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,...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Ingmar Nitze, Konrad Heidler, Sophia Barth, Guido Grosse
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	deep learning image segmentation permafrost thaw semantic segmentation disturbances computer vision Science Q
Acceso en línea:	https://doaj.org/article/b13024976aac4c3ebe8ec61149dc2bb5
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:b13024976aac4c3ebe8ec61149dc2bb5
record_format	dspace
spelling	oai:doaj.org-article:b13024976aac4c3ebe8ec61149dc2bb52021-11-11T18:53:12ZDeveloping and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps10.3390/rs132142942072-4292https://doaj.org/article/b13024976aac4c3ebe8ec61149dc2bb52021-10-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/21/4294https://doaj.org/toc/2072-4292In 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.Ingmar NitzeKonrad HeidlerSophia BarthGuido GrosseMDPI AGarticledeep learningimage segmentationpermafrost thawsemantic segmentationdisturbancescomputer visionScienceQENRemote Sensing, Vol 13, Iss 4294, p 4294 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	deep learning image segmentation permafrost thaw semantic segmentation disturbances computer vision Science Q
spellingShingle	deep learning image segmentation permafrost thaw semantic segmentation disturbances computer vision Science Q Ingmar Nitze Konrad Heidler Sophia Barth Guido Grosse Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps
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-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.
format	article
author	Ingmar Nitze Konrad Heidler Sophia Barth Guido Grosse
author_facet	Ingmar Nitze Konrad Heidler Sophia Barth Guido Grosse
author_sort	Ingmar Nitze
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	MDPI AG
publishDate	2021
url	https://doaj.org/article/b13024976aac4c3ebe8ec61149dc2bb5
work_keys_str_mv	AT ingmarnitze developingandtestingadeeplearningapproachformappingretrogressivethawslumps AT konradheidler developingandtestingadeeplearningapproachformappingretrogressivethawslumps AT sophiabarth developingandtestingadeeplearningapproachformappingretrogressivethawslumps AT guidogrosse developingandtestingadeeplearningapproachformappingretrogressivethawslumps
_version_	1718431720679669760

Developing and Testing a Deep Learning Approach for Mapping Retrogressive Thaw Slumps

Ejemplares similares