In a warming Arctic, permafrost-related disturbances, such as retrogressive thaw slumps (RTS), are becoming more abundant and dynamic, with serious implications for permafrost stability and bio-geochemical cycles on local to regional scales. Despite recent advances in the field of earth observation, many of these have remained undetected as RTS are highly dynamic, small, and scattered across the remote permafrost region. Here, we assessed the potential strengths and limitations of using deep learning for the automatic segmentation of RTS using PlanetScope satellite imagery, ArcticDEM and auxiliary datasets. We analyzed the transferability and potential for pan-Arctic upscaling and regional cross-validation, with independent training and validation regions, in six different thaw slump-affected regions in Canada and Russia. We further tested state-of-the-art model architectures (UNet, UNet++, DeepLabv3) and encoder networks to find optimal model configurations for potential upscaling to continental scales. The best deep learning models achieved mixed results from good to very good agreement in four of the six regions (maxIoU: 0.39 to 0.58; Lena River, Horton Delta, Herschel Island, Kolguev Island), while they failed in two regions (Banks Island, Tuktoyaktuk). Of the tested architectures, UNet++ performed the best. The large variance in regional performance highlights the requirement for a sufficient quantity, quality and spatial variability in the training data used for segmenting RTS across diverse permafrost landscapes, in varying environmental conditions. With our highly automated and configurable workflow, we see great potential for the transfer to active RTS clusters (e.g., Peel Plateau) and upscaling to much larger regions.
In this paper, we examine the influence of the 27 October 2012, M w 7.8 earthquake on landslide occurrence in the southern half of Haida Gwaii (formerly Queen Charlotte Islands), British Columbia, Canada. Our 1350 km 2 study area is undisturbed, primarily forested terrain that has not experienced road building or timber harvesting. Our inventory of landslide polygons is based on optical airborne and spaceborne images acquired between 2007 and 2018, from which we extracted and mapped 446 individual landslides (an average of 33 landslides per 100 km 2). The landslide rate in years without major earthquakes averages 19.4 per year, or 1.4/100 km 2 /year, and the annual average area covered by nonseismically triggered landslides is 35 ha/year. The number of landslides identified in imagery closely following the 2012 earthquake, and probably triggered by it, is 244 or an average of about 18 landslides per 100 km 2. These landslides cover a total area of 461 ha. In the following years-2013-2016 and 2016-2018-the number of landslides fell, respectively, to 26 and 13.5 landslides per year. In non-earthquake years, most landslides happen on south-facing slopes, facing the prevailing winds. In contrast, during or immediately after the earthquake, up to 32% of the landslides occurred on north and northwest-facing slopes. Although we could not find imagery from the day after the earthquake, overview reconnaissance flights 10 and 16 days later showed that most of the landslides were recent, suggesting they were co-seismic.
<p>Remotely sensed point clouds provide&#160;detailed structural data of landscapes and ecosystem characteristics. Especially in the analysis of forests and topography, this data type has proven its ability to derive relevant quantitative parameters&#160;such as biomass or subsidence rates.&#160;Arctic and boreal permafrost ecosystems are severely affected by climate change and resulting vegetation shifts,&#160;environmental&#160;disturbances, and permafrost thaw which lead&#160;to rapid changes in these northern environments that can be detected and characterized with point cloud datasets.&#160;In recent decades, the amount of point clouds acquired and generated in high-latitude regions by terrestrial (TLS), mobile (MLS), unmanned aerial system (UAS)&#160;based (ULS), up to airborne-based (ALS)&#160;LiDAR&#160;(Light detection and ranging) and Structure from Motion (SfM) has steadily increased. Multi-temporal datasets are available for a wide range of observation targets.</p> <p>The characteristics of the point clouds such as the extent of the area covered as well as the point density and thus the level of detail differ depending on the sensor, method, and the acquisition specifications. To use point cloud data for topographic, morphological, and forestry analysis, segmentation and classification of the point cloud into specific components such as individual trees, stems, foliage, or terrain features&#160;is essential. This is a time-consuming manual process and not feasible when addressing large datasets. Several previous analyses showed the potential for machine learning-based semantic segmentation of a single point cloud type, e.g., terrestrial LiDAR (TLS) with identical acquisition mode and sensor. We aim at an automated segmentation of different point cloud types generated by i)&#160;TLS, MLS, ULS and ALS&#160;as well as ii) SfM using (multi)spectral UAS and airborne image data to enable an analysis of Arctic and boreal permafrost ecosystems. Thereby, we will focus on the following questions:</p> <p>1) How can we reduce the time consuming process of labeling the point clouds?</p> <p>2) Can we train a model for segmentation using all point clouds or does transfer learning lead to better results?</p> <p>3) To what level of detail can we accurately segment and classify the different point&#160;cloud types?</p> <p>With this automated segmentation and classification, we aim to open up the possibility of exploiting the information contained in the multitude of point cloud data for a variety of ecological research applications.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.