Aquatic habitats are closely linked to surrounding terrestrial environments via reciprocal subsidies. Much of the research on these subsidies has been focused on streams, while subsidies across aquatic-terrestrial boundaries of other aquatic systems, like temporary ponds, have received little attention. Temporary ponds are seasonally inundated aquatic habitats and patterns of energy transport across transient aquatic-terrestrial boundaries are not well understood. To address the lack of information regarding these cross-habitat subsidies, I quantified: 1) leaf litter inputs, 2) amphibian egg inputs, 3) terrestrial insect inputs, 4) amphibian metamorph emergence, and 5) aquatic insect emergence for eight temporary ponds. Terrestrial insect inputs to ponds represented substantial high-quality subsidies to ponds that are generally unaccounted for in similar studies. While larger ponds produced greater total fluxes to terrestrial habitats, smaller ponds were often more productive per unit area. Therefore, a mosaic of small ponds may produce greater or equivalent subsidies and may be of greater value to terrestrial food webs than a single large pond. Resource subsidies of energy and nutrients can be transported via physical forces or biotic processes, such as animal migration or emigration. I quantified net nutrient fluxes associated with amphibian migrations across aquatic-terrestrial boundaries of eight temporary ponds in Illinois. I measured carbon (C), nitrogen (N), and phosphorus (P) subsidies from forests to ponds, in the form of eggs, and the reciprocal subsidies from ponds to forests via juvenile emigration. Juvenile emergence biomass did not resemble egg biomass, as not all species bred successfully in each pond, resulting in variability in the magnitude of nutrient fluxes both across ponds and species. The forest was not always the recipient system of net nutrient fluxes. Hydroperiod, trophic interactions, and species composition explained some dynamics of N and P subsidies. Most studies have focused on the magnitude of cross-boundary fluxes, but the impact of a subsidy is mediated by the size of the flux as well as the quality and relative abundance of similar resources in the recipient habitat. Long-chain polyunsaturated fatty acids (LC-PUFAs) are necessary for proper physiological function, are unevenly distributed across the landscape, and animals differ greatly in their ability to synthesize them de novo. This creates the potential for limitation and increases their possible importance as a subsidy. I examined LC-PUFA tissue concentration and export in eight species of emerging amphibian metamorphs across eight temporary ponds in a wetland complex. I found that tissue concentrations and export of LC-PUFAs varied across species, but were generally within the ranges of several freshwater fish and aquatic insects. Anurans exported higher amounts of LC-PUFAs than salamanders, largely due to the higher emergence biomass of anurans. Further, I explored fatty acid profiles and immune function of wolf spiders in wetland and upland habitats. Wetland spiders had higher tissue levels of aquatic-derived LC-PUFAs and elevated immune function compared to upland spiders. These patterns suggest aquatically derived nutrients, such as LC-PUFAs, may benefit organisms exploiting freshwater habitats.

I conducted surveys of amphibian, reptile, and small mammal communities surrounding 4 isolated, upland and 6 stream-connected temporary wetlands on Tombigbee National Forest and Noxubee National Wildlife Refuge in north Mississippi from May 2001 - March 2006. Trap captures yielded 17 amphibian species (n = 11,142), 21 reptile species (n = 541), and 10 small mammal species (n = 472). Upland pools supported greater diversity of Ambystomatid salamanders, anurans, lizards, and mice (Peromyscus spp.), than floodplain pools. Factors including landscape position of pools, proximity to alternate water sources, and barriers to dispersal potentially influenced faunal communities of temporary wetlands. Infrared-triggered cameras were used to monitor mammalian activity surrounding ephemeral wetlands to determine potential depredation of pitfall traps. Raccoons (Procyon lotor) accounted for most images (35% total images). I assessed relative effectiveness of sampling techniques to capture and retain herpetiles in presence of depredation. I tested pitfall trap, funnel trap, and pitfall trap combined with exclusion cover along driftence arrays. Overall, pitfall traps of both designs produced more captures than funnels, and excluded-pitfalls yielded greater captures than un-excluded pitfalls for most amphibians. Overall mortality rates were <2% of total captures with anurans accounting for most (63.30%) mortality. I submit that in long-term studies, pitfall traps with exclusion are prudent to limit sampling bias and mortality occurring with depredation of captured herpetofauna. During trapping, pit-traps of both designs yielded incidental captures of small mammals. Capture rates for small mammals were similar in un-excluded pitfall traps and excluded pitfall traps. Southern short-tailed shrews (Blarina carolinensis) and mice accounted for 93% of total captures and suffered 76% and 52% mortality, respectively, potentially due to exposure, starvation, flooding, and/or depredation. Additionally, one species listed as rare in Mississippi was captured during herpetofaunal surveys, oldfield mouse (Peromyscus polionotus). Due to potential impact of pitfall trapping on small mammals, I recommend that researchers either alter trapping methods to address non-target hazards (frequency of checking traps, providing shelter) or work cooperatively using an integrated survey approach for herpetiles and small mammals to limit trap mortality.

Cette thèse porte sur l'étude de la relation Homme-Mare temporaire dans la région de Benslimane. Une approche pluridisciplinaire intégrant l'écologie et la sociologie a été adoptée, afin d'apporter des réponses adaptées aux problématiques liées à ces habitats et de permettre leur développement durable. Le premier volet a pour but d'évaluer les facteurs clés déterminant la richesse de la végétation des mares de cette région. Les résultats obtenus révèlent l'influence prédominante des facteurs locaux, avec un effet significatif des facteurs régionaux et de la vulnérabilité des mares aux usages, sur l'expression de la richesse floristique. Les activités anthropiques affectent différemment la richesse en espèces Préférentielles. Le second volet se focalise sur les causes de régression des mares, en couplant l'évaluation des menaces à une enquête socio-économique. Les résultats montrent la prédominance d'une perception anthropocentrique des mares temporaires. Ce constat traduit le haut niveau de menace qui pèse sur les mares de cette région. L'évaluation de ces menaces, qui indique que 22 % des mares sont vulnérables et que 23 % sont menacées de destruction à court terme, soulignent le besoin urgent de nouvelles politiques environnementales et d'approches innovantes de gestion des mares temporaires. Enfin, le troisième volet s'intéresse à l'impact du pâturage sur la végétation des mares. Les résultats montrent que cette perturbation affecte la végétation des mares, à l'échelle régionale et locale. Ces résultats sont interprétés comme résultant de l'effet sélectif des herbivores et la tolérance différentielle des espèces aux perturbations. Au terme de cette étude, la gestion intégrée des mares apparaît comme le moyen le plus approprié pour le développement durable de cet "éco-socio-système" complexe, en conciliant développement économique et bon état écologique des ressources, et en liant les questions environnementales, économiques et sociales. / This thesis focuses on the study of the relationship Man-Temporary pool. A multidisciplinary approach integrating ecology and sociology has been adopted to provide appropriate responses to the problems related to these ecosystems and to allow their sustainable development. The first section aims at evaluating the key factors determining the vegetation richness within regional pools. The obtained results reveal the predominant influence of local factors, with a significant effect of regional factors and the vulnerability of pools to use, on the expression of the vegetation. Human activities affect differently the richness in preferential pool species. The second section focuses on the major causes of the regression of pools, by coupling threats assessment and a socio-economic survey. The results point out the anthropocentric perception of pools. This translates the high level of threat concerning the temporary pools of Benslimane region. The thr eat assessment, which shows that 22 % of pools are vulnerable, and 23 % are threatened of destruction of the short term, highlights the urgent need for new environmental policies and innovative approaches of temporary pools management. Finally, the third section concerns the impact of grazing on the vegetation of temporary pools. The results demonstrate that grazing affects the vegetation of pools, both at the regional and the local scale. These results are interpreted as a result of the selective influence of herbivores, and the differential tolerance of species to disturbance. At the end of the present study, the integrated management of pools appears as the most appropriate means for the sustainable development of this complex eco-socio-system, by combining the economic development and the ecological status of resources, and by linking environmental, economic and social issues.