This study was carried out to characterize ponds and aquaculture systems, and also to determine both the potential and actual impacts of pond aquaculture effluents on receiving stream quality in the Ashanti and Brong Ahafo regions of Ghana. Water, fish and macroinvertebrate samples were collected from upstream, downstream and nearby reference streams of, and questionnaires administered to, 32 farms. Total settleable solids were higher in ponds than reference streams (p = 0.0166); suspended solids was higher in ponds than reference streams (p = 0.0159) and upstream (p = 0.0361); and total phosphorus was higher in ponds than reference (p = 0.0274) and upstream (p = 0.0269). Total nitrogen was most clearly higher in ponds than all other locations: p = 0.0016, 0.0086 and 0.0154 for the differences between ponds and reference, upstream, and downstream respectively. BOD5 level was also higher in ponds than all locations (p = 0.0048, 0.0009, and 0.0012 respectively). Also, non-guarding fish species were more abundant in reference streams than downstream (p = 0.0214) and upstream (p = 0.0251), and sand-detritus spawning fish were less predominant in reference streams than upstream (p = 0.0222) and marginally less in downstream locations (p = 0.0539). A possible subsidy-stress response within study streams was also observed. Hence, ponds are potential sources of these water quality variables to receiving streams. Effluent-receiving streams, generally, were not much different from reference streams in terms of most the metrics of community structure and function used in the comparisons. Hence, even though receiving streams in Central Ghana may not be severely impacted by aquaculture effluents at the moment, the management of pond effluents will determine the scale of future impact. Vegetable, cereal, and livestock farming could serve as additional sources of fecal streptococci and coliform bacteria and nutrient-enrichment within the study area, besides aquaculture, and so these industries must also be included in efforts to minimize pollution of these streams. / Master of Science

Salinization of freshwaters by human activities is of growing concern globally. Salt pollution can cause adverse effects to aquatic biodiversity, ecosystem function, ecosystem services, and human health. In many regions of the world, and in coal-mining-influenced streams of the temperate forests of Appalachia USA, specific conductance (SC), a surrogate measure for the dissolved major ions composing salinity, has been linked to decreased diversity of benthic macroinvertebrates. However, assessments used to reach this conclusion have generally not accounted for temporal variability of salinity, as most studies use "snapshot" SC data collected concurrently with biological data at a single point in time. Effective management of salinization requires tools to accurately monitor and predict salinity while accounting for temporal variability. To improve those tools, I conducted analyses of 4.5 years of salinity and benthic macroinvertebrate data from 25 forested headwater streams spanning a gradient of salinity where non-salinity stressors were minimized. My objectives were to: 1) model the annual pattern of salinity, 2) determine if salinity measures derived from continuous data are more precise than snapshot SC as predictors of aquatic biology, and 3) quantify response to salinity of the benthic macroinvertebrate community. A sinusoidal model of the annual cycle of SC using daily measurements for 4.5 years revealed that salinity naturally deviated ± 20% from annual mean levels, with minimum SC occurring in late winter and maximum SC occurring in late summer. The pattern was responsive to seasonal dilution as driven by catchment evapotranspiration dynamics. Alternative discrete sampling intervals can approximate the pattern revealed by continuous SC data if sampling intervals are ≤ 30 days. Continuous SC variables did not significantly improve precision for prediction of benthic macroinvertebrate metrics (p > 0.1) as compared to snapshot SC using generalized additive mixed models. Results suggest that snapshot SC is a capable predictor of benthic macroinvertebrate community structure if sampling is carefully timed. However, continuous SC data can quantify chronic salt exposure, which supports a hypothesis to explain how temporal variability of field-based observations of salt sensitivity of benthic macroinvertebrate taxa may be influenced by life stage. Benthic macroinvertebrate community structure diverged from reference condition as salinity increased, with stronger relationships in Spring than in Fall. Intra-seasonal variation in community structure was also revealed across sampling dates. Non-Baetidae Ephemeroptera were most sensitive to salinity, with richness and abundance lower than reference at SC > 200 =µS/cm in Spring based on snapshot SC. Equivalent effects were predicted by mean monthly SC of 250-300 µS/cm from the prior Fall. Continuous conductivity monitoring may improve assessment of salinity effects because they can describe life-cycle exposure, which may aid investigations of mechanisms driving field-based observations of benthic-macroinvertebrate community alteration. / Ph. D. / Freshwater ecosystems around the world are at risk of contamination from salt pollution resulting from a variety of human activities. All natural freshwaters contain low levels of dissolved minerals, or salts, the combined concentration of which is referred to as salinity. Activities such as crop irrigation, road de-icing, and mining can cause salt pollution in streams and rivers, and excessive salinity can be toxic to many aquatic organisms. In many regions of the world, including in coal-mining-influenced streams of Appalachia USA, elevated salinity has been linked to decreased diversity of benthic macroinvertebrates, which are primarily aquatic insects, a group critical to healthy stream ecosystems. However, assessments used to reach this conclusion have generally not accounted for annual variability of salinity, as most studies use “snapshot” salinity data collected concurrently with biological data at a single point in time. Effective management of salinity impacts requires tools to accurately monitor and predict salinity while accounting for annual variability. Toward improving those tools, I conducted analyses of 4.5 years of salinity and aquatic insect data from 25 small central Appalachian mountain streams spanning a gradient of salinity. My objectives were to: 1) characterize the annual pattern of salinity using high-frequency salinity data, 2) determine if high-frequency salinity data is better than snapshot data for predicting aquatic insect diversity, and 3) measure the response to salinity of the aquatic insect community and identify salinity levels associated with insect biodiversity loss. High-frequency (daily) data revealed that salinity exhibited a predictable cyclic annual pattern with seasonal deviations of ± 20% from annual average salinity levels. Minimum salinity occurred during late winter and maximum salinity occurred in late summer. Lower-frequency salinity data can approximate the annual pattern if sampling interval is ≤ 30 days. Snapshot salinity was equally capable as high-frequency data of predicting aquatic insect diversity provided that snapshot salinity sampling is carefully timed. Diversity of many aquatic insects, especially mayflies, declined with increasing salinity, with stronger relationships in Spring than in Fall. Variation in diversity measures was also somewhat related to sample timing within seasons. Alteration of aquatic insect communities was evident at total salt concentrations levels of approximately 130 – 200 parts per million, depending on time of year. Efforts to manage salinity impacts to aquatic life may be improved by integrating knowledge of annual salinity patterns with how aquatic insects respond to salt pollution.

La signature isotopique du plomb (Pb) et du mercure (Hg) a été étudiée dans les lichens épiphytes des Pyrénées-Atlantiques. Un échantillonnage intégré et spatialisé à l’aide d’un système d’information géographique (SIG) a été mis en place afin de prendre en compte la diversité du territoire (occupation des sols, forêts, agriculture, industries et zones urbaines). A méso-échelle (département des Pyrénées-Atlantiques), la composition isotopique permet de différencier les zones impactées par les activités anthropiques mais aussi de discriminer l’origine de la contamination qu’elle soit de type industrielle ou urbaine. A l’échelle locale, la signature isotopique du Hg et Pb dans des lichens et des mousses prélevés dans la forêt d’Iraty (zone frontalière Franco-espagnole) a permis d’évaluer le type de dépôts accumulés dans ces différents bioaccumulateurs atmosphériques et le potentiel de ce nouvel outil pour le suivi des dépôts atmosphériques dans les écosystèmes éloignés. Enfin, sur la zone mercurifère d’Almadén (Castella la Mancha, Espagne) la signature isotopique du Hg dans les lichens et les sédiments apparait comme un outil pertinent pour étudier le transfert du Hg provenant de l’extraction minière vers les hydrosystèmes et l’atmosphère. / The isotopic signature of lead (Pb) and mercury (Hg) was measured in epiphytic lichens from Pyrénées-Atlantiques. An integrated and spatialized sampling was developed using geographical information system (GIS) to take into account the diversity of the territory (land-use, forest, agriculture, industries and urban areas). At meso-scale (county, Pyrenees-Atlantiques) the isotopic composition allows to differenciate areas impacted by anthropegenic activities and discriminate the origin of the contamination that it of industrial or urban type. At local scale, the isotopic signature of Hg and Pb in lichens and mosses sampled in Iraty forest (Franco-Spanish border) allowed to evaluate the kind of atmospheric deposits in these atmospheric bio-monitors and the potential of this new tool for the monitoring of atmospheric depositsin remote ecosystems. Finally, on the mercurifere area of Almadén (Castella la Mancha, Spain) the isotopic fingerprint of Hg in lichens and sediments appears to be a relevant tool to study the transfer of Hg from mining activities to hydrosystems and atmosphere.

Le développement des approches robustes et non paramétriques pour l’analyse et le traitement statistique de gros volumes de données présentant une forte variabilité,comme dans les domaines de l’environnement et de la génétique, est fondamental.Nous modélisons ici des données complexes de biologie appliquées à l’étude du comportement de bivalves et à l’analyse de liaison génétique. L’application des mathématiques à l’analyse du comportement de mollusques bivalves nous a permis d’aller vers une quantification et une traduction mathématique de comportements d’animaux in-situ, en milieu proche ou lointain. Nous avons proposé un modèle de régression non paramétrique et comparé 3 estimateurs non paramétriques, récursifs ou non,de la fonction de régression pour optimiser le meilleur estimateur. Nous avons ensuite caractérisé des rythmes biologiques, formalisé l’évolution d’états d’ouvertures,proposé des méthodes de discrimination de comportements, utilisé la méthode des shot-noises pour caractériser différents états d’ouverture-fermetures transitoires et développé une méthode originale de mesure de croissance en ligne.En génétique, nous avons abordé un cadre plus général de statistiques robustes pour l’analyse de liaison génétique. Nous avons développé des estimateurs robustes aux hypothèses de normalités et à la présence de valeurs aberrantes, nous avons aussi utilisé une approche statistique, où nous avons abordé la dépendance entre variables aléatoires via la théorie des copules. Nos principaux résultats ont montré l’intérêt pratique de ces estimateurs sur des données réelles de QTL et eQTL. / The development of robust and nonparametric approaches for the analysis and statistical treatment of high-dimensional data sets exhibiting high variability, as seen in the environmental and genetic fields, is instrumental. Here, we model complex biological data with application to the analysis of bivalves’ behavior and to linkage analysis. The application of mathematics to the analysis of mollusk bivalves’behavior gave us the possibility to quantify and translate mathematically the animals’behavior in situ, in close or far field. We proposed a nonparametric regression model and compared three nonparametric estimators (recursive or not) of the regressionfunction to optimize the best estimator. We then characterized the biological rhythms, formalized the states of opening, proposed methods able to discriminate the behaviors, used shot-noise analysis to characterize various opening/closing transitory states and developed an original approach for measuring online growth.In genetics, we proposed a more general framework of robust statistics for linkage analysis. We developed estimators robust to distribution assumptions and the presence of outlier observations. We also used a statistical approach where the dependence between random variables is specified through copula theory. Our main results showed the practical interest of these estimators on real data for QTL and eQTL analysis.

Moravian-Silesian Region is a part of an area so called „small black triangle". It is an area of higher pollution caused by some kind of industrial zone. This region belongs to the most polluted atmosphere in the Czech Republic. The pollution is caused by various factors but the most negative one is metallurgic industry but also mine-mining, transport, and small local furnaces. To analyze pollution, the instrumental neutron activation analysis with analyzing of moss samples were used. The moss has rudimentary root system so it takes nutrients as well as heavy metals from the atmosphere. All irradiation has performed in the Joint Institute for Nuclear Research (JINR) in the Frank Laboratory of Neutron Physics (FLNP) at the reactor IBR-2M. Two kinds of irradiation have been performed as short-term irradiation as long-term irradiation. Short-term irradiation took about 3 mins and then short-lived isotopes were measured for about 15 mins. The long-lived isotopes were measured twice after about 4 days for 30 mins and after 23 days for 90 mins after 3 days of irradiation. For long term irradiation, the cadmium-screened irradiation channel was used. For analyzation and data processing the software Genie 2000 as well as software developed at FLNP JINR were used. In sum, 43 samples of the terrestrial mosses (Hypnum cupressiforme, Pleurozium schreberi and Brachythecium rutabulum) were collected in autumn of 2015 in which the concentration of 38 elements were determined, including heavy metals. As a result, the factor analysis have been performed as well as maps analyzation.

Polychlorinated Biphenyls (PCBs) are a group of 209 ‘man-made’ chlorinated organic compounds that were widely used in the 20th century for a variety of industrial uses. PCBs were first commercially produced in the 1929 and were manufactured until the 1980s when their use was phased out due to environmental and human health risks. However, due to their widespread use and persistence they are ubiquitous in the environment and remain a contaminant of concern. The structural properties that determine the persistence of PCBs in humans were therefore elucidated by statistical analysis of data from the National Health and Nutrition Examination Survey (NHANES). PCBs with chlorine bonding in the 2,5-and 2,3,6- positions (and 2- in di- and tri-chlorinated congeners) were rapidly biotransformed and so can be classed as episodic congeners whereas PCBs with chlorine bonding in the 2,3,4-, 2,4,5-, 3,4,5-, and 2,3,4,5- positions were more resistant to biotransformation and can therefore be classed as steady state congeners. A fundamental requirement of using PCB signatures for environmental fingerprinting is an effective analytical method capable of producing high resolution signatures from biological samples. An extraction and clean-up method was developed that was successfully applied to different biological matrices (blood and tissues). A two dimensional gas chromatography with time of flight mass spectrometry (GCxGC-ToFMS) method was designed and optimised to provide a congener specific method capable of identifying 200 out of the 209 PCBs, with detection limits in human serum in the range of 1 to 10 ng g-1 lipid. The extraction and detection methods were used to determine the source of PCB contamination and age date exposure in workers at a transformer dismantling plant. A total of 84 different PCB congeners were identified in the sera of 30 workers with concentrations of the 7 indicator PCBs ranging from 1.2 - 39 μg g-1 lipid. Analysis of PCB signatures was able to distinguish recent from prolonged exposure and also identified an additional source of inhalation exposure in a subgroup of workers. Analysis of 12 different tissue types obtained from the common guillemot (Uria aalge) suggested a high degree of perpetuation between the PCB signature in different tissue types. This shows that comparative assessments can be undertaken between animals using different tissue types and that small (1 g) samples of blood can be used as a non-lethal sampling technique. The regional provenance of 25 wrecked Leach’s storm petrels (Oceanodroma leucorhoa) was also determined using PCB signatures. Results from GCxGC-ToFMS analysis revealed distinctively different PCB signatures in birds from Canada and Europe. The findings reported in this thesis enhance our understanding of PCB signatures in the environment and show how they can be used effectively to age date and identify the source of exposure in humans and animals.

Understanding the short-and long-term impacts of a biomonitoring and exposure project and reporting personal results back to study participants is critical for guiding future efforts, especially in the context of environmental justice. The purpose of this study was to evaluate learning outcomes from environmental communication efforts and whether environmental health literacy goals were met in an environmental justice community. We conducted 14 interviews with parents who had participated in the University of Arizona's Metals Exposure Study in Homes and analyzed their responses using NVivo, a qualitative data management and analysis program. Key findings were that participants used the data to cope with their challenging circumstances, the majority of participants described changing their families' household behaviors, and participants reported specific interventions to reduce family exposures. The strength of this study is that it provides insight into what people learn and gain from such results communication efforts, what participants want to know, and what type of additional information participants need to advance their environmental health literacy. This information can help improve future report back efforts and advance environmental health and justice.

The objective of my project was to determine variability and representability of mercury in the urban environment of Umeå in northern Sweden, based on applying the methods of forest moss biomonitoring (Swedish Environmental Protection Agency, ICP Manual). Mercury (Hg) is a commune pollutant in urban environments release to the atmosphere by anthropogenic activities. Industrial, traffic and incineration activities are the main sources of this element. Mercury is easily transported through the atmosphere and cycle through terrestrial and aquatic ecosystems, trending to bioaccumulate in organisms. The aims of the study are: (1) determine the representability and variability of the method in a specific urban environment, based on more intensive analyses of a green area within the city boundaries of Umeå, northern Sweden. (2) influence of site-specific conditions on the concentration of mercury in mosses. (3) City-scale variability in relation to national forest moss biomonitoring data (IVL.se). Results of urban environment measurements do not differ much respect the values of mercury concentration obtained sampling mosses far from the city, but it is subject to many factors that can alter results of the study.  Most of these are meteorological factors and the difficulty of find green zones close to cities with the suitable conditions to find mosses and perform a property sampling process avoiding throughfall and litterfall. The conclusion is that the use of mosses is a representative and valuable method to obtaining information in an urban environment but is limited by mentioned factors.

Algae are important components of stream food webs and often used in biomonitoring assessments. Little is known regarding the factors that limit their abundance in streams of the VA Coastal Plain. The surficial geology of the Coastal Plain is predominately sandy deposits which comprise the dominant substrate in streams of this region. In a comparative study of five streams located near the VCU Rice Center, we quantified substrate composition, light availability, and nutrient concentrations to assess their relative importance in determining benthic algal abundance. The proportion of stream area comprised of hard substrates was a significant predictor of variation in benthic algal abundance (r²=0.66). An experimental component comparing algal colonization on artificial hard substrates (tile) to the natural substrate reinforced the importance of substrate stability. Hard substrates which included gravel and aggregated clay likely provided greater stability for algal colonization relative to sand and silt deposits, resulting in lower mortality from scouring and sedimentation. Incident solar radiation was a secondary factor affecting algal abundance with shaded streams exhibiting lower benthic chlorophyll. Where substrate and light conditions were favorable, relationships between benthic algal abundance and dissolved phosphorus concentrations were observed. Seasonal fluctuations were ameliorated by high light conditions and constant disturbances at sites lacking hard substrates which kept CHLa at consistently low levels. The mean proportion of FBOM C derived from benthic algae ranged from 10 to 24%. In spite of the consistently low observed benthic algal abundance at sandy unstable Coastal Plain streams, benthic algae are an important source of benthic organic matter.

Freshwater ecosystems are under increasing pressure from a variety of contaminants, including heavy metals from mining operations, which can have complex effects that are difficult to evaluate. To detect early warnings from elevated concentrations of metals, organisms are sought to be used as monitoring tools. For example, mouthpart deformities in Chironomid larvae have been proposed as a bioindicator of stress in aquatic environments. However, the frequency and cause of these deformations, and their sensitivity to different stressors remain uncertain. In this study, I evaluated the usefulness of mouthpart deformities as a tool to monitor the effects of heavy metals from mining in northern Sweden. To do this, the mouthparts of 3789 Chironomid individuals analyzed from 17 sites closely located to mining operations and tested against concentrations of metals and DOC in the water chemistry of lakes and rivers. The frequency of deformities ranged from 0.00 – 4.79 % across all sites. Metal concentrations ranged from ‘very low’ to ‘low’ based on biological effect risk assessments. Of these, copper (R2 = 0.73) and cobalt (R2 = 0.66) were found to be significantly correlated with frequency of deformities. Additionally, the occurrence of deformities declined with DOC concentration, this was a nonlinear relationship. Frequencies of deformities observed in this study were lower than what have been reported to similar studies. The result from this study, together with other studies, suggest that deformities in Chironomid larvae are sensitive to even low levels of certain metals and could potentially be a good biomonitoring tool for early warnings of contamination in freshwater environments.