Search

Glacier retreat is a strong indicator of climate change and global warming. The anthropogenic changes in the Earth's atmosphere are mostly to blame for the climate extremes and their consequences in the last few decades. The Himalayan region is no exclusion to the trend. As glaciers begin to retreat, the glacial lake starts to fill or form behind the natural moraine or ice dam in the glaciers. The sudden release of the water, known as the Glacial Lake Outburst Flood (GLOF), can release a large amount of water and sediment. There have been various destructive GLOFs recorded in Nepal since the 1960s. It is vital to understand the GLOF dynamics, geomorphology and historical events to mitigate the GLOF hazards in the region. An advanced approach based on remote sensing data and empirical evidence is more suitable to tackle these issues. This research investigated 11 among 30 past events recorded in the HKH region (Nepal) to establish the causes and triggering factors that led to the catastrophic failure, which helped establish the vulnerability assessment of these glacial lakes. This eventually led to creating a GLOF vulnerability assessment framework that is unique and useful to the communities. This research concluded that 40% of the GLOF events was due to the moraine dam failure. In the retrospective approach, 5 out of 11 glacial lakes scored a very high total vulnerability score (TVR), which suffered catastrophic events in the past. The TVR of the currently existing 21 potential dangerous glacial lakes (PDGL) in Nepal was also conducted using the proposed assessment framework that concluded the 7 very high, 4 high, 5 medium, and the rest are low. Hence, this assessment tool's reliability is very high. This research also concluded that there should integrated approach to climate change adaptation and hazard mitigations in the region.

The Bolivian Andes have experienced sustained and widespread glacier area reduction and volume loss in recent decades. This study finds that from 1986 to 2018 glacier areas have shrunk from 529 km2 to 281 km2 (49 %) in the Bolivian Cordillera Oriental. Glacier melting and recession has been accompanied by the development of proglacial lakes, which can pose a glacial lake outburst flood (GLOF) risk to downstream communities. Therefore, glacier bed topographies were extracted and illustrate the potential development of 68 future lakes. Eight of these lakes possess populations downstream. A simple geometric model (MC-LCP) was used to model GLOFs from these potential future lakes, illustrating that ~1100 to ~2900 people could be affected by flooding if these lakes were to appear and to burst. The rest of this work is dedicated on the estimation of the risk from current, already existing lakes. Multi- Criteria Decision Analysis (MCDA) was used to rapidly identify potentially dangerous proglacial lakes in regions around the world without existing tailored GLOF risk assessments, where a range of proglacial lake types exist, and where field data are sparse or non-existent. After testing the robustness of the MCDA model against a number of past GLOFs, it was applied to the Bolivian Cordillera Oriental. From the 25 lakes possessing populations downstream, 3 lakes were found to pose 'medium' or 'high' risk, and required further detailed investigation. Since no attempt has yet been made to model GLOF inundation downstream from these proglacial lakes, 2m resolution DEMs were generated from stereo and tri-stereo SPOT 6/7 satellite images to drive a hydrodynamic model (HEC-RAS 5.0.3) of GLOF flow. The model was tested against field observations of a 2009 GLOF from Keara, in the Cordillera Apolobamba, and was shown to reproduce realistic flood depths and inundation. The model was then used to model GLOFs from Pelechuco lake (Cordillera Apolobamba) and Laguna Arkhata and Laguna Glaciar (Cordillera Real). In total, ~1100 to ~2200 people could be directly affected by outburst flooding.

The courts have become an increasingly important venue for efforts to combat the impacts of climate change. Claims have brought on a range of legal grounds, but many climate lawsuits are based on the existence of a causal relationship between emissions of greenhouse gases and impacts that occur as a result. Such claims may be made to establish emitters' liability for climate change impacts, or to demonstrate that contributions to climate change cause harm, and therefore that greenhouse gas emissions should be curbed. Establishing causation in climate litigation requires synthesis of climate science and legal arguments. This thesis evaluates how legal arguments can align with developments in climate change attribution science, a subfield of climate science that can assess the contribution of anthropogenic influence to climate-related impacts. It then considers how attribution-science evidence has been used in past lawsuits, and responds to emerging scientific questions regarding the effects of climate change that have arisen in the courts. Legal scholars have identified attribution science as holding the potential to establish causation in the courts, but the analyses described in section 4 of this thesis found that the evidence presented in climate lawsuits often lags developments in climate science and that, in some cases, this may have hindered findings of specific causal relationships between greenhouse gases and their consequences. Responding to these findings, sections 5 and 6 report analyses that advance the state of knowledge in areas of attribution science that hold particular legal relevance. Section 5 studied anthropogenic influence on the glacial lake outburst flood hazard from Laguna Palcacocha in the Peruvian Andes, in the context of ongoing litigation (Lliuya v. RWE). It concluded that 85-105% (5-95% confidence interval) of the observed industrial-era warming in the region surrounding Laguna Palcacocha was attributable to anthropogenic climate change and that the overall retreat of the Palcaraju glacier since 1880 is entirely attributable to this warming. Moreover, it is virtually certain (>99% probability) that the observed glacier retreat cannot be explained by natural variability alone. Laguna Palcacocha has expanded in the wake of this glacier retreat, substantially increasing the outburst flood hazard to the downstream city of Huaraz. Finally, this thesis responds to the increased use of human rights-based arguments in climate litigation and develops methods that extend attribution science analyses from quantifying the effect of climate change on physical hazards to their impacts on health. These methods address limitations in existing approaches, including the representation of the climate-health relationship, choices in calculating counterfactual temperatures, differing approaches for assessing long-term trends and individual events, and estimating the effects of adaptation. The approach described in this section is then applied to quantify heat-related mortality attributable to climate change in the Canton of Zürich (Switzerland) over 1969-2018. We found a substantial burden of heat-related mortality attributable to climate change both within and outside of heatwaves. Overall, 1,7000 heat-related deaths were attributed to climate change, and adaptation was found to have avoided at least a further 700 deaths. The thesis spans climate science, glaciology, epidemiology, and legal analysis. Altogether, this research explains how science can strengthen legal causation arguments in climate litigation, and develops methods and analyses to this end. Scientific evidence has a central role to play in developing well-grounded climate lawsuits, and methodological developments can resolve questions raised in court.

Glacial retreat around the world, accelerated by climate change, has led to the formation of glacier lakes that present a risk of a glacial lake outburst flood (GLOF). GLOFs are sudden, catastrophic events that are impossible to predict. Communities in the path of a potential GLOF are now attempting to implement adaptation projects, yet no quantitative data or guidance is available to understand the benefits of adaptation projects or how to weigh these benefits against the cost of project implementation. The objective of this work is to develop a rational decision making methodology for GLOF risk management that incorporates available scientific information and the uncertainty surrounding the understanding of GLOF events. The decision making methodology consists of 1) identifying flooding scenarios, 2) evaluating the consequences of flooding scenarios, and 3) a nuanced (in terms of the inclusion of intangibles and probabilistic events) economic analysis of flood consequences and adaptation options. The methodology is applied to Lake Palcacocha in Peru and Imja Lake in Nepal to demonstrate the robustness of the methodology in light of different sources of uncertainty and data gaps. For Imja Lake it is concluded that lowering the lake 10 m is the best decision, from an economic standpoint. Nonetheless, the decision is sensitive to changes in the decision tree variables, which should be assessed for accuracy. At Lake Palcacocha it was determined that a GLOF would result in substantial damage to the city of Huaraz and the best decision is to lower the lake 30 m and install an emergency warning system (EWS). This decision is robust to large changes in the uncertain variables.

Debris-covered are ubiquitous in the Nepal Himalaya and significantly alter the glaciers response to climate change and have large implications on the development of glacial lakes. The thickness of the debris is largely heterogeneous over the course of the glacier thereby promoting ablation in areas of thin debris and retarding ablation in areas of thick debris. The debris thickness typically increases towards the snout of the glacier, but can be difficult to measure as field measurements are time-consuming and laborious. This body of work utilizes satellite imagery in conjunction with a debris-covered glacier energy balance model to reasonably estimate the spatial variations in debris thickness for glaciers in the Everest region of Nepal. Sub-debris ablation rates may be computed using the same energy balance model, but requires detailed information regarding the properties of the debris and the surface processes. Detailed field data was collected over the 2014 melt season on Imja-Lhotse Shar Glacier to estimate many of the debris properties. This data was also used to model the sub-debris ablation rates to develop an understanding of the critical properties (i.e., thermal conductivity, albedo, and surface roughness) and processes (i.e., accounting for the latent heat flux) required to accurately model the impact of the debris. The heterogeneous debris cover often causes higher melt rates upglacier, which diminishes the glacier’s topographic gradient thereby promoting glacier stagnation and the development of glacial lakes. These glacial lakes form behind terminal moraines comprising soil and loose boulders that are susceptible to fail causing a glacial lake outburst flood (GLOF). GLOFs can have devastating impacts on infrastructure and communities located downstream; however, assessing the risks associated with these floods has traditionally required detailed field campaigns that are difficult to perform as these glacial lakes are located in remote areas at high altitudes. This body of work develops a holistic hazard assessment using solely remotely sensed data to objectively characterize the threat of a GLOF. This hazard assessment provides valuable information concerning potential GLOF triggers that may be used to direct future field campaigns and ultimately the management actions associated with these glacial lakes.

Outburst floods are one of the most catastrophic natural hazards for populations and infrastructure. They are usually generated from storm runoff, rapid melting of glacial ice or man-made and natural dam breaks, such as river dikes, volcanic debris dams and landslide damns etc. Such high-magnitude sudden onset floods generally comprise of an advancing intense kinematic water wave that can induce considerable sediment transport. Therefore, the exploration and investigation of sediment-laden outburst floods cannot be limited solely to water flow but must also include the flood-induced sediment transport. Understanding the complex flow-bed interaction process in large (field) scale outburst floods is still limited, not least due to a lack of well-constrained field data, but also because consensus on appropriate modelling schemes has yet to be decided. In recent years, attention has focussed on the numerical models capable of describing the process of erosion, transport and deposition in such flows and they are now at a point at which they provide useful quantitative data. Although the "exact" measure of bed change is still unattainable the numerical models enhance and improve insights into large outburst flood events. In order to model and better understand heavily sediment-laden flows and resulting geomorphic impacts, this thesis adopts a layer-based conceptual model which separates the system into an active bed layer, a water sediment mixed sheet flow layer and a suspension layer. Correspondingly, a layer-based hydro-morphodynamic model is proposed fully considering both bedload and suspended load based on shallow water theory. The model system is primarily composed of a combination of three modules: (1) a hydrodynamic module; (2) a sediment transport module; and (3) a morphological evolution module. In the thesis, firstly, a robust hydrodynamic model is proposed and tested including addressing the source terms and wetting/drying issues for application to irregular beds. Then based on the robust hydrodynamic model, a layer-based morphodynamic model is developed and solved numerically by an advanced second-order Godunov-type finite volume method. A series of theoretical and experimental tests are applied to validate the model in terms of both hydrodynamic and morphodynamic aspects. The results of these tests show that the developed models can predict the hydrodynamic and morphodynamic process effectively with good agreement with theoretical and experimental results. To demonstrate a real application, a full-scale volcano-induced jökulhlaup or glacial outburst flood (GLOF) at Sólheimajökull, Iceland is reproduced by the proposed model. The simulation of the sediment-laden outburst flood is shown to perform well, with further insights into the flow-bed interaction obtained from the modelling output. These results are beneficial to flood risk management and hazard prevention and mitigation.

Eskers are ridges of stratified glaciofluvial material deposited in englacial, subglacial or supraglacial channels and ice-walled canyons. Eskers have been used to infer the dynamics and palaeo-hydrology of large ice sheets, despite observations of palaeo-esker sedimentary architecture lacking rigorous constraints on depositional timescale. This research aims to identify the hydrological, glaciological and sedimentary controls on the sedimentary architecture of single event outburst flood (jokulhlaup) eskers at SkeiOarârjOkull and Bering Glacier. These eskers formed during monitored outbursts, providing time constraints on the depositional events, thereby making the eskers ideal analogues for palaeo-eskers. GPR data was collected using a pulseEKKO Pro 1100 system at SkeiOarârjOkull and Bering Glacier during field seasons in 2006 and 2007. At SkeiOarãrjOkull grids of 100 MHz GPR lines were collected on the glacier surface and 200 MHz GPR grids were collected on all workable sections of an esker and ice-walled canyon fill generated by a jokulhlaup in November 1996. At Bering Glacier 200 MHz GPR grids were collected on workable sections of an ice-walled canyon fill and esker generated during outbursts in July-August and October 1994, respectively. Examination of the GPR data has allowed development of site-specific models for esker and ice-walled canyon fill deposition, providing the first detailed insight into the sedimentary architecture of single event jokulhlaup eskers. These models show that single high-magnitude jOkulhlaups can generate eskers with complexities previously unexpected for single events. The englacial position of an esker is controlled by the presence of structural weaknesses within the ice and the jokulhlaup release mechanism. Esker sedimentary architecture, on the other hand, is controlled by a complex interaction between hydrological, glaciological and sedimentary factors. The most fundamental control on jokulhlaup esker sedimentary architecture is conduit geometry, which determines the type of macroform from which the esker is composed. Thus, eskers deposited during jOkulhlaups should be made up of both wide ridges as composite macroforms in areas of conduit expansion and narrow ridges composed of upper-stage plane beds in constricted conduits. The smaller scale sedimentary features, which include antidunes and boulder clusters, as well as the frequency of erosional structures, are controlled by interactions between the flow conditions and sediment supply. The eskers generated during the 1996 SkeiOarãrhlaup and 1994 Bering Glacier outburst flood demonstrate sedimentary architectures that are similar to those identified in many palaeo-eskers described from previous studies. The research suggests that jOkulhlaups may have had a greater role in palaeo-esker sedimentary architecture than previously hypothesised.

In this thesis I develop novel methods and datasets, using airborne radio-echo sounding and high-resolution satellite-derived digital elevation models, to improve the understanding of subglacial lake distribution and activity beneath the Greenland Ice Sheet. First, I use 23 years of airborne radio-echo sounding data, totalling over 500,000 km, to conduct the first systematic ice sheet-wide survey of subglacial lakes beneath the Greenland Ice Sheet. Following classification of basal reflector properties, I detect over 50 new stable subglacial lakes beneath the Greenland Ice Sheet. Surface elevation data also revealed 2 previously unidentified collapsed ice-surface basins indicative of subglacial lake activity. The results indicate that subglacial lakes are more common than previously thought and differ from those found beneath the Antarctic Ice Sheet both in their size and spatial distribution. Next, I develop a new automated technique using statistical analysis of highresolution, satellite-derived surface elevation data to detect localised elevation changes which are characteristic of subglacial lake drainage events. I then apply this framework to the Greenland Ice Sheet within 50 km of the ice margin to classify 7 potential active subglacial lakes which could be used as candidates for future geophysical surveys and further monitoring to understand to understand the role they play within the subglacial hydrological system. Finally, using the results from the previous chapter, I utilise a variety of satellite datasets to analyse a unique active subglacial lake in Northern Greenland. From the observational record, the subglacial lake appears to have drained twice in the past 36 years, in 1990 and 2014, but each event with differing drainage characteristics. In particular, observations of downstream ice surface fracturing and excavation of ice blocks 20 m tall in 2014, suggests that drainage event caused a subglacial outburst flood which propagated to the ice surface. This is the first observation of supraglacial emergence of subglacial flood water in Greenland, indicating more varied drainage mechanisms that are currently not accounted for in numerical ice sheet models.

This dissertation addresses two aspects of climate change impacts on water resources in the Cordillera Blanca mountain range in Ancash, Peru: glacial lake outburst floods (GLOFs) and water availability. Peru is one of the countries most impacted by climate change, largely due to the abundance of glaciers that play an integral role in the water resources systems of the Peruvian Andes and the coastal region. A warming climate has resulted in the accelerated retreat of many of these glaciers in recent decades. The two greatest impacts of climate change on water security in the Cordillera Blanca are GLOFs and water scarcity during the dry season. This dissertation studies both of these facets of water security in the Cordillera Blanca, Peru. As new glacial lakes emerge and existing lakes continue to grow, they pose an increasing risk of GLOFs that can be catastrophic to the communities living downstream. In this work, particular emphasis is placed on the upper watershed processes that typically comprise a GLOF event. Dynamics of avalanche-generated impulse waves are investigated through three-dimensional hydrodynamic lake simulations of potential GLOF scenarios at Lake Palcacocha, Peru. At Artesonraju Glacier, an emerging lake has recently formed and continues to grow as the glacier retreats. Future lake volumes are projected from ground penetrating radar measurements of ice thickness. With these projections of future lake conditions, possible future hazard conditions are studied at vi Artesonraju, and a new analytical method is presented for calculating approximate overtopping volumes from avalanche-generated waves. Climate change impacts on water availability have been studied through the analysis of approximately 50 years of precipitation data from a weather station in the Cordillera Blanca. These data have been analyzed for trends and changes in variability in precipitation patterns. As a foundation for climate-resilient development, precipitation trends and changes in variability have been linked to possible impacts on agricultural projects. The results of the precipitation data analysis were compared to studies of local perceptions of climate change, and it was concluded that people’s perceptions of change in precipitation patterns often do not reflect the trends observed in the gauged data.

A multi-proxy age model protocol was established using Labrador Sea sediment cores U1305 (57°N,48°W) and U1302 (50°N,45°W). These cores provided centennial-scale surface and deep-water records spanning the last interglacial (153-100ka). Ice rafted debris (IRD) counts and meltwater-runoff proxy Ba/Ca were used to investigate ice sheet instabilities; planktonic foraminiferal assemblages, planktonic δ18O and multi-species Mg/Ca records were used to reconstruct temperature and salinity; and fluctuations in Atlantic meridional overturning circulation (AMOC) were inferred from benthic δ13C. Intense cold and ice rafting accompanied Heinrich events H11 (135–130ka) and H12 (140ka), as well as a previously unrecognised Heinrich event (H13; 150–148ka), during which benthic and planktonic δ13C values indicate reduced AMOC and nutrient-rich Antarctic bottom-water (AABW) incursions. During H11, three warm phases preceded cooling and massive IRD pulses. Strongly depleted U1302 planktonic δ18O accompanied all IRD-rich cold periods, particularly H11, attributed to sea-ice formation and enhanced meltwater input. No Younger Dryas-like event was observed. Last interglacial temperatures were 2–5°C warmer in U1305, but similar to modern values at U1302, suggesting differing latitudinal responses to increased insolation (129–126ka). A diatom mat deposited ~128.7–128.6ka in U1305 indicates a proximal subarctic convergence front. Palaeotemperature reconstructions reveal two last interglacial warm maxima. The first (128.5–126.5ka) had the most diverse foraminiferal assemblages and was coeval with continued Greenland ice-sheet melt. A previously reported red layer, linked to a Laurentide outburst flood, has been dated to ~126.5ka and marks the culmination of localised intra-interglacial cooling. Subsequent AABW incursion events suggest rapid AMOC weakening and instability. Mg/Ca results reveal a second warming (123–117ka) associated with a vigorous AMOC, and was attributed to a stronger Irminger Current. IRD, SST and isotopic evidence indicate a prominent cold event ~117ka, followed by surface cooling and ice rafting linked to the North Atlantic cold events C26–C23.

In recent decades, the downwasting of several debris-covered glaciers in the Himalaya has led to the formation of large and potentially hazardous moraine- dammed lakes. The frequency of Glacial Lake Outburst Flood (GLOF) events in the Himalaya has steadily increased since the 1970s and as global temperatures continue to rise this trend is set to continue in the future. Downwasting of the debris-covered Ngozumpa Glacier in the Khumbu Himal, Nepal, has resulted in the abandonment of the lateral and terminal moraine crests, leaving them standing several tens of metres above the glacier surface. The moraines have exerted a control on the drainage of meltwater from the glacier surface and have encouraged ponding of meltwater on the glacier surface. The present study examines the evolution of perched supraglacial ponds on the Ngozumpa Glacier and assesses how the growth of these ponds affects the rate of downwasting of the glacier surface. The expansion rates of perched ponds can be rapid, up to 21,609 m ² a⁻¹, but the growth of these ponds tends to be terminated when contact is made with the englacial drainage network. The thesis documents for the first time a complete cycle of perched supra glacial pond growth and drainage and also provides direct evidence for internal ablation during pond drainage, a process that has only been inferred in previous research. The western lateral moraine has dammed back drainage from the western tributary valleys, resulting in the formation of laterally-dammed lakes. The research presented here examines the processes and rates of paraglacial reworking of the Ngozumpa moraines in order to assess the stability and longevity of the moraine dam. Approximately 1 km from the Ngozumpa terminus a large Spillway Lake has formed. Meltwater from upglacier is channelled into the lake and exits the glacier surface through an over-spill channel cut down through the western lateral moraine. The level of the Spillway Lake is thereby controlled by the height of the spillway channel through the western lateral moraine. The rate of expansion of the Spillway Lake is lower than that of the perched ponds upglacier, but as the Spillway Lake continues to enlarge and surface downwasting of the glacier surface proceeds, the lake could enter a period of rapid and unstable growth. By analogy with other glaciers in the Khumbu Rimal, it is possible that a large and potentially hazardous lake will form on the Ngozumpa within the next two decades.

Barrier Islands comprise 10% of the Earth’s shorelines, fringing every continent except Antarctica. Despite their ubiquity, much about the medium to long-term evolution of these coastal systems remain poorly understood, mostly due to the destruction of the geologic record as barriers migrate landward under the influence of rising sea level. Even where modern barriers and related strandplain systems have prograded and regressed, leaving evidence of their former geometries in the form of relict shorelines, field investigations often require intensive labor and time commitments to interpret past evolution. In this work, several investigations are undertaken to use novel numerical modeling techniques coupled with field interpretation and comparison to gain insights into the evolution of barrier islands from relict geomorphic features preserved on the continental shelf seabed and the surfaces of modern barriers. Much of these efforts focus on ‘drowned’ barrier features, or the remnants of barrier islands left stranded on the shelf during landward migration that occurred in the late Pleistocene and early Holocene, as well as patterns of abandoned foredune ridges generated during late Holocene shoreline progradation. Among the most intriguing results herein is the possibility that the internal dynamics of barrier islands can lead to periodic backstepping and partial deposition of the barrier structure without the need to invoke changing environmental forcing. Moreover, it can be shown that combinations of internal dynamic state and environmental forcing from relatively sudden changes in rate of sea-level rise could lead to a rich suite of barrier retreat behaviors. This might explain the morphologies of drowned barrier features produced during glacial meltwater and outburst flood ‘pulses’ prior to 8 kya. More importantly, these insights may prove practical in modern systems, where anthropogenic climate change threatens barrier islands with historically unprecedented rates of sea-level rise.

The Himalayas of South Asia are home to many glaciers, which are retreating because of climate change and leaving in their wakes large lakes. These lakes are held in place by naturally deposited and inherently unstable moraine dams, which are at risk of overtop by impulse waves, commonly generated when avalanches or landslides enter the lake. When an overtopping event causes catastrophic failure, a glacial lake outburst flood (GLOF) occurs. The Imja glacier is experiencing the highest mass loss rate in the Mount Everest region, contributing to the expansion of Imja Tsho, a lake with several villages downstream. A GLOF from the lake can bring catastrophic damage to these villages, threatening both property and human life and creating a need to understand the processes that trigger these floods. The process chain for an avalanche-induced GLOF was modeled numerically, using the output of one subprocess as input into the next. First, the volume and momentum of various avalanches entering the lake were calculated using RAMMS. Next, the subsequent waves were simulated using BASEMENT software and validated with empirical equations to ensure the proper transfer of momentum from the avalanche to the lake. BASEMENT was then used to model moraine erosion and downstream flooding, which yielded a map of risk areas downstream. Moraine erosion was calculated for two geomorphologic scenarios: a site-specific scenario using field data, and a worst-case scenario based on past literature and applicable to lakes in the entire region. Neither case resulted in flooding outside the river channel at downstream villages. The worst-case scenario resulted in some moraine erosion and increased channelization of the lake outlet, which yielded greater discharge downstream but no catastrophic collapse. The site-specific scenario generated similar results, but with very little erosion and a smaller downstream discharge. Results indicated that Imja Tsho is unlikely to produce a catastrophic GLOF in the near future; however, some risk exists within the river channel, necessitating continued monitoring of the lake. Furthermore, these models were designed for ease and flexibility so that they can be adopted by a wide range of stakeholders and appropriated for other lakes in the region.