The increasing interest in the use of palacoenvironmental indicators and in particular diatoms to reconstruct past changes in sea level has highlighted the need for a more precise methodology that (a) provides quantitative reconstructions, and (b) is applicable to a wide range of sedimentary environments. Despite the widespread and increasing recent interest in the use of diatoms as indicators of estuarinep alaeoenvironmentsa nd sea-levelc hange,e xisting interpretative models, basedo n simple classification of taxa into freshwater, brackish or marine forms, provide only _qualitative estimates of past conditions. Resulting palaeoenvironmental reconstructions are at best crude, offering some indication of past mean sea level height, and at worst erroneous, as they fail to consider the effect of post-mortem transport and other taphonomic processes. This study aims to address these problems for the coast of Britain by developing a more robust quantitative method for using diatoms as indicators of estuarine palaeoenvironments and sea-level change. More specifically, it aims to develop a quantitative predictive model (transfer function) that relatesd iatom assemblagec omposition to salinity, habitat, depositional environment and tidal level around the coast of Britain. This is done through the collection and analysis of a training data set of diatom assemblagesa nd environmental variables (salinity class, elevation, grain size, habitat type and sediment organic content) from 25 sites around the coast of Britain. Qualitative and quantitative relationships within the diatom assemblagesa nd between the diatom assemblagesa nd coastale nvironmentalv ariables are explored using TWINSPAN and canonical correspondencea nalysisr espectively. The key environmental variables driving diatom assemblagesin the intertidal environment are shown to be elevation, salinity and sediment particle size. Habitat type and site location also explain a significant amount of variation in the diatom data, suggestingr egional differences in diatom assemblagesn ot accounted for by geornorphological and sedimentological differences between sites. The final transfer function for inferring normalised tidal height has a root mean squared error (RMSE) of 0.26, and a squared correlation (1-2) between observed and diatom-inferred normalised tidal height of 0.61. Corresponding error estimates under cross-validation by leave-one-out are 0.34 and 0.35 for RMSEjack and rjack 2 respectively. The poor performance of the model in comparison to published regional transfer functions is concluded to be due to the mergt:i In g of data from a large number of sites over a large geographical area. Such merging - has apparently introduced a large amount of noise into the diatom / elevation relationship, and is probably related to the increased heterogeneity and interaction of sediment typeý and elevation, and to the observed regional overprint in the diatom assemblages. Analogue matching to infer Sample habitats from the diatom data performs with a success Z): rate of 59%. Further merging of the habitat types based on ecologically similarities increases the success rate to 82%. The correct scale of trade off between coverage of palaeoenvironments, fossil diatom species 4): and reduction of regionality in the modem training dataset is an issue that needs further research before this model may be applied to core material to assist in palaeoenvironmental reconstructions.

The Eocene Nummulitique (Lutetian to Priabonian) has been studied in the external chains of the French Alps in Haute Savoie and Haute Provence. The Nummulitique unconformably overlies the Mesozoic passive margin succession and represents the onset of sedimentation in the Alpine Foreland Basin which formed due to lithospheric flexure caused by the advance of the Alpine orogeny. The base of the formation is marked by a regional erosional unconformity that developed during subaerial exposure of the Alpine foreland. The Nummulitique may be divided into two informal members: the lower Infranummulitique, a succession of terrigenous carbonates, and the overlying Nummulitic Limestone, a shallow marine carbonate ramp succession. The Infranummulitique is composed of terrigenous carbonates thought to have been derived from the uplifted and eroding foreland which were redeposited in local depocentres due to the topography on the erosion surface. The Infranummulitique can be divided into four facies associations: i) a lenticular conglomerate/nodular marl deposited from ephemeral streams, ii) a sheet conglomerate deposited in a marginal marine fan delta, iii) a Cerithium marl deposited in a brackish water coastal plain/lagoon and iv) a Microcodium wackestone deposited from coastal marine channels. The Nummulitic Limestone is marked by the appearance of the first fully marine foraminifera and a change from terrigenous to autochthonous carbonate sedimentation on a low-energy ramp dominated by larger benthonic foraminifera. The inner-ramp is represented by the deposition of bioclast shoals (packstones and grainstones) dominated by either calcareous red algae, Nummulites or peloids. The middle-ramp is dominated by mud-rich wackestones with a fauna of flat foraminifera, with local winnowed accumulations attributable to storm reworking. The outer ramp and basin are represented by mudstones and marls with a sparse benthos. The Nummulitique shows a marked cyclicity within an overall deepening upwards succession which is interpreted to be the combined effects of tectonic basin subsidence and high-frequency (4th order) eustatic sea-level variations. As the basin developed, the eustatic signature producing the small-scale cyclicity was successively overprinted by accelerating basin subsidence which controlled the stratigraphy of the underfilled foreland basin. Initially, the carbonate productivity is able to keep pace with the relative sea-level changes and the ramp prograded into the basin. The combination of accelerating subsidence rates and nutrient and detrital influx from the approaching orogenic wedge reduced the carbonate productivity and the ramp drowned, leading to pelagic marl deposition. The drowning surface and small-scale cyclicity have been used to correlate between measured sections within each field area, but problems occur in correlating between areas due to the migration of the foreland basin producing a diachronous sedimentary succession, which shows a similar development around the Alps, regardless of the age of the sediments. This diachroneity is evident in the two study areas with similar sediments, cycles and key surfaces developed at different stages of the basin development. The similarity in the successions demonstrates that the early sedimentation in the French Alpine Foreland Basin was controlled primarily by flexural subsidence.

Graduation date: 1997

Thesis (Ph. D.)--University of Hawaii at Manoa, 1995. / Includes bibliographical references (leaves 181-188). / Microfiche. / xvii, 188 leaves, bound ill. 29 cm

Thesis (Ph.D.)--University of Wollongong, 2004. / Typescript. Includes bibliographical references: leaf 237-270.

Thesis (M.S.)--Naval Postgraduate School, 1993. / "December 1993." Includes bibliographical references (p. 89-90).

Thesis (Ph. D.)--University of Wisconsin--Madison, 1983. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 273-279).

Thesis (M.E.E.)--University of Delaware, 2006. / Principal faculty advisors: Kenneth E. Barner, Dept. of Electrical and Computer Engineering; and Mohsen Badiey, Dept. of Marine and Earth Studies. Includes bibliographical references.

Thesis (M. Phil.)--University of Hong Kong, 2005. / Title proper from title frame. Also available in printed format.

Thesis (M.S.) in Earth Sciences--University of Maine, 2006. / Includes vita. Includes bibliographical references (leaves 109-119).