DATA-DRIVEN ANALYSIS AND MAPPING OF THE POTENTIAL DISTRIBUTION OF MOUNTAIN PERMAFROST

• Main Author: Deluigi, Nicola
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: Université de Lausanne, Faculté des géosciences et de l'environnement 2018
• Subjects: mountain permafrost mapping; environmental modelling; machine learning; spatial data mining; cartographie du permafrost de montagne; modélisation environnemen- tale; apprentissage automatique; exploration de données spatiales; cartografia del permafrost di montagna; modellizzazione ambientale; apprendimento automatico; estrazione di dati spaziali; permafrost
• Online Access: https://serval.unil.ch/notice/serval:BIB_F417FD0D4407; https://serval.unil.ch/resource/serval:BIB_F417FD0D4407.P001/REF.pdf; http://nbn-resolving.org/urn/resolver.pl?urn=urn:nbn:ch:serval-BIB_F417FD0D44072

In alpine environments, mountain permafrost is defined as a thermal state of the ground and it corresponds to any lithosphere material that is at or below 0°C for at least two years. Its degradation is potentially leading to an increasing rock fall activity and sediment transfer rates. During the last 20 years, knowledge on this phenomenon has significantly improved thanks to many studies and monitoring projects, revealing an extremely discontinuous and complex spatial distribution, especially at the micro scale (scale of a specific landform; tens to several hundreds of metres). The objective of this thesis was the systematic and detailed investigation of the potential of data-driven techniques for mountain permafrost distribution modelling. Machine learning (ML) algorithms are able to consider a greater number of pa- rameters compared to classic approaches. Not only can permafrost distribution be modelled by using topo-climatic parameters as a proxy, but also by taking into ac- count known field permafrost evidences. These latter were collected in a sector of the Western Swiss Alps and they were mapped from field data (thermal and geoelectrical data) and ortho-image interpretations (rock glacier inventorying). A permafrost dataset was built from these evidences and completed with environmental and mor- phological predictors. Data were firstly analysed with feature relevance techniques in order to identify the statistical contribution of each controlling factor and to exclude non-relevant or redundant predictors. Five classification algorithms, belonging to statistics and machine learning, were then applied to the dataset and tested: Logistic regression (LR), linear and non-linear Support Vector Machines (SVM), Multilayer perceptrons (MLP) and Random forests (RF). These techniques inferred a classifica- tion function from labelled training data (pixels of permafrost absence and presence) to predict the permafrost occurrence where this was unknown. Classification performances, assessed with AUROC curves, ranged ...