This thesis investigates the prehistoric diet of humans and two of their key commensals on Aitutaki using stable carbon and nitrogen isotopes. This technique gives us new insight into the diet of these three groups and results are considered in the context of the development of agricultural systems on Aitutaki, as well as in light of the wider context of cultural developments and environmental change in the Cook Islands and Polynesia as a whole. Ultimately, it allows us to more fully understand the complex interactions between humans and the two largest commensals in order to evaluate the utility of these animals as proxies for humans in dietary analyses. The results indicate that the prehistoric human diet on Aitutaki can be characterised as mixed, dominated by terrestrial plants and marine protein with lesser amounts of terrestrial protein. While the amount of protein eaten by the individuals was very similar, they varied in how much marine or terrestrial protein they ate. The pigs had a slightly more terrestrial diet, with a greater emphasis on plant foods. Their protein intake was more variable and terrestrial in nature than the humans’. Dogs had a higher trophic level, more marine-oriented, diet than either humans or pigs. Like the humans, their diet contained more variation in the source of their protein. Overall, however, the diets of all three groups were very similar. While the overall nature of the human, pig and dog diets, and their relationships to one another, remained relatively constant over time, beginning in the 14th century they show a decline in the amount of fish consumed and had a more terrestrial diet overall. This confirms trends seen in the archaeofaunal assemblages, and throws light on the relationship between environmental change and human subsistence practices in East Polynesia. The similarity of the pig and dog diets to human diet, and the fact that the relationship remained constant over time, demonstrates the usefulness of the Aitutaki commensal animals as proxies for humans in stable isotope analysis. However, the specifics of that relationship vary by species and by place. While the individual dietary variability provides us with new ways of looking at dietary change within populations, it also demonstrates the importance of obtaining as large an assemblage for analysis as possible in order to ensure that samples are representative of the population as a whole.

Abstract The Late Devonian-Early Carboniferous Drummond Basin of central Queensland, Australia is one of the oldest units of the New England Fold Belt. Samples of authigenic clay minerals have been collected from across the Drummond Basin and analysed using mineralogic, isotopic and geochronologic techniques. X-Ray diffraction analysis on 36 samples selected from six exploration wells was carried out on volcaniclastic and argillaceous sandstones and mudstones of the Drummond Basin sequence in order to ascertain authigenic illite content. Sixteen samples were found to be suitable for K-Ar dating purposes, containing abundant illite and illite/smectite mixed layers (I/S). K-Ar age dating was carried out on these selected samples yielding three age groups at 300, 250 and 200 Ma. Temperature of formation was estimated for the authigenic illite and illite/smectite minerals, based on illite crystallinity and illite content of illite/smectite mixed layers. Two groups of temperatures have been delineated, with a high temperature assemblage of between 275¨¬C-320¨¬C and a low temperature assemblage of approximately 100¨¬C. When compared with the age groups, it is seen that the older event was typically hotter. Stable isotope compositions for the authigenic clay minerals were analysed with ¥äD and ¥ä18O values of between -68¢¶ to -117¢¶ and +7.5¢¶ to +14.4¢¶ (SMOW) respectively. Isotopic compositions for the fluid in equilibrium with the clay minerals were calculated using palaeotemperatures estimated from the clay mineralogy and range between -90¢¶ and -41¢¶ and -0.6¢¶ and +8.7¢¶ (SMOW) for hydrogen and oxygen respectively. These compositions are more depleted in deuterium and more enriched in 18O than those previously reported for basins in the region. The data also produced three groups of fluid compositions, which correlate with the age distributions delineated by the K-Ar geochronology. The first fluid composition, which correlated with the Permo-Carboniferous age (~300 Ma) is interpreted to result from a mixing of Permo-Carboniferous meteoric water with highly evolved formation water trapped within the Drummond Basin since deposition of the basin. The regional extensional event at the Permo-Carboniferous boundary, which initiated the opening of the Bowen Basin to the east of the Drummond Basin allowed meteoric water to percolate deep into the basin, enabling the growth of illite rich clay minerals at this time. The second fluid group is considered to represent the evolution of Late Triassic meteoric water migrating through the Drummond Basin during another extensional period in the region at approximately 200 Ma. This event precipitated a new period of clay mineral growth in the basin (reflected in the K-Ar dates), raised the geothermal gradient in the region and extended the crust, forming new sedimentary basins within the New England Fold Belt. The third assemblage of fluid compositions were found to be very similar to the Late Triassic group, with ¥äD and ¥ä18O values plotting very close to the Late Triassic meteoric water evolution trend. This is interpreted as representing the same fluid as that which produced the Late Triassic compositions and therefore the 250 Ma ages may represent partial resetting of older ages by the Late Triassic thermal episode. However, if it is assumed that these K-Ar ages are not old ages partially reset by a thermal event, the stable isotope data could represent a mixing of Mid-Late Permian meteoric water with formation water trapped in the basin. This event may represent the regional compressive Hunter-Bowen event and may be responsible for dewatering the basin before the Late Triassic extensional episode.

Thesis (Ph.D.)--McMaster University (Canada), 1988. / Source: Dissertation Abstracts International, Volume: 62-13, Section: A, page: 0000.

Η διατριβή αυτή μελετά τα σταθερά ισότοπα στον υετό της Ελλάδας. η γεωγραφική κατανομή τους είναι η θεωρητικά αναμενόμενη, ενώ οι βαθμίδες με τη θερμοκρασία και το υψόμετρο και η τοπική μετεωρική γραμμή συμφωνούν με τη βιβλιογραφία για την ευρύτερη περιοχή. Η επίδραση της πορείας της αέριας μάζας, δεν είναι ευδιάκριτη λόγω της επίδρασης τοπικών διεργασιών κλασμάτωσης. Επίσης, η ισοτοπική σύσταση των πηγών ακολουθεί την κατανομή της σύστασης του υετού και είναι γενικά απεμπλουτισμένη σε σχέση με αυτόν. Ελέγχεται η καταλληλότητα των δεδομένων για κλιματολογική θεώρηση των ισότοπων του υετού στην κεντρική και ανατολική Μεσόγειο. Οι υπάρχουσες τάσεις σπανίως είναι σημαντικές και δεν έχουν χωρικά οργανωμένες μορφές. Επομένως, έχει νόημα ο υπολογισμός κλιματικών τιμών και η ανάπτυξη, με χρήση συνδυασμού προτύπων παλινδρόμησης και γεωστατιστικών μεθόδων, πλεγματικών κλιματικών ισοτοπικών δεδομένων και πλεγματικών ισοτοπικών χρονοσειρών. Τέλος, ερευνάται η ικανότητα των συνοπτικών ταξινομήσεων να οδηγούν σε ισοτοπικά διακριτές κλάσεις, αναπαράγοντας, ταυτόχρονα, σημαντικό μέρος της παρατηρούμενης ισοτοπικής μεταβλητότητας. Γενικά οι ταξινομήσεις βελτιστοποίησης με σχετικά μεγάλο αριθμό κλάσεων ικανοποιούν τα ανωτέρω. επιχειρείται η ανασύσταση ισοτοπικών χρονοσειρών χρησιμοποιώντας μέσες τιμές, σταθμισμένες βάσει των συχνοτήτων εμφάνισης των συνοπτικών κλάσεων και των υψών υετού. Η βέλτιστη επίδοση επιτυγχάνεται από εποχικές ταξινομήσεις με κλάσεις που αναπαράγουν τις βασικές ατμοσφαιρικές κυκλοφορίες, ακόμα και αν αυτό δεν συνεπάγεται βέλτιστους στατιστικούς δείκτες. / this thesis studies the stable isotpes in precipitation over greece. the geographical distribution is the one theoreticaly expected, while the temperature and altitude gradients, and the local meteoric line are onstistent with the literature for the area. the influence of air mass trajectory is not clear due to the effect of local fractionation processes. also, the composition of springs follows that of rain and is generaly depleted compared to it. the suitability of existing central and eastern mediterranean data, for climatological purposes is examined. existing trends are spatially inconsistent and rarely significant. thus, it is possible to calculate climatic values and obtain, using a combination of regression models and geostatistical methods, gridded isotopic climatologies and time series. finally, the ability of synoptic classifications to provide isotopicaly distinct classes, while at the same time reproducing the observed isotopic variability, is assessed. optimization classifications with a relatively large number of classes fulfill these criteria. reconstruction of isotope time series using mean values weighted according to synoptic class frequency and rain amount is possible. the best results are obtained from seasonal classifications with classes representing the basic atmospheric circulations, even if this does not lead to optimized statistical indices.

The discharge of groundwater into surface water bodies is a hidden, but significant pathway for the input of water and matter into lakes, rivers, estuaries and the coastal sea. Since groundwater is most often characterized by higher levels of nutrients or heavy metals, its discharge has often a crucial effect on the surface water body´s chemistry and the ecosystem health as well as on the related ecosystem service supply. For instance, groundwater-derived nutrient inputs are essential to fuel primary productivity, but if critical thresholds are exceeded groundwater-derived nutrient inputs can cause eutrophication, which may trigger harmful algal blooms or the creation of oxygen minimum zones – a serious threat to aquatic life. This thesis focuses on quantifying submarine and lacustrine groundwater discharge by applying environmental tracer based methods with emphasis on radionuclide (radon and radium isotopes) and stable water isotope (δ18O, δ2H) techniques. These tracers are suitable for determining groundwater discharge as they show distinct concentration and isotope ratio gradients between groundwater and the receiving surface water. Four studies are presented in this thesis: (1) The quantification of the response delay of the mobile radon detector RAD7 applied for radon-in-water mapping. The response delay of the mobile radon-in-air detector RAD7 is determined for two detection set-ups (radon extraction via RADaqua and via a membrane module) as well as for a range of water flow rates. For the membrane module the response delay is less pronounced compared to the RADaqua. For instance, at a water flow rate of 1 l min-1 the peaks of the instruments recordings lag behind the radon-in-water concentrations by ~10 min for the membrane module and by ~18 min for the RADaqua. Further, it was demonstrated that faster water flow rates decrease the response delay. An algorithm is presented that allows the inverse calculation of radon-in-water concentrations from RAD7 records for the described detection set-ups and water flow rates. Thus, it allows a more precise localization of radon-in-water anomalies and, consequently a more precise localization of groundwater discharge areas. (2) Determination of submarine groundwater discharge into a large coastal bay (False Bay, South Africa) SGD consists generally of two components: (a) fresh terrestrial SGD (FSGD) driven by the inland hydraulic gradient and (b) seawater re-circulation (RSGD) through the coastal aquifer driven by seaward effects such as tidal pumping. A bay-wide radon mapping resulted in identification of a SGD site, where subsequently detailed investigations were conducted. At this SGD site a salt and a radon mass balance were applied consecutively for determining FSGD and total SGD, respectively. RSGD was inferred from the difference between FSGD and total SGD. For the radon mass balance, new approaches for calculating the radon degassing and mixing loss were proposed. The tracer mass balance revealed median FSGD of 2,300 m³ d-1 or 0.9 m³ d-1 per m coastline and median RSGD of 6,600 m³ d-1 or 2.7 m³ d-1 per m coastline. The FSGD rate was validated using (a) a hydrological model for calculating the groundwater recharge rate and (b) a groundwater flow model for delineating the subsurficial FSGD capture zone. This validation supported the tracer based findings. The relevance of this study is foremost the presentation of new methodological approaches regarding the radon mass balance as well as the validation of FSGD under consideration of hydrological and hydrogeological information. (3) Differentiation of fresh and re-circulated submarine groundwater discharge in an estuary (Knysna Estuary, South Africa) Knysna Estuary is a more complex system than False Bay since besides seawater, FSGD and RSGD also river water mixes within the estuary. Both FSGD and RSGD were differentiated by applying a mixing analysis of the estuary water. For this purpose, an end-member mixing analysis (EMMA) was conducted that simultaneously utilizes radon and salinity time series of estuary water to determine fractions of the end-members seawater, river water, FSGD and RSGD. End-member mixing ratio uncertainty was quantified by stochastic modelling (Monte Carlo simulation) under consideration of end-member characterization uncertainty. Results revealed highest FSGD and RSGD fractions in the estuary during peak low tide. Median fractions of FSGD and RSGD were 0.2 % and 0.8 % of the estuary water near the mouth over a 24 h time-series. In combi-nation with a radon mass balance median FSGD of 46,000 m³ d-1 and median RSGD of 150,000 m³ d-1 were determined. By comparison to other sources, this implies that the SGD is a significant source of dissolved inorganic nitrogen (DIN) fluxes into the estuary. This study demonstrates the ability of EMMA to determine end-member fractions in a four end-member system under consideration of end-member uncertainty. Further, the importance of SGD for the water and DIN budget of Knysna Estuary was shown. (4) Quantification of groundwater discharge and water residence time into a groundwa-ter-fed lake (Lake Ammelshainer See, Germany). The presented approach utilizes the stable isotopes of water (δ18O, δ2H) and radon for determining long-term average and short-term trends in groundwater discharge rates. The calculations were based on measurements of isotope inventories of lake and groundwater in combination with climatic and isotopic monitoring data (in precipita-tion). The results from steady-state annual isotope mass balances for both δ18O and δ2H are consistent and reveal an overall long-term average groundwater discharge that ranges from 2,800 to 3,350 m³ d-1. These findings were supported by the good agree-ment of the simulated annual cycles of δ18O and δ2H lake inventories utilizing the de-termined groundwater discharge rates with the observed lake isotope inventories. However, groundwater discharge rates derived from radon mass balances were signifi-cantly lower, which might indicate a distinct seasonal variability of the groundwater discharge rate. This application shows the benefits and limitations of combining δ18O/δ2H and radon isotope mass balances for the quantification of groundwater con-nectivity of lakes based on a relatively small amount of field data accompanied by good quality and comprehensive long-term meteorological and isotopic data (precipitation). This thesis presents important methodological achievements with respect to radon and stable water isotope mass balances, uncertainty quantification, geochemical differentia-tion between FSGD and RSGD and validation of FSGD. Further, first SGD estimates are reported for False Bay and Knysna Estuary in South Africa. / Der Austritt von Grundwasser in Oberflächengewässer stellt einen unsichtbaren Ein-tragspfad von Wasser und Stoffen in Seen, Flüsse, Ästuare und das küstennahe Meer dar. Die Konzentrationen vieler Stoffe wie beispielsweise von Nährstoffen und Schwermetallen ist im Grundwasser im Allgemeinen signifikant höher als in Oberflächengewässern. Daher können selbst volumetrisch verhältnismäßig kleine Grundwasseraustritte entscheidenden Einfluss auf Wasserchemie und den Gesundheitszustand des aquatischen Ökosystems haben, womit Auswirkungen auf die Bereitstellung von Ökosystemleistungen verbunden sein können. Beispielsweise sind grundwasserbürtige Nährstoffeinträge eine entscheidende Steuergröße für die Primärproduktivität. Überschreiten diese grundwasserbürtigen Nährstoffeinträge jedoch einen Schwellenwert, kann es zur Eutrophierung des Oberflächengewässers kommen. Dies wiederum kann toxische Algenblüten oder die Entstehung von Sauerstoffminimumzonen zur Folge haben und das aquatische Leben bedrohen. Diese Dissertation beschäftigt sich mit Methoden zur Quantifizierung von Grundwas-sereinträgen in den küstennahen Ozean, Ästuare und in Seen. Dabei stützt sich diese Arbeit primär auf Umwelttracer, vor allem auf Radionuklide (Radon- und Radium-Isotope) sowie die stabilen Isotope des Wassers (δ18O, δ2H). Diese Umwelttracer sind für die untersuchten Systeme in besonderer Weise geeignet, da zwischen Grundwasser und Oberflächenwasser ein ausgeprägter Gradient hinsichtlich Konzentration bzw. Isotopensignatur besteht. Vier Einzelstudien stellen den Kern dieser Arbeit dar: (1) Die Quantifizierung der Antwortverzögerung des mobilen Radon-Detektors RAD7, an-gewendet für die Radon-in-Wasser-Kartierung. Die Antwortverzögerung des mobilen Radon-in-Luft-Detektors RAD7 wurde für zwei Messanordnungen (Radonextraktion via RADaqua und via Membranmodul) sowie für einen Bereich von Wasserdurchflussraten bestimmt. Für die Radonextraktion via RADaqua ist die Antwortverzögerung stärker ausgeprägt als für das Membranmodul. Bei einer Wasserdurchflussrate von 1 l min-1 treten die Peaks der aufgezeichneten Werte ~10 min nach den Radon-in-Wasser Peaks auf, während die Verzögerung bei Radonextraktion via RADaqua ~18 min beträgt. Weiterhin wurde eine Reduktion der Antwortverzögerung mit zunehmenden Wasserdurchflussraten beobachtet. Der vorgestellte Algorithmus ermöglicht in Kombination mit den berechneten Radontransfer-Koeffizienten die inverse Modellierung der Radon-in-Wasser-Konzentrationen, basierend auf den RAD7-Messwerten. Dies ermöglicht beispielsweise eine genauere Lokalisierung von räumlichen Radon-in-Wasser Anomalien und folglich eine präzisere Bestimmung von Grundwasseraustrittsstellen. (2) Quantifizierung untermeerischer Grundwasseraustritte in eine große Meeresbucht (False Bay, Südafrika) Untermeerische Grundwasseraustritte (“Submarine Groundwater Discharge” – SGD) bestehen aus zwei Komponenten: (a) Süßwasser-SGD (“Fresh SGD” – FSGD) angetrieben durch den meerwärtsgerichteten hydraulischen Gradienten, und (b) re-zirkuliertem SGD („re-circulated SGD“ – RSGD), verursacht durch Prozesse wie gezeitengesteuerte Infiltration von Meerwasser in den Aquifer. Eine Radon-Kartierung entlang der gesamten Küstenlinie der Bucht führte zur Lokalisierung von SGD, woraufhin dort vertiefende Untersuchungen durchgeführt wurden. In diesem Bilanzgebiet wurden eine Salz- und eine Radon-Massenbilanz durchgeführt, um FSGD bzw. Gesamt-SGD zu bestimmen. RSGD wurde aus der Differenz von FSGD und SGD abgleitet. Für die Radon-Massenbilanz wurden neue Ansätze für die Berechnung der Radon-Entgasung in die Atmosphäre und des Radon-Mischungsverlustes mit küstenfernerem Wasser präsentiert. Die Tracer-Massenbilanzen ergaben einen FSGD-Median von 2.300 m³ d-1 bzw. 0,9 m³ d-1 pro Meter Küstenlinie und einen RSGD-Median von 6.600 m³ d-1 bzw. 2,7 m³ d-1 pro Meter Küstenlinie. Die FSGD-Rate wurde mit Hilfe eines hydrologischen Modells zur Abschätzung der Grundwasserneubildungsrate und eines Grundwasserströmungsmodells zur Abgrenzung des unterirdischen Einzugsgebiets des Bilanzraums bestimmt. Diese unabhängige Methode bestätigte die Tracer-basierten Ergebnisse. Die Bedeutung dieser Studie besteht zuvorderst in der Vorstellung neuer methodischer Ansätze bei der Radon-Massenbilanzierung sowie in der Validierung von FSGD unter Berücksichtigung hydrologischer und hydrogeologischer Daten. (3) Unterscheidung von FSGD und RSGD in einem Ästuar (Knysna Ästuar, Südafrika). Das Knysna-Ästuar ist hinsichtlich der Bestimmung von SGD im Vergleich zur False Bay ein komplexeres System, da sich neben Meerwasser, FSGD und RSGD auch Flusswasser in signifikanten Mengen im Ästuar mischt. FSGD- und RSGD-Anteile wurden anhand der chemischen Zusammensetzung des Ästuarwassers unterschieden. Für diesen Zweck wurde eine End-Member-Mischungsanalyse (EMMA) auf Grundlage von Radon- und Salinitätszeitreihen des Ästuarwassers durchgeführt. Durch ein Optimierungsverfahren wurde die Mischung der End-member Meerwasser, Flusswasser, FSGD und RSGD für jeden Zeitschritt mit dem Ziel der bestmöglichen Übereinstimmung mit den gemessenen Radon- und Salinitätszeitreihen bestimmt. Die Unsicherheit in der Bestimmung der End-member-Anteile wurde durch stochastische Modellierung (Monte-Carlo-Simulation) quantifiziert. Die höchsten Anteile von FSGD und RSGD traten bei Niedrigwasser auf. Die mittleren Anteile von FSGD und RSGD betrugen in der Nähe der Ästuarmündung 0,2 % und 0,8 % während einer 24-stündigen Zeitreihenmessung. Diese Informationen führten in Kombination mit einer Radon-Massenbilanz zur Bestimmung eines mittleren FSGD von 46.000 m³ d-1 sowie eines mittleren RSGD von 150.000 m³ d-1. Diese Ergebnisse implizieren unter Einbeziehung weiterer Daten, dass SGD ein bedeutender Pfad für den Eintrag von gelöstem anorganischem Stickstoff (DIN) in das Knysna-Ästuar darstellt. Diese Studie zeigt das Potenzial einer EMMA für die Bestimmung der Anteile von vier End-membern unter Nutzung von zwei gemessenen Variablen und unter Berücksichtigung der End-member-Unsicherheit. Außerdem wurde die Bedeutung von SGD für das Wasser- und DIN-Budget des Knysna-Ästuars aufgezeigt. (4) Quantifizierung von Grundwasseraustrittsrate und Wasserverweilzeit eines grundwas-sergespeisten Sees (Ammelshainer See, Deutschland). Der vorgestellte Ansatz nutzt die stabilen Isotope des Wassers (δ18O, δ2H) und von Ra-don für die Bestimmung des mittleren langfristigen sowie der aktuellen Grundwas-seraustrittsrate. Die Berechnungen beruhen auf Abschätzungen des Isotopeninventars anhand von Feldmessungen, der Isotopensignatur des Grundwassers sowie ergänzen-den Klima- und Isotopen-Daten (Niederschlag). Die Ergebnisse einer stationären Isoto-pen-Massenbilanz für δ18O und δ2H sind übereinstimmend und ergaben einen langfristigen mittleren Grundwasseraustritt von 2.800 bis 3.350 m³ d-1. Dieses Ergebnis wurde für die Modellierung des jährlichen Zyklus des Isotopeninventars im See benutzt, welches mit den gemessenen Isotopenwerten konsistent ist. Die auf Grundlage einer Radon-Massenbilanz abgeleiteten aktuellen Grundwasserzutrittsraten lagen im Gegensatz dazu deutlich niedriger, was jedoch nicht notwendigerweise einen Widerspruch darstellen muss, sondern vielmehr ein Hinweis auf eine möglicherweise ausgeprägte saisonale Variabilität des Grundwasseraustritts darstellen kann. Diese Studie zeigt Möglichkeiten und Grenzen der Anwendung von einer Kombination aus δ18O/δ2H- und Radon-Massenbilanzen für die Bestimmung der Grundwasseranbindung von Seen mit einem vergleichsweise geringen Messaufwand unter Nutzung qualitativ hochwertiger und umfangreicher Klima-und Isotopen-Daten (Niederschlag). Diese Dissertation präsentiert wichtige methodische Fortschritte hinsichtlich der An-wendung von Radon- und stabilen Isotopen-Massenbilanzen, der Quantifizierung von Unsicherheit, der Unterscheidung von FSGD und RSGD anhand geochemischer Daten und der Validierung von FSGD. Außerdem wurden erstmals SGD-Raten für Standorte in Südafrika (False Bay und Knysna-Ästuar) vorgestellt.

Depuis la fin de la dernière période glaciaire, chaque région du globe a connu une évolution du niveau marin relatif - ou niveau marin « ressenti » - qui lui est propre, principalement conditionnée par la conjonction de la remontée globale « eustatique » du niveau marin et des dynamiques isostatiques, responsables des mouvements verticaux de la surface terrestre. Ces derniers, encore en cours aujourd’hui, jouent un rôle majeur dans l’augmentation actuelle du niveau marin. Malgré d’importants travaux menés au Royaume-Uni au cours des dernières décennies, la connaissance de l’évolution du niveau marin relatif holocène et des dynamiques isostatiques sur la façade Atlantique ouest-européenne demeure largement imparfaite, notamment en ce qui concerne les côtes françaises. De par sa position géographique, le Finistère se pose comme un jalon idéal afin de renforcer les connaissances dans ce domaine. A ce titre, cette thèse a eu pour principaux objectifs d’obtenir une courbe fiable de remontée du niveau marin relatif holocène pour le Finistère et d’évaluer l’apport de nouvelles connaissances à ce sujet pour la compréhension des dynamiques isostatiques régionales. Dans un premier temps, un important travail de terrain a été réalisé sur différents sites répartis tout autour de la péninsule Finistérienne. Une nouvelle méthode multi-proxies, développée à partir de l’utilisation novatrice d’indicateurs géochimiques et microstratigraphiques, a permis de reconstituer l’évolution du niveau marin au cours des 8000 dernières années à partir des séquences sédimentaires prélevées. Nos résultats démontrent que le Finistère a connu une remontée continue du niveau marin relatif à des rythmes décroissants durant l’Holocène. Cette évolution a provoqué de profonds bouleversements dans l’organisation et le fonctionnement des systèmes sédimentaires côtiers dont les implications sont discutées. En particulier, de nouvelles hypothèses sont proposées quant à l’origine de la baisse pluri-métrique du niveau marin jusqu’alors invoquée autour de 3000 B.P. Dans un second temps, nos résultats ont été confrontés aux reconstitutions du niveau marin obtenues dans le sud-ouest de l’Angleterre ou dans le sud du golfe de Gascogne, ainsi qu’aux résultats produits par des modèles géophysiques de réajustement isostatique. Ces comparaisons ouvrent de nouvelles perspectives quant à la compréhension de l’amplitude et de la chronologie des dynamiques isostatiques holocène à l’échelle du nord-ouest de l’Europe. / Since the Last Glacial Maximum, the different regions of the world underwent specific relative sea-level evolutions, principally under the joint influences of the worldwide “eustatic” sea-level rise and regional isostatic dynamics. The latter, still going on today, appear to play a major role in the contemporaneous sea-level rise. Despite of the several studies made on the subject in the UK within the last decades, only little knowledge appears available on the relative sea-level evolution and on the isostatic dynamics at the scale of the Western Europe Atlantic façade, and particularly on the French Atlantic coasts. Due to its geographical position, the Finistère region stands out as a potentially ideal region for providing new material on the subject.This PhD work aimed to produce a new and reliable Holocene relative sea-level curve for the Finistère region. At first, sedimentary sequences were retrieved from several sites located all around the Finistère peninsula. In order to reconstruct RSL from these sequences, an innovative multi-proxy method was developed, conjointly using geochemical and microstratigraphic indicators. Our results bear out that, in the Finistère region, the Holocene RSL followed a continuous rise at progressively decreasing rates during the last 8000 yrs B.P. This evolution led to massive disruptions in the organization and functioning of the coastal sedimentary systems which are discussed. In particular, new hypotheses are proposed to explain the pluri-metric drop in RSL that was formerly invoked ca. 3000 B.P. The results we obtained were then compared to the RSL data available in the south-western UK or the south of the Bay of Biscay, along with glacio-isostatic adjustment geophysical models. These comparisons provide new enlightments on both the amplitude and the chronology of isostatic dynamics at the scale of the north-western Europe during the Holocene.

Le régime alimentaire est un élément clé de l’écologie des espèces et dépend à la fois des besoins nutritionnels des individus et des ressources alimentaires disponibles. Ainsi, les stratégies alimentaires peuvent varier en fonction de l’âge, du sexe ou de l’état physiologique des individus, mais également de paramètres environnementaux et, en particulier, de la saisonnalité. Reconstruire le régime alimentaire d’espèces cryptiques ou difficiles à observer en milieu naturel nécessite l’emploi de marqueurs capables de détecter les variations du régime alimentaire sur le court terme. L’analyse des patrons de micro-usures dentaires et des ratios d’isotopes stables du carbone et de l’azote constitue une approche pertinente pour distinguer le régime alimentaire entre différentes espèces ou entre populations. Dans le cadre de cette thèse, nous avons évalué le pouvoir prédictif de ces deux types de marqueurs sur les variations saisonnières et interindividuelles du régime alimentaire dans une population naturelle de mandrills (Mandrillus sphinx) sur laquelle des données comportementales étaient disponibles. Conformément aux comportements alimentaires observés, les patrons de micro-usures dentaires et les signatures isotopiques varient en fonction de la saison, de l’âge et du sexe des individus. L’ensemble de nos résultats nous indique que la diversité importante d’aliments consommés à faible fréquence par les mandrills ne peut être détectée par les marqueurs analysés qui semblent être, en revanche, pertinents pour identifier des changements significatifs de consommation d’aliments majoritaires tels que les fruits et les feuilles. Enfin, les signatures isotopiques varient en fonction du cycle reproducteur des femelles et de leur rang de dominance, nous indiquant que l’utilisation de tels marqueurs peut permettre la mise en évidence d’aspects de la physiologie des individus non reportés par des observations comportementales. Les résultats présentés dans cette thèse apportent des données de référence essentielles pour la reconstruction du régime alimentaire d’espèces éteintes partageant des traits communs avec le mandrill, telles que des primates généralistes de forêts tropicales. / Diet is a critical factor of species’ ecology and depends on both individuals’ nutritional needs and availability of their food resources. As such, feeding strategies may vary with age, sex or physiological condition of the individuals but also depend on environmental factors and, in particular, seasonality. In cryptic species and those that are difficult to observe in their natural habitat, diet reconstruction requires the use of proxies with the ability to detect dietary variations over the short term. Analyses of dental microwear patterns along with carbon and nitrogen stable isotopes constitute a relevant approach to discriminate between diets of different species or populations. In the framework of this thesis, we assessed the predictive power of these two proxies on seasonal and inter-individual variations of diet in a natural population of mandrills (Mandrillus sphinx) for which behavioural data was available. According to the feeding behaviours observed, dental microwear patterns and isotopic signatures vary depending on the season as well as individual’s age and sex. Taken together our results indicate that the large diversity of rarely consumed food items cannot be detected by the proxies analysed, which are however, relevant to identify significant changes in the consumption of major foods such as fruits and leaves. Finally, isotopic signatures vary across females’ reproductive statuses and dominance ranks, indicating that the use of those proxies may allow the highlighting of some aspects of individuals’ physiology not detected by behavioural observations. The results presented in this thesis provide a baseline data essential for dietary reconstruction of extinct species sharing similar traits with mandrills, such as generalist primates from tropical forests.

O Distrito Mineiro de Artigas, no Uruguai, compreende uma das maiores jazidas de ametista e ágata de que se tem conhecimento e é comparável às jazidas da região do Alto Uruguai, próximas ao município de Ametista do Sul, Brasil. A mineralização ocorre em geodos parcialmente preenchidos por ágata, quartzo incolor, ametista e ± calcita, que ocorrem nesta seqüência, da borda para o centro das cavidades. As rochas hospedeiras da mineralização são basaltos andesíticos da Formação Arapey, equivalente à Formação Serra Geral, no Brasil. Na área de estudo ocorre uma seqüência de seis derrames, de composição entre basalto e andesito. As estruturas de resfriamento são dos tipos I e II, que diferem pela ausência e presença de disjunção colunar, respectivamente, na porção interna do derrame. Destes seis derrames dois são do tipo I (maciço na zona central) e portadores de minério. Com base em observações de campo de estruturas rúpteis associadas às zonas mineralizadas, foi elaborado um modelo epigenético para a mineralização de ametista em geodos em basalto. Neste modelo, os geodos são formados após a solidificação da lava, por deformação ou dissolução do basalto, depois de alterado para argilominerais do grupo da esmectita pela interação com fluidos hidrotermais. O preenchimento ocorre após a abertura e pelo mesmo fluido que abriu a cavidade. O fluido é meteórico, de baixa salinidade e tem composição próxima a de água pura. A temperatura do fluido é < 200°C, com base na mineralogia de alteração da rocha hospedeira. A temperatura de cristalização dos minerais no interior dos geodos não ultrapassa 70°C. Esta temperatura foi calculada com base nos coeficientes de partição do oxigênio entre quartzo e água e entre calcita e água para um fluido com assinatura isotópica de T18O de -5 ‰. A intensa alteração hidrotermal nas rochas portadoras do minério e nos derrames, no âmbito do Distrito Mineiro de Artigas, corrobora para eventos epigenéticos relacionados com a mineralização. Isótopos de T34S nas encaixantes do minério indicam percolação de fluido em grande escala, e assinatura isotópica compatível com as unidades sedimentares subjacentes ao pacote vulcânico. A integração dos dados obtidos sugere que os geodos de ametista foram formados após a solidificação da lava e os minerais que preenchem os geodos cristalizadados por um fluido de origem meteórica. / The Artigas Mining District, in Uruguay, is a world-class amethyst deposit, and is comparable with those in Alto Uruguai region, Brazil (Ametista do Sul). The mineralization occurs as geodes partially filled by agate, colorless quartz and amethyst, ± calcite, which occur in this sequence from the rim to the inner part of the cavities. The host rocks are andesitic basalts from the Arapey Formation. The study area comprises a sequence of six lava flows, including basaltic to andesitic composition. The cooling structure of the flows is type I and type II; those are differentiated by the inner part of the flow. Type II has columnar joints, whereas type I is massive. Two out of six flows are mineralized and structured as type I flows. Based on field observations of brittle failures associated with the geode zone, an epigenetic model is elaborated for the mineralization of amethyst in geodes hosted by basalts. In this model, the geodes are formed after the flow solidifies. The cavities are formed by deformation or dissolution, after the basalt was altered to clay minerals (smectite group) by hydrothermal fluids. The filling occurred after the cavity was formed by the same fluid. The fluid is close to pure water in composition, of meteoric origin, with low salinity. The temperature is <200°C, based on alteration mineralogy in the host rock. The crystallization temperature is <70°C. The temperature is calculated based on fractionation coefficient of oxygen isotopes between quartz and water and between calcite and water. The value of T18O of the mineralized fluid is assumed to be -5 ‰. The intensive alteration in the host rocks and in lava flows within the mining district, favors the epigenetic hypothesis related to the mineralization processes. T34S isotopes on the host rock, agree with large scale hydrothermal fluid percolation. The data obtained suggest that the amethyst geodes were formed after the lava solidified and was filled by meteoric hydrothermal fluids.

Stomatal development and its regulation by environmental conditions (light, CO2 concentration) and physiological processes in the leaf of higher plants were investigated. The study was based on the assumptions that stomatal development should be regulated by signals coming from both external environment and leaf interior, and that the signal should be transduced from cotyledons to leaves. Transgenerational effect in stomatal development was also studied. Molecular and physiological approaches were applied to reveal the relationship between leaf environment, stomatal development, stomatal function and leaf physiology.

Made available in DSpace on 2015-06-17T19:34:31Z (GMT). No. of bitstreams: 0 Previous issue date: 2014-06-27. Added 1 bitstream(s) on 2015-06-18T12:48:40Z : No. of bitstreams: 1 000830242_20160627.pdf: 61065 bytes, checksum: fdcafce30d5addb2d0e1ddf71f1589de (MD5) Bitstreams deleted on 2016-06-27T11:55:31Z: 000830242_20160627.pdf,. Added 1 bitstream(s) on 2016-06-27T11:56:11Z : No. of bitstreams: 1 000830242.pdf: 1630719 bytes, checksum: 3bcc71c070b80050d03529c7d3469985 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / O hambúrguer é um produto cárneo industrializado e potencial veículo de bactérias produtoras ou não de toxinas nas diversas fases de seu processamento, desde sua origem até as fases de transformação, armazenagem, transporte e distribuição para o consumo. Além disso, seus requisitos de composição são susceptíveis à fraude pelo uso excessivo de extensores protéicos. Estes ingredientes auxiliam na diminuição nos custos de formulação dos produtos finais e maximizam os lucros de indústrias fraudulentas. A fiscalização pelos órgãos de vigilância é dificultada pela carência de metodologias analíticas apropriadas. Dentro deste contexto, tornase necessária a avaliação do aspecto higiênico-sanitário dos hambúrgueres através das análises microbiológicas e a determinação e quantificação da proteína de soja. Foram analisadas 110 amostras de hambúrgueres de carne bovina de 11 marcas comerciais brasileiras através da técnica dos isótopos estáveis de Nitrogênio -15 e Carbono-13 e segundo a metodologia microbiológica descrita pela Instrução Normativa n° 62 de 26 de agosto de 2003. Todas as 11 marcas comerciais brasileiras apresentaram contagem bacteriana dentro do limite (5x103 UFC/g) para Coliformes a 45°C, enquanto os valores para Staphylococcus coagulase positiva estavam acima dos padrões permitidos (5x103 UFC/g) na marca H (amostras 1 e 7). As marcas B (amostra 5), D (amostra 5), E (amostra 5) e G (amostra 4) também apresentaram contagens acima do estipulado (3x103 UFC/g) para Clostridium sulfito redutor a 46°C e a amostra 6 da marca L foi positiva para Salmonella spp. Apenas 3 amostras da marca G encontraram-se dentro dos 4% de adição de proteína não cárnica estipulados pela IN n° 20 de 31/07/2000. Todas as outras 107 amostras das 11 marcas comerciais burlaram a legislação específica em vigor / The burger is a manufactured meat product and potential vehicle of bacteria producers or not of toxins in various stages of processing, from its origin to the stage of transformation, storage, transport and distribution to consumption. Moreover, their composition requirements are susceptible to fraud by excessive use of protein extenders. These ingredients help in reducing the cost of formulation of the final products and maximize profits of fraudulent industries. The supervisory agencies monitoring is hindered by the lack of appropriate analytical methodologies. Within this context, it is necessary to evaluate the hygienic- sanitary aspects of burgers through microbiological analysis and the determination and quantification of soy protein. Were analyzed 110 samples of beef hamburgers of 11 brazilian trademarks by the technique of stable isotopes of Nitrogen-15 and Carbon -13 and second microbiological methodology described by Normative Instruction n° 62 of August 26, 2003. All 11 brazilian trademarks presented bacterial counts within the limit (5x103 CFU/g) for coliforms at 45°C , while the values for Staphylococcus coagulase positive were above the allowed patterns (5x103 CFU/g) in brands H (samples 1 and 7). Brands B (sample 5) , D (sample 5) , E (sample 5) and G (sample 4) also had scores above the stipulated (3x103 CFU/g) for Clostridium sulphite reducing to 46°C and sample 6 of brand L was positive for Salmonella spp. Only 3 samples of G brand were within 4% addition of non-meat protein stipulated by IN 20 07/31/2000. All other 107 samples from 11 trademarks cheated the specific legislation