Validering av modellerad skredkänslighet i finkornig jordart | Validation of Modeled Landslide Susceptibility in Fine Grained Soils in Sweden

Skred är en av de naturliga processer som formar landskapet omkring oss. De kan dock, om de sker ibebyggda områden, orsaka stor skada på byggnader, infrastruktur och utgöra en fara för människors liv.Det är därför viktigt att kunna identifiera områden där skred potentiellt kan ske så att lämpliga åtgärderkan vidtas i tid. En modell har tagits fram av Statens geotekniska institut (SGI) och Sveriges geologiskaundersökning (SGU) för att ge en första indikation på var skred kan ske. Modellen baseras på en algoritmav Tryggvason et al. (2015) där områden som består av finkorniga jordarter och där lutningsförhållandetär minst 1:10 betraktas som skredkänsliga. Syftet med denna studie var att undersöka till vilken gradmodellens aktsamhetsområden sammanfaller med tidigare skred från SGI och SGUs databaser. 90 - 94% av de tidigare skreden överlappade med modellerade aktsamhetsområden. Möjliga anledningar till attskred hamnat utanför dessa områden undersöktes även. Att de befinner sig i en jordart som inte betraktassom skredkänslig och att lutningen numera är lägre än tröskelvärdet för modellen var de huvudsakligaanledningarna till detta.

A landslide is a type of mass movement down a slope. It is a natural part of the evolution of the landscapearound us but can cause extensive damages on buildings, infrastructure and pose a threat to human livesif they occur in populated areas. It is therefore important to know which areas that are prone to landslidesso that appropriate measures can be taken in time. It is possible to calculate how stable a certain soil isby taking samples of it and testing it in the lab. In these tests it is determined how sensitive the soil is tovibrations, a higher water content and/or if it is remolded. However, this takes time and require a lot ofwork. Since not all soil types are equally sensitive these tests do not have to be performed on everyslope, but it is important that the most sensitive areas are not overlooked. For this reason, a model hasbeen developed that displays areas where the slope stability might need to be examined prior, forexample, larger infrastructure projects are started. From previous studies it has been found thatlandslides mostly occur in fine grained soils, like silt and clay, and where the slope is steeper than 5.7degrees. Areas that consist of any of these soil types and has a slope over this threshold are consideredpotentially sensitive to landslides in the model. The purpose of this study was to assess how well themodel is at identifying areas that might be prone to landslides. This was done by determining how manyof previous landslides, that are registered in two databases, that fall within the areas marked aspotentially sensitive. Why some landslides were located outside of these areas was also examined. Themain reasons were that the soil type the landslide occurred in is not considered sensitive by the modelor the inclination of the slope have changed since the landslide occurred. 90 - 94 % of the previouslandslides were found to be located within areas that the model point out as potentially sensitive.