Using data-driven algorithms for semi-automated geomorphological mapping.

In this paper, we compare the performance of two data-driven algorithms to deal with an automatic classification problem in geomorphology: Direct Sampling (DS) and Random Forest (RF). The main goal is to provide a semi-automated procedure for the geomorphological mapping of alpine environments, using a manually mapped zone as training dataset and predictor variables to infer the classification of a target zone. The applicability of DS to geomorphological classification was never investigated before. Instead, RF based classification has already been applied in few studies, but only with a limited number of geomorphological classes. The outcomes of both approaches are validated by comparing the eight detected classes with a geomorphological map elaborated on the field and considered as ground truth. Both DS and RF give satisfactory results and provide similar performances in term of accuracy and Cohen's Kappa values. The map obtained with RF presents a noisier spatial distribution of classes than when using DS, because DS takes into account the spatial dependence of the different classes. Results suggest that DS and RF are both suitable techniques for the semi-automated geomorphological mapping in alpine environments at regional scale, opening the way for further improvements.

Permafrost is warming at a global scale.

Permafrost warming has the potential to amplify global climate change, because when frozen sediments thaw it unlocks soil organic carbon. Yet to date, no globally consistent assessment of permafrost temperature change has been compiled. Here we use a global data set of permafrost temperature time series from the Global Terrestrial Network for Permafrost to evaluate temperature change across permafrost regions for the period since the International Polar Year (2007-2009). During the reference decade between 2007 and 2016, ground temperature near the depth of zero annual amplitude in the continuous permafrost zone increased by 0.39 ± 0.15 °C. Over the same period, discontinuous permafrost warmed by 0.20 ± 0.10 °C. Permafrost in mountains warmed by 0.19 ± 0.05 °C and in Antarctica by 0.37 ± 0.10 °C. Globally, permafrost temperature increased by 0.29 ± 0.12 °C. The observed trend follows the Arctic amplification of air temperature increase in the Northern Hemisphere. In the discontinuous zone, however, ground warming occurred due to increased snow thickness while air temperature remained statistically unchanged.

Data-driven mapping of the potential mountain permafrost distribution.

Existing mountain permafrost distribution models generally offer a good overview of the potential extent of this phenomenon at a regional scale. They are however not always able to reproduce the high spatial discontinuity of permafrost at the micro-scale (scale of a specific landform; ten to several hundreds of meters). To overcome this lack, we tested an alternative modelling approach using three classification algorithms belonging to statistics and machine learning: Logistic regression, Support Vector Machines and Random forests. These supervised learning techniques infer a classification function from labelled training data (pixels of permafrost absence and presence) with the aim of predicting the permafrost occurrence where it is unknown. The research was carried out in a 588km2 area of the Western Swiss Alps. Permafrost evidences were mapped from ortho-image interpretation (rock glacier inventorying) and field data (mainly geoelectrical and thermal data). The relationship between selected permafrost evidences and permafrost controlling factors was computed with the mentioned techniques. Classification performances, assessed with AUROC, range between 0.81 for Logistic regression, 0.85 with Support Vector Machines and 0.88 with Random forests. The adopted machine learning algorithms have demonstrated to be efficient for permafrost distribution modelling thanks to consistent results compared to the field reality. The high resolution of the input dataset (10m) allows elaborating maps at the micro-scale with a modelled permafrost spatial distribution less optimistic than classic spatial models. Moreover, the probability output of adopted algorithms offers a more precise overview of the potential distribution of mountain permafrost than proposing simple indexes of the permafrost favorability. These encouraging results also open the way to new possibilities of permafrost data analysis and mapping.