Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach
Seasonal or degradational thaw subsidence of permafrost terrain affects the landscape, hydrology, and sustainability of permafrost as an engineering substrate. We perform permafrost thaw sensitivity prediction via supervised classification of a feature set consisting of geological, topographic, and...
Published in: | Canadian Journal of Earth Sciences |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Canadian Science Publishing
2022
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1139/cjes-2021-0117 https://cdnsciencepub.com/doi/full-xml/10.1139/cjes-2021-0117 https://cdnsciencepub.com/doi/pdf/10.1139/cjes-2021-0117 |
id |
crcansciencepubl:10.1139/cjes-2021-0117 |
---|---|
record_format |
openpolar |
spelling |
crcansciencepubl:10.1139/cjes-2021-0117 2024-04-07T07:50:33+00:00 Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach Oldenborger, Greg A. Short, Naomi LeBlanc, Anne-Marie 2022 http://dx.doi.org/10.1139/cjes-2021-0117 https://cdnsciencepub.com/doi/full-xml/10.1139/cjes-2021-0117 https://cdnsciencepub.com/doi/pdf/10.1139/cjes-2021-0117 en eng Canadian Science Publishing http://www.nrcresearchpress.com/page/about/CorporateTextAndDataMining Canadian Journal of Earth Sciences volume 59, issue 11, page 897-913 ISSN 0008-4077 1480-3313 General Earth and Planetary Sciences journal-article 2022 crcansciencepubl https://doi.org/10.1139/cjes-2021-0117 2024-03-08T00:37:51Z Seasonal or degradational thaw subsidence of permafrost terrain affects the landscape, hydrology, and sustainability of permafrost as an engineering substrate. We perform permafrost thaw sensitivity prediction via supervised classification of a feature set consisting of geological, topographic, and multispectral variables over continuous permafrost near Rankin Inlet, Nunavut, Canada. We build a reference classification of thaw sensitivity using process-based categorization of seasonal subsidence as measured from differential interferometric synthetic aperture radar whereby categories of thaw sensitivity are reflective of ground ice conditions. Classification is performed using a neural network trained on both dispersed and parcel-based reference data. For Low, Medium, High, and Very High thaw sensitivity categories, generalized classification accuracy is 70.8% for 20.6 km 2 of dispersed training data. In all cases, the majority classes of Low and Medium thaw sensitivity are predicted with higher accuracy and more certainty, while the minority classes of High and Very High thaw sensitivity are underpredicted. Minority classes can be combined to improve accuracy at the expense of a reduced level of discrimination. The two-class problem can be classified with an accuracy of 81.8%, thereby effectively distinguishing between stable and unstable ground. The method is applicable to similar Low-Arctic permafrost terrain with geological and topographical controls on thaw sensitivity. However, generalized accuracy is reduced for parcel-based training, indicating that reference samples are not totally representative for inference beyond the parcel, and any deployment of the network to other geographical regions would benefit from full or partial retraining with local data. Article in Journal/Newspaper Arctic Ice Nunavut permafrost Rankin Inlet Canadian Science Publishing Arctic Nunavut Canada Rankin Inlet ENVELOPE(-91.983,-91.983,62.734,62.734) Canadian Journal of Earth Sciences |
institution |
Open Polar |
collection |
Canadian Science Publishing |
op_collection_id |
crcansciencepubl |
language |
English |
topic |
General Earth and Planetary Sciences |
spellingShingle |
General Earth and Planetary Sciences Oldenborger, Greg A. Short, Naomi LeBlanc, Anne-Marie Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach |
topic_facet |
General Earth and Planetary Sciences |
description |
Seasonal or degradational thaw subsidence of permafrost terrain affects the landscape, hydrology, and sustainability of permafrost as an engineering substrate. We perform permafrost thaw sensitivity prediction via supervised classification of a feature set consisting of geological, topographic, and multispectral variables over continuous permafrost near Rankin Inlet, Nunavut, Canada. We build a reference classification of thaw sensitivity using process-based categorization of seasonal subsidence as measured from differential interferometric synthetic aperture radar whereby categories of thaw sensitivity are reflective of ground ice conditions. Classification is performed using a neural network trained on both dispersed and parcel-based reference data. For Low, Medium, High, and Very High thaw sensitivity categories, generalized classification accuracy is 70.8% for 20.6 km 2 of dispersed training data. In all cases, the majority classes of Low and Medium thaw sensitivity are predicted with higher accuracy and more certainty, while the minority classes of High and Very High thaw sensitivity are underpredicted. Minority classes can be combined to improve accuracy at the expense of a reduced level of discrimination. The two-class problem can be classified with an accuracy of 81.8%, thereby effectively distinguishing between stable and unstable ground. The method is applicable to similar Low-Arctic permafrost terrain with geological and topographical controls on thaw sensitivity. However, generalized accuracy is reduced for parcel-based training, indicating that reference samples are not totally representative for inference beyond the parcel, and any deployment of the network to other geographical regions would benefit from full or partial retraining with local data. |
format |
Article in Journal/Newspaper |
author |
Oldenborger, Greg A. Short, Naomi LeBlanc, Anne-Marie |
author_facet |
Oldenborger, Greg A. Short, Naomi LeBlanc, Anne-Marie |
author_sort |
Oldenborger, Greg A. |
title |
Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach |
title_short |
Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach |
title_full |
Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach |
title_fullStr |
Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach |
title_full_unstemmed |
Permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach |
title_sort |
permafrost thaw sensitivity prediction using surficial geology, topography, and remote-sensing imagery: a data-driven neural network approach |
publisher |
Canadian Science Publishing |
publishDate |
2022 |
url |
http://dx.doi.org/10.1139/cjes-2021-0117 https://cdnsciencepub.com/doi/full-xml/10.1139/cjes-2021-0117 https://cdnsciencepub.com/doi/pdf/10.1139/cjes-2021-0117 |
long_lat |
ENVELOPE(-91.983,-91.983,62.734,62.734) |
geographic |
Arctic Nunavut Canada Rankin Inlet |
geographic_facet |
Arctic Nunavut Canada Rankin Inlet |
genre |
Arctic Ice Nunavut permafrost Rankin Inlet |
genre_facet |
Arctic Ice Nunavut permafrost Rankin Inlet |
op_source |
Canadian Journal of Earth Sciences volume 59, issue 11, page 897-913 ISSN 0008-4077 1480-3313 |
op_rights |
http://www.nrcresearchpress.com/page/about/CorporateTextAndDataMining |
op_doi |
https://doi.org/10.1139/cjes-2021-0117 |
container_title |
Canadian Journal of Earth Sciences |
_version_ |
1795665273118785536 |