Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps
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,...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
Multidisciplinary Digital Publishing Institute (MDPI)
2021
|
Subjects: | |
Online Access: | https://epic.awi.de/id/eprint/54883/ https://epic.awi.de/id/eprint/54883/1/remotesensing-13-04294.pdf https://doi.org/10.3390/rs13214294 https://hdl.handle.net/10013/epic.214c5215-4dab-41fd-aa57-131f5892ef8d |
id |
ftawi:oai:epic.awi.de:54883 |
---|---|
record_format |
openpolar |
spelling |
ftawi:oai:epic.awi.de:54883 2024-03-24T08:59:43+00:00 Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps Nitze, Ingmar Heidler, Konrad Barth, Sophia Grosse, Guido 2021-10-26 application/pdf https://epic.awi.de/id/eprint/54883/ https://epic.awi.de/id/eprint/54883/1/remotesensing-13-04294.pdf https://doi.org/10.3390/rs13214294 https://hdl.handle.net/10013/epic.214c5215-4dab-41fd-aa57-131f5892ef8d unknown Multidisciplinary Digital Publishing Institute (MDPI) https://epic.awi.de/id/eprint/54883/1/remotesensing-13-04294.pdf Nitze, I. orcid:0000-0002-1165-6852 , Heidler, K. , Barth, S. and Grosse, G. orcid:0000-0001-5895-2141 (2021) Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps , Remote Sensing, 13 (21), p. 4294 . doi:10.3390/rs13214294 <https://doi.org/10.3390/rs13214294> , hdl:10013/epic.214c5215-4dab-41fd-aa57-131f5892ef8d EPIC3Remote Sensing, Multidisciplinary Digital Publishing Institute (MDPI), 13(21), pp. 4294-4294, ISSN: 2072-4292 Article isiRev 2021 ftawi https://doi.org/10.3390/rs13214294 2024-02-27T09:55:26Z 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-ofthe- 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. Article in Journal/Newspaper Arctic Banks Island Herschel Island Kolguev lena river permafrost Alfred Wegener Institute for Polar- and Marine Research (AWI): ePIC (electronic Publication Information Center) Arctic Canada Herschel Island ENVELOPE(-139.089,-139.089,69.583,69.583) Tuktoyaktuk ENVELOPE(-133.006,-133.006,69.425,69.425) Remote Sensing 13 21 4294 |
institution |
Open Polar |
collection |
Alfred Wegener Institute for Polar- and Marine Research (AWI): ePIC (electronic Publication Information Center) |
op_collection_id |
ftawi |
language |
unknown |
description |
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-ofthe- 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. |
format |
Article in Journal/Newspaper |
author |
Nitze, Ingmar Heidler, Konrad Barth, Sophia Grosse, Guido |
spellingShingle |
Nitze, Ingmar Heidler, Konrad Barth, Sophia Grosse, Guido Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps |
author_facet |
Nitze, Ingmar Heidler, Konrad Barth, Sophia Grosse, Guido |
author_sort |
Nitze, Ingmar |
title |
Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps |
title_short |
Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps |
title_full |
Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps |
title_fullStr |
Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps |
title_full_unstemmed |
Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps |
title_sort |
developing and testing a deep learning approach for mapping retrogressive thaw slumps |
publisher |
Multidisciplinary Digital Publishing Institute (MDPI) |
publishDate |
2021 |
url |
https://epic.awi.de/id/eprint/54883/ https://epic.awi.de/id/eprint/54883/1/remotesensing-13-04294.pdf https://doi.org/10.3390/rs13214294 https://hdl.handle.net/10013/epic.214c5215-4dab-41fd-aa57-131f5892ef8d |
long_lat |
ENVELOPE(-139.089,-139.089,69.583,69.583) ENVELOPE(-133.006,-133.006,69.425,69.425) |
geographic |
Arctic Canada Herschel Island Tuktoyaktuk |
geographic_facet |
Arctic Canada Herschel Island Tuktoyaktuk |
genre |
Arctic Banks Island Herschel Island Kolguev lena river permafrost |
genre_facet |
Arctic Banks Island Herschel Island Kolguev lena river permafrost |
op_source |
EPIC3Remote Sensing, Multidisciplinary Digital Publishing Institute (MDPI), 13(21), pp. 4294-4294, ISSN: 2072-4292 |
op_relation |
https://epic.awi.de/id/eprint/54883/1/remotesensing-13-04294.pdf Nitze, I. orcid:0000-0002-1165-6852 , Heidler, K. , Barth, S. and Grosse, G. orcid:0000-0001-5895-2141 (2021) Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps , Remote Sensing, 13 (21), p. 4294 . doi:10.3390/rs13214294 <https://doi.org/10.3390/rs13214294> , hdl:10013/epic.214c5215-4dab-41fd-aa57-131f5892ef8d |
op_doi |
https://doi.org/10.3390/rs13214294 |
container_title |
Remote Sensing |
container_volume |
13 |
container_issue |
21 |
container_start_page |
4294 |
_version_ |
1794399592614199296 |