En Lorraine, le drainage fréquent des sols argileux favorise les transferts des intrants agricoles (pesticides et engrais) vers les eaux de surface. Dans ce cas, les bandes enherbées réglementaires (BCAE de la PAC) installées en bordure de cours d’eau sont court-circuitées et ne jouent plus leur rôle épurateur vis-à-vis des eaux de surface. Ainsi, des zones tampons humides artificielles (ZTHA) sont préconisées pour prendre le relais des bandes enherbées. Généralement, ces ZTHA sont installées à l’échelle du bassin versant ou du groupe de parcelles et sont donc de taille conséquente, les rendant concurrentielles de la surface agricole utile (SAU). L’objectif de ce projet de recherche est donc d’évaluer l’efficacité de deux ZTHA de taille réduite, installées dans la bande enherbée entre la parcelle drainée et le ruisseau. Pour cela, une approche multi-échelles intégrant des expériences de laboratoire en batch, des suivis dynamiques en bacs pilotes et un monitoring à la parcelle a été mise en place. Afin d’évaluer l’efficacité des ZTHA au cours des saisons de drainage 2012/13 et 2013/14, le suivi des teneurs en nitrate et de 79 pesticides a été réalisé dans les eaux en entrée et en sortie des ZTHA, ainsi que dans diverses matrices environnementales (sols, sédiment, paille, végétaux). Pour les deux dispositifs, la réduction du flux de nitrate s’est montrée stable, comprise entre 5,4 et 10,9 %, alors que la réduction du flux de pesticides s’est montrée très variable, comprise entre -618,5 et 100 %. Cette variabilité a été expliquée par quatre comportements distincts des molécules. Les fortes valeurs négatives d’efficacité ont été expliquées par une contamination des dispositifs à la suite d’épisodes de ruissellement connexes pour des molécules solubles telles que le 2,4-MCPA ou l’isoproturon. Pour ces deux molécules, l’étude de sorption a montré leur faible affinité avec les matrices environnementales, facilitant leur transport par les eaux de ruissellement et de drainage. D’autres molécules, telles que le boscalide ou l’OH-atrazine, ont présenté des abattements proches de 0. Pour le boscalide, ce comportement a été montré au laboratoire, en accord avec ses fortes capacités d’adsorption et de désorption observées lors des études de sorption et en bacs pilotes. Enfin, certaines molécules ont montré des abattements, allant de 9,9 à 100 %, en lien avec leur faible persistance (2,4-MCPA, clopyralid, mésosulfuron-méthyl,…) ou leur forte affinité pour les matrices environnementales (diflufenicanil, propiconazole, propyzamide,…). Ces ZTHA de taille réduite, installées dans les bandes enherbées, sont donc efficaces pour limiter la contamination des eaux de surface par les eaux de drainage agricole. Ainsi, la multiplication de ces dispositifs à l’échelle de la parcelle dans le paysage agricole permettrait de réduire la contamination des eaux de surface par les pesticides et les nitrates tout en préservant la SAU. Toutefois, l’efficacité de ces ZTHA restant limitée vis-à-vis de certains pesticides présentant un Koc faible et une DT50 élevée, leur installation ne doit pas remettre en cause la réduction en amont des apports phytosanitaires sur les parcelles / In Lorraine, the drainage of clay soils contributes to transfer of agricultural inputs (pesticides and fertilizers) towards surface waters. In this case, the regulatory grass strips (CAP regulation) installed along rivers are bypassed and no longer effective in purifying surface water. Thus, constructed wetlands (CWs) are recommended to take over grass cover bands. Generally, these CWs are installed at watershed scale or at that of several plots and they are therefore quite huge, making them competitive in the cultivated land. The objective of this project is to assess the effectiveness of two small CW, installed in the grass cover band between the drained plot and the river. For this, a multi-scale approach was set up integrating laboratory experiments in batch, pilot study taking into account the water dynamics and monitoring at field scale. The effectiveness of CW was assessed during the 2012/13 and 2013/14 drainage seasons. The nitrate and 79 pesticides concentrations were monitored at inlet and outlet of CWs and in various substrates (soils, sediment, straw and plants). For both devices, the removal efficiencies measured for nitrate were stable and ranged from 5.4 to 10.9 % of inlet amounts while the reduction of pesticide flows was very variable, ranging from -618.5 to 100 %. This variability was explained by four distinct behaviors. The high negative efficiency values were attributed to neighboured runoff events after heavy rains, for soluble molecules such as 2,4-MCPA or isoproturon. For both molecules, the sorption study showed low affinity for environmental substrates, facilitating their transport by runoff and drainage water. Other molecules such as boscalid or OH-atrazine showed effectiveness close to 0 %. For boscalid, this behavior has been observed in laboratory sorption and pilots studies, in agreement with its strong adsorption and desorption capacities. Finally, some molecules have shown reductions ranging from 9.9 to 100 %, in agreement with their low persistence (2,4-MCPA, clopyralid, mesosulfuron-methyl, ...) or their high affinity for environmental substrates (diflufenican, propiconazole or propyzamide). These small CWs, located in the grass cover bands, are effective in minimizing the contamination of surface waters by agricultural drainage water. Thus, multiplying the number of these devices at the plot scale in the agricultural landscape would reduce the contamination of surface waters by pesticides and nitrates while preserving the cultivated land. However, the effectiveness of these CWs are limited for pesticides with a low Koc and a long DT50; thus their installation in plot must not call into question the reduction in the use of pesticides

The increasing demand for and limited supply of fresh water necessitates an understanding of how human actions affect aquatic ecosystems. Anthropogenic impacts to these ecosystems occur in many forms including eutrophication, invasive species removals, and hydrologic alterations. Ground water extraction is one such action that can dramatically impact wetland hydrology and is increasing in occurrence globally as clean surface water resources are exhausted. Despite the importance of ground water extraction to meet human demand, little information is available concerning the response of vegetation communities to chronic ground water extraction. Over extraction is known to result in reduced water levels and duration, resulting in a shift towards more upland tolerant species; however, detailed information concerning the response of the individual species comprising these communities and how wetlands shift along with pumping regime remains unavailable. The following dissertation combines historical hydrology and ground cover vegetation data with recent monitoring to describe how ground cover (herbaceous species) and canopy (tree species) vegetation respond to fluctuations in hydrology and ground water extraction. Ground cover communities were extremely diverse with a total of 103 species being sampled in the historical ground cover vegetation dataset. Juncus repens was the most widely distributed species and was observed in 36% of all samples. The 29 species most widely observed in the ground cover strata (height) displayed relatively narrow ranges of preferred water depth and duration with Amphicarpum muhlenbergianum being found in the driest areas and Pontederia cordata and Ludwigia repens the wettest. In general species found in shallower water depths also tended to be found in locations with shorter hydroperiods, although woody species tended to found in areas with relatively shallow water depths with extended hydroperiod. Ground cover vegetation is extremely useful as an indicator of recent hydrology, although the hydrologic preference of the species in the current study does not reflect the assumed ecology of the species utilized by Florida Administrative Code 62-340.450. Additional research to validate and improve the accuracy of this classification system is required. When ground water extraction volumes in well fields was significantly reduced, ground cover communities were responsive, as was indicated by Permanova results (Before After Control Impact). All ground cover at wetlands located within well fields became more indicative of wetter conditions while control wetlands responding only to climate and weather all became drier. In contrast, several well fields displayed reductions in water levels and hydroperiod following extraction reductions. The shift in ground cover community indicates that ground water extraction has not produced an alternative stable state and restoration of these ecosystems is possible through alterations in ground water extraction volumes alone. As ground water extraction volumes were increased, tree communities responded by displaying increased occurrance of non-Taxodium sp. trees, mortality of wetland tree species, and light availability. All wetlands remained dominated by mature Taxodium sp. regardless of the amount of ground water impact indicating that each wetland has not yet shifted into a new community type as a result of non-Taxodium tree encroachment; however, recruitment and mortality patterns of both Taxodium and non-Taxodium species indicate this may occur in the future. Changes in light availability at the wetland floor associated with tree species is likely providing an additional feedback mechanism on ground cover communities. Results from this dissertation indicate that vegetation communities are extremely responsive to changes in hydrology and have shown significant changes associated with ground water extraction. These changes may not be permanent; however, and alterations in extraction volumes and timing can provide changes in vegetation communities even after decades. Routine long term monitoring should be conducted, in addition to critical assessments of current extraction volumes, to assess the current status of vegetation ecosystems and allow for individuals to best manage aquatic resources for all uses.

The Green Swamp Preserve is a large geographic area that has sustained many changes since Europeans settled in Florida. There has been little published research on the impacts of anthropogenic activity on this system. This thesis research seeks to document more recent changes in the Green Swamp and to evaluate the effects of various human activities on the system. The study period is from 1985 to 2015. For this time period changes in land use and landcover were examined using neural network classifications. Changes in vegetation health were evaluated by examining Normalized Difference Vegetation Index (NDVI) and Green Vegetation Index differences. Field site visits were made to document current conditions at thirty sample locations within the study area. Changes in land use and landcover and vegetation health were evaluated in relation to anthropogenic activities such as proximity to pollution sources, conservation lands and restoration sites. Statistical analysis was conducted to determine if statistically significant clustering occurred in these changes and if present geographically weighted regressions were performed to determine if a significant spatial relationship existed between the clustering and the various human activities. WAP data showed an overall decline in wetland health at the assessment sites and showed a trend of lower wetland health at sites within 2 Km of pollution sources, specifically petroleum tank contamination sites and state roads. The statistically significant clustering identified in land use landcover changes from 1985 to 2015 were in relation to changes from field, forested and wetland landcover types to built environments. Spatial relationships were identified between the proximity of petroleum tank contamination sites, state roads and solid waste facilities and clustering of NDVI decreases from 1985 to 2015. NDVI increases in the study area from 1985 to 2015 also showed statistically significant clustering in relation to conservation lands and lands purchased by the Southwest Florida Water Management District for environmental protection. These preliminary findings suggest that human activities may have influenced changes in the health of the Green Swamp. Further, more extensive research is suggested to confirm these findings.

Please read the abstract in the section 00front of this document / Dissertation (MSc (Botany))--University of Pretoria, 2006. / Plant Science / unrestricted

This thesis examines the role of environmental non-government organizations in relation to the role and responsibility of government for wetland conservation in British Columbia. The role of a small community-based environmental organization called the Burns Bog Conservation Society in decisions about Burns Bog, a large privately-owned wetland located in the Lower Mainland, is analyzed. Drawing from a broad range of literature on environmentalism and conservation, the thesis begins with an historical account of the environmental movement, giving a broad view of environmental conservation and the role of environmental organizations. A typology of roles and analytical framework is subsequently derived from a study of environmental non-government organizations involved in managing water resources in British Columbia's Fraser River Basin. Three main roles, advocacy, supplemental and transformative, are discussed as they relate to activities prior to and during the Burns Bog Analysis, a provincially established land use study of Burns Bog. This research found that a small community-based environmental organization, through advocacy, can ensure that ecological integrity is not neglected as a result of poor planning and decision-making. The case study demonstrates that a small environmental organization can pressure governments to broaden their view of land-use and environmental issues, assuring more informed decision-making. An environmental organization can also supplement the regular roles and responsibilities of government, first by supplying a service that government is unable or unwilling to provide, and second by participating in and legitimizing consensus-based decision-making processes. The case study demonstrates how a small environmental organization can provide and review information and participate in creating options and solutions in land-use decisions. Finally, an environmental organization can influence, overtime, the fundamental restructuring of government planning and decision-making processes and transform the way society thinks and operates. The thesis concludes that environmental non-government organizations have an important role to play in ensuring the recognition and conservation of British Columbia's wetlands in government planning and decision-making processes; in informing and educating government and the public about ecological systems and their values; and in counter-balancing strong economic, political and development interests. / Applied Science, Faculty of / Community and Regional Planning (SCARP), School of / Graduate

As habitat loss and fragmentation are major causes of decline in animal species, studying habitat requirements in these species is a key component of their recovery. I investigated the relationship between landscape composition and habitat use of Blanding’s turtles, Emydoidea blandingii, a freshwater turtle threatened by habitat loss and road mortality on most of its Canadian range. In 2010, I conducted a radio-telemetry survey of 44 Blanding’s turtles in southern Québec, Canada, and modelled their home range size from land cover proportions measured at many spatial scales. I also used data from a visual survey conducted in 2008 and 2009 to model wetland occupancy of the species at the landscape scale. Home range size of the Blanding’s turtle was significantly correlated to landscape composition, and the proportions of agriculture, open water and anthropogenic lands had the strongest relationships with home range size. However, those relationships were weak and the models were unable to predict home range size accurately. At the landscape scale, land cover and road density poorly predicted probability of occurrence, and Blanding’s turtles occupied wetlands in both disturbed and natural sites. Management of the species should focus on protecting sites of occurrence with high wetland density, low road density, and sufficient suitable habitat to cover their seasonal movement patterns.

Glyphosate-based herbicides are the dominant pesticide on the market and are utilized worldwide in both the agricultural and forestry industries. Their prevalence comes at a time when concern over the potential effects of pesticide application in amphibian spawning grounds is growing. The primary goal of this thesis was to determine if the glyphosate-based herbicide WeatherMax® has the potential to disrupt sexual differentiation and growth in the wood frog (Lithobates sylvaticus) in a pulse exposure at the predicted maximal environmental concentration (PMEC) of 2.88 mg acid equivalent per liter. This was carried out in laboratory, mesocosm and in-situ field exposures, in an attempt to determine how a potential disruption might vary between experimental environments. In this study, tadpoles from three split-wetlands targeted at the PMEC for WeatherMax were found to display no significant change in survival or growth, however gene expression of several genes involved in steroidogenesis during sexual differentiation (cyp19, cyp17, star, foxl2) were found to be affected. The effects on these genes appeared to be dependant on the exposure concentration of WeatherMax in each wetland, which varied even though all three wetlands were meant to target the PMEC. The wetland that was measured as having the highest herbicide concentration (PMEC 13) was found to have a female biased sex ratio. The results found in the field varied from those found in the more artificial exposures. In the laboratory the PMEC of WeatherMax experienced complete mortality, whereas in the mesocosms survival was not significantly affected. Sex ratios were unaffected in the laboratory, however at the PMEC there was a significant male bias in the mesocosms. The discrepancies in the results obtained from the different exposure types highlights the importance of real world exposures. That the same concentration that caused complete mortality in the laboratory caused sublethal effects in the field is of importance as it denotes that these endpoints may not be easily investigated in these synthetic exposures. This project is a part of the Long Term Experimental Wetlands Area (LEWA) and contributes to the body of information amassed therein on the impact of a glyphosate-based herbicide on amphibians in a wetland ecosystem.

This research project evaluated the overall hydroperiod and effects of monitoring well design parameters on observed levels of saturation in created wetlands with high-clay subsoils at the Cedar Run 3 mitigation bank site in Prince William County, Virginia. Three complete replications of an electronic central array and an associated surrounding array of manually monitored wells and piezometers were installed. The electronic arrays contained a U.S. Army Corps of Engineers (USACOE) standard monitoring well, as well as piezometers and tensiometers at three depths. The manually monitored well + piezometer arrays (3 per location; 9 total) consisted of 12 variants of screen types and filter pack materials, well diameter, and unlined bore holes. The site exhibited a complex seasonal hydroperiod ranging from ponded winter conditions to deep (< -50 cm) summer dry down. The site also exhibited epiaquic (perched) conditions following summer and fall precipitation events. Apparent water levels in deep (> 1 m) piezometers exhibited an unusual hydroperiod with highest levels in summer. Differences in well/piezometer diameter, design, and packing texture/fit produced surprisingly different apparent water levels that varied from ~ 4 to over 28 cm during both the winter ponded periods and summer subsoil water table flux periods. Thus, one important finding is that relatively simple differences in well designs can have dramatic effects on observed water levels. Overall, the standard USACOE appeared to be relatively accurate for predicting saturation levels during ponded periods, but nested piezometers are preferred and more accurate for the drier summer and fall. / Master of Science

Groundwater seepage to surface water is an important process to peatland ecosystems; however, the processes controlling seepage zone distribution and magnitude are not well understood. This lack of process-based understanding makes degraded peatland ecosystems difficult to restore and problematic for resource managers developing a sustainable design. Degraded peatlands, particularly abandoned cranberry farms, often have drainage ditches, applied surface sand, and decreased stream sinuosity to artificially lower the water table and support agriculture. These modifications disconnect the surface and groundwater continuum, which decreases thermal buffering of surface water significantly. The combination of a decreased influx of thermally buffered groundwater, a naturally low surface gradient, minimal canopy, and strong solar input causes surface water temperature extremes that degrade ecosystem health. Through strategically incorporating the natural processes to restore groundwater discharge to restored surface streams, surface water temperature extremes will be buffered promoting a healthy, resilient wetland ecosystem. Therefore, it is critical to understand the spatial hydrogeologic constraints that induce groundwater seepage. Here we examine the spatial relationship between surficial groundwater seepage and the subsurface hydrogeologic structure within a mineraltrophic peatland environment. We use multiple field methods to develop a process-based conceptual model of the ground water seepage development at the site; these methods include geophysical, thermal, and isotopic techniques. The results indicate that there are two distinct forms of groundwater discharge to the peatland platform: diffuse lower-flux marginal seepage and discrete higher-flux interior seepage. Both types of groundwater discharge develop through interactions with subsurface peatland basin structure, specifically when the basin slope is perpendicular to the regional groundwater gradient. These observations also allow insight into the formation of the groundwater discharge through time. The strong correlation between the subsurface basin structure and surficial groundwater expression will allow resource managers to more efficiently locate groundwater seepage on large, complex sites, and develop comprehensive management and restoration strategies for these critical ecosystems.

During recent decades the brownification through leaching of organic carbon into Northern hemisphere water bodies has become apparent. This is considered a problem because of the implications it has on ecosystems and freshwater sources used for drinking water. It could be beneficial if wetlands can be used for reducing brownification but also detrimental if wetlands instead increase it as construction of wetlands is common practice for treating agricultural runoff in Sweden.  This study compares inlet and outlet absorbance in five wetlands in Kalmar County in Sweden and finds that one removes colour, two increases colour and two does not change the colour of the water at all. The study also compares absorbance values with variables concerning flow and temperature and concludes that a higher absorbance at inlet increases removal similar to how nutrient removal in wetlands work. Inlet absorbance is also dependent on flow with higher flow resulting in higher absorbance. Periods of no flow did not contribute to any changes in absorbance, but this could be because of a limited sample pool. Future recommendations for research include a more extensive look into periods of no flow and affecting variables to ensure wetlands do not contribute to brownification.