Jökulhlaup (glacial outburst floods) present a recurring natural hazard to the inhabitants and infrastructure along Iceland’s south coast. The high-energy, debris-rich floods are capable of eroding, transporting and depositing vast quantities of sediment; and in some cases, can completely re-shape sandur topography within a single flood event. They can be triggered by drainage of ice-dammed lakes or by meltwater released during sub-glacial volcanic eruptions; or a combination of both. This research reconstructs a Tenth Century jökulhlaup at Sólheimajökull; a southern outlet glacier of Mýrdalsjökull ice cap overlying Katla volcano. The flood permanently altered sandur topography and marked an important shift in changing flood routes and resultant flood deposits at the site since. Collection and synthesis of field evidence provides detailed information about the jökulhlaup, the route it took, the contemporaneous glacier margin, and environmental conditions at the time. As Sólheimasandur’s last major jökulhlaup, this event is important, not just in terms of scale and magnitude (comparable with the largest floods known to have occurred at the site); but also because of its geomorphic and sedimentary signature. This research records and compiles the remnant features of the Tenth Century flood, some of which are still evident at Sólheimasandur today. The outcome is a template for linking jökulhlaup sources, pathways and deposits, focusing in particular on ice- proximal flood routeways as a determining control over the nature of flood deposit. Geomorphic, sedimentological and tephrostratigraphic techniques have been used to recreate the origin, mechanics, and impacts of this flood through analysis of its deposits. In the same manner the regional Holocene jökulhlaup chronology is refined for Sólheimasandur, supported by a dating framework of volcanic eruptions and glacial fluctuations. An innovative 3D visual simulation, VolcVis, has been created, which models the field site as it might have looked in the Tenth Century, based on field data collected in reconstructing the Tenth Century flood. The ability of the prototype simulation to assist in palaeo-geomorphic reconstruction, is tested by visually simulating the Tenth Century volcanic jökulhlaup for the first time. The model distills complex and often spatially disparate field data into an interactive, user-friendly tool, with wider applicability in communicating results to the general public, aiding hazard mitigation authorities and informing the designation of evacuation routes. Identification of the pathways, magnitudes and frequencies of past jökulhlaup at Sólheimajökull allows a greater accuracy in predicting contemporary flood routes, when other major subglacial eruptions occur in this region of Iceland. The value of this research is timely given recent volcanically-generated jökulhlaup at Eyjafjallajökull (2010) and Grímsvötn (2011); illustrating how heightening awareness of past (and present) glacio-volcanic hazards in this region of South Iceland ensures better preparedness for future events.

Lake Tuborg is a large lake on west-central Ellesmere Island, Nunavut. Part of the lake is meromictic, and contains trapped saltwater below about 55 m depth. The lake receives meltwater and sediments from multiple sources, including snowmelt and glacier melt. A lake process study was undertaken from 2001-2003 at Lake Tuborg that involved obtaining profiles of water temperature, salinity, transmissivity, and dissolved oxygen. Networks of short and long sediment cores were also obtained throughout the lake. During the last year of monitoring the lake and its sediments, a large catastrophic drainage of an ice-dammed lake occurred (a jökulhlaup). This was the largest jökulhlaup witnessed in Canada since 1947. Detailed measurements of lake conditions before, during, and after the jökulhlaup allowed the responses to be measured in great detail. The lake drained by floating its ice dam, an extremely rare drainage style in the Canadian High Arctic. The basin of Lake Tuborg closest to jökulhlaup inflow filled with fresh, cold and turbid water. A sill separates this basin from the larger more distal meromictic basin, and this sill effectively blocked turbidity currents from entering this basin. Conclusions from this phase of research include (1) salinity and temperature in the saltwater basin were minimally affected by the jökulhlaup, and (2) at a deep, distal location, an identifiable thick, coarse-grained, non-erosive deposit was produced by the jökulhlaup. The above conclusions allowed the varved sedimentary record to be examined for similar deposits in the past, with the assumption that similar deposits could be found in the long core record, the sediments could be dated, and that previous jökulhlaup deposits would also be nonerosive. Varve-thickness counting, Cesium-137 dating, and particle size analyses showed that prior to 1960, no similar events occurred in roughly the last thousand years. In addition, only three large jökulhlaups have occurred in the last thousand years, all of which occurred after about 1960. This significantly improves the understanding of the history of the lake, the surrounding glaciers, and the paleoclimate of the region. The lake bottom deposits that were sampled before, during and after the 2003 jökulhlaup were extraordinarily unique. A major part of the work of characterizing these deposits involved determining the size of their constituent particles. Image analysis of sedimentary particles using backscattered electron microscope imagery is a method to determine particle size at extremely high resolution. This tool improves on existing techniques since it automates the process of statistically processing images, quantifies the percentage of disturbances on images, and allows for extremely small measurement windows relative to particle size by implementing special particle counting rules.

<p>Our understanding of the climate of northern Sweden during the late Holocene is largely dependent on proxy-data series. These datasets remain spatially and temporally sparse and instrumental series are rare prior to the mid 19th century. Nevertheless, the glaciology and paleo-glaciology of the region has a strong potential significance for the exploration of climate change scenarios, past and future. The aim of this thesis is to investigate the 19th and 20th century climate in the northern Swedish mountain range. This provides a good opportunity to analyse the natural variability of the climate before the onset of the industrial epoch. Developing a temporal understanding of fluctuations in glacier front positions and glacier mass balance that is linked to a better understanding of their interaction and relative significance to climate is fundamental in the assessment of past climate.</p><p>I have chosen to investigate previously unexplored temperature data from northern Sweden from between 1802 and 1860 and combined it with a temperature series from a synoptic station in Haparanda, which began operation in 1859, in order to create a reliable long temperature series for the period 1802 to 2002. I have also investigated two different glaciers, Pårteglaciären and Salajekna, which are located in different climatic environments. These glaciers have, from a Swedish perspective, long observational records. Furthermore, I have investigated a recurring jökulhlaup at the glacier Sälkaglaciären in order to analyse glacier-climate relationships with respect to the jökulhlaups. </p><p>A number of datasets are presented, including: glacier frontal changes, in situ and photogrammetric mass balance data, in situ and satellite radar interferometry measurements of surface velocity, radar measurements, ice volume data and a temperature series. All these datasets are analysed in order to investigate the response of the glaciers to climatic stimuli, to attribute specific behaviour to particular climates and to analyse the 19th and 20th century glacier/climate relationships in northern Sweden.</p><p>The 19th century was characterized by cold conditions in northern Sweden, particularly in winter. Significant changes in the amplitude of the annual temperature cycle are evident. Through the 19th century there is a marked decreasing trend in the amplitude of the data, suggesting a change towards a prevalence of maritime (westerly) air masses, something which has characterised the 20th century. The investigations on Salajekna support the conclusion that the major part of the 19th century was cold and dry. The 19th century advance of Salajekna was probably caused by colder climate in the late 18th and early 19th centuries, coupled with a weakening of the westerly airflow. The investigations on Pårteglaciären show that the glacier has a response time of ~200 years. It also suggests that there was a relatively high frequency of easterly winds providing the glacier with winter precipitation during the 19th century.</p><p>Glaciers have very different response times and are sensitive to different climatic parameters. Glaciers in rather continental areas of the Subarctic and Arctic can have very long response times because of mass balance considerations and not primarily the glacier dynamics. This is of vital importance for analyzing Arctic and Subarctic glacier behaviour in a global change perspective. It is far from evident that the behaviour of the glacier fronts today reflects the present climate.</p>

Our understanding of the climate of northern Sweden during the late Holocene is largely dependent on proxy-data series. These datasets remain spatially and temporally sparse and instrumental series are rare prior to the mid 19th century. Nevertheless, the glaciology and paleo-glaciology of the region has a strong potential significance for the exploration of climate change scenarios, past and future. The aim of this thesis is to investigate the 19th and 20th century climate in the northern Swedish mountain range. This provides a good opportunity to analyse the natural variability of the climate before the onset of the industrial epoch. Developing a temporal understanding of fluctuations in glacier front positions and glacier mass balance that is linked to a better understanding of their interaction and relative significance to climate is fundamental in the assessment of past climate. I have chosen to investigate previously unexplored temperature data from northern Sweden from between 1802 and 1860 and combined it with a temperature series from a synoptic station in Haparanda, which began operation in 1859, in order to create a reliable long temperature series for the period 1802 to 2002. I have also investigated two different glaciers, Pårteglaciären and Salajekna, which are located in different climatic environments. These glaciers have, from a Swedish perspective, long observational records. Furthermore, I have investigated a recurring jökulhlaup at the glacier Sälkaglaciären in order to analyse glacier-climate relationships with respect to the jökulhlaups. A number of datasets are presented, including: glacier frontal changes, in situ and photogrammetric mass balance data, in situ and satellite radar interferometry measurements of surface velocity, radar measurements, ice volume data and a temperature series. All these datasets are analysed in order to investigate the response of the glaciers to climatic stimuli, to attribute specific behaviour to particular climates and to analyse the 19th and 20th century glacier/climate relationships in northern Sweden. The 19th century was characterized by cold conditions in northern Sweden, particularly in winter. Significant changes in the amplitude of the annual temperature cycle are evident. Through the 19th century there is a marked decreasing trend in the amplitude of the data, suggesting a change towards a prevalence of maritime (westerly) air masses, something which has characterised the 20th century. The investigations on Salajekna support the conclusion that the major part of the 19th century was cold and dry. The 19th century advance of Salajekna was probably caused by colder climate in the late 18th and early 19th centuries, coupled with a weakening of the westerly airflow. The investigations on Pårteglaciären show that the glacier has a response time of ~200 years. It also suggests that there was a relatively high frequency of easterly winds providing the glacier with winter precipitation during the 19th century. Glaciers have very different response times and are sensitive to different climatic parameters. Glaciers in rather continental areas of the Subarctic and Arctic can have very long response times because of mass balance considerations and not primarily the glacier dynamics. This is of vital importance for analyzing Arctic and Subarctic glacier behaviour in a global change perspective. It is far from evident that the behaviour of the glacier fronts today reflects the present climate.

A great number of tourists visits each day the Fox Glacier in New Zealand and are subjected todangers such as jökulhlaups. This phenomena can occur suddenly and in a destructive way when abuilt up of water suddenly bursts and releases large amounts of water. This is well known to happen atthe near by Franz Josef Glacier which has similar settings to Fox Glacier, but sparse previous researchon the subject indicates different results of whether jökulhlaups are present or not by Fox Glacier. Thisthesis therefore aims to study the conditions for jökulhlaups at the glacier. This is done by remotesensing and field observations of the sandur in the Fox Glacier Valley, qualitative interviews andliterature studies with a comparison with Franz Josef Glacier. The results shows that there has beenjökulhlaups from an ice dammed lateral lake which is also the most likable way for jökulhlaups tohappen again, although there are other potential ways for it to happen due to the large amount of heavyrain that this region gets.