemical and biological (chlorophyll a, qualitative and quantitative aspects of phytoplankton and zooplankton) variables of water and sediment were also monitored. The tadpoles caused an increase in the concentration of nutrients and a decrease of dissolved oxygen concentration and plankton density. In the contaminated treatments, as an indirect effect of the nutrient increase and the mortality of species of Cladocera and Copepoda, the density of phytoplankton and some tolerant species of rotifers increased. In the treatment with the highest concentration of the pesticide, all the tadpoles died. As a result of the death of the tadpoles, there was nutrient enrichment in the water, with an associated increase in the density of some phytoplankton species, reaching a total density of 96,919 ind mL-1. These results suggested that direct and indirect effects occurred on plankton community and water quality through the addition of the contaminant and the tadpoles. We conclude that the mesocosms might contribute to the understanding of these effects and provide subsidies for management of freshwater ecosystems impacted by human activities.

L'écologie a vu se développer, depuis une vingtaine d'années, une vision systémique se concentrant sur les interactions entre composantes de l'écosystème. L'association de ces composantes fait émerger des propriétés globales et agit en retour sur les propriétés des composantes. Ces propriétés sont dites émergentes. Leur étude a été appliquée aux réseaux trophiques de différents écosystèmes, pour la plupart littoraux, par des approches numériques visant à construire et analyser des modèles statiques. Une telle approche fait le choix de négliger certains aspects : structures spatiales, dynamiques temporelles, forçage des processus, dans l'objectif de favoriser la prise en compte d'un maximum de compartiments dans une vision synthétique. Ce travail repose entièrement sur les résultats acquis sur les différents compartiments et processus, avec comme objectif d'en faire une synthèse, de les confronter, et de dégager des propositions d'étude de nouveaux processus ou compartiments peu renseignés. De ma thèse à aujourd'hui, j'ai travaillé sur des questions relatives aux réseaux trophiques planctoniques avec, dans chaque situation, un contexte appliqué et un objectif fondamental. Celui-ci consiste à comprendre le fonctionnement de ce réseau planctonique. L'axe directeur utilisé pour discuter de la nature de ces différents réseaux trophiques planctoniques est essentiellement celui de Legendre et Rassoulzadegan (1996) reposant sur un gradient entre deux états : le réseau microbien et le réseau herbivore classique. On a ainsi pu dégager de nouveaux modes de fonctionnement des réseaux trophiques planctoniques et ces propriétés dégagées viennent enrichir la réflexion menée sur le problème appliqué soulevé dans chaque milieu. Suite à mon recrutement à l'Université de La Rochelle, la demande du LBEM était que je travaille sur les vasières intertidales tempérées, en couplant benthos et colonne d'eau. Le choix de conserver une approche en modèles statiques afin de prendre en compte au maximum la diversité fonctionnelle imposait de modifier les outils numériques pour les adapter à cet écosystème macrotidal. Les innovations apportées ont principalement concerné la modification de l'algorithme de l'analyse inverse par une approche de Monte Carlo, le couplage spatial par les flux de transport et des saisons, le développement d'indices basés sur les chaînes de Markov (liens lents ou rapides entre compartiments). L'application de ces méthodes à l'ensemble des données acquises sur le site de Brouage dans les Pertuis Charentais a permis de dégager les caractéristiques suivantes : i) rôle primordial du réseau microbien benthique : microphytobenthos, bactéries hétérotrophes, foraminifères, ii) rôle de cul-de-sac de la matière organique par les foraminifères (la place exacte des foraminifères dans le réseau reste à déterminer), iii) la nématofaune est la plaque tournante du réseau, iv) le réseau est importateur de détritus et exportateur de matière de « qualité », au sens de l'exergie. A partir de ce modèle, des scénarios ont pu être simulés sur des scenarii de culture et la conséquence d'une chute des productions primaires sur la nourriture disponible. Le modèle couplé en boites spatiales a permis de dégager 3 modes de fonctionnement distincts en fonction de la hauteur sur la zone intertidale. Outre ce travail sur la vasière de Brouage un travail de comparaison inter – sites a été initié par l'établissement d'un modèle à deux saisons sur la vasière de l'Aiguillon, au nord des Pertuis Charentais. Le fonctionnement trophique de cette vasière est très proche de celui de Brouage. Il s'en démarque surtout par un rôle moindre des bivalves cultivés et une forte exportation de matière par les poissons brouteurs (mulets). Outre une indication sur l'évolution naturelle des écosystèmes, les propriétés émergentes des réseaux trophiques peuvent être indicatrices d'un état d'évolution sous la pression de l'Homme. Il est alors nécessaire de développer des indicateurs d'écosystèmes capables d'évaluer les changements des environnements côtiers. L'utilisation des indices d'analyse des réseaux devrait permettre de passer à une vision systémique, s'intéressant aux interactions plutôt qu'aux stocks. Seule une comparaison étendue de nombreux réseaux en utilisant toujours les mêmes techniques permettra d'aller plus loin, en cherchant à observer vers quelles propriétés évoluent les systèmes, naturellement et sous pression anthropique.

The main theme of this thesis is the interactions of animals with the environment and each other. The thesis was written within the framework of a biomanipulation project “Pikeperch in Himmerfjärden”. With the aim to investigate possible trophic pit-falls, give the manipulation the best possible start, and find ways to monitor the progression of the manipulation. In Paper I the diet of the invader cladoceran Cercopagis pengoi is analysed with stable isotopes; conducted prior to stocking. C.pengoi has a preference for large copepods, indicating possible competition with fish. Paper II investigates the behavioural differences between pikeperch fingerlings reared in different environments (pond vs. tank). Results suggest that fish reared in semi-natural ponds are more likely to survive directly after stocking. In Paper III and IV, the diel vertical migrations (DVM) of copepods are in focus. In Paper III the migrations of two copepod species: Acartia spp. and Eurytemora affinis are studied over season and life stage. The amplitude of migration was found to increase with ontogeny for both species, indicating evasion of visual predators. Paper IV examines the varying migratory patterns of adult female E. affinis finding that these animals migrate more actively when feeding conditions deteriorate and growth decreases. The overall conclusions of the thesis are that behavioural, not only direct trophic interactions are key when studying ecosystems. / <p>At the time of the doctoraldefense, the following papers were unpublished and had a status asfollows: Paper2: In press. Paper 4: Manuscript.</p>