Search

The Loch Lomond Stadial (LLS) refers to the abrupt return to severe cold conditions that occurred in Britain, between 12.9 and 11.7 ka, subsequent to the retreat of the last (Late Devensian) British-Irish Ice Sheet. This period has long been associated with the regrowth of glaciers in upland areas of Britain and left a wealth of geomorphological evidence in the landform record. However, previous research on these glaciers has largely comprised localised case studies, producing a fragmented and spatially inconsistent dataset. This thesis draws together the published geomorphological evidence for Loch Lomond Stadial glaciation to build a coherent picture of the extent, style and dynamics of glaciation during the stadial. Geomorphological mapping of glacial landforms associated with this period is compiled from the published literature to create a map and geographic information systems database of over 95,000 features. The evidence used to produce this map is critically assessed in the most comprehensive review of the Loch Lomond Stadial to date and is used to identify conceptual themes, common to the geomorphology in multiple sub-regions within Scotland, England and Wales. Persisting uncertainties, particularly regarding the extent and timing of Loch Lomond Stadial glaciation, are discussed and recommendations of future research to address these are made. Building on this review, the glacial geomorphological map is then used to construct five glacial landsystem models which reflect the style of Loch Lomond Stadial glaciation; the cirque/niche glacier landsystem, the alpine icefield landsystem, the lowland piedmont lobe landsystem, the plateau icefield landsystem and the ice cap landsystem. Use of these models to classify the Loch Lomond Stadial glacial geomorphology reveals the spatial distribution of each landsystem. Three styles of glacier retreat are represented by the glacial geomorphology. It is demonstrated that both landsystem and retreat style reflect the combined importance of pre-existing topography and palaeoclimate. Given the paucity of dating constraints on Loch Lomond Stadial landforms, the thesis pilots the use of a relative dating technique using soil chronosequences to differentiate between Loch Lomond Stadial and older moraines in the English Lake District. The results of this study highlight the potential of this technique to discriminate between Loch Lomond Stadial and pre-Loch Lomond Stadial moraines in Britain, although further work is required.

This thesis explores continental weathering patterns in glacial and non-glacial river catchments in Iceland and Greenland. Specific attention is placed on characterizing the relative iron (oxyhydr)oxide export rates of glacial and non-glacial catchments to the ocean. Total element concentration and iron stable isotope measurements indicate that chemical weathering differences do exist between glacial and non-glacial catchments. The differences appear primarily related to variations in soil formation and organic matter availability between the environments. Physical, rather than chemical, differences between glacial and non-glacial catchments however dominate the relative differences in (oxyhyr)oxide export rates. Glacial and non-glacial river sediments from otherwise analogous terrains contain about the same quantity of iron (oxyhyr)oxide on a weight normalized basis. This equates to glacial rivers exporting far more iron (oxyhyr)oxides on a discharge weighted basis, because glacial rivers contain far higher suspended sediment concentrations that non-glacial rivers. Existing research shows that organic carbon accumulation and burial in marine environments scale directly with iron (oxyhydr)oxide accumulation. This means that shifts in continental weathering over glacial-interglacial cycles drive further changes in marine carbon burial creating a global climate feedback loop.

Groups of closely spaced minor moraines allow for observations of moraine formation and ice-marginal fluctuations on short timescales, helping to better understand glacier retreat and predict its geomorphological effects. Some minor moraines can be classified as annual moraines given sufficient chronological control, which implies a seasonal climatic driver of ice-marginal fluctuations. This leads to moraines being utilised as very specific, short-term records of glacier fluctuations and climate change. This research is common in lowland, maritime settings, but remains sparse in high-mountain settings. This study presents the detailed geomorphological and sedimentological results of minor moraines at two high-mountain settings in the European Alps. Geomorphological investigations included mapping and measurements through field observations and remotely-sensed imagery. Detailed sedimentological investigations followed excavation of moraines and include multiple scales of observation and measurements to support interpretations of sediment transport and deposition. Additionally, ground-penetrating radar data were collected in one foreland. Minor moraines at Schwarzensteinkees, Austria, formed as push or combined push and freeze-on moraines in two groups between approximately 1850 and 1930. The existence of a former proglacial lake appears to have exerted a strong control on moraine formation. Modern minor moraines at Silvrettagletscher, Switzerland, exist primarily on reverse bedrock slopes and have formed since approximately 1850 through push, freeze-on, and controlled moraine mechanisms. The presence of these bedrock slopes, and in some areas englacial debris septa, appear to exert the primary controls on moraine formation. The foreland of Gornergletscher, Switzerland, has been revisited using aerial imagery to assess if moraines are still forming annually, and this has been confirmed. These findings show a range of mechanisms responsible for moraine formation, which are then compared to previously published research on minor moraines to elucidate any common drivers of minor and annual moraine formation globally. This includes a global database of forelands where minor moraines have been studied, created as part of this research and presented as a table and Google Earth file, both easily accessible and freely available online, for use by other researchers when exploring similar topics.

The reconstruction of marine-based paleo ice sheet dynamics can reveal long-term ice sheet activity, and in turn provide constraints on the response of modern marine ice sheets (e.g. the West Antarctica Ice Sheet) to climate change. The marine-terminating Hebrides Ice Stream (HIS) flowed across the western Scottish shelf during the last glacial maximum (3024 ka) and drained a large portion of the northern sector of the British Irish-Ice Sheet (BIIS), affecting its stability. This thesis aims to examine how the HIS evolved and interacted with the changing climate and the underlying landscape after 27 ka. The work is subdivided into: a) an investigation of modern high-resolution bathymetry data coupled with seismic data with the aim of reconstructing deglacial dynamics; b) the analysis of Pb isotopic composition in sediment cores on the shelf in order to locate glacial sediment provenance; c) a study of Hebrides shelf core sedimentology and microfaunal assemblage to reconstruct Lateglacial paleoenvironmental changes. A three-stage deglacial pattern, where topography played a critical role, is defined: i) ice stream margin retreat punctuated by standstills, ii) topography-controlled fjordic retreat, with evolution from a coherent ice-sheet to separate fjord tidewater glaciers, and iii) a stabilisation at the transition from tidewater to land-based ice margins. Between 21 and 15 ka, fine-grained sediments transported by meltwater plumes were the product of erosion of Neoproterozoic basement, while the coarse-grained sediments were instead sourced from island igneous rocks. These results indicate prevailing sediment input from NW Scotland. Lateglacial sediment deposition was strongly influenced by shelf currents and shows wide variation. Therefore, the seismic and sedimentological interpretations need to be considered only on a local scale. Glacimarine sandy deposits in the Muck Deep region support a prolonged glacial occupancy until the latest stages of GS-1 (12.8-11.7 ka), and are at odds with recent studies indicating earlier glacial retreat. The thesis demonstrates the complex interactions between BIIS evolution, subglacial landscape and ocean dynamics. The outcome of this research can be useful to inform future numerical reconstructions.

This study examines how longitudinal foliation develops in glaciers and ice sheets in a wide range of topographic, climatic, and dynamic settings, at a variety of spatial scales. Study locations include four valley glaciers in Svalbard (Austre Brøggerbreen, Midtre Lovénbreen, Austre Lovénbreen, and Pedersenbreen), a valley glacier in Canada (Sermilik Glacier), and seven outlet glaciers in Antarctica (Hatherton Glacier, Taylor Glacier, Ferrar Glacier, Lambert Glacier, Recovery Glacier, Byrd Glacier, and Pine Island Glacier). Detailed structural mapping of the valley glaciers from satellite imagery and field-based measurements were used to document the formation of longitudinal foliation in small-scale ice masses. These findings were ‘up-scaled’ and applied to much larger glaciers and ice streams. Longitudinal foliation develops in concentrated bands at flow unit boundaries as a result of enhanced simple shear. However, longitudinal foliation is not directly observable from satellite imagery at the surface of larger-scale valley glaciers. The longitudinal structures visible at the surface of larger-scale glaciers form at flow unit boundaries and are composed of bands of steeply dipping longitudinal foliation; however, they appear as individual linear features on satellite imagery as a result of the comparatively low spatial resolution of the imagery. The persistence of flowlines in the Antarctic Ice Sheet through areas of crevassing and net ablation (blue-ice areas) suggests that they are the surface representation of a threedimensional structure. Flowlines are therefore inferred to be the surface expression of flow unit boundaries composed of bands of steeply dipping longitudinal foliation. The survival and deformation of flowlines in areas of ice flow stagnation indicates that flowlines form in their initiation zones and not along their entire length. Furthermore, these ice stagnation areas indicate that flowlines record past ice dynamics and switches in ice flow.

This research investigates the impact and timing of glaciations on the western Irish shelf and their sedimentary record. A multiproxy approach including seismic interpretation, marine sediment core analysis, radiocarbon dating, geomorphological analysis and Cosmogenic Nuclide (CN) exposure dating of landforms and erratics onshore is used to assess the glacial dynamics on the western margin of the British- Irish Ice Sheet (BUS) during the last glacial period. The combination of the methodologies and materials used and subsequent correlation with proxy climate records allows for an interpretation of the forcing mechanisms associated with the waxing and waning of the BUS in this region. Analysis of seismic profiles from the western Irish shelf reveals for the first time the internal stratigraphy of the large morainic complex, situated between ~52.5°N and ~54°N, which has been deposited by a grounded ice sheet on the mid-shelf. An early Quaternary shift in the style of sediment deposition at the outer Donegal Bay margin is associated with the onset of glaciations on the western Irish shelf. Lithofacies interpretations and radiocarbon dating of marine sediment cores suggest an asynchronous retreat pattern of the last BUS from north to south on the western Irish shelf. This study bridges the knowledge gap between previous investigations of the region and establishes a shelf-wide correlation of grounded ice sheet retreat on the western Irish shelf based on sediment core analysis and radiocarbon datings for the first time. The palaeoenvironmental changes during ice sheet retreat are assessed by an increased percentage of benthic foraminifera assemblages associated with temperate environments from a core at the outer shelf. The synthesis of CN exposure ages of erratics onshore, radiocarbon dated sediments from offshore and mapped glacial features allows for chronological constraint of a glacial readvance into Donegal Bay that flowed offshore from north county Mayo after 17.6 cal ka BP after the main-Last Glacial Maximum (LGM) ice sheet had retreated eastward towards the Sligo coast. The forcing mechanism for this ice margin readvancing into Donegal Bay is therefore interpreted as a glacidynamic response associated with debuttressing during general ice retreat, i.e. the retreat of an ice lobe eastward towards the Sligo coast that acted as a buttress to the ice lobe with a perpendicular ice flow pattern advancing from northern county Mayo into Donegal Bay; thus resulting in the ice margin readvance of the debuttressed ice lobe from northern county Mayo. Finally, dated deglacial offshore and onshore moraines and erratics available to this study are used to build a time-constrained-reconstruction of the dynamic behaviour of the western margin of the BIIS from the LGM to its deglaciation during the Woodgrange Interstadial (Irish terminology/equivalent to the Bolling-Allered Interstadial).

Despite a long history of research, dating back to the 1800s, we have only a generalised overview of the flow patterns and behaviour of the former Newfoundland Ice Sheet with much of its basic properties unknown or contested. While a substantial body of work has been carried out this has primarily used traditional field mapping approaches with geomorphic and sedimentary features recording local to regional scale ice sheet events being extrapolated up to ice sheet scale to produce ice sheet models. This evidence, from fragmented study areas often misses the larger scale behaviour of the ice sheet which has been revealed in palimpsest glacial landscapes revealing reorganisation of the ice sheet bed or bedform preservation by cold based ice. The problems associated with this approach has been alleviated by using the more contemporary "top-down" ice sheet interpretation in which remote sensing is used to map individual glacial landforms across the full ice sheet bed. Detailed landform maps allow for unravelling of complex flow patterns and geometries recorded in superimposition and cross-cutting bedform assemblages, thereby revealing ice flow pathways, ice divide positions and ice sheet configuration at snapshots in the ice sheets evolution and from which ice sheets can be reconstructed. This thesis applies this top-down eomorphological-data-driven approach to reconstruct the former Newfoundland Ice Sheet using a new database of greater than 165, 000 individually mapped subglacial and ice marginal glacial landforms derived from a systematic island-wide remote sensing mapping programme. This new subglacial landform record, comprising of > 50,000 glacial lineations and > 107,000 ribbed moraines, has identified a palimpsest landscape of superimposition, crosscutting and bedform preservation of more complexity than has been previously recognised. This has revealed a dynamic ice sheet continually modifying its bed during periods of shifting flow geometry, significant restructuring in ice sheet configuration, large-scale migration of primary ice dispersal centres in the order of 140km, with periods of active warm-based bedform development separated by periods of high bedform preservation. These subglacial patterns record spatial and temporal fluctuation in thermal dynamics with bedform distribution patterns also recording a disparate pattern of thermal dynamics with cold-based sluggish flow persisting for longer in the south and southeast while other sectors experienced warm-based active ice flow for much of the last glacial episode. The newly documented subglacial landforms of glacial lineations and ribbed moraines provide the primary units for the glacial inversion technique which follows a formalised framework of known glaciological principles to extract the information glacial landforms provide. A flowset approach was used to group the large number of these landforms into summary units. These represent discrete ice flow events with the characteristics of component landforms providing key information on the spatial, temporal and glaciodynamics of their development. These flowsets were used in conjunction with a reconstructed retreat pattern that was derived from the new mapping of ~ 5000 ice marginal landforms of eskers and moraines and alongside numerical dating constraints from the literature and published radiocarbon databases. These provided the full evidence base from which a twelve-phase model of the last Newfoundland Ice Sheet has been produced. This new model has significantly advanced the knowledge of the behaviour of the former Newfoundland Ice Sheet and importantly brings its response patterns into more apparent correlation with major climatic events which drove abrupt change in ice sheet behaviour during the last glacial episode. The reconstruction is therefore an important framework for further investigation of the role and response patterns of ice sheets in the wider climate system which ultimately may lead to better understanding as to how contemporary ice sheets may respond to future climate change.

Spatiotemporal variability in supraglacial debris properties have the potential to affect estimations of mass balance for debris-covered glaciers. This thesis is concerned with identifying the extent to which debris properties vary in space and time, and the role of these variabilities in estimations of specific mass balance. The research presented uses a combination of methods, including statistical analysis of field data, mapping and classification of thermal and optical remotely sensed data, and numerical modelling. Near-surface debris temperature was measured in the field to investigate short term spatial and temporal variability in debris properties and its influence on debris temperature over a monsoon season. The strongest correlation between timeseries of near-surface debris temperature and meteorological controls was with air temperature, with lesser correlations between rate of change in air temperature and incoming shortwave radiation. Subtle spatial variability was also identified, controlled by site aspect and slope, grain lithology and size, and potentially moisture content and thermal conductance of the bulk debris layer. The occurrence of spatiotemporal variability in supraglacial debris distribution was identified on Baltoro Glacier over a sub-decadal timescale, considered to be primarily due to differences in input of debris to the glacier system through mass movement events. The importance of variability in debris properties was investigated using a surface energy balance and heat transfer model, where the influence of debris thickness, albedo and aerodynamic roughness length was tested. The modelling results, although not directly comparable to mass balance estimates for Khumbu Glacier, showed a 223% increase in total specific mass balance for Khumbu Glacier’s debris-covered area over a monsoon season when a spatially variable debris layer was included. Including spatially variable albedo and aerodynamic roughness length along with debris thickness reduced estimates of specific mass balance, although were still higher than the base line model simulation. Consequently, this thesis confirms the occurrence of spatially and temporally variable supraglacial debris properties, over seasonal and sub-decadal periods, and that such variability is influential for estimates of glacier mass balance.

The mid-latitudes of Mars host numerous ice-related landforms that bear many similarities to terrestrial ice masses. One particular landform has a strong resemblance to valley or debris-covered glaciers found on Earth and have subsequently become known as ‘glacier-like forms’ (GLFs). GLFs have detailed surface morphologies consistent with recent deposition and viscous deformation of ice, but there is still uncertainty regarding their formation, current and former volume, and dynamic evolution. Specifically, this thesis presents new observations and results that assess the current and former volume and dynamics of Mars’ mid-latitude GLFs. The findings presented in this thesis suggest that at the broadest level Mars’ midlatitudes appear to preserve a complex and spatially heterogeneous record of glaciation and that GLFs appear to be the manifestation of a more localised period of alpine glaciation. GLFs represent an active component of the near-surface water budget of Mars, locking away an estimated global equivalent water layer of between 3 ± 1 and 10 ± 3 mm. Evidence of recession and mass loss was identified in approximately one-third (n = 436) of the GLF population, suggesting that these landforms once contributed a larger volume of water to the near-surface water budget of Mars. Assessment of environmental and topographical controls over current ice volume and ice-mass loss revealed that their distribution and size is unlikely to be purely controlled by insolation forcing. Instead it is suggested that regional to local meteorological and topographical conditions also play an important role in GLF ice accumulation and/or preservation, with variation in physical environments providing microclimates favourable for accumulation and/or preservation of ice. The emerging picture shows that Mars’ GLFs appear to have been dynamically active, and that they have played an important role in altering the surface landscape of Mars through erosion, transport and deposition of material.

Glaciers occupy 11% of Earth’s total surface and represent a significant but, as yet, poorly characterised ecosystem. As late as the mid-90s glaciers had been regarded as microbiologically sterile environments but there has since been major progress characterising diversity and functioning of glacier microbiota. The supraglacial environment has to-date been prioritised, but crucially the subglacial microbiota remains generally unknown despite their central importance in geochemical cycling. The dark and oligotrophic conditions typical to subglacial environments in general, and sediment entrained basal ice, in particular, are likely to select for chemolithotrophic carbon fixers that, by definition, will enable diverse heterotrophic microbial community development. Therefore, the basal ice microbiota, are likely to play fundamental roles in mineral weathering and geochemical cycling not only within basal ice but also subsequent foreland soil formation upon release. The main aim of this thesis is the first integrated geo-microbiological characterization, of geomorphologically distinct basal ice facies targeting an Icelandic temperate glacier, Svínafellsjökull. Here we show, via novel culture-dependent and -independent next generation molecular rRNA gene marker (16S and ITS) phylogenetics, that basal ice facies harbour a rich and diverse community of bacteria (Proteobacteria and Acidobacteria) and fungi (Ascomycota and Basidiomycota). An abundance of chemolithotrophic species (Thiobacillus, Gallionella, Nitrosospira) characterise the basal ice microbiome that is directly supported in the identified geochemical status of basal ice. The presence of reduced nitrogen species in the ice matrix and iron- and sulphur-rich minerals in basal ice sediment provides added functional support of the predominance of chemolithotrophs in basal ice. Based on total basal ice cell enumeration and export estimates (~1016 cells yr-1) it is clear that chemolithotrophic and heterotrophic microbial communities identified in foreland soils originate from basal ice, as opposed to supraglacial sources. Modelling highlights the importance of microbial activities on the geo-chemistry of the basal ice, e.g. oxidising minerals, acidifying the environment, and increasing the carbon content within the sediment. Once released, basal ice-derived microorganisms can survive the psychrophilic to mesophilic temperatures of the foreland and identified isolates from basal ice and the foreland were affiliated with species that play important roles in weathering and soil formation highlighting the functional importance of the basal ice microbiota both, in glacial and periglacial systems.

Despite the importance of ice streaming to the evaluation of West Antarctic Ice Sheet (WAIS) stability, we know little about mid-to long term changes in grounding line migration, ice streaming and ice accumulation in the upper Institute Ice Stream (IIS) catchment. In this thesis ground penetrating radar (GPR) and airborne radio-echo sounding (RES) methods have been employed to investigate the subglacial topography, internal stratigraphy and Holocene flow regime of the upper IIS catchment, in and around Horseshoe Valley. High resolution step-andcollect mode GPR was employed to assess the continuity of a Blue Ice Area (BIA) horizontal ice core climate record at Patriot Hills, where analysis has revealed two unconformities in the otherwise conformable 30,000 year climate sequence. By combining these data with airborne RES returns and pre-existing ice sheet models it is suggested that these unconformities represent periods of erosion, occurring as the former ice surface was scoured by katabatic winds in front of Liberty and Mable Hills. Snow_Blow simulations suggest that katabatic winds have scoured the leeward slopes of these mountain ranges for over 10,000 years. This temporal stability can account for the large volume of BI moraine deposits in Horseshoe Valley, where compressive BI flows promote glacial erosion and near-surface debris entrainment through freeze-on processes at the ice/bed interface and compressive thrust faulting. By investigating thicker ice flows in the upper IIS catchment and the Evans Ice Stream, this thesis has also analysed debris entrainment mechanisms at depth, where clasts are incorporated into the ice flow by englacial stratigraphic folding and shearing at the glacial thermal boundary, governed by spatial and temporal changes in ice flow, ice temperature and sediment availability. Mid-to long term changes in ice flow in the wider IIS catchment have been investigated from airborne RES transects, revealing internal layer buckling, and therefore former enhanced ice-sheet flow in three distinct tributaries of the IIS. Buckled ice layers throughout the slow flowing ice in the Independence Trough and the fast-flowing ice in the Ellsworth Trough suggest that enhanced ice flow through these topographically confined regions was the source of ice streaming and iceflow reconfiguration during the mid-to-late Holocene. Although buckled layers also exist within the slow-flowing ice of Horseshoe Valley, a thicker sequence of surface-conformable layers in the upper ice column suggests slowdown more than 4000 years ago, indicating that enhanced flow switch off here cannot be attributed to late-Holocene ice flow reorganisation. The dynamic nature of ice flow in the IIS and its tributaries suggests that ice stream switching and mass change may have been regular during the Holocene, and that these changes may characterise the decline of the WAIS in this area. These results have important implications for our understanding of ice-sheet dynamics and the response of the ice sheet to climate change and provides explanations for fluctuations in debris entrainment and transportation processes in Antarctica.

The Late Devonian and Mississippian record the transition from Greenhouse into a cooler global climate punctuated by glaciations. Evidence for glaciation in the latest Devonian is primarily derived from diamictites deposits in South America. The event was broadly coincident with eustatic changes and mass extinction associated with the Hangenberg Crisis. To understand the record of glaciation in more detail a stratigraphic, palynostratigraphic and chemostratigraphic field-study has been undertaken near Lake Titicaca, Bolivia. Three primary sections were measured; Chaguaya, Villa Molino and CR5. These record a 1.1 km thick Devonian-Carboniferous (D/C) boundary interval. Glacigenic strata are represented by the Cumaná Formation, dated to within the lepidophyta-nitidus (LN) miospore zone. Additional sections were measured along a strike-parallel topographic ridgeline, which revealed that the Cumaná Formation consists of a single, preserved advance of glacigenic strata into a marine shelfal environment. It is associated with incised palaeo-relief, which is >7 km in length and 110 m deep. The D/C boundary in the study area is defined on the loss of Retispora lepidophyta and a collapse in marine phytoplankton diversity during initial post-glacial transgression. Hence, palynological extinctions represent a proxy for the D/C boundary in South America that is tied to wider palaeoclimatic and glacioeustatic changes. A 2 ‰ positive carbon isotope excursion is associated with palynological extinctions at the D/C boundary. Palynofacies were unable to quantify the marine vs. terrestrial ratio of preserved organic matter as amorphous organic matter was preferentially lost during standard palynological processing techniques. However the excursion is likely related to a reduction in the supply of terrestrial and marine organic matter at a time of environmental stress. These combined results provide a high-palaeolatitude glacial perspective on debates regarding mass-extinction, eustatic and environmental change around the D/C boundary. The glacigenic Cumaná Formation is interpreted to represent the glacioeustatic control sea-level falls observed globally at the D/C boundary.

Sea level is an excellent proxy for past climate change as it represents the combined impact of changing temperatures and ice volumes through time. Reconstructing a record of global ice volume change is complex as the growth and loss of high volume ice sheets results in a spatially varying pattern of sea-level change. This is known as glacial isostatic adjustment (GIA) and means that all past sea-level indicators are effectively relative sea level (RSL) indicators. Each indicator is relative to a particular position on the Earth’s crust and requires a GIA correction to reconstruct global mean sea level (GMSL). Generating a GIA correction requires an appropriate ice volume and distribution history. As no field-constrained global ice history exists beyond the last glacial maximum we create five different global ice-loading histories to investigate a range of potential ice volume and dispersal scenarios through the last interglacial. Within this thesis we develop a methodology for inclusion of coral taxon depth-habitat relationships in the uncertainties associated with fossil coral reconstructed relative sea levels. We test our ice histories against the coral dataset, and find the best matches to the coral dataset come from ice histories that contain a longer interglacial and / or reduced ice volume through the interglacial than is currently found in many continuous records of sea level. We model the GIA response of the Hanish and Camarinal Sills, and Rosh Hanikra on the Israeli coast to determine how two continuous RSL curves, for the Red Sea and Gibraltar respectively, and the temporally discrete RSL indicators relate to GMSL. Our analysis reveals sensitivities that may be used to constrain the evolution of a past Eurasian ice sheet.

The last two glacial terminations represent the most recent, and best documented, periods of Earth warming in the geological record. During these terminations atmospheric CO\textsubscript{2 }rose by approximately 100 ppm and global mean temperatures increased by 4-6\textsuperscript{o}C. Whilst the driver for these deglaciations ultimately derives from changes in the insolation forcing at the edge of the atmosphere, feedbacks within the Earth\textquoteright s climate system act to amplify these small external forcings tipping the Earth from a cold glacial climate state to a warm interglacial climate state. A key question in Quaternary climate science is understanding which feedbacks are important in regulating global climate on glacial-interglacial timescales. On this topic, the Southern Ocean has long been considered to be an important player in regulating atmospheric CO\textsubscript{2 } on glacial-interglacial timescales. This thesis investigates some of the hypothesised drivers of changes in atmospheric CO\textsubscript{2 } on glacial-interglacial timescales by generating high-resolution multi-proxy records from the Southern Ocean spanning the last two glacial terminations. In particular, I focus on changes in the structure, circulation and biological productivity within the sub-Antarctic zone. A change in the deep ocean density structure has been hypothesised to have resulted in the release of CO\textsubscript{2 } from the deep ocean. Centennial records from the sub-Antarctic are used to reconstruct deep and intermediate water density for the first time. I demonstrate that timing of the major breakdown in the density gradient of the ocean significantly lagged the breakdown in the chemical gradient, suggesting that changes in the deep ocean density structure were not the major driver of the deglacial rise in atmospheric CO\textsubscript{2 }. Changes in the density structure of the Southern Ocean likely had significant implications for global circulation. In particular, the flow of low salinity water through the Drake Passage is thought to be important in setting the strength and geometry of Atlantic Overturning Circulation. Drake Passage through-flow speed was reconstructed from two sites in the central and northern margins of the Antarctic Circumpolar Current downstream of Drake Passage. These records suggest a very different structure of Antarctic Circumpolar flow through Drake Passage during glacial periods, and evidence significant changes in ocean temperature as a result of pronounced reductions in Drake Passage through-flow. The strength of the biological pump has long been identified as an important player in regulating atmospheric CO\textsubscript{2 }. In particular, a strong glacial increase in sub-Antarctic productivity has been observed at open ocean sites in the South Atlantic and Indian Ocean. However, the glacial-interglacial changes in productivity in sub-Antarctic shelf settings are less well-documented. The new high-resolution records presented here from the sub-Antarctic southwest Atlantic suggest a significant change in the CaCO\textsubscript{3}:C\textsubscript{org} ratio which likely has implications for the surface ocean\textquoteright s ability to uptake CO\textsubscript{2 }.

During the mid- to late Neogene (20 - 2.5 million years ago), episodic retreat of the Antarctic Ice Sheet (AIS) coincided with periods of higher-than-present atmospheric CO2, indicating ice sheet sensitivity to climatic conditions similar to those projected for the coming decades. Understanding Antarctic climate and vegetation during such a period of AIS retreat is crucial for our fundamental understanding of high latitude environments in warmer-than-present climate scenarios. This thesis presents a detailed geochemical study of sediments and plant fossils from the terrestrial Sirius Group of Oliver Bluffs, Transantarctic Mountains, located at 85 °S today and during the Neogene. Biomarker analysis of the sediments show strong evidence for a warmer Antarctica, where summer temperatures reached 5 °C. These relatively favourable conditions sup- ported a low diversity mixed vegetation. In contrast to the macrofossil record, there is geochemical evidence for conifers, suggesting that Antarctic vegetation was strongly controlled by local environmental variability. The warmer conditions are associated with a dynamic carbon cycle, evidenced by anomalously high and variable atmospheric 13C and possibly linked to atmospheric CO2 levels. Precipitation isotopes are reconstructed from plant compound isotope analysis of the fossils, and indicate markedly different hydrological cycling. This result is supported by climate modelling experiments which suggest that Antarctic hydrological cycling is most strongly governed by the extent of the ice sheet rather than by greenhouse gas radiative forcing. This thesis presents a new approach to exploring Antarctic climate and vegetation and provides important novel information on this crucial region of the world.

New Zealand fills a large geographic gap in the global glacial record, with landforms and near-surface deposits preserving a remarkable footprint of Quaternary glaciation. As one of the few land masses in the Southern Hemisphere, the record of glacial geomorphology is of great importance for research into the natural behaviour of the Earth’s climate system. This thesis presents a 3D simulation of the New Zealand Southern Alps glaciers at the Last Glacial Maximum (LGM, c. 30 to 18 ka) in an attempt to constrain glacial geometry of that period. To achieve this, the REVOLTA (Reconstruction of Volume and Topography Automation) model was developed, a Python script tool for ArcGISTM that requires just a DEM of glacier bed conditions and the down-valley extent of glaciation as initial inputs. Ice thickness is initially estimated at points along an automatically generated centreline network based on the perfect-plasticity rheology assumption, taking into account a valley side drag component of the force balance equation. Distributed ice thickness is subsequently interpolated using a glaciologically correct algorithm. Results indicate a total LGM ice volume of 6771.9 km3, in good correspondence of previous studies using a climate-driven ice dynamics approach. Combined with an estimate of contemporary ice volume (50.67 km3), this result reinforces the notion that New Zealand has lost almost the entirety (99.25 %) of its glacial ice since the LGM, although this volume has contributed to just 17.02 mm of global sea level rise. Analysis of the LGM distributed ice thickness output shows a large number of nunataks and exposed ridges in the central Mt. Cook and northern regions, with a localised icefield in the Fiordland area. LGM Equilibrium Line Altitudes (ELAs) automatically calculated using the Accumulation Area Ratio (AAR) method reveals an average lowering of 1074 m from present, with those to the west of main divide 461m lower than those to the east on average. LGM climatic conditions were estimated using the ELAs and scaled versions of contemporary temperature and precipitation distributions, suggesting a temperature reduction of between 5.6°C to 10.3°C and precipitation change of +4.3 % to +100.4% from present. When considering these new estimates in conjunction with critically evaluated previous evidence, an average LGM cooling of 6.5 °C to 8°C is proposed, a refinement on the wide range of previously published values. Importantly, there is large spatial variability between catchments, with eastern regions experiencing significantly greater cooling and greater precipitation increases (or less decrease) than their western counterparts. Increased westerly circulation and reduced sea level altering the relative position of the orographic barrier is a suggested potential mechanism for the predicted precipitation pattern changes, whilst increased southerly flow bringing cool air up the east coast is a possible cause of the temperature change differential predicted. The proportion of precipitation falling as snow or rain was also found to be an important factor when considering New Zealand LGM conditions, with up to 50% estimated to be falling as rain at the LGM ELA, with a strong east-west differential. Input dataset generation for REVOLTA resulted in several important research outcomes. A DEM approximating LGM bed conditions was created, using a variety of novel techniques to modify the existing DEM. These included the estimation of contemporary ice thickness distribution using the VOLTA (Volume and Topography Automation) model for removal from the DEM, merging offshore and lake bathymetry and considering Holocene in-fill sediments. Furthermore, an in-depth review of the most up to date literature and datasets regarding the lateral extent of LGM glaciation was also carried out, generating an updated ‘outline’ of LGM glaciation.

Glaciers and ice caps are recognised as an important component of the global carbon cycle. Carbon within glacial systems exists in organic and inorganic forms, across supraglacial, englacial and subglacial realms. It is often difficult to detach cryospheric carbon cycling from hydrology, with the transfer of carbon between glacial inventories relying upon meltwater flows. Classical glacial hydrology consists of distributed drainage delivering delayed flow meltwaters, throughout the accumulation season, superseded by quick flow, aerated channelized drainage during increased ablation. It is upon this template that most existing studies have addressed the dynamics of carbon within glaciated catchments. However, Icelandic glacial systems provide an opportunity to investigate the role of subglacial volcanism in driving carbon dynamics. Hydrochemical properties of Sόlheimajökull bulk meltwaters indicate untraditional redox conditions, with discharge of reduced, anoxic meltwaters in Summer, when expansion of subglacial drainage intersects the Katla geothermal zone. This unique hydrological regime generates profound effects upon the solute flux from the glacier, particularly with regard to the carbon budget. Dissolved inorganic carbon dynamics are dominated by weathering of basaltic bedrocks and accessory hydrothermal calcites, fuelled by subglacial geothermal proton supply. Widespread basal anoxia during summer facilitates methanogenesis resulting in large quantities of methane being discharged from beneath the glacier (flux range between 9,179 to 22,551 tonnes per year). Evidence suggests subglacial microbial acetoclastic methanogenesis is responsible with δ13C and δD CH4 values of ~60‰ and -320‰ respectively, supported by laboratory identification of methanogenesis in Sόlheimajökull subglacial sediments. The organic counterpart to the carbon cycle is invoked to serve as the energy source for microbial metabolism. Such direct measurements of subglacial methane have rarely been achieved at contemporary ice margins. This study therefore provides an exciting opportunity to identify methane sources and carbon cycling in areas subjected to subglacial volcanism and to consider these within the broader context of global carbon dynamics.

The Last Glacial Maximum (LGM) Patagonian Ice Sheet (PIS) provides the unique potential to reconstruct the former meridional position and precipitation rates of the Southern Westerly Wind Belt (SWWB) due to its considerable latitudinal distance perpendicular to this global wind system. To use this potential, a high-resolution 3D reconstruction of the LGM PIS would be crucial, but knowledge about even the exact planform of this biggest Quaternary ice sheet on the Southern Hemisphere, excluding Antarctica, is still very poor. The widely accepted limits of the ‘Finiglacial’ ice sheet extent, mapped over 80 years ago by Caldenius (1932), has been identified as the LGM extent at many locations (e.g. Kaplan et al., 2004, 2008a,b; Singer et al., 2004; Hein et al., 2009). However, apart from the broadly correct assignation of the ‘Finiglacial’ limit to the LGM, its resolution remains poor, imperfect, and outdated, despite some later adaptations (e.g. Hollin and Schilling, 1981a; Clapperton, 1993) with small improvements. All attempts to provide 3D reconstructions (Hollin and Schilling, 1981a; Hulton et al., 1994, 2002) date back almost 15 years and did not match the poorly constrained glacial geomorphological record. This study presents a novel attempt to cast all relevant absolute dating locations of glacial landforms into a refined spatio-chronological context of the LGM extent of the PIS. Based on this newly set spatio-chronological context, a geographical information system (GIS)-based high-resolution LGM extent reconstruction, both in planform and in 3D, is presented within this study. In contrast to its predecessors, the applied GIS-based modelling approach enables a grade of reconstruction resolution for an ice sheet, that is as big as Spain and that is otherwise found only in local valley-scale glacier reconstructions. The spatial resolution of the GIS-based 3D ice-sheet reconstruction surface lies at 100 metres, as does the implemented topographic data. This leads to a spatial ice-sheet surface resolution that is at least 200 times higher than any previous attempt to cover the entire LGM PIS. The 3D reconstruction of the LGM PIS using the newly defined extent is in good agreement with observations on vertical ice extents (Boex et al., 2013) and displays, for the first time, a realistic interaction with the complex Patagonian topography. The area of the LGM PIS is, with ca. 504,500 km2 (±8.5%) roughly a fourth of the area of Greenland and represents 554,500 km3 (±10%) of ice or 1.38 metres (±10%) of sea-level equivalent. The now available 3D reconstruction provides a starting point that, after minor revisions, could potentially allow gauging accumulation rates along up to 2,300 km perpendicular to the SWWB and, thus, inferring valuable palaeo-climatic insights from the LGM.

Recent climatic amelioration during the 20th and 21st centuries has stimulated the recession of glaciers world-wide. Moraines developed by valley glaciers provide a sedimentary record of their past response to climatic forcing. Despite the use of moraines for understanding the character and behaviour of former glaciers, our understanding of moraine development is lacking largely due to limited opportunities to study active moraine formation. This thesis reports on internal structure and sedimentary composition of lateral-frontal moraine at two Arctic (Austre Lovénbreen, Svalbard and Isfallsglaciären, Sweden) glaciers and one Alpine (Schwarzberggletscher, Switzerland) glacier and aids understanding of their palaeo-environmental significance. The internal structure and sedimentary architecture of Arctic lateral-frontal moraine is documented using ground penetrating radar (GPR) and via shallow excavation. Lateral-frontal moraine at both Arctic sites are shown to contain buried ice within their lateral zones, but not within their frontal zones, although the volumetric content of the buried ice and potential origin varies between sites. Frontal zones are therefore likely to be better preserved in the geomorphological record following complete de-icing, whereas lateral zones, which may also be subject to de-icing and external censoring from slope processes, may be poorly preserved. As the internal structure is dissimilar across Arctic sites, it is argued that the processes involved in the development of landforms by Arctic polythermal glaciers vary between high-Arctic and continental Scandinavian settings. Arctic lateral-frontal moraine are also distinct from those found at Alpine sites which are organised into stacked diamicton units that dip away from the lateral margin of the glacier. The sedimentary signature of Arctic and Alpine lateral-frontal moraine are investigated and compared. All moraines investigated exhibit clast-form gradients which is interpreted to relate to the relative significance and spatial variation of ‘active’ and ‘passive’ debris transport mechanisms within Arctic and Alpine valley glacier landsystems. However, the climatic, glaciological, and topographic regimes in which the moraines form influence the resulting character of the landform. The evolution of a degrading ice-cored moraine at Austre Lovénbreen is investigated using repeat photogrammetric topographic surveys. Relict glacier ice is undergoing moderate rates of degradation, predominately via down-wastage and could potentially be preserved as an archive of former high-Arctic glacier characteristics. This thesis also contributes to the wider body of knowledge in relation to the use of unmanned aerial vehicles (UAVs) and ‘Structure-from-Motion’ (SfM) photogrammetry for geomorphological research. The multi-technique approach employed by this research has allowed for the glaciological significance of subsets of lateral-frontal moraine (‘Controlled’, ‘Østrem’ and ‘Alpine’ type) within in glaciated valley landsystem to be better understood. Conceptual models accounting for landform development are presented and aid Quaternary studies that seek to identify and use lateral-frontal moraine for dating past glacier activity and determining palaeo-glacier characteristics.

Accurate forecasting of the Greenland Ice Sheet's contribution to global sea-level rise requires detailed knowledge of the processes in action at marine-terminating glaciers and their sensitivity to the controls that govern them. Both ice velocities and supraglacial-melt have increased significantly over the last decade, while recent research suggests that the dynamic ice-loss can be attributed to oceanic rather than atmospheric forcing. Warm, subtropical-originating Atlantic water has been identified as a primary driver of mass loss across the marine sectors of the Greenland Ice Sheet. Subglacial drainage networks inject fresh-water at depth producing a buoyant upwelling at the glacier front, renewing the Atlantic water in contact with the ice and thus enhancing the submarine-melt. To investigate these processes, their mutual forcing, and seasonal variability, a time-series of oceanographic, glaciological, and atmospheric measurements was collected at the front of Store Glacier, a major marine terminating outlet in Uummannaq bay (West Greenland). These data revealed a submarine-melt rate of 1.9±0.5md⁻¹ during winter months, which is five times larger than previous assumptions, and is attributed to the year-round contribution of basal-melt to the subglacial discharge of fresh water. A novel method using repeated 3D scanning of the submerged part of the glacier front permitted us for the first time to directly measure a summer submarine-melt rate of 3.4±0.7md⁻¹. Together, summer and winter submarinemelt contributed to 14% of the annual frontal ablation and up to 25% during summer months. Measurements from inside two upwelling plumes provided insight to their physical characteristics as well as their impact on the entire glacier front. These results highlight the need to reconsider the assumptions used in current model parametrisations of the ocean-glacier interface when evaluating the response of the Greenland-ice sheet to a warmer climate.

Uncertainty over future sea level rise is of great concern to society. One useful analogue for future sea levels is the Last Interglaciation (~130-118ka), when global sea level was around 6m higher than present. While most of the expected components to this are well simulated, the magnitude and location of the Antarctic contribution is poorly understood. Here, I perform a variety of ice sheet sensitivity tests to determine constraints on the extent/geometry of the Antarctic ice sheet during the Last Interglaciation in comparison with Antarctic ice core δ180 data and far-field relative sea level data, raised beaches and fossil corals from Patagonia and Australia. The effect of changes in solid earth deformation at ice core sites are simulated using a glacial isostatic adjustment model, and show minimal impact from plausible ice sheet collapses at any considered data sites, and that the ongoing deglacial uplift effects are likely to form a significant component of solid earth elevation around the Ross and Weddell Seas. I use the coupled atmosphere-ocean climate model, HadeM3, to simulate 2kyr snapshot runs for each ice sheet scenario (132-118ka). Precipitation seasonality is very poorly simulated over inland East Antarctica. Modelled changes in temperature between LIG and preindustrial show most similarity to ice core 8180 data when changes in precipitation seasonality are not accounted for. In the event of collapse of any of the marine based West Antarctic, Wilkes, and Aurora basins, cyclonic conditions would dramatically alter local climate, which is not observed in any data, indicating the ice sheet to be in similar to present geometry. Finally, I investigate the effect of glacial-isostasy on the nearest far-field relative sea level data, from Patagonia and Australia. Australian data suggests that there would have to be continual gradual sea level input throughout the Interglaciation. Although a unique fingerprint of West Antarctic collapse might be seen in Patagonian data, uncertain tectonic uplift and deglaciation history means this would be practically impossible to determine. Overall, I am able to determine that the Wilkes or Aurora subglacial basins did not significantly retreat. There remains no data suggesting a West Antarctic collapse in the Last Interglaciation, although the magnitude of global sea level strongly implies some Antarctic contribution.

During the last 20 years, West Antarctica has experienced enhanced ice discharge to the ocean due to loss of buttressing from melting and collapsing ice shelves. On the other hand, increases in precipitation have been reported in East Antarctica in line with an expected wetter atmosphere in a warming climate. The big questions that still lie ahead are therefore: (i) Will the enhanced precipitations in East Antarctica compensate the dynamic mass losses observed in West Antarctica in the future? (ii) And what will be the resulting contribution to sea level rise (SLR)? To answer those questions we need to have a firm grip on the present day mass balance (MB) of Antarctica and on the mechanisms that govern both the surface mass balance (SMB) and the ice discharge (D) into the ocean. This thesis investigates the MB of Antarctica using the input-output method (10M) allowing for a direct diagnoses of local, regional and global MB in Antarctica. It does this for both the Antarctic Ice Sheet (AIS) and the ice shelves. Because the mass imbalance of AIS is of the order of 5-10% of both accumulation and attrition terms of the mass budget (~2000 Gt yr-1), all glaciers around Antarctica as well as each assumption made require precise attention. This thesis starts with a chapter exploring the grounding zone (Chapter 2), and then goes on to the actual mass balance calculations of the AIS in Chapter 3 and of the Antarctica ice shelves in Chapter 4. The Grounding lines (GL) of Antarctica have been widely studied using various techniques at a local and regional scale. In recent years GL datasets aiming for circumpolar coverage have been published using different approaches. However these datasets still bear unexplained discrepancies of up to tens of kilometres in numerous places around Antarctica. In Chapter 2 four recent datasets are compared which track either the surface break of slope (h) or the inward limit of tidal flexure (F) as proxies for the grounding point (G). From visual examination and from a particle tracking scheme (PTS), it is found that all GL datasets agree within 1-2 km on slow moving ice and on the sides of fast flowing features (FFFs). However it is confirmed that h, obtained from photogrametry or photo clinometry, is not a reliable proxy in central parts of FFFs because of multiple breaks-in-slope and artefacts. It is further confirmed that the most reliable methods to map G in such places are those tracking F. In addition, a gravitational driving stress (td) is computed from a 1 km Antarctic digital model elevation (DEM) and leads to the finding that driving stress mapping (DSM) supports dynamic approaches in grounding line location. This reconciles static and dynamic grounding line methods by showing that they map the same features providing that altimetry is used rather than imagery for static methods. Guided by these analyses a new, up-to-date, and complete grounding line of Antarctica is compiled. The potential of DSM is successfully tested on a grounding line migration case study in West Antarctica. To investigate the grounding zone around Antarctica and its ice dynamics, DSM is further used. DSM allows to map sharp transitions across G for slow moving ice, as well as complicated transitions on fast flowing features (FFFs). Complicated transitions on FFFs contradict the idea of there being an ideal transition occurring at G, whereby the ice flow regime switches from basal drag-dominated to lateral drag-dominated. Rather, it is found that acceleration occurs upstream of G and that deceleration occurs downstream of G. This changes the understanding of the grounding zone ice dynamics, where ice was believed to accelerate at G due to loss of basal drag. Using DSM in combination with ice penetrating radar (lPR), reported and new ice plains (i .e. lightly grounded areas) are detected and mapped. They extents cover ~55,000 km2 around the Ross, the Filchner-Ronne, and the Larsen C ice shelves. These findings have implications for our understanding of ice sheet stability since ice plains are particularly prone to grounding line migration and can stretch up to ~300 km inland of G. In Chapter 3 the MB of the AIS is assessed using the input-output method (lOM). The grounding line fluxes (GLF) and 5MB are estimated for 110 drainage basins covering the whole AIS. The GLF is computed using up to date grounding lines and additional radar ice thicknesses data compared to previous 10M studies. 5MB values are re-evaluated in light of new drainage basins defined from an ice velocity field rather than from topography. 5MB is taken as the 30 years mean of three regional climate models. Due to a number of improvements in the GLF methodology, an unprecedented 94% of the ice sheet area is surveyed, i.e. an increase of + 13% from the latest 10M study. Un-surveyed areas are accounted for using mass trends (MT) from a Bayesian hierarchical modelling solution from the RATES (Resolving Antarctic ice mass TrEndS) project. The integrated AIS mass balance is -63 ± 83 Gt yr- I and divides into -22 ± 28, -62 ± 45, and 22 ± 64 Gt yr- I for the Antarctic Peninsula (AP), the West Antarctic Ice Sheet (WAIS), and the East Antarctic Ice Sheet (EAIS), respectively. The integrated MB is therefore a lower 10M estimate compared to previous 10M studies and reconciles the 10M with the other MB methods of satellite gravimetry and altimetry. Because the stability of the AIS is intimately linked to the stability of ice shelves, Chapter 4 finally focuses on the mass balance of ice shelves around Antarctica, giving the partition between calving fluxes (CF) and basal mass balance (BMB), the main processes by which Antarctic ice is lost. Before this study, iceberg calving had been assumed the dominant cause of mass loss for the Antarctic ice sheet, with previous estimates of the calving flux exceeding 2,000 Gt yr- I . More recently, the importance of melting by the ocean had been demonstrated close to the grounding line and near the calving front. So far, however, no study had reliably quantified the calving flux and the BMB (the balance between accretion and ablation at the ice-shelf base) for the whole of Antarctica. The distribution of fresh water in the Southern Ocean and its partitioning between the liquid and solid phases was therefore poorly constrained. Here, a first estimate of the mass balance components for all ice shelves in Antarctica is produced using calving flux and grounding-line flux from satellite and airborne observations, modelled ice-shelf snow accumulation rates, and a regional scaling that accounts for un-surveyed areas. The total CF is 1321 ± 144 Gt yr- I and the total BMB is -1454 ± 174 Gt yr-1 . These numbers mean that about half of the ice-sheet surface mass gain is lost through oceanic erosion before reaching the ice front, and that the calving flux is about 34% less than previous estimates derived from iceberg tracking. In addition, the fraction of mass loss due to basal processes varies from about 10 to 90 % between ice shelves. A significant positive correlation between BMB and surface elevation change is found for ice shelves experiencing surface lowering and enhanced discharge. It is therefore suggested that basal mass loss is a valuable metric for predicting future ice-shelf vulnerability to oceanic forcing.

Pine Island Ice Shelf (PIIS), in the Amundsen Sea, is losing mass due to warm ocean waters melting the ice from below. The glacial meltwater appears as a warmer and more saline water mass (with lower O2 concentration) than theWinterWater. Tracing meltwater pathways from ice shelves is important for identifying the regions most affected by the increased input of this water type. Water mass characteristics (temperature, salinity, O2 concentration) are used to calculate glacial meltwater fractions (MW). The observations from the Amundsen Sea show a plume of MW travelling away from PIIS along ¾ = 27.7 kg m¡3, out to the continental shelf edge. We investigate the reliability of the interpretation of the observations as a signature of MW. Physical and biological processes can affect the calculated apparentMWby causing variations in the water mass characteristics. In theWeddell Sea, iceberg meltwater was found to enhance biological productivity. In the Amundsen Sea, the biological productivity was seen to artificially decrease the apparentMWsignature. We analyse the effects of these processes on the reliability of the calculated meltwater fractions using a modified one-dimensional ocean model. The model simulates the effects of an increase in sea ice production and an influx of Lower CDW, as well as biological activity. These processes are found to result in an observation that can conventionally be interpreted as a meltwater signature, similar to the plume observed at the continental shelf edge. Recommendations are made to improve the reliability of MW calculations, including the identification of a ‘pseudo’-CDW endpoint and to increase the uncertainty associated with the O2 concentrations. A meltwater pathway leading to the west of PIIS, along the coastline, is observed. This has implications for water mass characteristics further to the west and ultimately AABW formation in the Ross Sea.

Since reaching their LIAMs, Himalayan glaciers have generally undergone a period of retreat, evident from large moraines left at former ice limits. Currently, however, detailed assessments of Himalayan glacier fluctuations over the past century are limited and fail to compare spatially or temporally to records available in Central Europe, North America and Scandinavia. Consequently, the variability and magnitude of glacial change across the Himalayas, which is a key indicator of climatic change in this region, is yet to be fully understood. Against a background of poor data availability, Corona imagery and historic GLIMS glacier outlines now offer an opportunity to assess glacier extent for regions of the Himalayas pre-1980. Corona imagery, acquired by a US space-borne reconnaissance mission operational from 1960 to 1970, represents a particularly unique dataset offering high resolution imagery (~1.8 m) with stereo-scopic capabilities. Utilising Corona imagery, there is an opportunity to produce detailed maps of Himalayan glacier extent and extract ice surface elevation estimations, in some instances, for the first time. Despite having been de-classified in 1995, the use of Corona data in the Himalayas has been neglected, mainly because of orthorectification challenges related to its unique geometric distortions. Hence, there remains a need to develop a low cost and easily replicable method of accurately orthorectifying Corona imagery enabling its use as a large-scale glacier mapping tool in the Himalayas. In response to this need, Corona images are orthorectified in this study through the use of: (1) a non-metric photogrammetry approach; and (2) horizontal and vertical reference data acquired from ortho-ASTER imagery and the freely available ASTER GDEM. By comparing glacier measurements derived from Corona imagery, GLIMS data and more contemporary ASTER data, changes in glacier area, length and in some instances volume, between the 1960/70s and early 2000s, were quantified for glaciers selected within four study areas located in Uttarakhand, India and Central Nepal. Importantly, this cross-regional glacier change dataset both complements and enhances current Himalayan records. Most notably, results indicate that glaciers selected in the Bhagirathi and Pindar/Kali basins, Uttarakhand, reduced in area by a relatively small 7.97±0.29% and 7.54±0.26%, respectively. Contrastingly, glaciers selected in the more easterly located Seti and Trisula basins reduced in area by 29.78±0.2% and 50.55±0.08%, respectively. Comparisons of Corona DEM (derived from Corona stereo-pairs) and ASTER Global DEM elevations at the terminus regions of four glaciers revealed extensive surface lowering, ranging from 87±27 m to 142±27 m. For Corona processing, the methods applied were shown to orthorectify Corona images to an accuracy that allows comparable glacier outlines to be delineated, further demonstrating the mapping potential of this dataset. However, for Corona DEM extraction, the use of ASTER spatial control data was shown to be inadequate and the presence of large vertical errors in the DEMs generated hindered the measurement of glacier volume change. For this purpose, it is therefore recommended that the methods developed are tested with the use of very high resolution spatial control data.

Tidewater glaciers form a significant drainage catchment of glacierised areas, directly transporting meltwater from the terrestrial to the marine environment. Surface melt of glaciers in the Arctic is increasing in response to warmer atmospheric temperatures, whilst tidewater glaciers are also exposed to warmer ocean temperatures, stimulating submarine melt. Increased freshwater discharge not only freshens fjord waters, but also plays a key role in glacimarine sedimentary processes, transporting sediment to glacial fjords. Despite this, the temporal evolution of meltwater production, storage and release from tidewater glacier systems at seasonal and interannual time scales is poorly understood. This leaves large uncertainties in the predictions for future sea level rise, ocean circulation and the impacts on the marine ecosystem. This study focuses on Kronebreen, a tidewater glacier which flows into the head of Kongsfjorden, north west Svalbard. Surface melt produces freshwater runoff, which is discharged from the grounding line as a buoyant, sediment laden plume, which spreads laterally across the surface water. This supraglacial melt is the dominant freshwater source, contributing an order of magnitude more freshwater to Kongsfjorden, than direct submarine melting of the ice face. Calibration of MODIS band 1 satellite imagery with in situ measurements of Total Suspended Solids and spectral reflectance, provides a method to quantify meltwater and sediment discharge. Plume extent has been determined for each cloud free day, from June to September, 2002 - 2013. Analysis of plume extent with atmospheric temperature and modeled surface runoff, gives a source to sea insight to meltwater production, storage and discharge. The extent of the plume changes in response to meltwater; larger plumes form when discharge increases. These results reveal that meltwater discharge into Kongsfjorden lags atmospheric temperature, the primary driver of meltwater production, by over a week during June and July. This is reduced to only 1 - 2 days in August and September, indicating a decline in meltwater storage as the ablation season progresses, and the development of more efficient glacial drainage. Sediment plumes respond to meltwater production, making them a valuable tool for quantifying meltwater discharge from a tidewater glacier. Insights to glacier hydrology can also be obtained when surface processes are also considered. This furthers the understanding of tidewater glacier hydrology, which is valuable for improving the accuracy of sea level rise predictions.

An improved understanding of the processes controlling the dynamics of the Greenland Ice Sheet is needed to enable more accurate determination of the response of the ice sheet to projected climate change. Meltwater produced on the ice sheet surface can penetrate to the bed and cause ice motion to speed up through enhanced basal sliding. However, the importance of coupled hydro-dynamics both to current ice sheet motion and future stability over the coming century is unclear. This thesis presents observations from the south-west Greenland Ice Sheet which improve our understanding of coupled hydro-dynamics. It commences with an investigation of the response of ice motion to exceptional meltwater forcing during summer 2012. Simultaneous field observations of ice motion (by GPS) and proglacial discharge show that, despite two extreme melt events during July 2012 and summer ice sheet runoff 3.9 s.d. above the 1958– 2011 mean which resulted in faster summer motion, net annual motion was slower than in the average melt year of 2009. This suggests that surface melt-induced acceleration of land-terminating regions of the ice sheet will remain insignificant even under extreme melting scenarios. The thesis then examines spatial variability in ice motion, in relation to an inferred subglacial drainage axis, using GPS and satellite radar observations from a land-terminating margin up to 20 km inland where ice is 800 m thick. Whilst spatial variability in subglacial drainage system configuration is found to control ice motion at short timescales, the proportional contribution of summer motion to annual motion is almost invariant. The structure of the subglacial drainage system does not therefore appear to significantly influence spatial variations in net summer speedup. Lastly, observations are made by applying feature tracking to 30 years of optical satellite imagery in a ~170 by 50 km area along the ice sheet margin (where ice reaches ~850 m thick) to examine whether coupled hydrology-dynamics affects inter-annual ice motion. Hydro-dynamic coupling resulted in net ice motion slowdown during a period of clear climate warming. Further increases in meltwater production may therefore reduce ice sheet motion. The thesis concludes that at land-terminating margins of the Greenland Ice Sheet, (1) larger annual meltwater volumes do not result in faster annual ice motion; (2) the detailed structure of the subglacial drainage network appears unimportant to the role of summer motion in determining annual motion; and (3) atmospheric warming over several decades has been accompanied by a slowdown in ice motion. As such, hydro-dynamic coupling is unlikely to form a significant positive feedback between surface melting and ice motion in response to projected climate warming. The wider relevance of these findings to tidewater systems requires further investigation.

The cryosphere impacts global climate in various ways. Snow and ice have a higher albedo than land or the open ocean and therefore affect the total reflectivity of the earth. Sea ice forms an insulating layer over the polar oceans controlling both heat and water vapour fluxes between the atmosphere and polar ocean. Ice sheets hold around 77% of Earth's freshwater reserves, and recent increases in ice loss from the Earth's ice sheets are cause for concern. This thesis develops numerical tools that can be used for the study of various ice processes such as ocean -- sea ice interaction, ice sheet and glacier dynamics. A coupled ocean - sea ice model is developed, using the open source, unstructured, adaptive ocean model Fluidity and a finite element sea ice model developed at the Alfred Wegener Institute of Polar and Marine Research, FESIM. The tightly coupled model is verified and validated through a series of tests, demonstrating its dynamical capabilities. The sea ice dynamics are a model of Elastic-Viscous-Plastic rheology, as described in Hunke and Dukowicz. The thermodynamic parameterisation is similar to the 1D simplest model of Parkinson and Washington which is based on the zero-layer approach of Semtner. Furthermore, a new computational framework for carrying out ice sheet simulations is presented. A thermo-mechanical, non-linear, full-Stokes model is used to carry out the exercises of the Ice Sheet Model Intercomparison Project for higher--order models (ISMIP--HOM). The results presented here show that Fluidity compares favourably with other ice sheet models. Further tests are performed to demonstrate the use of dynamic adaptive remeshing in lowering the computational cost of models compared to their structured, fixed-mesh counterparts, by focusing resolution only where and when required. Finally, initial simulations of the full Greenland ice sheet are performed demonstrating the potential utility of adaptive meshes for large-scale, full-Stokes modelling.

The main aim of this research project was to improve our understanding of the diversity, function and ecology of glacial microbiomes. Snow and ice algae are critical players in supraglacial habitats and form extensive blooms in spring and summer. Here I present results on the diversity and the function of snow and ice algae on 21 glaciers in 4 Arctic settings: Greenland, Iceland, Svalbard and Sweden. For the first time, I have evaluated the full microbial community composition (i.e., algae, bacteria, archaea) in the main supraglacial habitats, namely green snow, red snow, biofilms, dirty ice, and cryoconite holes. I have cross-correlated these data with metabolic analyses (i.e., metabolomics, pigments, fatty acids) and critical physico-chemical parameters. I found that snow and ice algae were the first communities to appear after the onset of melting and they showed positive net photosynthetic rates indicating accumulation of organic matter. Furthermore, for the first time I have described these communities in Iceland. My data reveal that red pigmented snow algae are cosmopolitan, and independent of location specific geochemical and mineralogical factors. Only six taxa made up >99% of the algal communities: two uncultured Chlamydomonadaceae, Chloromonas polyptera, Chloromonas nivalis, Chloromonas alpina and Raphidonema sempervirens. In contrast, the composition of green snow varied between the studied locations with higher relative abundance of Raphidonema sempervirens and Microglena sp. in Svalbard, and Chloromonas polyptera in Sweden. Furthermore, I show that green and red snow are not successive stages but two independent phenomena with different adaptation strategies. In all sites, bacteria were mostly represented by the phyla Bacteriodetes, Proteobacteria and Cyanobacteria. The bacterial community composition varied between the different habitats on the phylum level, whereas on the class level they also showed strong biogeography. Archaea showed overall low species diversity. The synthesis of pigments and fatty acids in snow and ice algae were mainly driven by nitrogen and less so by phosphorus limitation. This is especially important for pigments which cause a darkening of glacial surfaces. I show that snow and ice algae dramatically decrease surface albedo which will eventully result in higher melting rates of glaciers.

In this thesis, I use synthetic aperture radar (SAR) and radar altimeter data to make new observations of Antarctic and Greenland ice sheet grounding lines. I use ERS SAR data acquired between 1992 and 2011 to map the Petermann Glacier grounding line on 17 occasions using quadruple difference interferometric SAR (QDInSAR). Over the 19-year period, the grounding line position varied by 0.5 km, on average, with no significant trend over time. Although tidal forcing explains a fraction (34 %) of the movement, localised variations in the glacier thickness could explain it all were they to alter the glaciers hydrostatic balance as they advect downstream – a hitherto unconsidered possibility that would reduce the accuracy with which changes in grounding line position can be detected. Next, I developed a new technique for detecting grounding lines using differential range direction offset tracking (DRDOT) in incoherent SAR data. I then applied this technique to a sequence of 11 TerraSAR-X images acquired in 2009 over Petermann Glacier. The DRDOT technique is able to reproduce the shape and location of the grounding line with an estimated lateral precision of 0.8 km and, although this is 30 times poorer than QDInSAR, provides a complementary method given the paucity of coherent SAR data. Finally, I developed another new method for detecting the grounding line as the break in ice sheet surface slope computed from CryoSat-2 elevation measurements. I then applied this technique to map grounding lines in the sectors of Antarctica buttressed by the Filchner-Ronne, Ekström, Larsen-C, and Amundsen Sea ice shelves. The technique is able to map the grounding line to within 4.5 km, on average, and, although this is far poorer than either QDInSAR or DRDOT, it is computationally efficient and can succeed where SAR-based methods fail, offering an additional complementary approach.

Patagonia has one of the most extensive and well preserved glacial geomorphic records of anywhere else in the world. This study provides empirical constraints of Patagonian Ice Sheet (PIS) configuration and dynamics during the last two glacial cycles and the Holocene at 49°S. In particular a chronology of palaeo-ice surface elevations, thickness changes and ice-thinning rates is developed. Cosmogenic surface exposure dating across eight mountain transects at Hielo Patagonico Sur (HPS) outlet glaciers' Viedma and Chico, and from eastern facing mountain valley glaciers, is utilised to reconstruct PIS ice-surface elevation changes through time. The earliest dated glacier ice thickness marker falls into MIS6 (Marine Isotope Stage). The last glacial cycle is characterised by continuous ice-surface elevation lowering from a maximum at MIS5a to the Antarctic Cold Reversal (ACR). Large scale PIS cover occurred between ca. 40-47 kyrs, in both the HPS outlet glaciers and the eastern mountain valleys. This regional ice-cover considerably pre-dates the global Last Glacial Maximum (gLGM). A gLGM vertical ice surface expression is absent from the study area. A minor re-advance and/or stillstand at ca. 18 kyrs is recorded by the valley glaciers, but is not evident at the HPS outlet glaciers. Glacier Viedma records a prominent ice elevation during the ACR with rapid thinning at the end of the climate perturbation. Ice-surface lowering from the ACR limit continues with acceleration toward the late-Holocene. The PIS palaeo-ice elevation and thinning record presented here is sensitive to internal PIS dynamics, local and global climate forcing acting on different timescales, with glacier response to climate perturbation strongly linked to glacier catchment size and distance from the main source of precipitation at 49°S.

Many of the molecules in interstellar space are in the solid state, in so-called interstellar ices. The research in this thesis is focused around two key astronomical questions; How is ice distributed in star forming regions? How is ice affected by - or affecting - star formation processes? I provide answers to these questions through the analysis of submillimeter and near-infrared observations. The observations analysed in this thesis consist of (partially published) archival data acquired mainly with the AKARI and Herschel space telescopes, and the ground-based ESO/VLT. To facilitate the reduction and analysis of some of this data two major software packages (ARF2 and Omnifit) were created with the Python programming language The operation of both packages is fully documented in the thesis appendix. The study of methanol ice prevalence in star-forming regions found that methanol ice can be found towards many more lines of sight than previously reported, and that its abundance relative to water ice can vary between a few to ~40%. I also confirm that methanol very likely exists mixed in a water-rich ice component, a result consistent with our current understanding of methanol ice formation. Proof was found of high-temperature chemistry forming water in the warm postshock gas of YSOs. In this same region it was found that up to 99% of the methanol is being destroyed as it is sputtered from the surfaces of dust grains into the gas phase. A novel analysis technique of slitless AKARI near-infrared spectroscopy yields an unprecedented number of water ice column density estimates towards background star lines of sight covering 12 separate 10' x 10' fields of view in as many molecular clouds. A moderate correlation is found between water ice column density and dust optical depth at 250 microns, with the correlation potentially varying from cloud to cloud.

Ice shelves play an important role in the mass balance of an ice sheet, by providing a link between the ocean and ice. Melting at the base of an ice shelf can play a vital role in its mass balance and stability. Topographic channel features have been found on the base of ice shelves, and have been found to alter melting, however the mechanism behind this alteration is unknown. Petermann Glacier is a major outlet glacier in North West Greenland, draining approximately 6% of Greenland Ice Sheet. It terminates in a long, thin ice shelf, constrained within a high-walled fjord. The ice shelf has pronounced longitudinal channel features on its base, which limited observations suggest direct ocean currents in a mixed layer of ocean and melt waters, focusing melt in these regions. Petermann Glacier underwent two large calving events in 2010 and 2012, and the impact of these events, or possible further calving events, on basal melting is unknown. Using the MITgcm to model the ocean cavity beneath an idealised ice shelf, this thesis discusses the impact of basal channels on interactions at the ice base and circulation within the cavity. This is supplemented with a modelling investigation into the interactions beneath Petermann Glacier, and the impact of recent calving events. The inclusion of channels was found to have a stabilising effect on the ice shelf by decreasing the mean basal melt rate, caused by the refocusing, and decrease in intensity of, the meltwater layer flow beneath the ice shelf. This stabilisation and resulting 'survivor bias' explains why channels are commonly found on the base of warm water ice shelves. The model of Petermann Glacier found similar melt patterns to observational studies, however with a lesser magnitude. The calving events of 2010 and 2012 removed areas of ice shelf with low melt rates, resulting in little impact on the overall volume of ice removed through ocean melting, though further calving would vastly reduce the volume of ice melted. One consequence of calving is the increase in melting-induced undercutting at the ice front, leading to the potential for enhanced secondary calving.

Ice flow produces morphological features at the bed and on the surface of ice sheets. These ‘flow signatures’ provide us with an insight into the mechanisms, history and characteristics of ice sheet flow. In this thesis I examine the characteristics of basal and surficial ice sheet flow signatures, as well as possible links between them. The first chapter introduces ice sheet flow signatures. At the bed, a suite of landforms known as subglacial bedforms are found. The surface of an ice sheet is home to longitudinal surface structures (LSSs) and transverse surface structures (TSSs). Whilst the two environments are mostly considered in isolation, links between the sets of flow signatures found at each have been suggested. Section A deals with basal flow signatures. Chapter 2 asks whether subglacial bedforms are patterned. Drumlins are found to be regularly placed within the landscape, and likely grow or shrink over time. Chapter 3 examines whether subglacial bedforms conform to a size and shape continuum. By collating and analysing a dataset of 96,900 measurements of size and shape it is found that 3 continua of subglacial bedforms exist: flutes, lineations and ribs. The latter two are joined by an understudied class of quasi-circular bedforms. Section B deals with surficial flow signatures. In Chapter 4 I present and analyse a map of the LSSs of the Antarctic Ice Sheet. The morphology, spatial distribution and glaciological context of LSSs leads to the proposal of a model for their formation. Chapter 5 presents the first systematic study of TSSs. Mapping and analysis reveals that TSSs are regularly spaced, differ little in their morphology between ice streams and are most likely stationary. Section C compares basal and surficial ice sheet flow signatures. In Chapter 6, morphological comparisons, a case study of the Rutford Ice Stream, analysis of ice penetrating radar and examination of ice flow modelling lead to the conclusion that the majority of basal and surficial flow signatures are separate entities. Chapter 7 concludes the thesis and provides suggestions for future research.

Groundwater-surface water exchange significantly impacts proglacial hydrology and ecology. This study applies a multidisciplinary approach to investigate groundwater-surface water exchange in the proglacial zones of two retreating glaciers in SE Iceland. Mapping of decadal changes in the extent of proglacial groundwater seeps in the large outwash plain of Skeiðarársandur has shown a 97% decline, as well as substantial falls in groundwater levels. Field and laboratory measurements suggested high spatial variability in hydraulic conductivity at the Skaftafellsjökull foreland. The highest hydraulic conductivity was measured in areas underlain by glaciofluvial deposits whilst the lowest hydraulic conductivities were associated with glacial tills and lacustrine deposits. Precipitation was identified as an important control on groundwater levels on various temporal scales. Automated monitoring of meltwater and groundwater levels also identified fluctuations in meltwater level as an important control on hydraulic heads, whose importance on groundwater levels has been observed during various flow regimes. The close connection between meltwater and groundwater levels suggest high meltwater-aquifer exchange. However, high meltwater-aquifer exchange is contested by significantly different geochemical and isotopic composition of groundwater and meltwater. Hydrogeological flux estimates suggest high spatial variability in groundwater seepage into the Instrumented Lake, which was attributed to the high variability in hydraulic conductivity around the lakeshores. These are also supported by high –resolution temperature mapping at the lake bed, which suggested that groundwater upwelling in the fine-grained lakeshore took place at discrete locations. This study suggests climate and glacier margin fluctuations as primary controls on proglacial groundwater-surface water exchange. It also highlights the importance of groundwater contributions to water quality and ecology, with groundwater-fed bodies possibly sustaining important ecological niches. However, proglacial groundwater-fed features are transient and are threatened by changes in precipitation and glacier retreat. Further declines in groundwater-fed hydrological systems are therefore projected to adversely impact proglacial groundwater-surface water interaction.

Glaciers cover approximately 10% of land surface, but are rarely considered active components of global biogeochemical cycles. Research over the past two decades has revealed that active and widespread hydrological systems are present at the bed of the Antarctic and Greenlandic Ice Sheets. This has driven recent concern in the biogeochemistry of subglacial environments. The ocean waters surrounding the ice sheets harbour highly productive, ecologically and economically important waters. Only recently has an ice sheet influence, via the export of nutrients, been considered. A high-resolution hydrological dataset and dissolved and particulate geochemical record, from multiple ablation seasons of two contrasting Greenland Ice Sheet catchments were collected. These were used to infer subglacial biogeochemical weathering processes, and associated nutrient cycling. Estimates of glacial bioessential nutrient fluxes to downstream environments revealed the potential for significant export of iron and phosphorus to the near-coastal euphotic zone. Labile particulate nutrient species dominate total nutrient export due to the high concentrations of extremely fine-grained and highly reactive suspended sediments in meltwaters. Meltwater filterable iron and phosphorus concentrations were found to be temporally variable over an ablation season due to an evolving subglacial meltwater source coupled with evolution of subglacial drainage. More remote regions of the subglacial environment are progressively flushed over the course of an ablation season, releasing waters rich in nutrients compared to major ions. "Outburst events", driven by the drainage of supraglacial lakes into the subglacial environment, were found to be an especially important mechanism in driving pulses of stored, nutrient rich meltwaters downstream. It is demonstrated that dissolved macronutrient (nitrogen, phosphorus and silica) and solute export from the Greenland Ice Sheet are annually dynamic, with significantly higher fluxes in warmer melt years. This is likely to be driven through more exhaustive basal flushing by supraglacial meltwaters, and by progressively more efficient linking of remote regions of the subglacial environment with downstream drainage. Changes in nutrient export from the Greenland Ice Sheet with enhanced melting under future climate warming scenarios are therefore likely.

Over a tenth of the Earth is covered by microbially-dominated glacial ecosystems. The microbial activity on glaciers and ice sheets is postulated to have regional, potentially global, impacts on environmental biogeochemical nutrient cycling. Their activity is also believed to affect glacier surface albedo, potentially enhancing the effects of climate change in these environments. However, the magnitude and implications ofthe glacial microbial dynamics are still unknown. This PhD research aimed to elucidate the microbial dynamics within different glacier surface habitats on the Greenland Ice Sheet, together with their impacts on glacier organic matter production, nutrient export and albedo reduction. A variety of original laboratory and field studies were designed to address these aims. Microorganisms on glacier surfaces were found to be responsible for significant amounts of organic matter production, accumulation, recycling and export from glacier surfaces. The microbial communities were shown to be stable throughout a melt season, following a fast community turnover after initial snow melt. They could also be seeded by small amounts of aeolian microorganisms. The organic matter the m icrobes produced and transformed was substantial enough to reduce glacier surface albedo by -16%, equivalent to - 18 Gt yr-I of potential melt increase for the Greenland Ice Sheet, in the current climate. Furthermore, biogeochemical nutrient cycling by glacial microorganisms was successfully applied to fertilising nutrient-deprived sediment, for potential applications in countries struggling with barren soils and potentially for terraforming other planets. The results ofthis body of work thus have significant implications for predicting the impacts of climate change and anthropogenic pollution on glacial environments; future estimates of nutrient export from glaciers and the effects on downstream environments; and even for developing new soil fertilisation techniques for terrestrial and extraterrestrial use.

The Irish Ice Sheet has been an active area of research since the mid 19th century. Traditional approaches to reconstructing the Irish Ice Sheet have used the landform and sediment record to define its extent and formulate models of its geometry through the Pleistocene. Subglacial bedforms dominate the landform record and till is the dominant sediment. The bedforms demonstrate a complex ice flow history, however recent research has indicated that ice flow directions interpreted from subglacial bedform geomorphology could be misleading. It is therefore appropriate to consider other methods of investigating ice flow patterns that are independent of geomorphology. As the bedforms are primarily composed of till, an alternative approach to determining flow directions is to establish the provenance of the till using geochemistry. High density regional soil geochemical data sets comprising> 1 0,000 sample points are available for the north of Ireland. This thesis applies a multi variate statistical approach to the soil geochemistry to establish till provenance in the northern sector of the Irish Ice Sheet. Till provenance is determined for the entire region and also investigated in a range of palaeoglaciological contexts. Results are compared to recent models of flow history for the Irish Ice Sheet. Till provenance is also applied to determine the degree of glacial dispersal of gold mineralisation from known host regions in the north of Ireland. The thesis demonstrates that till in this sector of the Irish Ice Sheet shows a very strong geochemical link to local bedrock composition, with sharp geochemical transitions observed at lithological boundaries. However, subglacial transport and clear dispersal patterns are detected particularly at flow set scale in a range of palaeoglaciological contexts. Geochemical evidence is found for flow directions for each of the major phases of the history of the last Irish Ice Sheet and this evidence compliments existing models. No unequivocal glacial dispersal patterns are detected for gold, though the prospectivity of currently unexplored areas is highlighted. The research demonstrates that soil geochemistry makes a significant contribution to Quaternary research in Ireland which has not been previously recognised.

The Cairngorm Mountains contain an outstanding assemblage of glacial landforms from both the deglaciation of the last British Irish Ice Sheet and the Younger Dryas readvance. Glaciers are recognised as sensitive indicators of past and present climate change and, thus, these landforms provide information about past climate and glacier-climate interaction that can be used to contextualise the present climate change. Previous interpretations have left doubt over the extent and style of the Younger Dryas readvance. In addition, the pattern and timing of deglaciation in the southern Cairngorms and, particularly, how local and external ice masses interacted is unclear. New geomorphological mapping from aerial images and fieldwork has been compiled in a GIS for a 600km2 area of the Cairngorm Mountains. This has allowed a complex pattern of ice-dammed lakes and local and regionally sourced ice margins to be reconstructed during the retreat of the last British Irish Ice Sheet. The mapping has been combined with new cosmogenic surface exposure ages taken from areas of hummocky moraine previously subject to differing age interpretations. The effect of moraine denudation on apparent 10Be ages has been checked by inverse modelling of the 10Be concentration vs. boulder height. The results indicate more extensive Younger Dryas glaciation, with glacier reconstructions and equilibrium-line altitudes (ELAs) comparable with the surrounding areas. Reconstruction of both valley and plateau-fed glaciers are presented, with modelling of local topoclimatic factors, such as radiation, avalanche and snow drifting, combined with precipitation gradients, explaining most of the variation within the glacier ELAs. The geomorphological evidence and palaeoclimatic inferences are important, alongside a growing number of palaeoglaciological studies, in acting as evaluation areas for current numerical models of ice-sheet growth and decay.

The seismic stratigraphy and sedimentary architecture of the Canadian Beaufort Sea margin are investigated using a comprehensive grid of two-dimensional seismic reflection data. Three cross-shelf troughs, representing locations of former ice streams draining a 1000 km-long section of the Laurentide Ice Sheet (LIS), are examined: the Mackenzie, Amundsen Gulf and M'Clure Strait systems. These palaeo-ice streams operated during the last, Late Wisconsinan glacial maximum, as well as during a hitherto unknown number of earlier Quaternary glaciations. Their dynamics influenced past ice-sheet configuration and may have forced abrupt climate change through transport of ice and freshwater to the Arctic Ocean. The objectives of this work are to constrain the number of ice advances through each trough, to discuss the possible timing of these events, and to examine the impact of Quaternary glaciation on the continental shelf and slope. The implications of these data are discussed in relation to ice dynamics at the northwest LIS limit, the glacial history of the Canadian Beaufort Sea margin, and the geomorphological imprint of palaeo-ice streams. The number of Quaternary ice advances across the Canadian Beaufort Sea margin varies markedly between the Mackenzie Trough (two) and the Amundsen Gulf Trough (at least nine). The Mackenzie Trough was probably occupied by an ice stream during the Late Wisconsinan and either the Illinoian or Early Wisconsinan glaciation. The Amundsen Gulf ice stream was initiated earlier in the Quaternary. The architecture of the slope beyond the Mackenzie Trough reflects this comparatively short history of ice advance and lacks the progradational architecture and major glacialsedimentary depocentre that is characteristic of slopes seaward of high-latitude cross-shelf troughs. In contrast, trough-mouth fans ( of volumes -10,000 km.3 and -60,000 km3) are present beyond Amundsen Gulf and M'Clure Strait, respectively. The location of 75 High Arctic cross-shelf troughs is presented together with a synthesis of their key physiographic characteristics and available glacial-geological evidence of past occupation by ice streams. The dimensions and architecture of the three troughs on the Canadian Beaufort Sea margin are compared to those of other High Arctic troughs. The Amundsen Gulf and M'Clure Strait troughs were probably two of the most significant cross shelf troughs in the High Arctic, in terms of their dimensions and palaeo-drainage basin areas. A number of buried glacigenic landforms, including grounding-zone wedges and lateral moraines, are identified from the Canadian Beaufort Sea shelf, recording the former positions of still-stands or re-advances in the ice margin. The youngest sequence of sediment in the Amundsen Gulf Trough is interpreted to have been deposited by a subsidiary ice stream, the Anderson ice stream, subsequent to Late Wisconsinan ice retreat through the Amundsen Gulf. This provides evidence of dynamic ice-stream behaviour and the reorganisation of the north west sector of the LIS during the last de glaciation.

Predicting the long-term behaviour of present-day ice sheets is hampered by the short timescales of our observations and restricted knowledge of the subglacial environment. Studying palaeoice sheets can help by revealing the nature and amplitude of past centennial- to millennial-scale ice sheet change. This thesis uses glacial sediments and landforms to examine the evolution of the partly marine-based British-Irish Ice Sheet (BIIS) and its bed, in western Scotland. Three zones of the former BIIS are considered: ranging from a mountain ice cap, to a core area of the ice sheet, to a peripheral marine-terminating sector. The topography of the subglacial landscape was an important in uence on the location of dynamic and stable components of the ice sheet. At an ice cap scale, zones of glacier inception and retreat were linked to catchment elevation and size. At the ice sheet scale, the migration of ice divides and thermal boundaries were focused through corridors of low relief subglacial topography. The main west-east ice divide of the BIIS in central Scotland migrated by 60 km, 10% of the ice sheet's width, through one such corridor during the glacial cycle. A major change in the ow regime of the BIIS in western Scotland accompanied the development of a marine-based sector on the Malin Shelf. As the BIIS advanced to the shelf edge, ice ow was drawn westwards { orthogonal to the earlier, geologically controlled, ow pattern. Retreat of the BIIS from the shelf edge occurred at an average rate of 10 m a-1, but was punctuated by at least one episode of accelerated retreat at 100 m a-1. In each zone of the BIIS examined, a rich palimpsest landscape is preserved and the role of earlier glaciations in conditioning or priming the landscape is highlighted. Western Scotland in particular is dominated by features relating to a 'restricted' mountain ice sheet, suggested to have been the prevailing ice sheet mode during the Early and Middle Quaternary. Where the last BIIS was underlain by soft sediments, glacier movement at the bed was facilitated by a combination of basal sliding and a localised mosaic of shallow deforming spots, allowing landform and sediment preservation. In places, till deposition was focused over permeable substrates acting to seal the bed, promote lower e ective pressures, and enhance motion by basal sliding. The modern land surface in western Scotland provides an approximation for the relief of the former glacier bed, and can be used for conceptual palaeoglaciological reconstructions. Areas of focused postglacial deposition have, however, obscured parts of the ice sheet bed, with demonstrable implications for quantitative palaeoglaciological analyses. Methods to improve the representation of former ice sheet bed in these areas are discussed and may be pertinent to future palaeo-ice sheet modelling exercises.

Slope failures in infrastructure assets across the UK and Ireland have been observed in recent years. These failures can in many cases be attributed to progressive failure. The mechanisms which cause progressive failure are not understood and research is ongoing in many areas to understand them. One process under investigation is strength reduction by strain softening due to seasonal cycles in pore water pressure. The research presented here has incorporated field investigations, laboratory testing and numerical analysis to assess the geomechanical response of glacial till cuttings to climatically driven pore water pressure cycles. Three research sites were selected and their geotechnical and hydrogeological properties characterised through field investigation and laboratory testing. Alternative methods of in situ stiffness measurement have been used to establish the small strain stiffness to the formation for use in numerical simulations. A pore water pressure monitoring programme has been established across three study sites and provides a comprehensive dataset for use in this and future analyses. Pore water pressures are seen to be driven by climate controlled recharge. Advanced stress path testing was carried out in the laboratory to assess the possibility of strain softening. Reconstituted samples of glacial till were taken through extension and compression loading, followed by controlled pore water pressure cycles to simulate seasonal changes. From these tests shear strain was observed to continue with no increase in deviator load, indicating that strain softening maybe possible in this soil deposit. Validated numerical models of coupled stress-strain and pore water pressure have been compiled which represent the overconsolidated glacial till cuttings in this area. From this base climatically driven pore water pressure cycles were simulated. Shear strains are recorded as increasing with time with the greatest magnitude of change occurring at the toe of excavations.

'Large cut slopes (excavations) in glacial tillare a common occurrence along road and rail networks in Northern Ireland, due to unique drumlin topography in the province. Despite observed occurrences of delayed slope failures within till cuttings, their long-term performance is not well understood. Decreases in the stability of overconsolidated clay till slopes with time have been shown to be in response to pore water pressure (PWP) cycles induced by soil-climate interactions. This research has developed and presented an understanding of the mechanisms controlling PWP fluctuations within till slopes in response to climatic factors. Three cut slopes in till were instrumented to measure PWP variations, surface soil water content, suction, and shallow water table fluctuations. Field and laboratory tests were carried out to characterise the geotechnical and hydraulic properties of the till. The field instrumentation facilitated the identification of a number of characteristic controls on PWP variations, including the seasonal soil water balance and more rapid PWP responses due to gravity driven recharge and stress changes. A numerical soil water storage model was used to estimate the temporal variability in recharge to the till. This water balance accounts for PWP variation, in-situ hydraulic conductivity and fracture porosity within the till. The outputs from the water balance model were subsequently used as a boundary condition in a transient seepage model. This is coupled with a limit equilibrium analyses to assess variability in the factor of safety of the cuttings. Furthermore, a generic model was used to explore how variations in climatic and hydrogeologic conditions will affect long-term slope stability. The research highlights the control of the hydrogeologic setting and climatic conditions on PWP dynamics, and makes suggestions for how this understanding can be used in design, management and remediation measures of cut slopes to protect these assets in the future.

The British-Irish Ice Sheet (BIIS) is predicted to have deglaciated rapidly from ~ 18 ka, in response to rising sea level and temperature, similar forcings experienced by modern polar ice sheets. As the main conduits of ice mass loss, the reaction of ice streams to these forcings is thought to have been central in determining the mode and timing of this deglaciation. However, lack of understanding of ice stream influence on the glaciology and deglaciation of ice sheets limits confidence in ice sheet model predictions. NW Scotland is an area of the last BIIS predicted to have been dominated by ice stream onset conditions. This thesis presents results from a geomorphological and terrestrial cosmogenic nuclide (TCN) analysis study which resulted in the production of a composite ice-sheet thermal regime map and retreat chronology for the last BIIS in this region. Mapping and surface exposure dating suggest that the regional glaciology and landscape evolution was dominated by the presence of ice-stream onset zones during Greenland Stadial-2 (GS-2). Mountain top erratics were uplifted and transported to high elevation during GS-2, before 16.5 ka BP. By inference, mountain summits were covered by ice during maximal ice sheet conditions. The existence of sharp thermo-mechanical contrasts, developed in response to ice streaming, are proposed as the main controls on bedrock erosion and terrestrial sediment deposition. The interpretation of ‘trimlines’ in NW Scotland as englacial thermo-mechanical boundaries, is verified by the identification of ‘rip-offs’, a newly recognised geomorphic feature in the UK, and by quantitative demonstration of the increase in glacial erosion in the vicinity of these boundaries. Geomorphic and TCN data supports a conceptual model of thermal inversion following ice-stream cessation. The first description of ‘till tails’ in the UK provides insight into the glaciological organisation and thermal evolution of the BIIS. A dated (17.6 ka BP) terrestrial glacial limit on the north Sutherland coast indicates early ice retreat from the shelf and provides a minimum ii constraint on ice-stream cessation. This indicates rapid loss of ice extent and volume following shutdown of the Minch palaeo-ice stream. Major ice sheet reorganisation c. 15-16 ka BP is suggested by the correlation of some lateral margin ages with high elevation erratic deposition ages implying significant ice thinning and margin retreat prior to this time. Additionally, thinning of ~300 m is predicted for some areas prior to 14 ka BP.

Glacier surges are amongst the most dynamic of glaciological phenomena, but their controlling mechanisms remain incompletely understood. Surging glaciers are characterised by cyclical flow instabilities and the rapid transfer of ice to the ablation area, typically resulting in significant mass loss. The High-Arctic archipelago of Svalbard is one of several regions in the northern hemisphere which contain a high-density of surge-type glaciers, variously estimated to be between 13-90% of the total glacier population across the islands. Developing a better understanding of which of these figures, if either, is most realistic is important in the context of glacier dynamics and related contributions of small glaciers and ice caps to sea level change in the immediate future. This study presents detailed assessments of the margins of several known surge-type glaciers in Svalbard in order to update and improve the existing framework by which they are identified, and to provide a foundation for future reassessments of the surge-type glacier population based on distinct ice-sediment-landform assemblages. A range of techniques is utilised, including geomorphological and structural glaciological mapping, sedimentological analysis, basal ice descriptions, and stable isotope analysis. This work provides further insight into diagnostic indicators of surge behaviour preserved in basal ice sequences; provides links between surge dynamics and basal ice sequences, the glaciological structure and the landform record; and investigates the structural and tectonic development of surge-type glaciers. Based on this, surge landsystems are proposed for: (1) small valley glaciers, (2) large land-terminating glaciers, and (3) large tidewater glaciers. It is suggested that these three landsystems, with some variability, broadly characterise the geomorphology of the vast majority of known Svalbard surge-type glaciers and, in conjunction with structural glaciological and basal ice investigations where relevant, may allow previously unknown surge-type glaciers to be identified in the field, from aerial photographs, and on sea floor imagery. This work adds to the existing repertoire of modern analogues and the breadth of surging glacier landsystems, and provides a holistic basis for assessing possible palaeo-surge behaviour within the Quaternary record.

The Cenozoic represents the transition from the greenhouse world of the Cretaceous to the ice house world of today. Nonetheless, it was not a steady linear trend from an ice free world to an ice house, and there were several reversals along the way. One such reversal was the Mid Miocene Climatic Optimum (MMCO, 14-17 Ma). Material from the Ocean Drilling Program Site 926 on Ceara Rise in the Equatorial Atlantic Ocean is used to examine two intervals within the MMCO from 16.4-15.9 Ma and 15.7-15.3 Ma. Stable isotopes and trace element ratios in two species of benthic foraminifera, Cibicidoides wuellerstorfi, and Oridorsalis umbonatus, and one planktonic foraminifer, Globigerinoides trilobus are used to reconstruct multiple climatological parameters at 2-4 kyr resolution. Paired Mg/Ca and δ18O measurements from the benthic species examined suggest that sea level varied by as much as 40m during the MMCO. Sea level variability was accompanied by changes in sea surface salinity, as measured using δ18O and Mg/Ca from G. trilobus, which suggests that the Intertropical Convergence Zone (ITCZ) was responding to high-latitude forcing centred upon the Northern Hemisphere. The implication is that significant Northern Hemisphere ice sheets were present during the Middle Miocene, some 13 Ma before their currently held date of inception during the Pliocene. These changes in sea level were accompanied by large changes in benthic and planktonic carbonate saturation states (Δ[CO32-]), which are inferred to be representative of changes to global alkalinity and dissolved inorganic carbon. Changes in global export productivity, as evidenced by benthic foraminiferal accumulation rates and the planktonic-benthic gradient of δ13C, are inferred to be controlling the carbon cycle and atmospheric pCO2ATM during the MMCO. The ITCZ is also inferred to be controlling primary productivity at Ceara Rise. Additionally, the relationship between multiple trace element/Ca ratios and Δ[CO32-] are examined using core-tops taken from the Norwegian Sea. The relationship between Mg/Ca and Δ[CO32-] in C. wuellerstorfi is confirmed by expanding the available holothermal data. Mg/Ca in the benthic species Pyrgo murrhina appears to respond exclusively to changes in Δ[CO32-], and the relationship of U/Ca in P. murrhina to Δ[CO32-] is the opposite of that seen in C. wuellerstorfi and O. umbonatus.

The presence and configuration of subglacial water and sediment maintain the fast flow of arterial ice streams in Antarctica and airborne ice-penetrating radar data represent a potential resource of information about the ice-bed interface. In this thesis an original contribution to the exploration of Antarctic subglacial environments is made through the analysis of airborne surveys from Evans, Institute and Möller Ice Streams, West Antarctica. The primary approach employed is the derivation of bed-returned power (BRP), a proxy for ice-bed reflectivity, which is strongly influenced by the presence of liquid water. Estimating radar englacial attenuation (EA) accurately is a critical part of BRP analysis and a modelled approach is primarily used. BRP is derived across Evans Ice Stream and shows large-scale patterns relating to hypothesised hydrological and geological contrasts at the ice-bed. These results are developed to investigate the influence of: (1) adopted EA correction; (2) the influence of assigned ice dielectric properties in modelled EA; (3) subglacial roughness and (4) the spatial scale over which BRP is derived. Some areas of high basal drag can be detected with BRP analysis, indicating that variations in subglacial hydrology are responsible for their existence. The widely-used empirical method of estimating EA by relating ice thickness to uncorrected BRP is shown to be unreliable where ice properties change along a transect. Monte Carlo error analysis of modelled EA shows that poorly constrained ice dielectric properties also result in significant BRP uncertainty. BRP beneath Institute and Möller Ice Streams is derived on catchment- and local-scales over hypothesised subglacial features. Bungenstock Ice Rise is marked with a clear BRP signal but the locations of "active" lakes, as delineated by satellite altimetry, do not. The sensitivity of idealised flow paths to surface change and grid size are investigated. Potential future research directions regarding BRP analysis are discussed.

Over the last 50 years a significant increase in the atmospheric and upper ocean temperatures in the Antarctic Peninsula (AP) region has been observed. As a result major ice-shelves have retreated during the 20th century. In connection, glaciers have accelerated and an increased dynamic ice mass loss is observed, especially over the last decade. Despite these major changes, an exact quantification of ice mass changes of the AP, with its roughly 1000 glaciers, is not available. Almost no long-term (multi-decadal) glacier mass balance records for the AP exist and in-situ measurements are rare. On the other hand, the United States Geological Survey (USGS) and British Antarctic Survey (BAS) archives hold a large number of historic aerial stereo-photographs of the AP, dating back to the early 1940s. These images contain a valuable source of information and have been used to demonstrate widespread retreat of glaciers in this region. Less effort has been made so far to use this stereo-photography for the extraction of elevation data to compare it with recent elevation information to determine glacier volume change from which mass changes may be estimated. This dissertation seeks to close this research gap and to extend the number of mass balance records for the AP, by investigating, measuring, and analysing historical glacier elevation change in the AP using digital elevation models (DEMs) derived from USGS and BAS airborne (1948-2005) and ASTER spaceborne (2001-2010) stereo imagery. To ensure reliable and accurate measurements of surface elevation change, extracted DEMs need to be registered in a precise manner. The lack of ground control information in the AP is a major obstacle for this and can result in inaccurate absolute orientations of DEMs. If uncorrected, possible offsets between DEMs introduce significant error and i can lead to an over- or underestimation of glacier change. Thus, in order to precisely co-register corresponding historic and modern DEMs an iterative robust least squares surface matching algorithm was applied. The underlying surface matching approach was previously developed for small-scale coastal erosion studies at Newcastle University. Within the context of this work it has been successfully modified and improved to enable large scale glacier change assessment in areas of steep topography which is typical for the AP. For a total of 12 glaciers in the AP, located along the western coast between 64° and 71° S, DEMs from the historic archive stereo-imagery were successfully extracted and combined with DEMs derived from modern aerial and ASTER satellite imagery. The improved surface matching approach allowed precise co-registration of these DEMs and enabled the accurate measurement of glacier surface mass balance at the lower portion of the glaciers. Widespread frontal glacier surface lowering, of up to 50 m, has been observed on 12 glaciers with a mean lowering rate of 0.28 ± 0.03 m/yr over a period of 37 years (1970-2007). Higher rates, of up to 0.6 m/yr, were observed in the north-western Peninsula. Two glaciers which have multi-epoch coverage show a significantly larger-than-average lowering since about 1990. These results are in close correspondence with an increase in positive degree days over the last four decades and suggest that much of this lowering can be attributed to atmospheric forcing. However, the observed spatial and temporal variations in the lowering rates suggests that the pattern of surface change is not a simple one and that a regional upscaling is not straight forward. The glaciers represent only 1.2 % of all estimated glaciers in the AP and only the glacier fronts (~20 % of each glacier) were studied. Observations also show an elevation increase at some higher altitude locations within a few km of the glacier fronts, raising the potential that the lowering may have been at least partially compensated for by increased high-altitude accumulation.

In glacierised regions, suspended sediment fluxes are highly responsive to climate-driven environmental change and can provide important information regarding the relationships between glacier variations, climate and geomorphic change. As a result, understanding patterns of suspended sediment transport and their relationship with meltwater delivery is of critical importance. However, studies of glacial suspended sediment transport are often limited by interpreting patterns of suspended sediment transfer based on whole-season data, allowing precise patterns to become masked. This thesis aims to contribute to the understanding of suspended sediment transfer in glacierised basins through the investigation of patterns of suspended sediment delivery to the proglacial area of Storglaciären, a small polythermal valley glacier located in the Tarfala valley, Arctic Sweden. High temporal resolution discharge and suspended sediment concentration data were collected during two summer field campaigns at Storglaciären. Interpretations of suspended sediment transport data were made using diurnal hysteresis and sediment availability data, combined with suspended sediment shape and magnitude data classified by applying principal component and hierarchical cluster analyses. Analysis of the dominant discharge generating processes at Storglaciären was also conducted using principal component analysis, allowing patterns of discharge to be better understood. This was complemented by analysis of the structure and evolution of the glacier drainage system by linear reservoir modelling and flow recession analysis. The results suggest that patterns of discharge and suspended sediment transport at Storglaciären are complex, with distinct processes and magnitudes of transport evident at both proglacial outlet streams, Nordjåkk and Sydjåkk. These processes are intrinsically linked to meteorological variables, with both ablation-driven and precipitation-driven discharge exerting influence over patterns of suspended sediment transport in the proglacial area of Storglaciären.

Accurate forecasting of the Greenland Ice Sheet's contribution to global sea level change requires detailed knowledge of how ice ow responds to surface water inputs. Both ice velocities and surface melt have increased significantly over the last decade but recent research suggests that ice ow acceleration over the summer is regulated by the seasonal evolution of the subglacial drainage system. To investigate these and associated processes, a network of continuously-operating, dual-frequency global positioning system (GPS) receivers was deployed on a 140-km-long land-terminating transect in West Greenland, providing centimetre-precise, high-frequency records of ice motion. These data reveal that the enhanced summer ow regime is comprised of discrete, transient accelerations driven by the diurnal melt cycle, rapid in situ supraglacial lake drainage and rainfall/melt events. In 2010, a comprehensive array of instruments captured the rapid ( ~ 2 hour) drainage of a large supraglacial lake via a 3-km-long fracture, hydraulically-driven through km-thick ice. A further pronounced, widespread and sustained acceleration driven by rainfall and melt, observed in late August 2011, suggests that the predicted increase in cyclonic activity over Greenland may drive widespread off-season melt, rainfall and ow acceleration across the ice sheet. Together these events provide new insights into the basal hydrodynamic controls on ice sheet motion. Furthermore, observations of a persistent year-on-year acceleration in ice ow between 2009 and 2012 at a high elevation site located ~ 50 km inland of the equilibrium line support the hypothesis that the observed inland expansion of supraglacial lakes is driving faster ice ow at high elevations. These observations contrast with the prevailing self-regulation model and reveal that despite surface melt increasing water inputs to the bed are still insuffcient to develop effective subglacial drainage in the ice sheet's interior.