El presente trabajo consiste en el modelamiento hidrológico e hidráulico de la quebrada Cashahuacra ubicada en el distrito de Santa Eulalia con el fin de proponer la estructura de un sistema de alerta temprana que integre estos modelos para evitar la pérdida, primordialmente, de vidas humanas. Se presenta una síntesis de los principales eventos de flujo de detritos ocurridos en dicho lugar así como un marco teórico que aborda conceptos básicos relacionados al estudio de la hidrología e hidráulica. Para el modelo hidrológico, se llevó a cabo la caracterización de la quebrada Cashahuacra y se obtuvo sus parámetros principales, así también, fue necesario la adquisición de los datos históricos de precipitación. De esta manera, el modelo hidrológico fue realizado con el software HEC-HMS versión 4.2 en colaboración con ArcGIS versión 10.2.1. y su extensión HEC-GeoHMS. Así, se pudo determinar los caudales máximos bajo los periodos de retorno de 5, 10, 20, 50, 100, 200 y 500 años. El modelo hidráulico fue desarrollado con el software FLO-2D versión PRO. Tuvo como parámetros de entrada al hidrograma calculado por el modelo hidrológico, a la topografía digital del terreno, y a parámetros reológicos como la viscosidad y el esfuerzo de cedencia. Luego, fueron obtenidas las velocidades y profundidades máximas de los periodos de retorno mencionados líneas arriba. Finalmente, se ha integrado estas dos modelaciones en la estructura de un sistema de alerta temprana frente a huaycos y se ha sugerido la implementación de un radar meteorológico para la quebrada en estudio. / The present work consists of hydrological and hydraulic modeling of the Cashahuacra stream located in the Santa Eulalia district in order to propose the structure of an early warning system that integrates these models to avoid the loss, primarily of human lives. A synthesis of the main debris flow events occurring in that place is presented as well as a theoretical framework that addresses basic concepts related to the study of hydrology and hydraulics. For the hydrological model, the characterization of the Cashahuacra stream was carried out and its main parameters were obtained, as well as the acquisition of the historical precipitation data. In this way, the hydrological model was made with the HEC-HMS software version 4.2 in collaboration with ArcGIS version 10.2.1. and its HEC-GeoHMS extension. Thus, it was possible to determine the maximum flow rates under the return periods of 5, 10, 20, 50, 100, 200 and 500 years. The hydraulic model was developed with the FLO-2D PRO version software. It had as input parameters to the hydrograph calculated by the hydrological model, to the digital topography of the land, and to rheological parameters such as viscosity and yield strength. Then, the maximum speeds and depths of the return periods mentioned above were obtained. Finally, these two models have been integrated into the structure of an early warning system against hurricanes and the implementation of a meteorological radar for the stream under study has been suggested. / Tesis

Le massif du Cantal, localisé dans la province volcanique Cénozoïque du Massif Central (France), est le plus grand complexe volcanique Miocène d'Europe. Il est majoritairement constitué de brèches volcanoclastiques qui lui confèrent des dimensions impressionnantes et une morphologie surbaissée. Des observations de terrain, couplées à des datations K-Ar, nous ont permis de reconstituer l'évolution morpho-structurale, la dynamique éruptive et l'histoire volcanique du massif. Après avoir contraint dans l'espace et dans le temps une séquence stratigraphique identique tout autour du massif, nous montrons qu'il existe des évidences chrono-stratigraphiques, structurales et géomorphologiques en faveur de la formation d'une caldeira centrale de 8 x 10 km il y a environ 8 Ma occupée par un lac. L'étude morphologique et sédimentologique de la principale unité bréchique du massif, le Grand Écoulement Bréchique, nous a conduit à le définir comme l'un des plus importants debris flow syn-éruptifs connus au monde dont nous estimons un volume de l'ordre de 100 km3. Nous proposons que cet écoulement résulte d'une éruption sous-lacustre intracaldeira majeure. L'interaction entre un magma juvénile et l'eau du lac de caldeira, selon un dynamisme surtseyen, a initié la formation du debris flow qui s'est propagé sur les pentes externes du volcan à 360° jusqu'à plus de 25 km de sa source. Cette étude ouvre des perspectives d'une part quant aux risques volcaniques liés aux éruptions sous-lacustres intracaldériques majeures, et d'autre part quant à la possibilité de reconstituer la morphologie passée de la source d'un écoulement bréchique volcanoclastique âgé de plusieurs Ma, par l'étude de son dépôt selon des critères sédimentologiques et morphologiques.Dans un deuxième temps, la combinaison de nouvelles datations K-Ar et d'observations de terrain nous ont permis de reconstituer l'histoire volcanique du plateau du Cézallier, siège de la plus jeune activité volcanique de France métropolitaine, du plateau de l'Aubrac, et également la chronologie du volcanisme le long du Sillon Houiller. Nous avons pu ainsi établir les relations spatio-temporelles entre le massif du Cantal et les provinces volcaniques adjacentes révélant le rôle important des fractures héritées de l'Hercynien pour la remontée des magmas. Nous confirmons ainsi une migration spatio-temporelle vers le nord du volcanisme en Auvergne. Finalement, l'acquisition au cours de ce travail de 47 nouveaux âges K-Ar de 12,8 Ma à 9 ka complète la chronologie du volcanisme du Massif Central en précisant sa distribution spatio-temporelle.

This thesis seeks to analyze the occurrence and behavior of rainfall-triggered landslides in coastal British Columbia. In particular, it focuses on the analysis of landslide temporal and spatial distributions occurrence and their magnitudes, and considers the major factors that influence regional landslide behavior. Implicit in the research is the understanding that the landscape of coastal BC is managed, and that landslides, in addition to occurring naturally may be caused by, and certainly impact, resources that are important to humankind. Underlying each chapter is the rationale that by better understanding the causes of, and controls on landslide occurrence and magnitude, we can reduce the impacts and lower the associated risk. Statistical magnitude-frequency relationships are examined in coastal BC. Observations suggest that landslides in coastal British Columbia tend to a larger size until about 10,000 m2 in total area. At this point larger landslides are limited by landscape controls according to a power law. Probabilistic regional hazard analysis is one logical outcome of magnitude-frequency analysis and a regional mass movement hazard map for Vancouver Island is presented. Physiographic controls on statistical magnitude-frequency distributions are examined using a cellular automata based model and results compare favorably to actual landslide behavior: modeled landslides bifurcate at local elevation highs, deposit mass preferentially where the local slopes decrease, find routes in confined valley or channel networks, and, when sufficiently large, overwhelm the local topography. The magnitude-frequency distribution of both the actual landslides and the cellular automata model follow a power law for magnitudes higher than 10,000 m2 - 20,000 m2 and show a flattening of the slope for smaller magnitudes. The results provide strong corroborative evidence for physiographic limitations related to slope, slope distance and the distribution of mass within landslides. The physiographic controls on landslide magnitude, debris flow mobility and runout behavior is examined using detailed field and air photograph analysis. The role of slope on deposition and scour is investigated and a practical method for estimating both entrainment and runout in the field, as well as in the GIS environment, is presented. Further controls on landslide mobility, including the role of gullies and stream channels, roads and benches and intact forests, are considered. The role of landslides in controlling landscape physiography is also examined. In particular, it is determined that moderate-sized landslides do the most work transporting material on hillslopes, defined by a work peak, and that magnitude varies based on local physiography and climate. Landslides that form the work peak are distinct from catastrophic landslides that are themselves formative and system resetting. The persistence time for debris slides/debris flows and rock slides/rock avalanches is calculated over six orders of magnitude and an event is considered catastrophic when it persists in the landscape ten times longer than the population of landslides that form the work peak. A detailed case study examines meteorological controls on landslide occurrence and the role of extreme weather is considered. A critical onset of landslide triggering rainfall intensity is determined to be between 80 mm and 100 mm in 24 hours and wind is determined to result in increased local precipitation. The role of rain-on-snow is also evaluated and determined to be crucial to landslide occurrence. Finally, a conceptual model of landslide-induced denudation for coastal mountain watersheds spanning 10,000 years of environmental change is presented. Recent human impacts are calculated for landslide frequencies over the 20th century. The impact of logging during the last 100 years is unambiguous; logging induced landslides almost doubles the effect frequency of the wettest millennia in the last 10,000 years. This suggests that the impact of logging outpaces that of climatic change. Debris slides and debris flows are estimated to have resulted in a landscape lowering of 0.7 m across the Vancouver Island during the last 10,000 years.

Disaster impacts have grown significantly in the last half century. Additionally, in 2007 the number of urban residents surpassed that of rural populations generating a struggle for resources, inevitably leading to increased challenges in the achievement of reduction in both urban poverty and disaster vulnerability. Although the literature on disaster recovery suggests that a ‘build back better’ approach is now the accepted norm, there are still many cases worldwide where the reconstruction process actually rebuilds rather than reduces vulnerabilities. The literature on disaster risk reduction provides some basic principles for sustainable hazard mitigation, however, evaluation criteria for effective post-disaster response and reconstruction have yet to be developed. This research will enrich the ongoing debate about what ‘sustainable hazard mitigation’ entails and how it fits into broader development goals in less developed countries (LDCs). A case study examination of a socio-ecological system allows for the identification of the ways in which planning, policy, partnerships, and the like, can be used to reduce vulnerabilities in a post-disaster setting, thus, improving outcomes in future disastrous events. In the absence of a framework for evaluation of disaster risk reduction effectiveness in the literature, an Assessment of Post-disaster Risk Reduction Effectiveness (PDARRE) was created. Thirty three criteria were derived from the literature and ‘good practices’ to address common challenges and necessary actions for successful post-disaster reconstruction which results in reduced vulnerability. The selected case study is a debris flow disaster which decimated the northern coastal state of Vargas, Venezuela in 1999. Torrential rainfall exceeding 900mm fell on the Sierra El Avila Mountains over three days. The results from the PDARRE evaluation found an overwhelmingly poor response to the Vargas disaster, although some positive actions were also noted. Individual community members were not well-informed of the risks they faced living in Vargas and have still not been provided adequate capacity to reduce their vulnerability, nine years after the disaster. The creation of new institutions immediately following the debris flows led to slow decision-making and weak governance as new managers struggled to adapt to their new positions. In addition, poor communication across government institutions, lack of enforcement of zoning policies and an incomplete system of early warning compounded vulnerability and governance concerns. Long-term monitoring of post-disaster recovery and reconstruction has typically been left to local governments that often get distracted by economic pressures and changes. To assist with post-disaster efforts, PDARRE was created to monitor and evaluate effectiveness. The criteria for this assessment were derived from many sources and were organized into categories to assist local governments to see which areas of their disaster response system are weakest, and enable effective adjustments to their activities, consequently improving the entire disaster management system. Though other checklists and tools for post-disaster response activities do exist, I argue that these over-emphasize the immediate response activities and time-frame. The post-disaster context provides an opportunity to harness funding that can be directed at-risk and vulnerability reduction efforts. Consistent with the perspective of prominent international NGOs, this research is based on the prevailing belief that disaster management can be more successful if mainstreamed into broader sustainable development goals and activities. Similar to other disasters, the Vargas debris flow disaster was a convergence of unfortunate and dangerous circumstances. As disasters continue to grow in magnitude and increase in frequency, the importance of strong disaster management plans will be reinforced the world over. With a synthesis of poverty and vulnerability reduction strategies, disaster-affected communities can use the post-disaster context as an opportunity to achieve more sustainable livelihoods, increased equity and improved safety.

This thesis seeks to analyze the occurrence and behavior of rainfall-triggered landslides in coastal British Columbia. In particular, it focuses on the analysis of landslide temporal and spatial distributions occurrence and their magnitudes, and considers the major factors that influence regional landslide behavior. Implicit in the research is the understanding that the landscape of coastal BC is managed, and that landslides, in addition to occurring naturally may be caused by, and certainly impact, resources that are important to humankind. Underlying each chapter is the rationale that by better understanding the causes of, and controls on landslide occurrence and magnitude, we can reduce the impacts and lower the associated risk. Statistical magnitude-frequency relationships are examined in coastal BC. Observations suggest that landslides in coastal British Columbia tend to a larger size until about 10,000 m2 in total area. At this point larger landslides are limited by landscape controls according to a power law. Probabilistic regional hazard analysis is one logical outcome of magnitude-frequency analysis and a regional mass movement hazard map for Vancouver Island is presented. Physiographic controls on statistical magnitude-frequency distributions are examined using a cellular automata based model and results compare favorably to actual landslide behavior: modeled landslides bifurcate at local elevation highs, deposit mass preferentially where the local slopes decrease, find routes in confined valley or channel networks, and, when sufficiently large, overwhelm the local topography. The magnitude-frequency distribution of both the actual landslides and the cellular automata model follow a power law for magnitudes higher than 10,000 m2 - 20,000 m2 and show a flattening of the slope for smaller magnitudes. The results provide strong corroborative evidence for physiographic limitations related to slope, slope distance and the distribution of mass within landslides. The physiographic controls on landslide magnitude, debris flow mobility and runout behavior is examined using detailed field and air photograph analysis. The role of slope on deposition and scour is investigated and a practical method for estimating both entrainment and runout in the field, as well as in the GIS environment, is presented. Further controls on landslide mobility, including the role of gullies and stream channels, roads and benches and intact forests, are considered. The role of landslides in controlling landscape physiography is also examined. In particular, it is determined that moderate-sized landslides do the most work transporting material on hillslopes, defined by a work peak, and that magnitude varies based on local physiography and climate. Landslides that form the work peak are distinct from catastrophic landslides that are themselves formative and system resetting. The persistence time for debris slides/debris flows and rock slides/rock avalanches is calculated over six orders of magnitude and an event is considered catastrophic when it persists in the landscape ten times longer than the population of landslides that form the work peak. A detailed case study examines meteorological controls on landslide occurrence and the role of extreme weather is considered. A critical onset of landslide triggering rainfall intensity is determined to be between 80 mm and 100 mm in 24 hours and wind is determined to result in increased local precipitation. The role of rain-on-snow is also evaluated and determined to be crucial to landslide occurrence. Finally, a conceptual model of landslide-induced denudation for coastal mountain watersheds spanning 10,000 years of environmental change is presented. Recent human impacts are calculated for landslide frequencies over the 20th century. The impact of logging during the last 100 years is unambiguous; logging induced landslides almost doubles the effect frequency of the wettest millennia in the last 10,000 years. This suggests that the impact of logging outpaces that of climatic change. Debris slides and debris flows are estimated to have resulted in a landscape lowering of 0.7 m across the Vancouver Island during the last 10,000 years.

Disaster impacts have grown significantly in the last half century. Additionally, in 2007 the number of urban residents surpassed that of rural populations generating a struggle for resources, inevitably leading to increased challenges in the achievement of reduction in both urban poverty and disaster vulnerability. Although the literature on disaster recovery suggests that a ‘build back better’ approach is now the accepted norm, there are still many cases worldwide where the reconstruction process actually rebuilds rather than reduces vulnerabilities. The literature on disaster risk reduction provides some basic principles for sustainable hazard mitigation, however, evaluation criteria for effective post-disaster response and reconstruction have yet to be developed. This research will enrich the ongoing debate about what ‘sustainable hazard mitigation’ entails and how it fits into broader development goals in less developed countries (LDCs). A case study examination of a socio-ecological system allows for the identification of the ways in which planning, policy, partnerships, and the like, can be used to reduce vulnerabilities in a post-disaster setting, thus, improving outcomes in future disastrous events. In the absence of a framework for evaluation of disaster risk reduction effectiveness in the literature, an Assessment of Post-disaster Risk Reduction Effectiveness (PDARRE) was created. Thirty three criteria were derived from the literature and ‘good practices’ to address common challenges and necessary actions for successful post-disaster reconstruction which results in reduced vulnerability. The selected case study is a debris flow disaster which decimated the northern coastal state of Vargas, Venezuela in 1999. Torrential rainfall exceeding 900mm fell on the Sierra El Avila Mountains over three days. The results from the PDARRE evaluation found an overwhelmingly poor response to the Vargas disaster, although some positive actions were also noted. Individual community members were not well-informed of the risks they faced living in Vargas and have still not been provided adequate capacity to reduce their vulnerability, nine years after the disaster. The creation of new institutions immediately following the debris flows led to slow decision-making and weak governance as new managers struggled to adapt to their new positions. In addition, poor communication across government institutions, lack of enforcement of zoning policies and an incomplete system of early warning compounded vulnerability and governance concerns. Long-term monitoring of post-disaster recovery and reconstruction has typically been left to local governments that often get distracted by economic pressures and changes. To assist with post-disaster efforts, PDARRE was created to monitor and evaluate effectiveness. The criteria for this assessment were derived from many sources and were organized into categories to assist local governments to see which areas of their disaster response system are weakest, and enable effective adjustments to their activities, consequently improving the entire disaster management system. Though other checklists and tools for post-disaster response activities do exist, I argue that these over-emphasize the immediate response activities and time-frame. The post-disaster context provides an opportunity to harness funding that can be directed at-risk and vulnerability reduction efforts. Consistent with the perspective of prominent international NGOs, this research is based on the prevailing belief that disaster management can be more successful if mainstreamed into broader sustainable development goals and activities. Similar to other disasters, the Vargas debris flow disaster was a convergence of unfortunate and dangerous circumstances. As disasters continue to grow in magnitude and increase in frequency, the importance of strong disaster management plans will be reinforced the world over. With a synthesis of poverty and vulnerability reduction strategies, disaster-affected communities can use the post-disaster context as an opportunity to achieve more sustainable livelihoods, increased equity and improved safety.

A series of deposits, located along the southern flanks of Grand Mesa, Colorado, and extending to the south, are problematic, and the processes related to emplacement are not understood. The overall area is dominated by two landform systems, Grand Mesa, which supported a Pleistocene ice cap, and the North Fork Gunnison River drainage. Thus, one has to ask: Are these deposits the result of the melting of the ice cap or are they fluvial terraces associated with the evolution of the ancestral Gunnison River? The goal of this research was to map the areal extent of the deposits and to interpret the formation and climatic significance in understanding the evolution of the Pleistocene landscape in the region. An extensive exposure, parallel to State Highway 65 near Cory Grade, was used for detailed description and sampling. Three additional exposures, ~10 to 20 km (~6 to 12 mi) were used to extend the areal extent of sampling. The study area was mapped using aerial photography and traditional field mapping aided by GPS. From the field work, a detailed stratigraphic column, including lithology and erodability, was constructed. Vertical exposures of the deposits were described, mapped, and recorded in the field and using detailed photo mosaics. Samples were collected from each stratum of the deposits for grain-size, shape, and sorting analyses. Five distinct depositional facies were identified. Sieve analysis on collected samples shows that four distinct grain-sizes occur in the outcrops; coarse sand, very-coarse sand, granule, and pebble and boulder. Mean grain-sizes range from 0.0722 to 0.9617, -0.0948 to -0.9456, -1.0566 to -1.9053, and -2.0050 to -3.4643, respectively. Glacio-fluvial depositional environments were identified and supported with observations of sedimentary structures and clast composition. Two major environments of deposition are recorded in the deposits; fluvial deposits from glacial outburst floods, and debris flow deposits. Imbrication of clasts in the strata suggests the flow came from the direction of Grand Mesa to the north. Facies and subsequent sequences were constructed to portray evidence that supports the glacio-fluvial mode of deposition.

Glacial floods and mass movements of ice, rock or debris are a significant hazard in many populated mountainous regions, often with devastating impacts upon human settlements and infrastructure. In response to atmospheric warming, glacial retreat and permafrost thaw are expected to alter high mountain geomorphic processes, and related instabilities. In the Aoraki/Mount Cook region of New Zealand's Southern Alps, a first investigation of geomorphic hazards associated with glacial change is undertaken and is based primarily on the use of remote sensing and Geographic Information Systems (GIS) for mapping, modelling, and analysing related processes and terrain. Following a comprehensive review of available techniques, remote sensing methods involving the use Advanced Spaceborne Thermal Emission and Radiometer (ASTER) imagery were applied to map glacial ice, lakes and debris accumulations in the Aoraki/Mount Cook region. Glacial lakes were mapped from two separate classification techniques using visible near infrared wavelengths, capturing highly turbid and clearer water bodies. Large volume (10⁶– 10⁸ m³) proglacial lakes have developed rapidly over recent decades, with an overall 20 % increase in lake area recorded between 2002 and 2006, increasing the potential for large mass movement impacts and flooding from displaced water. Where significant long-term glacial recession has occurred, steep moraines have been exposed, and large talus slopes occupy formerly glaciated slopes at higher elevations. At the regional-scale, these potential source areas for debris instabilities were distinguished from surrounding bedrock slopes based on image texture variance. For debris and ice covered slopes, potentially unstable situations were classified using critical slope thresholds established from international studies. GIS-based flow routing was used to explore possible intersections between zones of human use and mass movement or flood events, assuming worst-case, probable maximum runout distances. Where glacial lakes are dammed by steep moraine or outwash gravel, primarily in cirque basins east of the Main Divide, modelled debris flows initiated by potential flood events did not reach any infrastructure. Other potential peri- and para-glacial debris flows from steep moraines or talus slopes can reach main roads and buildings. The direct hazard from ice avalanches is restricted to backcountry huts and walking tracks, but impacts into large glacial lakes are possible, and could produce a far reaching hazard, with modelled clear water flood-waves capable of reaching village infrastructure and main roads both east and west of the Main Divide. A numerical modelling approach for simulating large bedrock failures has been introduced, and offers potential with which to examine possible lake impacts and related scenarios. Over 500 bedrock slope failures were analysed within a GIS inventory, revealing distinct patterns in geological and topographic distribution. Rock avalanches have occurred most frequently from greywacke slopes about and east of the Main Divide, particularly from slopes steeper than 50°, and appear the only large-magnitude failure mechanism above 2500 m. In the schist terrain west of the Main Divide, and at lower elevations, other failure types predominate. The prehistoric distribution of all failure types suggests a preference for slopes facing west to northwest, and is likely to be strongly influenced by earthquake generated failures. Over the past 100 years, seismicity has not been a factor, and the most failures have been as rock avalanches from slopes facing east to southeast, particularly evident from the glaciated, and potentially permafrost affected hangingwall of the Main Divide Fault Zone. An initial estimate of permafrost distribution based on topo-climatic relationships and calibrated locally using mean annual air temperature suggested permafrost may extend down to elevations of 3000 m on sunny slopes, and as low as 2200 m on shaded slopes near the Main Divide. A network of 15 near-surface rock temperature sensors was installed on steep rock walls, revealing marginal permafrost conditions (approaching 0 °C) extending over a much larger elevation range, occurring even where air temperature is likely to remain positive, owing to extreme topographic shading. From 19 rock failures observed over the past 100 years, 13 detachment zones were located on slopes characterized by marginal permafrost conditions, including a sequence of 4 failures that occurred during summer 2007/08, in which modelled bedrock temperatures near the base of the detachments were in the range of 1.4 to +2.5 °C. Ongoing monitoring of glacial and permafrost conditions in the Aoraki/Mount Cook region is encouraged, with more than 45 km2 of extremely steep slopes (>50°) currently ice covered or above modelled permafrost elevation limits. Approaches towards modelling and analysing glacial hazards in this region are considered to be most applicable within other remote mountain regions, where seismicity and steep topography combine with possible destabilizing influences of glacial recession and permafrost degradation.

River blockages formed by rock avalanches appear to pose a higher hazard potential than other landslide dams, given the extreme run-out distances and volumes of rock avalanche deposits. Recent research has identified rock avalanche deposits to have internal sedimentology consisting of a coarse surficial material (carapace) and a finer fragmented interior (body) potentially of critical importance to rock-avalanche dam stability. Physical scale modelling of overtopping failure and breach development in rock avalanche dams was used to quantify the influence of this sedimentology on critical breach parameters, and their prediction using existing embankment dam breach technologies. Results from this study indicate that the time to failure for rock avalanche dams is approximately twice that observed for homogeneous dams due to the armouring properties of the carapace; and that peak discharge is not significantly affected by sedimentology. While application of empirical, parametric, dimensional and physically based models indicated that uncertainty associated with predicted dam break discharges could range from ±19% to ±107%, no modelling technique was able to simulate the armouring phenomenon adequately. Comparison of actual and simulated breach evolution shows linear assumptions of breach depth and width development (as observed in homogeneous dams) to be incorrect. In the context of hazard management, the results suggest that empirical regression relationships should be used for rapid assessment of potential dam break flood magnitude.

Glacial floods and mass movements of ice, rock or debris are a significant hazard in many populated mountainous regions, often with devastating impacts upon human settlements and infrastructure. In response to atmospheric warming, glacial retreat and permafrost thaw are expected to alter high mountain geomorphic processes, and related instabilities. In the Aoraki/Mount Cook region of New Zealand's Southern Alps, a first investigation of geomorphic hazards associated with glacial change is undertaken and is based primarily on the use of remote sensing and Geographic Information Systems (GIS) for mapping, modelling, and analysing related processes and terrain. Following a comprehensive review of available techniques, remote sensing methods involving the use Advanced Spaceborne Thermal Emission and Radiometer (ASTER) imagery were applied to map glacial ice, lakes and debris accumulations in the Aoraki/Mount Cook region. Glacial lakes were mapped from two separate classification techniques using visible near infrared wavelengths, capturing highly turbid and clearer water bodies. Large volume (10⁶– 10⁸ m³) proglacial lakes have developed rapidly over recent decades, with an overall 20 % increase in lake area recorded between 2002 and 2006, increasing the potential for large mass movement impacts and flooding from displaced water. Where significant long-term glacial recession has occurred, steep moraines have been exposed, and large talus slopes occupy formerly glaciated slopes at higher elevations. At the regional-scale, these potential source areas for debris instabilities were distinguished from surrounding bedrock slopes based on image texture variance. For debris and ice covered slopes, potentially unstable situations were classified using critical slope thresholds established from international studies. GIS-based flow routing was used to explore possible intersections between zones of human use and mass movement or flood events, assuming worst-case, probable maximum runout distances. Where glacial lakes are dammed by steep moraine or outwash gravel, primarily in cirque basins east of the Main Divide, modelled debris flows initiated by potential flood events did not reach any infrastructure. Other potential peri- and para-glacial debris flows from steep moraines or talus slopes can reach main roads and buildings. The direct hazard from ice avalanches is restricted to backcountry huts and walking tracks, but impacts into large glacial lakes are possible, and could produce a far reaching hazard, with modelled clear water flood-waves capable of reaching village infrastructure and main roads both east and west of the Main Divide. A numerical modelling approach for simulating large bedrock failures has been introduced, and offers potential with which to examine possible lake impacts and related scenarios. Over 500 bedrock slope failures were analysed within a GIS inventory, revealing distinct patterns in geological and topographic distribution. Rock avalanches have occurred most frequently from greywacke slopes about and east of the Main Divide, particularly from slopes steeper than 50°, and appear the only large-magnitude failure mechanism above 2500 m. In the schist terrain west of the Main Divide, and at lower elevations, other failure types predominate. The prehistoric distribution of all failure types suggests a preference for slopes facing west to northwest, and is likely to be strongly influenced by earthquake generated failures. Over the past 100 years, seismicity has not been a factor, and the most failures have been as rock avalanches from slopes facing east to southeast, particularly evident from the glaciated, and potentially permafrost affected hangingwall of the Main Divide Fault Zone. An initial estimate of permafrost distribution based on topo-climatic relationships and calibrated locally using mean annual air temperature suggested permafrost may extend down to elevations of 3000 m on sunny slopes, and as low as 2200 m on shaded slopes near the Main Divide. A network of 15 near-surface rock temperature sensors was installed on steep rock walls, revealing marginal permafrost conditions (approaching 0 °C) extending over a much larger elevation range, occurring even where air temperature is likely to remain positive, owing to extreme topographic shading. From 19 rock failures observed over the past 100 years, 13 detachment zones were located on slopes characterized by marginal permafrost conditions, including a sequence of 4 failures that occurred during summer 2007/08, in which modelled bedrock temperatures near the base of the detachments were in the range of 1.4 to +2.5 °C. Ongoing monitoring of glacial and permafrost conditions in the Aoraki/Mount Cook region is encouraged, with more than 45 km2 of extremely steep slopes (>50°) currently ice covered or above modelled permafrost elevation limits. Approaches towards modelling and analysing glacial hazards in this region are considered to be most applicable within other remote mountain regions, where seismicity and steep topography combine with possible destabilizing influences of glacial recession and permafrost degradation.