Models of the evolution of the modern Transantarctic Mountains, the uplifted flank of the West Antarctic rift system, require constraints on the timing, magnitude, and spatial pattern of surface uplift, rock uplift, denudation, and faulting. This study presents a model of Cenozoic landscape development for the Shackleton Glacier area (85°S, 176°W) by integrating structural geologic, geomorphologic, and apatite fission-track thermochronologic (AFTT) methods. Cenozoic denudation (up to 5-6 km) began ∼50 Ma near the Ross Ice Shelf coast and migrated inland by escarpment retreat, as evidenced in the AFTT and geomorphologic record. Dissected planation surfaces are scattered at elevations from 500 m to almost 4000 m above sea level and represent episodes of relative tectonic quiescence between periods of uplift. The fault structure of the range front is consistent with an interpretation of dextrally transtensional kinematics, which apparently accommodates a smaller transcurrent component than the greater West Antarctic rift system.

This project described the vegetation of a stretch of the Parana River and investigated which natural or human-imposed factors might be controlling its characteristics, its capacity for biodiversity support and its potential to provide an economic resource. This information is of value in the planning of management strategies aiming to conserve biodiversity and develop sustainable ways in which the floodplain resources can be utilised. Central to this study were the surveys of vegetation and environmental characteristics of aquatic, terrestrial and transitional habitats of the Parana floodplain near Porto Rico. These produced extensive data sets which helped to reveal the types of vegetation-environment relationships structuring the floodplain plant communities. To complement this investigative approach, three aspects of the functioning of floodplain vegetation were chosen for closer study. These were the impacts of livestock grazing on wetland and island vegetation, competitive interactions between pairs of free-floating aquatic plant species and the role of aquatic macrophytes in contributing carbon to aquatic food webs. Three major community types were identified in the aquatic habitats of the floodplain, one which included Eichhornia azurea in mixture with several free-floating and emergent species, a second in which E. azurea was strongly dominant and a third comprised purely of submerged species. Two strongly contrasting broad vegetation communities were identified at bank and shore environments. Polygonum and Ludwigia species were important in one group and Poaceae, creepers, woody plants and ferns in the other group. Sub-groups of these communities could also be suggested, but these are less distinct. Most of the floodplain sites supported a Poaceae-creeper community type with the remaining sites supporting a community indicated by Polygonum species. The Poaceae group was comprised of a number of sub-communities in which the importance of Poaceae relative to other species varied. The aquatic vegetation communities differed structurally with contrasting community biomass, canopy height, canopy cover, species richness and stem density. They were associated with waterbodies with different water depth and pH and different sediment nitrogen and phosphorus contents. Water flow rate category and underwent light availability also differed between the sites that tended to support the different vegetation types. The two major bank and shore vegetation communities differed in canopy cover, in the soil nitrogen, phosphorus and calcium levels with which they were associated and in the steepness of the bank on which they tended to grow. Floodplain vegetation communities contrasted in species richness and differed in the soil nitrogen and calcium levels and river systems with which they were associated.

This thesis investigates the role that superglacial snowpack hydrology plays in mediating meltwater delivery to glacier systems. The movement of water through glaciers is of fundamental importance as a control on proglacial hydrograph amplitude and timing, subglacial and proglacial geomorphic processes, the hydrochemistry of glacial runoff, and glacier dynamics, and as such has been the subject of considerable research effort. Although studies in non-glacial environments have shown that meltwater waves are both dampened and delayed by passage through snow, the role of supraglacial snowcover in mediating water inputs to the rest of the glacier system ahs received limited attention in studies of glacier hydrology to date. It has been suggested, however, that the varying thickness, and ultimately removal, of the superglacial snowpack may play a role in controlling the timing and magnitude of ice velocity events. Despite this suggested importance there have been few field observations of the hydrological behaviour of supraglacial snowpacks or of the way in which this behaviour evolves during the melt season. A thorough assessment of the linkages between supraglacial snowpack conditions and glacier dynamic events has therefore not been possible. This study helps fill this gap in our knowledge by explicitly investigating the hydrological behaviour of the supraglacial snowpack at an alpine glacier and its evolution during the summer melt season. Field data was collected during two summer field seasons (2003 and 2004) at Haut Glacier d’Arolla, Valais, Switzerland. Dye tracing experiments were used as the primary method of obtaining information about water flow through the snowpack. Dye was used both qualitatively, to give a visual impression of flow patterns through the snowpack, and quantitatively, with return curves detected by fluorometer providing detailed information about rates of dye movement and dispersion through the snowpack. Physically-based modelling representations of water flow through snow also informed consideration of the characteristics of snowpack runoff. Experiments were designed to determine: i) the nature of water flow through the supraglacial snowpack; ii) if, and in what way, this evolves over the course of the melt season; and iii) what factors control water movement, and the importance of their roles. In order that links between supraglacial snowpack hydrology and other parts of the glacier system could be considered, season-long records of glacier dynamics, proglacial meltwater discharge, and water quality parameters indicating subglacial conditions were also collected.

Methodology The basis of the research was to identify those physical, chemical and geological attributes of slate which make it a good roofing material. These were often related to specific Scottish quarries in order to select those which showed the best potential for resumed production. However, as no fresh samples of Scottish slate were available, the work was based initially on slate of known and reliable quality from producing quarries elsewhere. The parameters examined were mineral composition, metamorphic grade, fabric and weathering properties. When the parameters most closely related to the quality of finished slates were established then the Scottish slate was examined for these same parameters. Mineral Composition Recent progress in the determination of the chemical formulae of the phyllosilicates of fine-grained material has enabled the mineral composition of slate to be calculated from its total oxide composition. Calculations based on representative samples from each group of quarries gave good results when taken from sites in close proximity, but application to samples from more remote sites gave a weaker correlation. Metamorphic Grade Metamorphic grade affects the crystallinity and the grain size of a material. Traditionally for sedimentary rocks both properties were evaluated by the sharpness of peaks in XRD analysis, using Full Width at Half Magnitude (FWHM) as the criterion. However, this method is not effective when applied to slate, and an alternative method of determining its crystallinity was developed, based on the intensities of peaks of the main minerals.

Eutrophication results in the loss of conservation and amenity value from shallow lakes. Efforts have been made to restore shallow lakes by reducing the external nutrient loading. Removing nutrient-rich inflows can reduce nutrient loading but may alter lake hydrology. This thesis is primarily aimed at investigating the effects of a nutrient-rich river on water chemistry and plankton by comparing six shallow (<3.5 m) lakes that are connected to and isolated from the River Erewash, in the Attenborough Nature Reserve, Nottinghamshire, U.K. Lakes that received River Erewash discharge had higher nutrient concentrations and higher phytoplankton biomass than those that were isolated from it. Turbid water was also found in a lake isolated from the River Erewash but with a nutrient-rich inflow stream. Lakes isolated from the inflows had abundant submerged macrophytes and clear water but other lakes were turbid and devoid of macrophytes. In lakes without inflows, cyanobacteria were proportionally more abundant. Lakes receiving nutrient-rich water were generally dominated by small chlorophytes and centric diatoms. Phosphorus concentrations had little effect on the phytoplankton. N and Si in the connected lakes, and zooplankton grazing and N in the isolated lakes, probably limited phytoplankton. A mesocosm experiment found chlorophyll-a concentrations were reduced by the addition of silica, and that the biovolume of dinophytes increased in the mesocosms without nitrate addition. Total zooplankton biomass did not change significantly between treatments. Summer floods reduced P concentrations in the connected lakes, suggesting that internal P loading was diluted and flushed out. Phytoplankton biomass was also lower during floods. Cryptophytes and diatoms dominated the phytoplankton and cyanobacteria were rare during flooding. A simulation of lake restoration by river diversion using mesocosms confirmed the importance of lake flushing for reducing internal loading. Diverting the River Erewash in order to reduce the nutrient loading to Attenborough Nature Reserve may be problematic in the short-term, because the lack of flushing may increase the effect of internal P loading and favour cyanobacterial growth. Reducing the external supply of N may further stimulate the dominance of cyanobacteria while P concentrations remain high.

It has become increasingly clear that river channel sediment dynamics must be taken into account within British flood risk management because changes in channel morphology resulting from sediment transfer can have an impact on channel flood capacity. It is also recognised that an understanding of catchment-scale sediment dynamics is desirable with respect to many other aspects of river management. However, despite this recognition, application of existing approaches that account for coarse sediment dynamics has been limited within British river management. Based on these considerations, this study aims to develop and substantiate a new approach that quantitatively accounts for catchment-scale coarse sediment dynamics in British rivers. These research efforts contribute to the activity of the Flood Risk Management Research Consortium (http://www.floodrisk.org.uk/) A review of the availability and accuracy of data sources useful to considerations of coarse sediment dynamics reveals that only discharge, channel slope, and channel width can be represented widely at the catchment-scale. As a result, none of the approaches currently available to account for coarse sediment dynamics were found to be both scientifically robust and practically applicable at the catchment-scale. This leads to the conclusion that the most suitable approach to account for coarse sediment dynamics at the catchment-scale in British rivers is a reach-based sediment balance model, using no more than slope, width and discharge data. A new reach-based sediment balance model, ST:REAM (Sediment Transport: Reach Equilibrium Assessment Method), is developed. It has several unique features including: representation of the entire catchment network; automatic delineation of the catchment network into functional reaches using a zonation algorithm; application of a new general formula for the prediction of bed surface material transport rate; and adoption of an assumption that makes it unnecessary to collect bed material size data. The outputs from ST:REAM are in the form of predicted Capacity Supply Ratios which compare the annual mass of sediment predicted to enter a reach with the annual mass of sediment predicted to leave it. Initial assessment of ST:REAM using two test catchments shows that it can produce a reasonable representation of observed, broad-scale sediment dynamics. The accuracy of its predictions decreases when attempting to incorporate downstream variability in bed material size into the model, and scale issues are encountered when attempting to increase the resolution at which reaches are identified by the zonation algorithm. ST:REAM has many potential applications within river management, but it is of most value when providing a broad-scale picture of predicted reach sediment balances throughout the drainage network. As well as the practical applications of ST:REAM, the research contained within this thesis has important theoretical implications, relating both to the insights it provides on catchment-scale sediment dynamics in particular and methodological and foundational developments in the field of sediment studies more generally. Online version lacks Appendices, which were submitted on CD-ROM accompanying printed version.

This study of slope instability in the Matlock region of Derbyshire uses, and tests the applicability of, a variety of techniques from reconnaissance through to site investigation. The aims have been to (i) provide a greater understanding of landsliding in this area and (ii) produce a critical assessment of the techniques employed. A procedure was developed for the application of Landsat-5 TM imagery to slope stability studies, however, such imagery was found to be of very limited use in the study area. Morphological and geomorphological mapping from aerial photographs and in the field proved to be of greatest use for supplying information on landslide location, morphology, type and recent activity, and also in identifying surface water conditions. These techniques are relatively rapid, require the minimum of equipment, and provide a large amount of relevant data in a short time. They are also applicable to both reconnaissance surveys and site investigations. The geomorphological survey of Oker Hill near Matlock revealed a large multiple rotational landslide. An ochre-precipitating spring is situated within this landslide. A similar spring occurs on Mam Tor landslide in north Derbyshire, and Vear (1981) has shown this to be evidence of pyrite oxidation of the shale, a process which is believed to lead to periodic slope failure. Analysis of core material from two boreholes through the Oker Hill landslide confirmed that pyrite has been lost from the failed material and that the present base of the weathered zone is in the pyritic shale. Pyrite oxidation is resulting in iron oxide staining, the loss of sulphur (presumably as sulphuric acid which will participate in further reactions) and in an overall deterioration of the rock quality of the shale.

Characterisation of the complex geomorphological and ecological structure of river channels into workable units of instream habitat is a key step in enabling the assessment of habitat for river management purposes. The research presented in this thesis uses a range of methodological approaches at a variety of spatial scales in order to improve the conceptual basis of habitat characterisation at the reach and sub-reach scale. An appraisal of published works is used in conjunction with an extensive analysis of habitat features for sites across the UK, and intensive field studies on the River Tern, Shropshire, to improve the conceptual basis and ecological validity of the 'physical biotope' as the basic unit of instream habitat. Physical biotopes demonstrate correlations with biologically functional habitat units at relatively broad scales, suggesting that assemblages of habitat units may provide the most appropriate level of simplification of aquatic habitat structure. A simplified, but more transferable classification using biotope assemblages is suggested, with potential application to a range of instream assessment and river design needs. Reach-scale field surveys reveal complex and dynamic relationships between channel hydraulics and morphology and highlight the influence of sampling design and hydrological context on the outcomes of rapid field surveys. A microscale research component addresses within-biotope variation at small scales by focusing on high frequency flow behaviour and sediment transport mechanisms which have, to date, been largely overlooked in biotope studies. This provides both detailed descriptions of hydraulic behaviour, and an indication of differences in internal spatial and temporal heterogeneity between biotopes, with implications for instream biota.

There is growing concern that the adoption of intensive agricultural land management practices in upland areas of the UK over the past 50-60 years may have affected hydrological responses and sediment transfer regimes in river catchments and could, therefore, be contributing to increased levels of flood risk and ecological disturbance. However, recent evidence from a research catchment at Pontbren in mid-Wales indicates that the implementation of a more sustainable livestock farming strategy could help to mitigate some of these impacts, raising the possibility that strategic land use planning could be used as a cost-effective, multi-functional river management option. The impacts of historical land use changes and land management practices on contemporary sediment dynamics in the study area are explored in this thesis through a system approach which acknowledges the importance of interrelationships between hydrological and geomorphological processes. Results from hydrological experiments and modelling exercises are used to inform analyses of spatial and temporal variation in sediment production and transfer from a variety of potential sources. Grazed, agriculturally-improved pastures were found to supply fine material to stream channels via both surface runoff and field drains. In particular, drain-derived sediment is likely to represent an important component of the total fine sediment yield in subcatchments where agricultural intensification has been widespread. Agricultural drainage ditches were also found to act as sources of sediment in such areas, along with eroding channel banks. Sediment production from bank sources may relate to historical changes in peak flows caused by agricultural intensification. Stream sediment yields are strongly related to differences in sediment supply from the aforementioned sources and could therefore be reduced by limiting mobilisation at the point of origin within the landscape. In terms of channel-derived material, this could be achieved through peak flow reductions associated with woodland and hedgerow restoration.

This study investigated contemporary floodplain sedimentation, interactions between sediment, vegetation, and agricultural land use, and the potential utility for a Bayesian Network Decision Support System (BNDSS) to assist farmers in making better decisions concerning agricultural land use. The research was performed around Bara Bania Mouza (village) under Daulotpur Uazila in Manikgong district of the Brahmaputra-Jamuna floodplain in Bangladesh. This area was selected because it is representative of the young and active floodplain, where the land is flooded and receives overbank sediment deposition every year. The research employed exploratory data analysis and Bayesian approaches to identify and investigate causal relationships among the variables and so support probabilistic inferences. The study investigated two distinctly different types of monsoonal flood: a bonna (an abnormally large flood that occurred in 2007) and a barsho (a normal flood that occurred in 2008). Data on landforms, flood hydraulics, sediment dynamics (suspended sediment concentrations and sediment accumulation rates), and vegetation, rain-fed flooding, land use and farmers knowledge on soil suitability and cropping were collected through field surveys. The results establish how flow and sediment dynamics contrast as a function of landform and demonstrate that the thickness and calibre of deposited sediment strongly influence farmers' decisions on which and how many crops to cultivate on a given plot. Natural vegetation (e.g. sun grass) and certain agricultural crops were shown to have huge potential for use in slowing floodwater and trapping coarse grain sediment particles in buffer stripes. Marked contrasts were also observed between the characteristics of sediment deposited by rain-fed and river water flooding. Questionnaires and semi-structured interviews revealed that although farmers have profound knowledge on soil types and crop associations their methods are crude and little or no science is involved in the investigation of soil and sediment properties. Despite this, farmers' estimates of soil properties proved to be reasonably accurate with the estimate of particle size differing by only <15% from the results of laboratory particle size analysis. This suggests that the farmers' methods do give reliable indications of key soil attributes, but that they could be improved if scientific information was integrated with their local knowledge. A Bayesian approach provides a means of achieving this and the BNDSS developed in this study was found to produce good results when compared to field observations and backward propagation indicated that for better decision making it is crucial to consider both physical and socioeconomic variables. The findings of the research reported in this thesis show that sedimentation has major impacts on agricultural land use dynamics in the Brahmaputra-Jamuna floodplain and that both natural vegetation and agricultural crops significantly influence sediment movement and the way that deposition is distributed over the floodplain. In a wider context, flood, sediment, vegetation and agricultural land use dynamics are controlled by complex set of both physical and human phenomena that are challenging to describe, integrate, analyse and interpret in a single study. In light of this, it is not surprising that the findings presented in this thesis highlight important gaps in knowledge that need to be addressed through further research.