| Abstract | Worldwide, the higher frequency of natural disasters observed during the last century and the demographic constant growth, which led to the demand in land-use areas and the intensification of
urbanisation, have increased the need to mitigate the risk of natural hazards. In the case of landslide hazards, inventorying and susceptibility mapping techniques are used, in urbanised and remote areas, when designing, planning, and building
infrastructures. In the Canadian Arctic, the presence of permafrost terrains is an important parameter to consider when planning the construction of linear infrastructures, like pipelines, because of their sensitivity to the soil thermal regime
variation, caused either by the effects of global climate warming or human activities.
In this perspective, NRCan initiated a regional landslide hazards mapping project to map landslide-prone areas in the Mackenzie Valley (MV). Worldwide needs
for energy led the oil and gas industry to explore frontier regions, such as the MV, where a new natural gas pipeline of 1,220 km in length has been proposed to connect the Mackenzie Delta reservoirs to the Alberta facilities. The study area of
approximately 24,000 km2 was defined along the Inuvik-Norman Wells section of the future gas pipeline. This initiative aims to 1) provide baseline knowledge on the types, regional distribution, and controlling factors of landslides and 2) develop a
landslide susceptibility mapping methodology.
To fill the knowledge gap, a preliminary landslide inventory was created that contains more than 1,800 features and covers approximately 40% of the study area. Landslide limits, points, and associated
attributes such as the landslide type were combined to create the MV Landslide Spatial Database - Version 1.0 (MVLSD). In 2007, the database will be published as an ESS Open File and a Keyhole Markup Language (KML) file, which can be visualised in
Google Earth ™. Preliminary results indicate an average density of one landslide per 5 km2 and show that the dominant landslide types are retrogressive thaw flows (28%) and active layer detachments (25%). About 46% of all landslides took place in
morainal deposits. Furthermore, a landslide density map and a preliminary analysis of forest fires, which are an important landslide controlling factor in a permafrost environment, will be presented. Finally, the developed landslide susceptibility
mapping qualitative parametric methodology will be described, which is based on spatial data analysis modelling. The Travaillant Lake-Thunder River area was selected as a test site where six layers were used as input in the model: surficial geology,
permafrost type, permafrost ice content, slope angle, slope aspect, and land cover. This methodology was proven effective as a good correlation was observed when superposing landslide locations from the MVLSD over the landslide susceptibility map. In
conclusion, landslide inventory, density, and susceptibility maps are important and useful planning tools to help defining appropriate and safe locations for future infrastructures. |