Search

The Hindu Kush–Karakoram Himalaya (HKKH) hold the largest ice mass outside of the poles. It is estimated that 1.9 billion people depend on the major rivers that are partially sustained by meltwater from these glaciers. Glacier mass loss has accelerated over the last decade and predictions indicate that this will continue over the next century, though vary widely and are based on relatively simplistic representations of climate and glacier behaviour. The aim of this thesis is to investigate the impact of climatic drivers on glacier mass balance, and how a robust representation of mass balance influences longer-term evolution of debris-covered Khumbu Glacier in the Nepal Himalaya. A cross-disciplinary modelling approach is adopted, utilising meteorological observations and regional climate models to aid downscaling of presentday and future climate, which is used to force a distributed energy balance model. The distributed mass balance results are integrated into a 3D higher-order ice flow model to study glacier evolution under historical and future atmospheric forcing. Energy balance modelling results demonstrate that the meteorological drivers of Khumbu Glacier mass balance vary substantially over space and time. The seasonality is more marked at lower elevations, where net shortwave radiation provides much of the energy available for melt, though net longwave radiation contributes appreciably during the monsoon. Despite temperatures remaining below freezing in the Western Cwm, ablation through sublimation is simulated. The calculation of a sub-debris ablation curve for Khumbu Glacier is applied to simulated clean-ice mass balance, improving glacier-wide and altitudinal ablation rates relative to observations. By 2100, temperature is found to increase by 1.8? to 3.9? depending on the emission scenario, with highest warming in the winter. Over the coming 80 years, precipitation increases by 15%, though varies substantially between models, demonstrating the high uncertainty regarding future precipitation trends. The value of forcing the ice flow model with distributed mass balance is seen in the simulated present-day velocities, particularly of the Khumbu icefall, which match closely with remote-sensing observations. The results indicate that the stagnant debris-covered tongue has already dynamically, and perhaps physically, detached from the active Khumbu Glacier. The significant recent warming means that the active section of Khumbu Glacier is committed to a 22% volume loss by 2100, and 34% by 2300. Future increased precipitation can act to offset the impact of moderate warming on glacier mass balance. However, under the most extreme warming scenario, mass loss is significant, particularly during the monsoon. Despite having the world's highest accumulation area, Khumbu Glacier may cease to exist by 2160.

Glaciers and ice caps are prominent features in the landscapes surrounding the Greenland Ice Sheet and in the archipelagos of the Arctic seas. Their dynamic response in mass and flow to climate variability makes them a visible and important expression of the changes that are affecting the Earth's climate system. Arctic glaciers and ice caps are losing mass rapidly, a trend sustained over recent decades and that is expected to continue into the 21st century driven by climatic change and exacerbated by the amplification of polar warming relative to the planetary mean. Mass imbalance of Arctic land ice is a source of major concern given its impact on global sea level rise, ocean circulation, regional hydrology, and on the local ecosystems and communities, which are also impacted by these changes, with increased ecological, socio-economic and geopolitical pressures afflicting the region. To effectively mitigate and adapt to these environmental changes, the large-scale, continuous monitoring of the Arctic cryosphere has become essential, and satellite geodesy is a critical tool for estimating glacier and ice cap mass balance. In this thesis, seven years of CryoSat-2 high-resolution swath interferometric altimetry were utilised to investigate changes in the volume of Arctic glaciers and ice caps outside of Greenland between 2010 and 2017. From these data, a pan-Arctic assessment of mass imbalance was produced, and losses partitioned into signals associated with atmospheric processes and glacier dynamics. The sampled satellite observations provide a detailed picture of the response of Arctic glaciers and ice caps to climate forcing, eliciting contrasting patterns of change that are driven by a combination of oceanic and atmospheric forcing and by internal instabilities (glacier surges). Results show that between 2010 and 2017, Arctic glaciers and ice caps lost 609 ± 7 gigatonnes of ice, contributing 0.240 ± 0.007 millimetres per year to global sea level rise. While surface ablation is responsible for 87 % of losses across the Arctic, dynamic imbalance is increasing in the Barents and Kara Sea region where it now accounts for almost half of the total ice loss. This dynamic imbalance is associated with a northward shift of Atlantic climate and is sustained by a range of climate-dynamic feedbacks and processes that have been amplified in areas adjacent to the archipelagos of the Eurasian Shelf. The thesis builds on these initial pan-Arctic findings to further investigate, in Chapter 5, the processes that shape the current response to forcing of the glaciers and ice caps in the Eurasian High Arctic, with a particular focus on glacier fluctuations in the Russian archipelagos of Novaya Zemlya and Severnaya Zemlya. Time series of surface elevation change between 2010 and 2018 in conjunction with concurrent climate anomaly trends provide insights into the contrasting cryospheric changes affecting these two regions, as well as their association with contemporary atmospheric and oceanic forcing. While both frontal ablation of tidewater glaciers and surface melt are found to jointly contribute to volume changes in Novaya Zemlya, ice dynamics are entirely responsible for the recent net ice loss from Severnaya Zemlya. A quasi-linear relationship between coupled ocean-atmosphere forcing and volumetric changes in Novaya Zemlya is found to be appropriate in explaining the overall regional trends in ice loss at decadal scale, in agreement with previous findings from tidewater glaciers in east Greenland. In Severnaya Zemlya, the role of ocean warming is found to be the key factor in driving dynamic ice loss, owing to the intrusion of warmer Atlantic Water circulating along the Eurasian continental margin. Together these results suggest that meteorological processes will continue to dominate glacier loss throughout the Arctic, while warming of the Arctic Ocean will increasingly affect the dynamic stability of tidewater glaciers in the Barents and Kara Sea regions, with unpredictable and possibly abrupt shifts in mass balance. In this region, as well as globally, ice dynamics represent the largest source of uncertainty in climate models predicting how land ice will respond to future climatic fluctuations. The findings from this thesis support the principle that linear relationships between environmental forcings and glacier change may be sufficiently accurate to parametrise ice loss in regions where synchronous, coupled ocean-atmospheric forcing prevails. The results therefore provide important support to current efforts aimed at estimating future ice loss, freshwater budgets, and sea level contributions from Arctic glaciers and ice caps in a changing climate.

With climate change as an increasingly important issue, the Greenland and Antarctic ice sheets have suffered rapid ice mass losses contributing to sea level rise. Iceberg calving is one of the main factors for ice mass loss and may also generate large tsunamis when icebergs calve into a water body, threatening the fishing and shipping industries and coastal communities. One of the most impressive tsunamis generated by iceberg calving with an amplitude of 45 to 50 m was observed at the Eqip Sermia glacier in Greenland in 2014. Herein, such tsunamis are called iceberg-tsunamis (IBTs). This study aims to investigate the generation and propagation of large IBTs. A novel numerical model is developed based on the Immersed Boundary Method (IBM) provided in a toolbox in the open source code Foam-extend 4.0. A new multiphase flow solver is implemented to solve the Reynolds-Averaged Navier-Stokes equations with the support of handling moving immersed boundaries, and it is then coupled with a motion solver to simulate Fluid-Structure Interaction (FSI) to determine the iceberg motion. Further, unique large-scale laboratory experiments were conducted in a 50 m x 50 m large basin. Two rigid blocks with densities of ~920 kg/m³ weighting up to 187 kg were used to mimic icebergs and to model the five idealised calving mechanisms: (A) capsizing, (B) gravity-dominated fall, (C) buoyancy-dominated fall, (D) gravity-dominated overturning and (E) buoyancy-dominated overturning. The analytical solution of a floating heaving sphere case is used to validate the newly implemented flow solver and the numerical radiated wave amplitudes show a maximum deviation of only 10.3% from the corresponding analytical solution. Further, the numerical model is used to simulate the two laboratory experiments from mechanisms B and D generating the largest and most dangerous IBTs. A laminar flow model is selected as the consideration of turbulence does not improve the accuracy. For IBT generation, the maximum vertical displacement in the fall case and the pitch angle in the overturning case are 24.3 and 19.9% larger than the corresponding laboratory measurements. During this stage of the IBT, the pressure force is the dominant force while the viscosity force can be neglected in the overall force balance. For IBT propagation, the leading IBT amplitude and wave height decay are captured with a maximum relative error of -41.2 and -29.3%, respectively. This large deviation is mainly due to the water splash involved in the laboratory experiments, which can not fully be modelled in the numerical simulations. If this splash is artificially removed, then the corresponding relative errors become -17.2 and -9.2%, respectively. Additionally, the numerical IBT train energy shows a good agreement with the laboratory results with a relative error of 1.0% in mechanism B, while the difference in mechanism D is much larger due to the oscillation of the iceberg. Further, the capability of this numerical model of simulating FSI is confirmed by the agreement of the contact forces of the icebergs acting on the glacier terminus with that in other studies. Smaller and less important IBTs generated in the remaining mechanisms A, C and E are also simulated. However, due to the nature of the IBM making it challenging to fully satisfy the continuity equation and the no-slip boundary condition around the immersed boundary when multiphase flows are involved, more work is required to also simulate these small waves, in the order of 10⁻³ m at laboratory scale, well. Nevertheless, the numerical model is capable of simulating large and dangerous IBTs as further demonstrated by the successfully reproduced 2014 Eqip Sermia case. Overall, this work demonstrates that the proposed numerical model has the potential to support IBT hazard assessment. Future work will focus on the improvement of the immersed boundary representation technique to numerically investigate IBTs involving additional and more complex calving mechanisms and scenarios.

Antarctica has long been isolated from its nearest landmasses and has been at least partially glaciated for the past ~35 million years. This long-term isolation combined with the fragmentation of habitats creates a unique environment for the examination of biogeographic patterns. Previously it was believed that the repeated full glaciation of Antarctica wiped out the pre-existing terrestrial fauna and flora and that the continent was, therefore, subsequently re-colonised after periods of glacial retreat, such as from the particularly extensive Last Glacial Maximum. However, recent molecular studies of a variety of taxonomic groups have shown that much of the current terrestrial life found on the continent is generally ancient and must have survived these glaciation events. The main body of Antarctica (East or continental Antarctica) and West Antarctica and the Antarctic Peninsula show differing geological histories and distinct biogeographic patterns in a variety of invertebrates including mites, springtails and nematodes, indicating distinct evolutionary histories. However, despite tardigrades being the most common and widespread invertebrate within Antarctica there has been limited study of their Antarctic biogeography. This thesis set out to investigate the evolutionary and biogeographic patterns of two eutardigrade species (Acutuncus antarcticus and Mesobiotus furciger). At the initiation of this study, A. antarcticus was believed to be a pan-Antarctic species, with a Gondwanan distribution, and M. furciger a globally common taxon and a Southern hemisphere species with a widespread Antarctic distribution. Individuals were sampled that originated from multiple locations across Antarctica. These individuals were then used to document a range of morphometric characters and underwent molecular extraction targeting the three gene regions of 18S, 28S and COX1. Further sequence data were added from GenBank. Phylogenetic analyses using both Maximum Likelihood and Bayesian methods showed that both species are likely to be species groups, with highly distinct lineages in the maritime and continental Antarctic regions. The levels of divergence detected strongly support cryptic speciation. This conclusion was further supported by morphological analyses, which identified significant differences between material sampled from each of these regions. Ancestral state and molecular clock analyses showed that both of these species groups had ancient Antarctic origins, requiring long-term regional survival within Antarctica, consistent with patterns identified in other invertebrate taxa. Morphological comparisons between individuals within single genetic clades obtained from different habitats also identified significant differences, supporting phenotypic plasticity in response to environmental conditions. This finding, not noted previously in tardigrades, has important taxonomic implications, as species identification relying on morphological characters alone may now result in miss-identification.

Subglacial hydrology has been linked to some of the most dynamic ice sheet behaviour and is likely a key regulator of the impacts of climate change on ice sheet mass balance. Evidence of palaeo-meltwater drainage on the beds of former ice sheets enables a large-scale assessment of the distribution and nature of the subglacial hydrological system, which helps to contextualise spatio-temporally limited observations from contemporary ice masses. In this thesis I develop a new fully-automatic mapping method to detect and map meltwater tracks. I use this as a starting point for holistic meltwater feature mapping (i.e. incorporating a range of erosional to depositional signatures) to investigate the nature and evolution of the subglacial drainage system across the former Keewatin sector of the Laurentide Ice Sheet. I map a range of meltwater features (i.e. eskers and associated assemblages, tunnel valleys and meltwater tracks) and propose the grouping of tunnel valleys and meltwater tracks under the term 'meltwater corridor'. I then propose a formation theory for meltwater corridors based on the repeated interaction between a conduit and the surrounding hydraulically-connected distributed drainage system driven by variable surface meltwater inputs. This is a process known to occur in contemporary settings and is able to account for the observations made from the holistic landform map. This provides new insight into the distribution and nature of the subglacial hydrological system over a large area. Results within also contribute to a better understanding of temporal relationships through the use of eskers beads and fans as a time-transgressive signature, provide a mechanism for sediment access and entrainment at the bed and links the presence of meltwater pathways to the formation and preservation of other subglacial bedforms (e.g. drumlins and ribbed moraines) and overlying ice sheet dynamics. This thesis has benefited enormously from the recent release of freely available, widespread high- resolution digital elevation data - the ArcticDEM - which has enabled the identification and mapping of meltwater drainage signatures.

Debris-covered glaciers comprise around a third of all glaciers in High Mountain Asia, and thus represent a critical source of meltwater for those living in the foothills of the region. Many of these debris-covered glaciers are situated at elevations exceeding 4,500 m above sea level and display a number of unique supraglacial features and mass balance characteristics arising from the spatially variable thickness of their supraglacial debris layers. Given their remoteness, these glaciers are difficult to access; field-based data are severely lacking, particularly for the subsurface environment, meaning both the parameterisation and validation of numerical modelling experiments are frequently based on assumptions and calculations. To address this gap, this research investigates the subsurface environment of a high-elevation debris-covered glacier, Khumbu Glacier, Nepal Himalaya, through a series of field-based experiments. In particular, the thermal regime, subsurface hydrology, and englacial debris content of Khumbu Glacier are targeted. The results reveal that Khumbu Glacier has a polythermal regime, with a notable proportion of temperate, and potentially warming, ice towards the terminus, which may support the presence of a persistent subsurface hydrology. The glacier's drainage system transports meltwater inefficiently during the early melt season, but has the capacity for more efficient transport with greater inputs. Meltwater transit through a large chain of linked supraglacial ponds is slow, which may promote enhanced surface ablation and, together with a warming thermal regime, may increase the glacier's sensitivity to future climatic warming through time. However, in contrast, the continued melt-out of a relatively high englacial debris content may act to suppress melt rates within only a few decades. Combined, the data presented in this thesis provide a suite of empirical observations that can reduce the current uncertainties inherent in contemporary numerical predictions of debris-covered glacier evolution and water resource availability within the region.

Predictions of the Antarctic ice sheet contribution to sea-level have large uncertainties. Confidence in projections relies on multi-model agreement, probabilistic calibrations to judge simulations by consistency with observations and a solid representation of observational uncertainties. Each of these factors require statistical considerations which can be challenging and even computationally prohibitive, especially since ice sheet models are often computationally expensive. Here we show that diverging results from two structurally very different ice sheet models can be fully explained by differences in the study designs. Furthermore we find that the spatial characteristics of ice thickness change observations can be used to improve probabilistic calibrations of simulations from the high-resolution ice sheet model BISICLES in the Amundsen Sea Embayment, West Antarctica. These spatial constraints reduce the 50-year sea-level contribution uncertainty interval by nearly 40%, compared to a calibration on total mass loss, and by nearly 90% compared to using no observations at all. Finally, we build a stochastic model to construct an ensemble of plausible bedrock topographies and show that measurement and interpolation uncertainties for Pine Island Glacier topography contribute substantially to predictive uncertainties. Our most optimistic and pessimistic 100-year projections are 4.7 ±-0.87 mm and 19.4 ±-5.15 mm sea-level contribution (mean and standard deviation), where the stated uncertainties originate solely from the bedrock. Together, this work improves our understanding of limitations in modelling the future of the Amundsen Sea Embayment and helps to substantiate and reduce predictive uncertainties. Most of the uncertainty quantification methods we adapt to ice sheets in this work can be applied to large ensemble, Antarctic-wide model studies where fast but exhaustive use of available information is crucial. The stochastic approach to bedrock uncertainty is, in its current form, limited to regional applications but is nevertheless a call for caution for the use of any single bedrock topography.

Cryoconite holes are miniature freshwater ecosystems on glacier surfaces worldwide that harbour a diverse microbial community. Microbial isolates from cryoconite holes show exceptional environmental tolerance: they are able to grow in a wide range of pH, high salinities and scavenge a wide range of organic carbon sources. All the isolates tested were facultative anaerobes The community resilience exceeds the 'normal' conditions encountered, and enables their functionality in microenvironments, and in transition environments caused by freeze-thaw. This study demonstrates for the first time that environmental conditions, in particular concentrations of oxygen, are spatially heterogenous at the microscale. Anoxic zones develop rapidly following disturbance and can be found inside cryoconite granules and in thick layers of cryoconite sediment. These microhabitats, with limited oxygen, create microniches for anaerobic metabolism. Facultatively anaerobic microorganisms isolated from cryoconite holes are capable of anaerobic biodegradation via fermentation, and anaerobic metabolism in the holes is dominated by fermentation. An excess of fermentation products, especially acetate, can be detected in defrosted cryoconite sediments and in long-term incubations, indicating incomplete metabolic pathways. Analysis of the microbial community structure revealed that terminal oxidisers are present in low abundance in some of the samples, but the terminal metabolic steps such as sulphate reduction or methanogenesis are not coupled with fermentation. This means that fermentation products could be released to downstream environments, but the extent and impact of this process requires further quantification. Anaerobic metabolism and anaerobic microorganisms are therefore important components of cryoconite ecosystem, which degrade organic matter and impact the glacier carbon cycle.

The focus of this thesis is former ice mass extent and evolution in the Wicklow Mountains, Ireland, during the Last Glacial-Interglacial Transition (LGIT; ~20-10 ka). Whilst previous research has established the existence of a local ice cap, confluent with the wider British-Irish Ice Sheet (BIIS) at the Last Glacial Maximum (LGM; ~24 ka BP), the nature of ice mass recession during the LGIT and subsequent glacier regrowth during the Younger Dryas (YD; 12.9-11.7 ka) requires evaluation. This thesis aims to produce the first regional assessment of the local palaeoglaciological record. The results of systematic geomorphological mapping, based on a combination of remote sensing and field techniques, are presented in the form of the first glacial geomorphological map of the Wicklow Mountains. The post-LGM evolution of local ice is examined based on an assessment of the geomorphological record, establishing the extent, style, and evolution of LGIT mountain glaciation during deglaciation. Using morphostratigraphic principles, two distinct phases of glaciation are recognised; (1) marginal cirque glaciation during the YD, and (2) sub-stages of more extensive topographically-restricted mountain glaciation earlier in the LGIT. A conservative approach based on an assessment of topographic setting, geomorphology, and radiation modelling indicates that the region hosted six small YD cirque glaciers. The glaciers had an average size of 0.39 km2 and covered a total area of 2.33 km2, based on their reconstruction using a combination of geomorphology and surface profile modelling. Reconstructed ELAs (Area-Altitude Balance Ratio; AABR 1.9 ± 0.81) range from 467 ± 9 m to 739 ± 8 m, with a regional average of 621 ± 9 m. Glacier-derived palaeoprecipitation calculations imply wetter conditions than at present. However, topoclimatic modelling demonstrates that local glacier mass balances were augmented by the redistribution of snow by wind and avalanching, and topographic shading, which artificially lowered ELAs. These processes were essential for marginal YD glacier initiation and survival. The style and pattern of wider LGIT ice recession is also established and identifies that local post-LGM ice did not recede and thin as one coherent ice mass. Instead, it separated into multiple topographically-controlled ice masses of different size (alpine icefield, plateau icefield, valley glaciers, and cirque glaciers) in coeval existence. Topography exerted a control on the pattern of local glacier recession and associated moraine distribution. A combination of topographic pinning points (i.e. topographic steps, lateral constriction, and tributary valley incursion), favourable slope gradients (0-15°), and changes in slope appear to have had an important influence on where moraines formed, implying an element of glacier stability at such locations. This provides further insight into non-climatic controls on moraine formation and presence/absence. This thesis delivers a more detailed understanding of post-LGM glaciation in the Wicklow Mountains. It provides clarification on the scale of YD glaciation and the impact of local topoclimatic processes on cirque glacier initiation and survival. It also provides insight into the significant role of topography in controlling retreat patterns and stability of ice masses, in response to climate amelioration throughout the wider LGIT.

The work presented within this thesis is focussed upon the interactions between glacial ice and rock debris across the cryosphere; the melt rate of debris-covered ice, the formation of ice sails and the movement of meteorites through Antarctic blue ice have all been investigated. I numerically modelled the melt rate of debris-covered ice and found that the surface temperature of a debris layer can be used to estimate the melt rate of any underlying ice - even when the thermal properties of the debris layer are unknown. This potentially removes the need for the majority of expeditions as it means that it is feasible to predict the melt rate of a debris-covered glacier using data from satellites and weather stations alone. Ice sails are substantial clean ice melt features which can be over 45 m high, last for 50 to 100 years and are distinctively shaped with sloped, flat sides. Due to the rarity of their emergence they had, until now, seldom been studied. By combining a mathematical model, analysis of satellite imagery, a review of historical sightings, and analysis of both pictorial and first-hand evidence, we gained valuable insights into their formation, persistence and decline. We found that, in order for ice sails to emerge, a debris-covered glacier must exist at a high altitude, be only very gently sloped (so that the debris layer remains thin for a long time) and have patchy and uneven debris cover. Two thirds of all meteorite finds have been discovered in Antarctica, concentrated in blue ice areas called meteorite stranding zones. Of these, the vast majority are stony, with only 0.7% being classified as iron - almost eight times less than elsewhere. It was hypothesised by Evatt et al that this deficiency can be explained by a hidden layer of iron meteorites that are trapped roughly 40 cm below the surface of the ice (see Evatt et al, 2016). However, by considering both the three-dimensionality of the problem and by modelling the attenuation of solar radiation through blue ice in a detailed manner, I was able to use numerical modelling techniques to predict that most iron meteorites, as well as some of the larger stony ones, lie less than 10 cm below the surface of the ice. This is a significant difference that will make detection and retrieval far easier during future meteorite recovery missions.

Understanding the effect warming has on ice sheets is vital for accurate projections of climate change. A better understanding of how the Antarctic ice sheets have changed size and shape in the past would allow us to improve our predictions of how they may adapt in the future; this is of particular relevance in predicting future global sea level changes. This research makes use of previous reconstructions of the ice sheets, ice core data and Bayesian methods to create a model of the Antarctic ice sheet at the Last Glacial Maximum (LGM). We do this by finding the relationship between the ice sheet shape and water isotope values. We developed a prior model which describes the variation between a set of ice sheet reconstructions at the LGM. A set of ice sheet shapes formed using this model was determined by a consultation with experts and run through the general circulation model HadCM3, providing us with paired data sets of ice sheet shapes and water isotope estimates. The relationship between ice sheet shape and water isotopes is explored using a Gaussian process emulator of HadCM3, building a statistical distribution describing the shape of the ice sheets given the isotope values outputted by the climate model. We then use MCMC to sample from the posterior distribution of the ice sheet shape and attempt to find a shape that creates isotopic values matching as closely as possible to the observations collected from ice cores. This allows us to quantify the uncertainty in the shape and incorporate expert beliefs about the Antarctic ice sheet during this time period. Our results suggests that the AIS may have been thicker at the LGM that previous reconstructions predict.

A significant portion of Norwegian glaciers are plateau icefields. These are highly sensitive to changes in climate because of their top-heavy hypsometry. A small rise in the equilibrium line altitude may cause the ablation area to expand significantly, leading to rapid icefield recession. This behaviour deserves particular research attention in light of the current warming of the Earth’s climate system. This doctoral thesis examines the response of the icefields Hardangerjøkulen in southern Norway, Svartisen just within the Arctic Circle, and Langfjordjøkelen in Arctic mainland Norway to climate warming since the Little Ice Age (LIA). These ice masses have been in retreat since they reached their maximum LIA positions between ~1750-1925. The glacial landform record of these icefields is used to reconstruct their maximum LIA extent. In addition, historical maps are used to reconstruct their ~1900 extent. These data sets are compared with data from available Norwegian glacier inventories of the present-day glacier extent, revealing substantial changes in icefield size. The greatest percentage decline occurred at Langfjordjøkelen in Arctic Norway, with an area reduction of 57 % between 1925 and 2018. Hardangerjøkulen in southern Norway lost 37 % of its original LIA area between 1750 and 2013, and the Vestre and Østre Svartisen icefields in northern Norway had lost 16 % and 23 % of their ~1900 areas by 2000, respectively. This research demonstrates how the LIA glacier extent can be employed as a valuable baseline to assess long-term glacier change.

Contribution to global mean sea level rise by ice sheets, ice caps and glaciers is accelerating. The total volume of water stored globally in terrestrial ice is estimated by a multitude of methods but principally by the interpolation of icethickness data. For the ice sheets and large Arctic ice caps, ice thickness is predominantly measured by airborne radio-echo sounding surveys which use radio waves to detect the bed of the surveyed ice mass. While such surveys are now extensive, large portions of ice masses are generally unsurveyed due to their size. In order to quantify ice thickness and subsequently ice volume over the entirety of an ice mass, interpolation of the input measurements is used. Throughout this whole process, uncertainties arise. Initially, from the radio-echo sounding (RES) survey and subsequently, in the interpolation. Compounding this is the absence of ground-truthing for measurements and interpolations due to the inaccessibility of ice mass beds. Hence, there is a requirement to find alternative means of quantifying uncertainty in ice thickness measurements and subsequently derived bed topography, and analyses made from these data to reduce the uncertainty in sea level change projections. This thesis develops and applies methods which aim to reduce uncertainty in ice thickness and bed topography datasets. Using high-resolution elevation data, this study exploits the likely similarity between currently ice-covered topography and formerly glaciated topography in the Arctic to generate datasets which provide alternative validation for ice mass bed topography. For the first time topographic error in RES surveying is quantified and corrections are formulated for treating future and historic ice thickness and bed topography data. Additionally, the propagation of these uncertainties through interpolations of bed topography is quantified and reduced, focussing on the Greenland Ice Sheet. Finally, the full suite of methods is applied to ice caps in the Canadian Arctic to generate, for the first time, ice cap wide topography for ice caps in the region that hold approximately a third of the freshwater outside of the continental ice sheets. By quantifying and reducing uncertainty in datasets of bed topography and ice thickness this thesis assesses the perceived stability of the continental ice sheets and large ice Arctic ice caps. From this, the implications of this for near and far term global mean sea-level rise are investigated.

There has been a progressive increase in the number and area of ice-marginal lakes along the western margin of the Greenland Ice Sheet (GrIS) since the late 1980s. Ice-marginal lake formation and growth have been widely associated with accelerated rates of mass loss and terminus recession at alpine glaciers, yet their impacts on the GrIS have remained unquantified. This thesis therefore investigated the influence of ice-marginal lakes on ice-margin dynamics in west Greenland at multiple spatial and temporal scales, using both established remote sensing techniques and the novel integration of time-lapse photography with Structure-from-Motion and Multi-View Stereo. A regional-decadal scale analysis of ice-margin change along a ~5000 km length of the GrIS revealed that lake-terminating ice-margins receded faster than their terrestrial counterparts between 1987 and 2015. In addition, the rate of recession at lake-terminating ice-margins accelerated over the study period and increasingly outpaced recession at terrestrial ice-margins. Altitude, latitude, lake area and the length of the lake – ice-margin interface were also identified as significant controls on rates of lake-terminating ice-margin recession. Local-seasonal scale ice-margin dynamics were investigated using the first continuous year-round volumetric record of calving at a lacustrine ice-margin. These data highlighted two distinct calving regimes; with melt-undercutting driving high calving rates under ice-free lake conditions, and force imbalances at the ice-cliff driving low calving rates when the lake was frozen. These results are important because they demonstrate that ice-marginal lakes are key regulators of ice-margin dynamics at the GrIS. The quantitative data derived through this study provide an empirical foundation upon which modelling efforts can incorporate the influence of ice-marginal processes. This is particularly pertinent given that rates of mass loss and recession at lake-terminating margins of the GrIS are likely to accelerate in coming decades in response to continued ice-marginal lake expansion and a lengthening melt season.

Glaciers in contact with proglacial lakes show accelerated mass loss rates through mechanical and thermal processes (Carrivick and Tweed, 2013). An inventory of 108 proglacial lakes in Arctic Sweden is presented, results show they have increased in number and area with glacier recession since the 1950s/60s. The retreat of 24 glaciers was mapped from satellite imagery between 2010 and 2018. A common assumption persists that smaller proglacial lakes remain at a uniform 1C. ASTER satellite data was analysed for suface temperatures of proglacial lakes and used to form a conceptual classification of proglacial lake thermal regime by stage of deglaciation. This thesis presents some of the first recorded proglacial lake temperatures from the front of a calving Arctic glacier. The innovative use of thermal infrared imagery at the ice front in July 2017 was supplemented with thermistor temperature surveys (using an ASV). Combined with evidence of rapid thermal undercutting and associated calving, this study provides some of the first direct evidence of proglacial lake temperatures directly impacting the retreat of an Arctic glacier.

Accurate projection of the future evolution of the Greenland and Antarctic ice sheets is a key challenge of glaciology and climate science. Ice sheet flow is dominated by ice streams; narrow corridors of fast flow bounded laterally by slower flowing ice, discharging the majority of ice from an ice sheet. These ice streams are particularly vulnerable to retreat, and their behaviour evolves along with the ice sheet as a whole. These interactions contribute to uncertainty in projections of ice sheet evolution. In part, this uncertainty can be addressed by examining the palaeo record, providing information on ice stream behaviour over thousands of years. This thesis presents a series of simulations of the British-Irish Ice Sheet using the latest generation BISICLES ice sheet model. Model simulations are used to determine the role of ice streams in the demise of the British-Irish Ice Sheet. First, BISICLES is used to examine the dynamical processes that control the retreat of a major ice stream of the British-Irish Ice Sheet, and this demonstrates vulnerability of the ice stream to Marine Ice Sheet Instability. Then a new basal sliding scheme is implemented coupled with thermo-mechanics, and this successfully models the placement and spacing of the majority of British-Irish Ice Sheet ice streams. Finally, simulations of the deglaciation style of the North Sea demonstrate the significant influence of the Norwegian Channel Ice Stream. Through these simulations, in combination with novel model-data comparison techniques, the considerable role of ice streams in the demise of the British-Irish Ice Sheet is shown. Ice stream evolution and interaction with other factors driving deglaciation needs to be adequately considered in the aim of projecting the future evolution of the Antarctic and Greenland ice sheets. Research such as presented here, modelling and reconstructing palaeo ice sheets, continues to advance this aim.

Supraglacial environments are vital to glacial and ice sheet systems as they collect atmospheric and windblown inputs, are the locus of meltwater production and are a key nutrient source to englacial, subglacial and downstream environments. They have also been found to host abundant and diverse microbial communities focused in three main habitats; snow packs, bare ice and cryoconite holes. Yet due to the harsh conditions of this environment, microbial activity was considered of little global importance, particularly to biogeochemical cycles. Recently, large scale blooms of supraglacial microbial communities, in areas such as the Greenland Ice Sheet Dark Zone, an annually reoccurring area of decreasing surface albedo due to large scale algal blooms along the west-coast, have begun to change this assumption. In an effort to enhance the current understanding of nutrient cycling in supraglacial environments and the influence of resident microbial communities, comprehensive datasets of snow, ice and meltwater from multiple ablation seasons were collected from the Greenland Ice Sheet Dark Zone. A snow incubation experiment was also conducted to investigate microbial nutrient cycling under simulated polar winter conditions. The data indicate microbially mediated nutrient cycling in supraglacial environments during both the polar winter and main growth season, having implications for nutrients released from the snow pack during spring thaw and nutrients later exported from the supraglacial environment. Results from the spring thaw on the Greenland Ice Sheet (GrIS) show the snow pack to be an important source of nitrogen to the supraglacial environment, which may act as a stimulant for the annual algal bloom initiation. After the development of the algal bloom, during the peak of the growth season, the data reveal bare ice surfaces to be areas of high dissolved organic matter accumulation as glacier algae cycle nutrients from the inorganic to the organic phase more efficiently then heterotrophs remineralize the organic matter. Glacier algae blooms also depict a potential retention ability for maintaining dissolved organic nutrients at the surface. As a result, nutrient export to downstream environments is likely to be pulsed rather than constant, as well as rich in labile, dissolved organic nutrients, potentially impacting chemoheterotroph and chemoorganotroph community activity in subglacial environments. Further, data from early ice melt during spring thaw on the GrIS indicate dissolved organic nutrients retained in the ice surface over winter from the previous glacier algal bloom, provide a further nutrient source for the following season algal bloom. As the GrIS is the largest contributor to cryospheric sea level rise, and glacier algae blooms have been shown to significantly increase melt rates on the GrIS, it is imperative to better understand how they utilize and cycle nutrients in these surface ice habitats, as their expansion has critical implications for global sea level rise.

This thesis-led research is informed by emotional and cognitive dissonance, arising in response to confronting the experiential magnitude of living in precarious ecological times. In Staying with the Trouble, Donna Haraway (2016) invites us to “stay with” the challenges of such unresolvable discomfort. Yet, what is the experience of attempting to do so, and how might this enable generative forms of worthwhile response? From my own experiences, I witness myself shifting between either feeling ‘too much’ or ‘too little’, a paradoxical dynamic leading toward sensations of paralysis, numbness and overwhelm. My research seeks to understand these personal experiences by positioning them within a wider context of Western cultural narratives. It asks, in what ways are such narratives supporting and perpetuating collective states of disconnection and detachment? My hypothesis is that a relationship exists between cultural frameworks of time and cultural frameworks of empathy that are limiting my capacity to experience myself as deeply interconnected and interdependent within the world. This lack of recognition, I will argue, has a detrimental ecological impact. I therefore propose an urgent reassessment of the normative constructions of time and empathy alongside an exploration of their creative reimagination, which I will offer in the form of two possible alternative narratives; Radical Empathy and Glacial Time. Radical Empathy is an amplified approach to relating with others which draws from, and exists in dialogue with ideas of “kinship” (Haraway 2016), the “suprasocial” (Manning 2013) and “solidarity” (Morton 2017). It forms part of a locus of theory rooted in queer and feminist thinking which is committed to posthuman and nonhuman concerns. Adapted from the work of John Urry (1994; 2011) Glacial Time is an alternative temporality that I am defining as a set of practices and attitudes toward thinking about time. It contributes toward an ecocritical exploration of time (Huebener 2018) that I will suggest opens new pathways for relating within the present, as well as offering new imaginations for orientating toward the future. In positioning time and empathy as political and ethical concerns, it is my aim to investigate how reductive binary groupings such as ‘us’ and ‘them’ can be sustained in ways which enable us to locate ‘others’ outside of our circles of care and concern. I will assess how dominant linear temporal frameworks further justify and reproduce these divisive relational configurations, maintaining the conditions for personal and collective dissonance in the process. I will further look for ways in which to challenge dominant forms of temporal logic which forge value systems centred around progress and productivity that bind together an understanding of life and liveliness in ways that I perceive to be harmful. I will then ask how a more radical approach to empathy might instead rewrite the mythologies of containment that create a bounded and separate self, and offer alternative ways in which to orientate toward each other. I will engage with Radical Empathy and Glacial Time by means of a “doing-thinking” methodology (Gibson-Graham 2008) based in staying-with my own experiences and drawing them into my research to facilitate and support embodied connection as much as possible. In addition to reflecting on elements of my own practice, I will present my experiences of other writers, artists and thinkers whom I understand to be enacting, performing or generating elements of Radical Empathy or Glacial Time in their work. I look to these practices as a means of finding ways through dissonance and paralysis toward an increased capacity for responsiveness and responsibility in meeting the world.

The Antarctic Peninsula Ice Sheet (APIS) is currently experiencing significant mass loss, contributing to global sea level rise. Increasing mass loss is primarily a response to reductions in the thickness and extent of ice shelves, triggering retreat, acceleration and drawdown of marine terminating glaciers. These changes have occurred in tandem with atmospheric and oceanic warming, which have had additional impacts on sea ice, the properties of water masses and marine productivity. A full assessment of these recent changes, together with the relative importance of key drivers for initiating grounding line and ice shelf retreat, is limited by the timescales over which we have recorded contemporaneous measurements of ice sheet change(s) and climate observations. Consequently, there remains a need to understand the behaviour of the APIS over a range of timescales so that we can fully understand future ice sheet change. This study utilises marine geological and geophysical data from Anvers-Hugo Trough (AHT), to improve our understanding of the glacial and environmental history of the western Antarctic Peninsula following the Last Glacial Maximum (25-19 kyr BP). Detailed multi-proxy analysis of sediment cores allows identification of sedimentary units deposited over the last deglaciation of AHT. Particular focus is paid investigating the variability in sediments deposited in the transitional (grounding line proximal to distal) environment in order to identify key depositional processes active in this setting and infer grounding line behaviour over deglaciation. Transitional sedimentary units are found to be deposited systematically; varying with marine processes, grounding line retreat rate, the presence/absence of an ice shelf, bathymetric position and across trough position. Quantitative assessment of absolute diatom abundance and species assemblages, aid reconstruction of the palaeoenvironmental conditions in AHT during and after deglaciation and allow the link between deglaciation and primary productivity to be assessed. These data indicate that the AHT experienced a period of enhanced diatom productivity following the last deglaciation of the trough, associated with meltwater discharge and nutrient input from sea and glacial ice melt. This led to deposition of laminated diatomaceous oozes, with their spatial distribution and accumulation influenced by seafloor bathymetry. This indicates that the distribution of laminated oozes does not necessarily reflect the spatial extent of surface productivity. The timing of deglaciation and post-glacial environmental change is inferred from 27 new radiocarbon dates. The grounding line had retreated from the mid-shelf as early as 15.84 cal. kyr BP, reaching Palmer Deep on the inner shelf shortly before ~12.85 cal. kyr BP. Peak, postglacial primary productivity on the mid to inner shelf followed grounding line retreat by ~4.6 kyr, indicating that deglaciation alone did not create this highly productive environment within AHT. Additional drivers of glacial and/or sea-ice melt, including high atmospheric and sea surface temperatures, were required to trigger this period of peak diatom productivity.

This thesis uses climate, glacio-hydrological models (GHMs) and groundwater models to advance understanding of: 1) twenty-first century climate change impacts on glacier-fed river flow regime and proglacial groundwater dynamics at the Virkisjökull Glacier Observatory in Iceland; and 2) uncertainties associated with model projections which underpin this understanding. The research is split into three studies. Study 1 tests a novel, signature-based Limits of Acceptability framework for constraining structural uncertainties in GHMs. The framework successfully identifies deficiencies in different melt and runoff-routing model structures, but cannot identify a population of acceptable model structures. Study 2 uses an ensemble of regional climate projections and GHMs to project changes in 25 characteristics (signatures) of river flow regime up to 2100. The results show that the magnitude, timing and variability of river flow are sensitive to climate change and that projection uncertainties stem from incomplete knowledge of future climate and glacio-hydrological processes. The dominant uncertainty source, however, is signature-specific. Study 3 includes a proglacial groundwater model into the climate-GHM model chain and shows that climate change will perturb intra-annual groundwater level timing and variability leading to changes in groundwater-surface water interactions. Uncertainties in groundwater projections primarily stem from future climate uncertainty.

In recent years, there is an increasing trend on the scientific, commercial and engineering activities in cold regions and polar oceans. The research on the interaction of water waves and structures in the region with ice cover has therefore received considerable attention. This thesis primarily considers two categories of such problems. On one hand, as a dominate feature of polar ocean, the existence of ice sheet can cause the working environment of ocean structures in icy water quite different from that in open water. When water waves propagate into the ice sheet, wave reflection and transmission occur at the ice edge, and the energy propagation is carried out in the form of hydroelastic waves due to the elastic deformation of the ice sheet. The hydrodynamic coefficients of ocean structures in waves under some typical ice conditions and their resonance features need to be investigated. The corresponding physical scenarios involve: the wave excited motion of a body floating on open water in a waterway confined between two semi-infinite ice sheets, the wave diffraction and radiation by a bottom-mounted vertical circular cylinder standing arbitrarily in a three-dimensional polynya, and the diffraction of hydroelastic waves by multiple vertical circular cylinders. On the other hand, the water surface in a channel may be frozen and covered by an ice sheet during severe winter in cold regions. The propagation of hydroelastic waves in channels is also quite different from free surface waves in the open water channel. Thus, an efficient solution procedure is developed to study hydroelastic waves propagating in a channel. The corresponding dispersion relationships, profiles of hydroelastic waves, and the strain distribution in the ice sheet are investigated.

An accurate characterization of the inaccessible subglacial environment is key to accurately modelling the dynamic behaviour of ice sheets and glaciers, crucial for predicting sea-level rise. The composition and water content of subglacial material can be inferred from measurements of shear wave velocity (Vs) and bulk electrical resistivity (R), themselves derived from Rayleigh wave dispersion curves and transient electromagnetic (TEM) soundings. Conventional Rayleigh wave and TEM inversions can suffer from poor resolution and non-uniqueness. In this thesis, I present a novel constrained inversion methodology which applies a Markov chain Monte Carlo implementation of Bayesian inversion to produce probability distributions of geophysical parameters. MuLTI (Multimodal Layered Transdimensional Inversion) is used to derive Vs from Rayleigh wave dispersion curves, and its TEM variant, MuLTI-TEM, for evaluating bulk electrical resistivity. The methodologies can include independent depth constraints, drawn from external data sources (e.g., boreholes or other geophysical data), which significantly improves the resolution compared to conventional unconstrained inversions. Compared to such inversions, synthetic studies suggested that MuLTI reduces the error between the true and best-fit models by a factor of 10, and reduces the vertically averaged spread of the Vs distribution twofold, based on the 95% credible intervals. MuLTI and MuLTI-TEM were applied to derive Vs and R profiles from seismic and TEM electromagnetic data acquired on the terminus of the Norwegian glacier Midtdalsbreen. Three subglacial material classifications were determined: sediment (Vs < 1600 m/s, 50 Ωm < R < 500 Ωm), permafrost (Vs > 1600 m/s, R > 500 Ωm) and weathered/fractured bedrock containing saline water (Vs > 1900 m/s, R < 50 Ωm). These algorithms offer a step-change in our ability to resolve and quantify the uncertainties in subsurface inversions, and show promise for constraining the properties of subglacial aquifers beneath Antarctic ice masses. MuLTI and MuLTITEM have both been made publicly available via GitHub to motivate users, in the cryosphere and other environmental settings, for continued advancement.

The high-mountain cryosphere forms water towers that are important for ecosystem services provision, supplying large populations living in mountains and the surrounding lowlands and producing potable water resources, and water for agriculture, industry and hydropower generation. However, continued glacier recession and mass loss is projected throughout the twenty-first century, and this raises major concerns regarding the future sustainability of cryospheric water resources. While glacier meltwater represents an essential drought-resilient freshwater resource in vulnerable drought-prone regions, little research has focused on the contribution made by runoff from rock glaciers. These are located widely throughout the high-mountain cryosphere and estimates of rock glacier water volume equivalent (WVEQ) vs glaciers suggests that the former may constitute increasingly important long-term water stores. Owing to the insulating effects of thick supraglacial debris cover, rock glaciers are climatically more resilient than glaciers; therefore, their relative importance versus glaciers may increase under future climate warming. Yet, while the hydrological role of debris-free glaciers and debris-covered glaciers has been the subject of much research, that of rock glaciers has received comparatively little attention. Given the need for strong climate adaptation in many of the world’s mountain regions, it is clear that a more comprehensive understanding of all components of the hydrological cycle in the high-mountain cryosphere is required. In this thesis, I develop the scientific understanding of rock glacier significance in deglacierizing mountains across a range of spatial scales (local, national, regional and global), with a specific focus on High Mountain Asia (HMA). The review chapter critically assesses the state of current scientific knowledge regarding the hydrological role of rock glaciers in high mountain systems and serves to form the context for the empirical chapters. The thesis has three key themes to which the empirical chapters are aligned: (1) the distribution and hydrological significance of rock glaciers at global scales, (2) the distribution and hydrological significance of rock glaciers at regional and national spatial scales (Himalaya and Nepalese Himalaya), and (3) advancing rock glacier evolutionary theory. (1) the thesis created a meta-analysis of existing systematic rock glacier inventories and compiled the first near-global rock glacier database (RGDB). The RGDB presented here includes >73,000 rock glaciers (intact = ~39,500, relict = ~33,500), which contain a WVEQ of 83.7 ± 16.7 Gt [~69–102 trillion litres]. Furthermore, the estimated ratio of rock glacier: glacier WVEQ is 1:456 globally. (2) the results of the meta-analysis described in (1) show that only ~9% of studies included in the RGDB cover the Hindu Kush Himalaya (HKH); therefore, I produced the first systematic rock glacier inventory for the (i) Nepalese Himalaya (national-scale), and (ii) Himalaya (regional-scale). In the former (i) I inventoried >6,000 rock glaciers, and these are estimated to contain a WVEQ of 20.90 ± 4.18 km3 (19.16 ± 3.83 Gt). For the Nepalese Himalaya estimated rock glacier: glacier WVEQ ratio is 1:9. In the latter (ii) ~25,000 rock glaciers have been inventoried. The total WVEQ is 51.80 ± 10.36 km3 (47.48 ± 9.50 Gt) with an estimated rock glacier: glacier WVEQ ratio of 1:24. The results of Theme 1 and 2 indicate that rock glaciers form considerable long-term water stores, which may become increasingly important as climatically-driven glacier recession and mass loss continues throughout the twenty-first century and beyond. (3) in order to understand debris-free glacier transition to rock glaciers I use in situ sedimentological data and kite aerial photography (KAP) data and develop a conceptual hypothesis to explain the key drivers of this process. The thesis suggests that sediment connectivity (i.e. the strength of the link between sediment sources and downslope landforms) is one such driver of these transition processes. As a consequence, I hypothesise that the presence of well-developed lateral moraines along glacier margins serves to reduce this connectivity, and thus reduce the likelihood of glacier-to-rock glacier transition occurring. The corelationships between rock glaciers and glacial, periglacial and paraglacial processes are also evaluated in the context of rock glacier origin and the changing influence these processes have upon rock glacier evolution through their lifecycle. Collectively, this research has shaped the understanding of the current and potential future role of rock glaciers in mountain hydrology and is the first to comprehend the distribution and hydrological significance of rock glaciers globally and in the Himalaya.

Organic matter is an important component of global biogeochemical cycles, which are sensitive to processes of landscape change. The worldwide retreat of mountain glaciers is one of the most profound examples of such change. Chilean Patagonia represents a unique study region, which is both relatively untouched by direct human activity and undergoing significant rates of deglaciation. The biogeochemical role of glaciers in this region is relatively understudied and this thesis offers the first regional scale assessment of the sensitivity of organic matter in rivers and fjords to glacier retreat. This assessment comprises three complementary studies. The first adopts a space-for-time approach, in which the effects of spatial variations in landscape properties on organic matter flux and composition are used to infer the effects of ongoing glacier retreat over time. It was found that glaciers drive higher catchment yields for particulate organic matter and supply dissolved organic matter rich in protein-like and aliphatic substances. However, the influence of glaciers is dampened by non-glacial sources and sinks in the downstream landscape. The second study used incubation experiments to quantify the proportion of bioavailable dissolved organic carbon in river water from selected catchments. These incubations showed that bioavailability in these catchments is lower than that observed in glacial rivers of the northern hemisphere. This may be explained by prior processing and specific intrinsic compositional factors but is not systematically related to catchment glacier cover. The third study presents the first molecular level analysis of dissolved organic matter in a glaciated fjord. This analysis showed that riverine organic matter may comprise a significant proportion of the total organic carbon pool, which is sensitive to changes in glacier melt inputs. The relationships between organic matter cycling and landscape change in Chilean Patagonia established in this thesis highlight some potential implications of deglaciation for global biogeochemistry.

The Greenland Ice Sheet (GrIS) has become the largest single mass contributor to global sea level. Since the mid-1990s the mass loss has been accelerating. The current consensus on the main drivers behind the shift in Greenland climate dynamics are: (1) A state change of the North Atlantic Oscillation towards an anticyclonic phase, (2) an above average temperature increase due to Arctic amplification and (3) an increase in absorbed shortwave radiation due to the melt-albedo feedback. However, thus far very few studies attempted to partition the GrIS surface energy budget (SEB) to find the main energy source behind the observed increase in surface melt. Furthermore, little is known about how much clouds have and will contribute to the recent and projected future GrIS melt increase. In this thesis we find, that downwelling solar radiation has been the main driver of the recent changes in the SEB over the GrIS. Furthermore, we establish a connection between the observed increase in high-pressure frequency, an unreported reduction in summer cloud cover and the recent mass loss acceleration over the GrIS. Additionally, we also show that cloud microphysics are the leading source of uncertainties in future projections of GrIS melt. Our analysis suggests that differences in cloud water phase distribution can add similar or larger uncertainties to projected Greenland melt than the choice of greenhouse gas emission scenario. Because our analysis identifies a great sensitivity of the GrIS to cloud conditions, we mark three key areas to increase our ability to model the Greenland sea level rise contribution: (1) Increasing the amount of cloud (microphysical) observations in the Arctic. (2) Improving the representation of clouds in (regional) climate models. (3) Considering the dynamic nature of the Greenland cloud radiative effect and its transient response in a warming climate.

Glacitectonics are found in all known glacial landscapes, but complex relationships exist at many scales between structures, landforms and processes. The deformation mechanisms forming glacitectonic complexes mainly occur at oscillating dynamic glacier margins. Understanding these processes offers insight of ice sheet dynamics through the influence of advance/retreat phases on sediment supply and deformation. However, understanding of the factors controlling sedimentation and deformation at former ice sheet margins is limited. A unique high-resolution seismic dataset, acquired for a large-scale windfarm development project is used to address the lack of understanding of the evolution of the Dogger Bank, the dynamics of the British-Irish Ice Sheet (BIIS) in the southern North Sea and potential interactions with the Fennoscandian Ice Sheet (FIS). This seismic dataset is used to characterise, describe and interpret detailed evidence of glacitectonic deformation in the Dogger Bank region. Traditional 2D interpretation of the high-resolution seismic data is compiled using a new 'domain' approach to offer insight into regional-scale glacitectonics. This data is integrated through a dense seismic survey network to generate a palaeo- land- surface map and a 3D model of deformation kinematics for the Dogger Bank region. This study identi fies a large moraine complex with wide-spread deformation up to a depth of approximately 40m below the present day sea-level, representing a LateWeichselian steady-state, oscillating ice sheet, with an active ice retreat from the north (FIS). This steady-state phase is followed by a signifi cant readvance event by the BIIS from the West, where deep deformation (up to 200m below the present-day sea level) at the western side of the Dogger Bank reflects a phase of a unstable ice advance.

Mountain development is dynamic and spans significant temporal (103 to 106 years) and spatial (single valley to mountain range) scales. Understanding and quantifying the fundamental processes which govern the growth and denudation of mountains is important as these provide context for and influence rates of modern climate change. This thesis investigates the glaciological drivers of mountain denudation with specific focus on the mountains of Wicklow, Ireland and the Pyrenees, France/Spain, and the coupling of glacial and hillslope erosion over the last glacial cycle. These mountain ranges represent end-members on a glaciological spectrum between (i) ice cap and (ii) ice field and alpine style glaciation. Chapter 3 introduces and tests novel geochronological techniques to permit widespread dating of glacial and hillslope deposits and to investigate sub-aerial weathering rates; a proxy for baseline erosional efficiency. In Chapter 4, these techniques are applied to glacial landforms, with results analysed in the context of recalibrated surface exposure ages (10Be, 36Cl). These data reveal complex patterns of ice advance/recession, with variable ice residence times and glacier configurations; complexity which places a first-order control on the magnitude and spatial focusing of erosion during glacial cycles. Chapter 5 presents a database of rock slope failure deposits (RSFs) to investigate the timing and significance of post-glacial slope instability. These data indicate that high-frequency, low magnitude RSFs (103 - 105 m3) can contribute significantly to mountain denudation, with a pulse in paraglacial erosion after deglaciation (127 - 148 mm ka-1), while the frequency and magnitude of interglacial RSFs is closely coupled to climatic drivers. In Chapter 6, cirque and geomorphometric analyses indicate that glacier configurations and the relative duration of maxima and marginal glacial conditions control mountain range morphology through the spatial focusing of erosion, with contrasting rates of valley and cirque incision between the studied regions. Chapter 7 assesses the long-term signature of glacial erosion, incorporating a range of geochronological, geomorphological and geomorphometric analyses. In turn, these data show that while glacial erosion has profoundly shaped the Wicklow Mountains, the morphological significance of glacial erosion in the Pyrenees wanes at large spatial scales and at low elevations; a trend which reflects the size, longevity and erosional efficiency of Quaternary glaciers. These results address the fundamental processes which control the efficiency, timing and focusing of mountain range denudation.

Recent work has highlighted the importance of the so-called “weathering crust” as a microbially and hydrologically active layer on glacier surfaces. However, this layer is yet to undergo investigation, with no estimates of water, microbial or nutrient fluxes through it to downstream freshwater and marine ecosystems. The mechanics of the weathering crust, and its role in transport and/or retention of particulate impurities at the glacier surface presents a research imperative. To investigate the eco-hydrology of glacier surfaces, this thesis presents a dataset collected at eleven sites in the Northern Hemisphere from the Canadian Arctic to the European Alps, collected between 2014 and 2016. To interrogate this dataset, the study develops and tests a novel logging piezometer which is used to calculate mean weathering crust hydraulic conductivity at all locations of 0.184 m d-1, equivalent to a sandstone, and meltwater velocities of 10-1 m d-1. This hydrologically poor aquifer, causes the storage of water at the surface for tens of day, providing an ideal medium for biogeochemical cycling. For microbial cell enumeration, a flow cytometry protocol is presented which is suitable for glacial environments providing accurate, reliable cell counts. Across the eleven sites, mean microbial cell concentration in weathering crust meltwater was revealed to be ≈ 104 cells mL-1 . It was unclear what controls exist upon cell concentrations in the weathering crust, however no links between weathering crust hydraulic conductivity, electrical conductivity or water temperature and cell concentrations were observed. Cellular particulate organic carbon flux (POC) form this active environment contributes a minimum of 1.1 Tg of cellular carbon per year to downstream freshwater and marine environments per year.

The aim of this thesis is to develop and test the luminescence dating of cobblesfrom glaciofluvial sediments. In recent years luminescence dating has increasingly been applied to date glaciofluvial sediments, but uncertainties about the degree of bleaching of the luminescence signal make dating challenging. Sub-surface luminescence measurements from cobbles from well-bleached environments are able to confirm that the cobbles were well-bleached at deposition. Having this confirmation when studying heterogeneously bleached environments would be a significant advantage. Bleaching experiments are undertaken to assess if numerical models of bleaching with depth are correct in nature. Measurements confirm that as the length of exposure increases the luminescence signal is reset to greater depths. Cobbles obtained from Orrisdale Head, Isle of Man, show significant sub-surface bleaching, with rock slices to depths of 12 mm into the cobble sub-surface having been completely bleached at deposition. Fading-corrected IRSL50 ages (20.7 ± 1.3 ka) agree with independent age control at the site. One cobble also appears to show both the advance (26.2 ± 0.3 ka) and retreat of the Irish Sea Ice Stream. A major advantage of applying luminescence dating to cobbles instead of sand-sized grains is that at depths of > 2 mm into the cobble 92 % of the dose rate comes from the cobble itself and this makes luminescence ages insensitive to water content. In further tests of this approach, cobbles from two locations in north Wales show limited sub-surface bleaching, however the IRSL50 ages from the surface slices agree with independent age control. Following the trial at locations with independent age control, cobbles are obtained from a deposit at Bridgwalton which marks the furthest extent of a separate ice lobe which occupied the Cheshire-Shropshire basin. The IRSL50 age (25.3 ± 1.6 ka) gives the first depositional age for this location and shows that the Last Glacial Maximum (LGM) at this site is synchronous with that observed for the IrishSea ice stream at the Scilly Isles. Luminescence dating of cobbles has an enormous potential in providing accurate and robust ages for glaciofluvial sediments that are challenging to date.

The majority of current ice core studies focus on analysing the inorganic component of atmospheric aerosol, trapped and preserved in the ice as a record of past atmosphere. However, this does not fully represent the make-up of atmospheric aerosol, which can be up to 50% organic. This thesis aims to develop the understanding and quantification of a number of these organic compounds in ice core samples. A novel and promising area of 'organics in ice' research lies within the groups of primary and secondary compounds released from the terrestrial and marine biospheres; these compounds may help us to form a record of past biosphere emissions, with implications for biological productivity and atmospheric chemistry. A small selection of studies obtaining new records from these types of organic compounds in ice have demonstrated this concept, for example lipid compounds in snow layers dating back 450 years in Greenland, oxidation products of isoprene and monoterpenes in ice up to 350 years old in Alaska, and carboxylic acids and inorganic ions between 1942-1993 from Grenzgletscher (Monte Rosa Massif) in the southern Swiss Alps. Compound concentrations were related back to Northern Hemisphere temperature, atmospheric transport pathways and intensities, and biomass burning signals respectively. There are many terrestrial and marine biogenic compounds not yet investigated in ice core samples. Thus we are presented with an almost untapped reservoir of new climate information. Therefore, it is timely to produce a method of analysis for a long list of the most promising of these compounds (herein 'target compounds'), namely fatty acids and secondary oxidation aerosol of terpenes (SOA), allowing quantification of these novel analytes in ice core samples to investigate the concept further. This project begins with an investigation in to the possible contamination sources of the target compounds in ice core samples. It attempts to quantify the threat of contamination throughout the drilling, storage and analyses processes. It finds there is substantial presence of organic compounds in media used during ice core processing, but the risk to target compounds is minor where clean-protocols are followed, and the threat is limited to outer surface ice of a solid ice core. Recommendations for ice core processing steps in preparation for organics analyses are outlined based on these results. A method of high performance liquid chromatography-mass spectrometry (HPLC-MS) including rotary evaporation preconcentration of samples is then optimised for detection of target compounds. The final method achieves good average recoveries of 80%, and reproducibility of 9% RSD. The method is reproducible on different instruments, based on an interlab comparison. An extension of this method, direct injection HPLC-MS analysis (where sample preconcentration is eliminated) is tested for the benefits of reducing sample volume and contamination introduced by preconcentration steps. The method is successful for SOA compounds (7% RSD) but not for fatty acids where background contamination is very high. The method of preconcentration HPLC-MS is then applied to samples from two ice cores; the marine-aerosol dominated Bouvet Island (sub-Antarctic) ice core, and the terrestrial-aerosol dominated Belukha Glacier (Russian Altai Mountains) ice core. Novel organic compounds are detected in both cores. Compound concentration time-series are investigated statistically by comparison to pre-existing inorganic compound records from the same cores and to historical climate data, and by application of back trajectory modelling using the Met Office's Numerical Atmospheric-dispersion Modelling Environment (NAME). In the case of Bouvet Island, the fatty acid oleic acid is found to have statistically significant correlations with the sea-ice marker methanesulfonic acid, and indeed with sea ice concentration during July to September in a geographical region extending westwards from the island along the maximum sea ice extent margin. The mechanism behind this correlation is suggested to be that of algal blooming during spring months, releasing oleic acid which is transported by strong westerly winds to the island. This highlights the prospects for a suite of marine biogenics in ice cores being used as sea ice markers. In the Belukha core, where sample records are available at sub-annual resolution, a suite of SOA compounds display summer-time peaks in concentrations. This seasonal variability is shown to be related to emission signals of these organic compounds, which is an exciting prospect for future work in improving our understanding of budgets of these aerosols in the atmosphere. The records detected demonstrate great promise for the use of organic compounds as environmental markers.

Accelerated mass loss from the Greenland Ice Sheet in recent decades has been concentrated around the coastal margins, where glacier ice fronts and the undersides of floating ice are in contact with warm ocean waters. The interaction of the ice sheet with a warming ocean leads to thinning of tidewater glaciers, which has been linked to increased glacier velocity, calving, retreat and subsequent mass loss. Our understanding of these ice-ocean interactions is limited, particularly regarding submarine melting of tidewater glacier ice fronts, which has been proposed as an initial trigger for glacier retreat and acceleration. Understanding the mechanisms promoting changes to tidewater glacier dynamics is critical, as they are currently absent from ice sheet models and present a large source of uncertainty in 21st century sea level rise predictions. This thesis develops two novel remote sensing techniques, to investigate both submarine melt rates under tidewater glacier floating ice tongues and iceberg freshwater fluxes from submarine melting, providing improved datasets and process understanding that can be used to constrain changes to the Greenland Ice Sheet in a warming world. The first result of this thesis is the development of a methodology to estimate submarine melting under floating ice tongues using satellite imagery. Submarine melt rates were derived by differencing along-flow ice tongue surface elevation, in combination with ice tongue velocity and changes in surface mass balance. Kangiata Nunaata Sermia (KNS), a large tidewater glacier in southwest Greenland, was used as a proof-of-concept study site. Mean submarine melt rates under the seasonal ice tongue at KNS in spring 2013 reach over 0.8 ± 0.3 m d−1, decreasing with distance down-fjord from the glacier grounding line and varying across-fjord. These variations in melt rate likely result from changes in ice tongue draft and fjord water temperature with depth, but may also reflect the strength of subglacial discharge plumes exiting beneath the glacier grounding line. Expanding upon these initial melt rate results, the thesis next explores the spatial and temporal variations in submarine melting at KNS from 2012 to 2014. Using the same methodology, derived melt rates vary significantly near the glacier grounding line, both spatially and temporally, with mean melt rates of 1.3 ± 0.6, 0.8 ± 0.3, and 1.0 ± 0.4 m d−1 across the ice tongue in 2012, 2013 and 2014, respectively. Areas of higher submarine melting correspond spatially with locations of subglacial discharge plumes and ice front calving activity observed during the melt season using time-lapse camera imagery, as well as to locations of modelled subglacial flow paths with large upstream catchment areas. These results suggest a dynamic subglacial hydrology system capable of rapidly re-routing subglacial discharge to different terminus outlets both within and between melt seasons. Furthermore, they provide an empirically-derived link between the presence of subglacial discharge plumes and areas of high spring submarine melting and calving along glacier termini. Moving to east Greenland, the final results chapter turns to Sermilik Fjord, where a new methodology for estimating freshwater fluxes from icebergs is developed. The amount, timing and location of meltwater produced via submarine melting of icebergs can significantly impact local fjord water circulation and heat budget, which has implications for glacier calving, retreat and acceleration in addition to nutrient fluxes and primary productivity. Previous methods for estimating iceberg meltwater fluxes have either been complex models, themselves reliant on limited field data and poorly constrained model parameters, or user-intensive, expensive remote sensing techniques. This thesis presents a simplified approach for deriving summer and autumn iceberg freshwater fluxes in 2017, using freely available Sentinel-2 satellite imagery to estimate iceberg velocity and seasonal changes in iceberg volume with distance down-fjord. Integrated 2-month full-fjord freshwater fluxes reach 1270 ± 650, 1200 ± 610, 3410 ± 1740, and 1150 ± 590 m3 s−1 for June-July, July-August, August-September and September-October, respectively. The estimated iceberg freshwater fluxes are highest across August and September, likely due to a combination of warming fjord water temperatures at depth as autumn approaches and increased calving into the fjord system in the months prior, and decrease with distance from the head of the fjord. The proportion of solid ice exiting the fjord averages 14% of the total ice volume calved into the fjord from Helheim Glacier, the largest tidewater glacier feeding into Sermilik Fjord, confirming that a significant volume of freshwater is released at depth along the length of the fjord. The volume of freshwater generated from iceberg melt is comparable to or greater than subglacial discharge volume throughout the year, and has important implications for fjord-scale circulation, submarine melt rates and primary productivity. In terms of future developments, this thesis presents two new methodologies focused on ice-ocean interactions: one to estimate submarine melt rates near tidewater glaciers and one to derive iceberg freshwater fluxes entering glacial fjords. These innovative methods have generated important new datasets in a challenging environment from which it has consistently proven difficult to derive estimates of submarine melt rates. The additional utility of these methods is their potential application to other fjord systems for constraining both fjord-scale and ice-sheet wide ice-ocean models, developments that are critical for understanding the sensitivity of the Greenland Ice Sheet and surrounding ocean basins to future climate change.

In this thesis, I develop and demonstrate a system for monitoring fluctuations in the speed of Greenland ice sheet outlet glaciers with high temporal frequency from imagery acquired by a range of satellite missions. This work is motivated by an ambition to utilise a new era of operational satellites to better understand how environmental changes are affecting the flow and mass of Greenland's outlet glaciers. First, I exploited the systematic and frequent acquisition schedule of the Sentinel-1 satellite constellation to track weekly variations in the speed of four fast-flowing, marine-terminating glaciers - Jakobshavn Isbræ, Petermann Glacier, Zachariæ Isstrøm and Nioghalvfjerdsfjorden - between 2015-2017. By combining the Sentinel-1 data with an eight-year time-series derived from TerraSAR-X, I produced a decadal record of variations in glacier flow. On a technical level, I was able to demonstrate the value of Sentinel-1's 6-day revisit time for glaciology, because it leads to an increase in the degree of correlation between consecutive images and also to improved tracking of movement near to the glacier calving fronts. On a scientific level, I was able to demonstrate that a strong correlation exists between iceberg calving events and glacier speedup, and to show for the first time that Jakobshavn Isbræ has begun to slow down. Next, I assessed the capability of the Sentinel-1 constellation to detect and chart seasonal changes in the speed of five slow-flowing glaciers situated in a 14,000 km2 land-terminating sector of central-west Greenland. These new measurements offer significantly improved spatial and temporal resolution when compared to previous missions, in all seasons. I was able to show that there are marked differences in the degree of seasonal speedup of the five glaciers - with summertime increases in ice flow ranging from 21 to 49 % - reinforcing the need for comprehensive monitoring and the challenges of making regional extrapolations. Thanks to the high temporal frequency afforded by Sentinel-1, I was also able to document for the first time the detailed spatial pattern of speedup persistence, and to show that short- lived peaks of melting match transient spikes in glacier velocity. Finally, I explored the added value and complementarity of the Sentinel-2 multi- spectral instrument (MSI) for tracking ice motion. I was able to combine measurements acquired by Sentinel-1 and Sentinel-2 to detect short-term changes in iceberg drift, iceberg calving, ice motion, and supraglacial lake area at Jakobshavn Isbræ. I also showed that measurements of glacier flow determined from both satellites are in good agreement, and that the spatial coverage they afford is greatest in opposing seasons, illustrating the promise of Sentinel-2 for glaciology.

Glacial runoff often contains nitrate far in excess of nitrogen (N) concentrations in snow. ‘Excess’ nitrate is likely produced by nitrifying bacteria in subglacial environments, yet the source of the additional N has not been identified. In this thesis, rocks are investigated as an N source to subglacial microbial ecosystems following their comminution by glacial erosion and leaching with meltwater. Weathering processes involving the release of N and micronutrients, iron (Fe) and manganese (Mn), from rock to glacial environments were evaluated using a combination of field studies in Adventdalen (Svalbard) and rock weathering experiments. Attempts were also made to constrain isotope fractionation during weathering to improve the use of stable isotopes to trace N provenance. The sedimentary geological succession underlying glaciers in the study area contained significant N (up to 0.21 wt. %), particularly in organic rich, siltstones and shales. Fe and Mn were largely found within oxide and carbonate minerals while significant pyrite bound Fe (max. 5.2 wt. %) was present in organic rich sediments. In simulated glacial weathering experiments, nitrogen was almost entirely liberated as ammonium with much of it absorbed to mineral surfaces rather than dissolved in solution. Pyrite oxidation liberated Fe and generated acidity which drove i) silicate dissolution releasing N and ii) carbonate dissolution yielding dissolved Mn. The largest N yields were associated with rocks containing abundant organic N which may have been liberated via oxidative weathering, protonation of amines and/or organic matter degradation by free radicals produced during crushing. Liberated ammonium was partly lost as ammonia gas when the solution pH exceeded 8 and also absorbed to sediment, particularly when clay minerals were abundant. Further experiments demonstrated an isotope enrichment in adsorbed ammonium of up to 10‰ δ15N relative to bedrock N, caused by fractionation from a combination of adsorption and volatilisation processes. In Svalbard, up to 89% of nitrate in glacial runoff was derived from non-snowpack sources. Solute was acquired into snowmelt via chemical weathering of rock talus in ice marginal moraines largely via coupled sulphide oxidation and silicate dissolution. When oxygen was available, ammonium released from rock was converted to nitrate via microbial nitrification. This assertion is supported by nitrate δ18O values of close to 0‰, vastly different from atmospheric oxygen. However, when flowpaths were anoxic, nitrate was absent likely due to microbial denitrification. Geological variations may control the magnitude of nutrient export by weathering in glacial catchments. For instance, N and Mn were particularly enriched in water from the Longyearbreen catchment which overlies shales enriched in these elements. Furthermore, clay minerals in bedrock significantly influences the distribution of N (as ammonium) between dissolved and particulate phases through ion exchange reactions.

In this thesis I develop novel methods and datasets, based on the processing of CryoSat-2 satellite radar altimeter data, to improve the understanding of retrieving measurements of surface elevation and elevation change over the Antarctic and Greenland ice sheets. First, I used 6 years of CryoSat-2 altimetry to create a model of the surface height of the Antarctic Ice Sheet and ice shelves. Posted at a resolution of 1 km, 94 % of the grounded ice sheet and 98 % of the floating ice shelves are observed, and the remaining grid cells North of 88 ° S are interpolated using ordinary kriging. Taking into account slope-dependent errors and the distribution of slopes across the ice sheet, I estimated the average accuracy of the DEM to be 9.5 m - a value that is comparable to, or better than that of other models derived from satellite radar and laser altimetry. Next, I developed a new technique to retrieve estimates of the depth distribution of radar backscatter from CryoSat-2 altimeter waveforms using a backscatter model. I then applied this model to chart spatial and temporal variatibility in radar backscatter and, for the first time, explicitly estimate radar penetration depth across the interior of the Greenland Ice Sheet. I then used this information to correct for artefacts in elevation trends derived from Cryosat-2 pulse-limited altimetry resulting from an episodic melt event which reset the radar scattering horizon. Incorporating the penetration depth into the surface height retrieval, I find improved agreement when compared to independent airborne laser altimeter data recorded over the same time period. Finally, I used CryoSat-2 altimetry to estimate seasonal elevation changes in the Greenland Ice Sheet. Using regional climate model simulations of height fluctuations due to surface process alone, I demonstrate that CryoSat-2 observations track elevation changes driven by melting and snowfall accumulation in the ice sheet ablation zone. I then mapped spatial and temporal variations in seasonal elevation change, demonstrating the ability of CryoSat-2 to monitor changes in Greenland which arise due to its meteorology.

Twelve published papers are submitted for this D.Sc. thesis. The papers were published in 1995 to 2009 when Roger Braithwaite was employed by the University of Manchester. The submitted papers relate to the general area of 'glaciers and sea level' and there is a strong theme that glacier conditions differ between regions, hence the title 'geography of glacier mass balance'. Paper no. 1 analyses degree-day factors from Greenland whereby glacier melting can be calculated from temperature data. Paper no. 2 discusses the effect of glacier mass balance changes on global sea level rise, including a simple theoretical treatment and estimation of the temperature-sensitivity of global sea level using the degree-day model. The latter depends on the combined geographies of glacier area and temperature-sensitivity of glacier mass balance. Papers 3, 4 and 5 analyse observed glacier mass balance from all over the world, and papers 6, 7 and 8 discuss the modelling of mass balance with the degree-day model. A trend of increasingly negative observed mass balance is only detectable towards the end of the 1995-2009 period. Papers 9, 10 and 11 discuss the climate at the glacier equilibrium line altitude (ELA) and propose a simplified approach to calibration of the degree-day model. Paper 12 returns to the topic of global sea level change with improved insights from papers 9, 10 and 11. Observed mass balance variability (papers 3, 4 & 5) and modelled temperature-sensitivity of mass balance (papers 6, 7 & 8) both show contrasts between maritime and continental climates with lower variability and sensitivity at high latitudes and in continental interiors, and higher variability and sensitivity at lower latitudes. The underlying global geography of mass balance variability and temperature-sensitivity is therefore controlled by the geography of annual precipitation.

Studies of past changes in the environment are essential to understand the speed and scale of modern-day climatic and environmental change. This research investigates the palaeoenvironmental changes throughout the last glacial and current interglacial period in Galway Bay, Western Ireland, using a multiproxy examination of geophysical data and twenty-two sediment cores. The methodologies used in this study include the interpretation of multibeam echo-sounder bathymetry and backscatter, altimetry, hydrological and seismic data, as well as the analysis and interpretation of the sediment cores' physical properties, X-radiographs, grain size analysis, microfossil content, geochemical signatures (micro-XRF) and radiocarbon dates. These data together allow for an interpretation of the driving forces behind the observed patterns in the distribution of sediments at the seafloor and sub-surface from the last glacial maximum until present day. Such data also provide field data, which can be used to constrain sea level in the bay after 19 ka cal BP. This is particularly valuable as the only available sea-level reconstructions for the area are derived from glacio-isostatic adjustment (GIA) models. Seismic data cover the mid and outer bay and reveal three distinct seismic units, representing bedrock, glacial till and post-glacial marine sediments. The till unit, overlying the bedrock, shows evidence of a glaciofluvially cut channel along the northern portion of the bay, constraining the relative sea-level depth in the bay during deglaciation to 58m below modern-day levels. This lowstand is in broad agreement with a GIA predicted sea-level position at 15 ka cal BP. The overlying marine sediment unit, in which prograding clinoforms and infilling are visible, represents the sea-level transgression since 15 ka cal BP. Shoreline deposits (rounded pebbles in a silty sand matrix, found in four cores) were deposited in the mid and outer Bay between 12.6 ka cal BP and 8.8 ka cal BP. The identification of such submerged palaeo-shore line sediments in depths of 39-43m reveals a late glacial to early Holocene slowstand. This identified slowstand lasted 2500 years longer than predicted by GIA based predictions. Core data suggests that, due to this delayed subsequent transgression, GIA derived RSL level predictions were not reached until ~4000 years BP. As well as providing important information for RSL reconstructions, the sediments and foraminifera within allow the identification of palaeoenvironmental conditions over time and space. In general, the pre- and early Holocene period is characterised by high sediment input in warm, productive, estuarine conditions with 2 separate inlets in the bay leading to two distinct environments: (1) a high energy beach environment transitioning to low energy, transitional waters in the north, and (2) a low energy, anoxic subtidal environment in the south. The mid Holocene period is characterised by the opening up of the bay and the presence of the gastropod Turritella sp. across the entire area. This represents a medium energy, warm, productive, environment in transitional waters. The late Holocene is typified by the disappearance of Turritella sp. and a transition to marine conditions. This represents a deepening, productive and high energy environment with warm and wet climatic conditions in the region. Interestingly though, global and regional climatic events such as the Younger Dryas, the 8.2 ka and 4.2 ka event have left no significant imprint on the sedimentary record. Overall, the data acquired from Galway Bay documents constant relative sea-level transgression, with a trend towards a wetter and marine- influenced environment after the initial deglaciation. The more recent Late Holocene sediments and current seabed topography record the influence of tidal currents, storm events and wave action, characteristic of the western Irish coastline. Results from this research can help validate both climate and sea-level models for the region, which in turn provide predictions for future environmental developments.

Supraglacial lakes in the ablation zone of the Greenland Ice Sheet (GrIS) often drain rapidly (in hours to days) by hydraulically-driven fracture (“hydrofracture”) in the summer. Hydrofracture can deliver large meltwater volumes to the ice-bed interface and open-up surface-to-bed connections, thereby routing surface meltwater to the subglacial system, altering basal water pressures and, consequently, the velocity profile of the GrIS. The study of rapidly draining lakes is thus important for developing coupled hydrology and ice-dynamics models, which can help predict the GrIS’s future mass balance. Remote sensing is commonly used to identify the location, timing and magnitude of rapid lake-drainage events for different regions of the GrIS and, with the increased availability of high-quality satellite data, may be able to offer additional insights into the GrIS’s surface hydrology. This study uses new remote-sensing datasets and develops novel analytical techniques to produce improved knowledge of rapidly draining lake behaviour in west Greenland over recent years. While many studies use 250 m MODerate-resolution Imaging Spectroradiometer (MODIS) imagery to monitor intra- and inter-annual changes to lakes on the GrIS, no existing research with MODIS calculates changes to individual and total lake volume using a physically-based method. The first aim of this research is to overcome this shortfall by developing a fully-automated lake area and volume tracking method (“the FAST algorithm”). For this, various methods for automatically calculating lake areas and volumes with MODIS are tested, and the best techniques are incorporated into the FAST algorithm. The FAST algorithm is applied to the land-terminating Paakitsoq and marine-terminating Store Glacier regions of west Greenland to investigate the incidence of rapid lake drainage in summer 2014. The validation and application of the FAST algorithm show that lake areas and volumes (using a physically-based method) can be calculated accurately using MODIS, that the new algorithm can identify rapidly draining lakes reliably, and that it therefore has the potential to be used widely across the GrIS to generate novel insights into rapidly draining lakes. The controls on rapid lake drainage remain unclear, making it difficult to incorporate lake drainage into models of GrIS hydrology. The second aspect of this study therefore investigates whether various hydrological, morphological, glaciological and surface-mass-balance controls can explain the incidence of rapid lake drainage on the GrIS. These potential controlling factors are examined within an Exploratory Data Analysis statistical technique to elicit statistical similarities and differences between the rapidly and non-rapidly draining lake types. The results show that the lake types are statistically indistinguishable for almost all factors, except lake area. It is impossible, therefore, to elicit an empirically-supported, deterministic method for predicting hydrofracture in models of GrIS hydrology. A frequent problem in remote sensing is the need to trade-off high spatial resolution for low temporal resolution, or vice versa. The final element of this thesis overcomes this problem in the context of monitoring lakes on the GrIS by adapting the FAST algorithm (to become “the FASTER algorithm”) to use with a combined Landsat 8 and Sentinel-2 satellite dataset. The FASTER algorithm is applied to a large, predominantly land-terminating region of west Greenland in summers 2016 and 2017 to track changes to lakes, identify rapidly draining lakes, and ascertain the extra quantity of information that can be generated by using the two satellites simultaneously rather than individually. The FASTER algorithm can monitor changes to lakes at both high spatial (10 to 30 m) and temporal (~3 days) resolution, overcoming the limitation of low spatial or temporal resolution associated with previous remote sensing of lakes on the GrIS. The combined dataset identifies many additional rapid lake-drainage events than would be possible with Landsat 8 or Sentinel-2 alone, due to their low temporal resolutions, or with MODIS, due to its inferior spatial resolution.

Increasing surface runoff from the Greenland Ice Sheet due to a warming climate not only accelerates ice mass loss by altering surface mass balance, but may also lead to increased dynamic losses. This is because surface melt draining to the bed can reduce ice-bed coupling, leading to faster ice flow. Understanding the impact of surface melt on ice dynamics is important for constraining the contribution of the Greenland Ice Sheet to sea level rise. The aim of this thesis is to numerically model the influence of surface runoff on ice velocities. Three new models are presented: an updated supraglacial hydrology model incorporating moulin and crevasse drainage, along with lake drainage over the ice surface via channel incision; an ice sheet model implementing a numerically efficient formulation of ice flow; an adjoint code of the ice flow model based on automatic differentiation. Together with a subglacial hydrology model, these represent the key components of the ice sheet system. The supraglacial hydrology model is calibrated in the Paakitsoq region. Model output shows the partitioning of melt between different drainage pathways and the spatial distribution of surface drainage. Melt season intensity is found to be a relevant factor for both. A key challenge for simulations applying a coupled ice-flow/hydrology model is state and parameter initialization. This challenge is addressed by developing a new workflow for incorporating modelled subglacial water pressures into inversions of basal drag. A current subglacial hydrology model is run for a winter season, and the output is incorporated into the workflow to invert for basal drag at the start of summer in the Russell Glacier area. Comparison of the modelled subglacial system to observations suggests that model output is more in line with summer conditions than winter conditions. A multicomponent model integrating the main components of the ice sheet system is developed and applied to the Russell Glacier area. A coupled ice-flow/hydrology model is initialized using the proposed workflow, and driven using output from the supraglacial hydrology model. Three recent melt seasons are modelled. To a first order, predicted ice velocities match measured velocities at multiple GPS sites. This affirms the conceptual model that summer velocity patterns are driven by transitions between distributed and channelized subglacial hydrological systems.

Retreating glaciers and ice sheets provide a significant contribution to sea level rise, which will affect future populations and their activities. Accurate sea level projections are needed in order to best inform policy makers, but these projections are limited by our understanding of dynamical change at marine-terminating glaciers. Terrestrial time-lapse photography has proved to be a viable approach for obtaining high-detail observational records, and is used here to examine signals of dynamical change at two tidewater glaciers in Svalbard. Photogrammetric measurements were extracted using PyTrx (`Python Tracking'), a new photogrammetry toolbox that has been developed here for deriving velocities (e.g. glacier surface velocity), surface areas (e.g. supraglacial lake area, surfacing plume area), and line distances (e.g. terminus profiles). PyTrx has been created as a Python-alternative photogrammetry software, and offers additional functionality to the typical monoscopic feature-tracking toolboxes that are currently available. Subglacial hydrology and its relation to basal sliding were examined at Kronebreen, Svalbard. The results revealed a difference in flow efficiency between the north and south regions of the glacier tongue, which influences spatial patterns in surface velocities. Long-term changes in ice flow were concluded to be controlled by the location of effcient and inefficient drainage, and the position of regions where water is stored and released. Changes in terminus conditions and calving processes were examined at Tunabreen, a surge-type tidewater glacier. Observations suggested that atmospheric forcing plays a larger role in terminus stability than previously considered, and it is likely that terminus dynamics at Tunabreen are the product of a unique interplay between oceanic and atmospheric forcing which are shaped by the glacier's surge-type nature. Additionally, calving activity at Tunabreen can be characterised as high-frequency, low-magnitude events, and a high proportion of its long-term calving activity can be attributed to the rate of under-cutting at the terminus. In all, these studies demonstrate that long-term changes in glacier dynamics are dictated by the small changes in basal and terminus conditions, and how they vary from year-to-year. Future research now needs to be directed towards understanding how small-scale processes vary over multiple melt seasons, in order to establish how they operate at longer timescales. PyTrx provides an appropriate basis to continue this work and expand the capabilities of the toolbox.

Ice streams are fast flowing outlet glaciers through which over 90% of the ice stored within the Antarctic Ice Sheet drains. The dynamic behaviour of ice streams is therefore crucial in controlling the mass balance of the ice sheet. Over the past few decades, Antarctica has been losing mass. Much of this mass loss has been focussed around coastal regions of the Antarctic Ice Sheet. Some of the most dramatic changes such as grounding-line retreat, acceleration and surface elevation change have been observed in Pine Island Glacier (PIG) and its neighbouring ice streams. This is of particular concern because these ice streams account for 10% of the discharge from the west Antarctic Ice Sheet and therefore have the potential to contribute significantly to global sea-level rise. One of the key challenges in accurately forecasting this future sea-level rise is improving understanding of processes occurring at the beds of ice streams. This requires detailed knowledge of the properties and dynamics of the bed. This thesis aims to address this knowledge gap by investigating the spatial and temporal characteristics of the bed of PIG using high-resolution geophysical data acquired in its trunk and tributaries and beneath the ice shelf. The thesis begins by analysing radar-derived high-resolution maps of subglacial topography. These data show a contrasting topography across the ice-bed interface. These diverse subglacial landscapes have an impact on ice flow through form drag, controlled by the size and orientation of bedrock undulations and protuberances. The next chapter provides a quantitative analysis of these landscapes using Fast Fourier analysis of subglacial roughness. This analysis investigates the roughness signature of subglacial bedforms and the how the orientation and wavelength of roughness elements determine their correlation with ice dynamic parameters. The slow-flowing inter-tributary site is found to have a distinct signature comparable to 'ribbed' patterns of modelled basal shear stress and transverse 'mega rib' bedforms. Roughness oriented parallel to ice flow with wavelengths approaching mean ice thickness are found to have the highest correlation with ice dynamic parameters. The temporal stability of PIG is analysed using repeat radar measurements. No significant change is observed over a period of 3-6 years with no evidence of rapid erosion or the evolution of subglacial bedforms as observed in previous repeat measurements of ice-stream beds elsewhere in Antarctica. This suggests that the widespread deforming till layer detected in extensive seismic reflection surveys is in steady state. Lastly, the thesis explores geomorphological evidence of twentieth-century grounding-line retreat beneath PIG Ice Shelf using high-resolution geophysical data acquired from autonomous underwater vehicle surveys. Evidence of erosion, deposition, meltwater flow and post-glacial modification is observed in fine detail. The observed distribution of sediment supported previous surveys indicating a geological transition coinciding with the ridge that acted as a former stable grounding-line location. Metre-scale resolution images of recently deglaciated ice stream beds were found to reveal bedforms that are not detectable with traditional offshore bathymetric surveys. Together these findings reveal the role of short wavelength topography as both an influence on, and product of fast ice stream flow. It also highlights the spatial diversity of subglacial environments and the need to focus future research on tying detailed observations of ice-stream beds with knowledge of basal properties over time.

Satellite altimetry has been traditionally used in the past few decades to measure elevation of land ice, quantify changes in ice topography and infer the mass balance of large and remote areas such as the Greenland and Antarctic ice sheets. Radar altimetry is particularly well suited to this task due to its all-weather year-round capability of observing the ice surface. However, monitoring of ice caps and ice fields - bodies of ice with areas typically smaller than ~ 10,000 km2 - has proven more challenging. The large footprint of a conventional radar altimeter and coarse ground track coverage are less suited to observing comparatively small regions with complex topography. Since 2010, the European Space Agency’s CryoSat-2 satellite has been collecting ice elevation measurements over ice caps and ice fields with its novel radar altimeter. CryoSat-2’s smaller inter-track spacing provides higher density of observations compared to previous satellite altimeters. Additionally, it generates more accurate measurements because (i) the footprint size is reduced in the along-track direction by means of synthetic aperture radar processing and (ii) interferometry allows to precisely locate the the across-track angle of arrival of a reflection from the surface. Furthermore, the interferometric capabilities of CryoSat-2 allow for the processing of the delayed surface reflections after the first echo. When applied over a sloping surface, this procedure generates a swath of elevations a few km wide compared to the conventional approach returning a single elevation. In this thesis, swath processing of CryoSat-2 interferometric data is exploited to generate topographic data over ice caps and ice fields. The dense elevation field is then used to compute maps of elevation change rates at sub-kilometer resolution with the aim of quantifying ice volume change and mass balance. A number of algorithms have been developed in this work, partly or entirely, to form a complete processing chain from generating the elevation field to calculating volume and mass change. These algorithms are discussed in detail before presenting the results obtained in two selected regions: Iceland and Patagonia. Over Icelandic ice caps, the high-resolution mapping reveals complex surface elevation changes, related to climate, ice dynamics and sub-glacial, geothermal and magmatic processes. The mass balance of each of the six largest ice caps (90% of Iceland’s permanent ice cover) is calculated independently for the first time using spaceborne radar altimetry data. Between October 2010 and September 2015 Icelandic ice caps have lost a total of 5.8± 0.7 Gt a ̄1, contributing 0.016± 0.002 mm a ̄1 to eustatic sea level rise. This estimate indicates that over this period the mass balance was 40% less negative than the preceding 15 years, a fact which partly reflects the anomalous positive balance year across the Vatnaj ̈okull ice cap (~ 70% of the glaciated area) in 2014/15. Furthermore, it is demonstrated how swath processing of CryoSat-2 interferometric data allows the monitoring of glaciological processes at the catchment scale. Comparison of the geodetic estimates of mass balance against those based on in situ data shows good agreement. The thesis then investigates surface elevation change on the Northern and Southern Patagonian Ice Fields to quantify their mass balance. This area is characterized by some of the fastest flowing glaciers in the world, displaying complex interactions with the proglacial environments (including marine fjords and freshwater lakes) they often drain into. Field observations are sparse due to the inaccessibility of these ice fields and even remotely sensed data are limited, often tied to comparisons to the topography in 2000 as measured by the Shuttle Radar Topography Mission. Despite gaps in the spatial coverage, in particular due to the complex topography, CryoSat-2 swath radar altimetry provides insight into the patterns of change on the ice fields in the most recent period (2011 to 2017) and allows to independently calculate the mass balance of glaciers or catchments as small as 300 km2. The northern part of the Southern Patagonian ice field displays the strongest losses due to a combination between ice dynamics and warming temperatures. In contrast Pio XI, the largest glacier on this ice field and in South America, is advancing and gaining mass. Between April 2011 and march 2017, the two ice fields combined have lost an average of 21.29± 1.98 Gt a ̄1 (equivalent to 0.059± 0.005 mm a ̄1 eustatic sea level rise), 24% and 42% more negative when compared to the periods 2000-2012/14 and 1975-2000. In particular the Northern Patagonian ice field, responsible for one third of the mass loss, is losing mass 70% faster compared to the first decade of the 21st century. These results confirm the overall strong mass loss of the Patagonian ice fields, second only to glaciers and ice caps in Alaska and the Canadian Arctic, and higher than High Mountain Asia, which all extend over areas ~ 5-8 times larger (excluding glaciers at the periphery of the Greenland and Antarctic ice sheets).

Ice-rafted detritus (IRD) across the North Atlantic provides an important archive for reconstructing the dynamics of adjacent ice margins during the Last Glacial Maximum (LGM). The complex relationship between ice sheet mass balance and IRD flux is still unclear; an increase in IRD could indicate both a positive or a negative mass balance. To address these uncertainties, the source of the IRD needs to be established. Determining the source of IRD for the British Irish Ice Sheet (BIIS) hitherto has only identified broad lithospheric provenances. This study aimed to link IRD within three adjacent deep ocean cores to individual ice streams draining the former BIIS using Xray Fluorescence (XRF) core scanning to define elemental geochemical provinces. Principal Component Analysis was used to establish five glacigenic ice stream end members from the former BIIS: 1) Irish Sea; 2) Celtic Sea; 3) Donegal; 4) Malin Sea and 5) Minch. These end members were then used to geochemically link the IRD in the deep ocean to ice stream source. All three deep ocean cores record BIIS IRD flux; contribution of IRD from the ice streams is governed by the presence of a marineterminating margin and the direction of the surface currents which are shown to be variable during deglaciation. Significant BIIS IRD contribution to the deep ocean occurs between 31-20 kyr BP. Analyses suggests a marine-terminating BIIS pre-Heinrich event 4 and provides offshore evidence for a smaller regionalised marine terminating margin ~14 kyr BP and during the Younger Dryas stadial. The data highlight shifts in the timing of ice stream vs. ice lobe contribution from adjacent sources (Hebrides Ice Stream vs. Donegal Bay ice lobe) and that IRD source in the Rosemary Bank has stronger affinities with the Hebrides Ice Stream than with the adjacent Minch Ice Stream, indicating northward iceberg drift along the NW sector of the margin during and following the Last Glacial Maximum.

Heterogeneous ice nucleation (IN) is one of the most ubiquitous phase transitions on earth and impacts a plethora of fields in industry (e.g. air transport, food freezing and harsh-weather operations) and science (e.g. freeze avoidance of animals, cryobiology, cloud research). Still to date, we are lacking reliable answers to the question: What is it at the molecular scale that causes an impurity to facilitate the freezing process of supercooled liquid water? In this thesis we make headway towards identifying such microscopic principles by performing computational studies combined with data-driven approaches. In chapter 3 we screen a range of model substrates to disentangle the contributions of lattice match and hydrophobicity and find that there is a complex interplay and an enormous sensitivity to the atomistic details of the interface. In chapter 4 we show that the heterogeneous setting can alter the polymorph of ice that forms and introduce the concept of pre-critical fluctuations, yielding new ideas to design polymorph-targeting substrates. Chapter 5 deals with the liquid dynamics before and during the nucleation event, an aspect of nucleation that mostly goes unrecognized. We show that the homogeneous nucleation event happens in relatively immobile regions of the supercooled liquid, a finding that opens new avenues to understand and influence heterogeneous nucleation by targeting dynamics rather than structure. Finally, Chapter 6 builds on the large amount of data created during this project in that we combine all previously simulated systems and devise a machine-learning approach to find the most important descriptors for their ice nucleation activity (INA). With this we identify new microscopic guidelines and demonstrate that the quantitative prediction of heterogeneous INA is in reach. The unveiling of a computational artifact that potentially affects many computational interface studies is also part of this thesis.

Ice shelves are important to the climate system as they control the release of fresh water from ice sheets into the ocean, with consequences for sea level rise and ocean dynamics. Channels modifying basal melt rates and structural integrity have been observed inscribed in the undersides of some ice shelves. Observations indicate that some channels run in a direction transverse to ice flow and it has been suggested that these form due to seasonal variations in ocean properties. This thesis analyses the effect of seasonal variability on ice shelves and meltwater plumes in the underlying ocean. A linear perturbation analysis on vertically integrated ice and plume models showed that seasonal forcing of subglacial discharge or ice flux can generate small ripples melted into the base of the ice shelf. These ripples did not develop into overdeepened channels, but the ripples caused by ice flux appear similar to basal terraces observed underneath some ice shelves. Code was developed to run 1-D nonlinear simulations with the vertically-integrated equations, producing similar results to the linear case. However, runs neglecting hydrostatic pressure gradients exhibited a feedback causing ice flux-generated ripples to grow into small proto-channels. A horizontally-integrated plume model was derived, incorporating the Coriolis force and transverse plume flow into a 1-D model which agreed well with a 3-D ocean simulation. Coupling this horizontally-integrated plume with a co-evolving ice shelf prevented proto-channels from forming. It appears unlikely that subglacial discharge or ice flux variations can give rise to observed transverse channels. A new approach was developed to predict the evolution of internal radar reflectors observed within ice shelves, using vertically-integrated models of ice flow. It is hoped this approach might have applications for inverse modelling and data assimilation.

Robust reconstructions of glacial retreat are required to elucidate controls on ice-sheet deglaciation and (a)synchronicity in the climate system. Such reconstructions can be derived through detailed geological investigations of glacial landforms and sediments. Patagonia, in southern South America, contains a well-preserved glacial record and occupies an important location regarding major Southern Hemisphere climate systems. However, resolving the history of ice-sheet deglaciation has proven problematic, owing to generally low-resolution geomorphological mapping, and a paucity of high-precision dating control. This thesis explores the deglacial history of the Lago General Carrera-Buenos Aires (LGC-BA) and Lago Cochrane-Pueyrredón (LC-P) ice lobes (46-48°S) of the former Patagonian Ice Sheet (PIS). The region is chosen because there are uncertainties associated with: (i) the deglacial configuration and evolution of retreating ice lobes and proglacial lake systems; and (ii) the sub-millennial timing of deglacial events. A new glacial geomorphological map is produced to reconstruct the spatio-temporal evolution of ice lobes and proglacial lakes. The elevation(s) of relict lake shorelines are examined to quantify glacio-isostatic adjustment and accurately assign former lake levels, lake outflows, and ice-margin positions, through the last deglaciation. These new reconstructions are used to target sites for high-resolution dating. A ~1000-year varve chronology is developed at LGC-BA after examining the sedimentary properties of laminated glaciolacustrine sequences. Tephrochronology anchors this record to the calendar-year timescale. The chronology indicates that PIS retreat commenced at 18,086 ± 214 cal a BP, and accelerated from 5 to 18 m yr-1 over the next ~1000 years. The new timing for deglaciation at LGC-BA allows an exploration of climate forcing mechanisms. Comparisons with palaeoclimatic records reveal a hemisphere-wide warming influence on LGC-BA deglaciation. Moreover, spectral analyses of varve thickness time-series demonstrate a likely teleconnection with tropical Pacific climate oscillations, and PIS sensitivity to short-term variations in melt-season temperature.

The well-preserved record of glacial sediment-landform assemblages in Scotland provides an excellent opportunity to reconstruct the extent and style of former glaciation. Despite a concerted research effort, there remain areas where glacial events are poorly constrained. This is exemplified by the Gaick, a dissected plateau in the Central Grampians, which has proven to be an enigmatic and controversial area. Previously-proposed models fail to adequately explain glacial events in the area, partly due to a paucity of detailed geomorphological, chronological and sedimentological investigations. This thesis presents the results of systematic studies of the sediment-landform record in the Gaick. These investigations, combined with the application of morphostratigraphic principles, have elucidated sediment-landform signatures indicative of multiple glacier fluctuations, specifically (i) interactions of local and regional ice lobes following unzipping during ice sheet deglaciation, (ii) a major stillstand of an extensive pre-Younger Dryas plateau icefield, and (iii) spatiallyrestricted plateau icefield glaciation during the Younger Dryas. The sediment-landform record also suggests a two-phased Younger Dryas advance, as has been found elsewhere in Scotland. The sediment-landform evidence was used to produce palaeoglaciological reconstructions for the three glacial phases. The clarity and completeness of the geomorphological record relating to the Younger Dryas, combined with ice surface profile modelling, allowed the three-dimensional reconstruction of a ~42 km2 plateau icefield. This reconstruction yielded an equilibrium line altitude value of 751 ± 46 m, which was used to derive a sea-level equivalent precipitation estimate of 826 ± 331 mm a-1. Taken together with glacier-derived precipitation estimates from across Scotland, this indicates a strong west-east precipitation gradient during the Younger Dryas. This thesis arrives at a more nuanced understanding of former glaciation in the Gaick, resolving discrepancies with previous conceptual models. In particular, this thesis demonstrates that the Gaick was an important ice dispersal centre throughout the Last Glacial-Interglacial Transition.

Like in other mountain ranges around the world (the Southern Alps, the Himalayas and the Rockies), glaciers in the European Alps are a major source of fresh water for human usage and actively contribute to sea-level rise. Among European Alpine glaciers, this study shows that 12% are debris-covered, representing 54% in ice volume. Debris-covered glaciers are those presenting a layer of rock debris on their surface. This debris layer has an insulation effect on the glacier, which influences the response to climate change of this type of glacier. However, little is known about their long-term historical evolution and their future behaviour in the face of climate change. By directly comparing debris-covered glaciers and clean-ice glaciers at several different spatial and temporal scales, this study shows the large impact that debriscovered glaciers have on the Alpine cryosphere. On a glacier-wide scale, this study shows that, over 200 years, a debris-covered glacier (Glacier Noir in the French Alps) has retreated 21% less than an adjacent clean-ice glacier (Glacier Blanc) under the same climatic conditions (a temperature rise of minimum 2°C since the 1960s). On the scale of the European Alps, this study shows that debris-covered glaciers will extend the water supply of 145 million inhabitants. By the middle of the 21st century, debris-covered glaciers will contribute more than 22% of the total glacier runoff from the Alps, with a slower release than clean-ice glaciers. The importance of debris-covered glaciers in terms of behaviour and quantity could change the way they are represented in numerical models and could also change forecasts of the fate of glaciers in the European Alps under future climate scenarios.

The Antarctic ice sheet oscillated between nearly deglaciated and near-modern proportions, sometimes within ~110 kyr, during transient glacial phases of the Oligocene-Miocene icehouse world (5.33-33.9 Ma). Ice sheet hysteresis is closely coupled with carbon-cycle feedback mechanisms that are strongly affected by ocean circulation. However, the absence of high-resolution records representing the North Atlantic end member, a key region driving modern climate variability and ocean circulation, hamper full understanding of Oligocene-Miocene ocean circulation and climate. In this thesis, I integrate palaeomagnetic, X-ray fluorescence (XRF), and environmental magnetic records to date and reconstruct current variability from contourite drift sediments, recovered at Integrated Ocean Drilling Program (IODP) Site U1406 (northwest Atlantic). The revised splice ensures stratigraphic continuity for the studied section. The presented novel palaeomagnetic data processing protocol, which is applicable to all weakly magnetised sediments, ensures the use of only high quality directional data for the construction of a reliable magnetostratigraphy between Chrons C6Ar – C9n (21-27 Ma). Astronomical tuning using XRF-based ln(Ca/K), a proxy for CaCO3 content, provides independent dates for the identified geomagnetic reversals. I determine the best age estimates of reversals using a set of criteria to reconcile the Oligocene-Miocene geomagnetic polarity time scale. Congruent obliquity-paced CaCO3 cycles at Ocean Drilling Program Site 926 (equatorial Atlantic) and IODP Site U1406 suggest that one corrosive bottom water mass influences the CaCO3 content of the boreal Atlantic. The location and occurrence of contourite drift sediments at IODP Site U1406 advocates that the Deep Western Boundary Current (DWBC) is the candidate for such corrosive bottom waters. DWBC variability around the Oligocene-Miocene Transition (~23 Ma) is traced using the detrital magnetic fraction, containing titanohaematite, recognised through detailed rock and environmental magnetic measurements. During the last ca. 1 Myr of the Oligocene, the decrease in CaCO3-corrected concentration and coercivity of titanohaematite indicates an invigorated DWBC. If these characteristics of the DWBC are substantiated, this will have major implications for CO2 reconstructions on orbital time scales and Oligocene-Miocene climate models.

The controls on glacial movement remain one of the most poorly understood elements of the glacial system, largely due to the inaccessibility of the subglacial environment. Here a geotechnical study is presented that investigates the behaviour of till under pore pressure controlled decreases in normal effective stress. This is supplemented with field data from subglacial wireless probes, dGPS, GPR and a UAV survey of the glacial foreland to provide a broad view of subglacial deformation at Skalafellsjökull, Iceland. The geotechnical laboratory investigation was undertaken on till sampled from an ice marginal location at Skalafellsjökull, close to the probe deployment site. This showed that the till adhered to the Mohr-Coulomb model, behaving plastically. In addition, a back pressure shear box was used to model pore pressure controlled reductions in normal effective stress. Linearly reducing normal effective stress by increasing back pressure resulted in episodic increases in horizontal strain rates. Small reductions in shear stress resulted in large strain rate reductions but dilation hardening did not occur. Dilation hardening has previously been suggested as a causal factor in stick-slip glacial motion and a reason for the apparent scale dependence of till rheology, but these experiments suggest it may not be as significant as previously thought. High spatial variability was found throughout the field investigation in probe, dGPS and mapping data from the UAV survey. dGPS deployed above in situ subglacial probes recorded hourly ice movement indicative of stick-slip motion but it was not possible to link this motion to subglacial processes via probe data. Pore pressure data suggested spatially variable coupling to the subglacial hydrological system and inconsistent forcing by surface melt. Tilt data showed consistent movement patterns and lacked regular up-glacier movements previously thought to indicate ice-bed decoupling and sliding. The lack of evidence for forcing of velocity variability from the in situ data suggests a global forcing mechanism at this site, probably longitudinal or lateral coupling to higher velocity areas of the glacier. The spatial variability of the subglacial data has similarities to the metre scale spacing of flutes on the foreland mapped with the UAV imagery, suggesting flute formation has potential to result in complex probe behaviour.

In this thesis, I explore the flow dynamics associated with ice shelves confined within channels and the buttressing they provide to grounded ice. Ice shelves are the floating extensions of ice sheets and act as the interface between the ice sheet and the ocean. They form when ice flows out from the interior of the ice sheet towards the coast and begins to float as the ice thins. Ice shelves are often found within a channel or pinned in place by stationary bedrock outcrops. The interest in their dynamics is motivated by the buttressing effect they provide to the grounded ice, which strongly controls the rate of ice discharge and thereby the contribution to sea-level rise. I use a combination of mathematical modeling, fluid-mechanical laboratory experiments and geophysical data analysis to develop an improved understanding of ice-shelf flow dynamics. Initially, geophysical data in the form of Antarctic ice-surface velocity data is analysed, producing maps of strain rate, shear rate and strain orientation for Antarctic ice shelves. This allows the geophysical setting and flow processes to be explored, particularly by identifying areas where resistance to ice flow is generated and regions of the shelf that make no contribution to buttressing. Using the geophysical data, I find good agreement between a theoretical scaling relationship for ice flow at the ice-shelf calving front and data from Antarctic ice shelves. I proceed to develop an idealized mathematical model of an ice shelf confined to flow in a channel. By assuming shear-dominated dynamics within the shelf, analytical solutions are obtained for steady-state ice-shelf thickness profiles in parallel and diverging channels. This model is developed further to include both shear and extensional stresses, from which numerical solutions for steady-state shelves are calculated. The results from these two models are then compared. It is found that shear stresses dominate the dynamics throughout the majority of the shelf, with adjustment regions at the upstream and downstream boundaries where extensional dynamics become important. Output from these models is also compared with geophysical data and it is observed that there is good agreement between several features of the thickness profiles and velocity fields. In addition to the geophysical data, comparisons are made with fluid-mechanical laboratory experiments designed to simulate the flow of an ice shelf in a channel. The advantage of performing experiments of this kind is that parameters such as the fluid rheology can be varied, allowing for direct comparison with a range of parameters in the mathematical models. From these experiments, surface velocity fields and thickness profiles are collected, which are used to make comparisons with the models. Clear differences are observed in the velocity and strain-rate fields produced using fluids with different rheologies, for which there is qualitative agreement with the output from the mathematical models.