A machine learning method for Arctic lakes detection in the permafrost areas of Siberia
ABSTRACTThermokarst lakes are the main components of the vast Arctic and subarctic landscapes. These lakes can serve as geoindicators of permafrost degradation; therefore, proper lake distribution assessment methods are necessary. In this study, we compared four machine learning methods to improve e...
Published in: | European Journal of Remote Sensing |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Taylor & Francis Group
2023
|
Subjects: | |
Online Access: | https://doi.org/10.1080/22797254.2022.2163923 https://doaj.org/article/73d3be4c3a1f4b18a512deac53b10acd |
id |
ftdoajarticles:oai:doaj.org/article:73d3be4c3a1f4b18a512deac53b10acd |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:73d3be4c3a1f4b18a512deac53b10acd 2023-05-15T14:57:42+02:00 A machine learning method for Arctic lakes detection in the permafrost areas of Siberia Piotr Janiec Jakub Nowosad Zbigniew Zwoliński 2023-12-01T00:00:00Z https://doi.org/10.1080/22797254.2022.2163923 https://doaj.org/article/73d3be4c3a1f4b18a512deac53b10acd EN eng Taylor & Francis Group https://www.tandfonline.com/doi/10.1080/22797254.2022.2163923 https://doaj.org/toc/2279-7254 doi:10.1080/22797254.2022.2163923 2279-7254 https://doaj.org/article/73d3be4c3a1f4b18a512deac53b10acd European Journal of Remote Sensing, Vol 56, Iss 1 (2023) Thermokarst lakes subpolar areas Landsat MERIT DEM supervised classification Siberia Oceanography GC1-1581 Geology QE1-996.5 article 2023 ftdoajarticles https://doi.org/10.1080/22797254.2022.2163923 2023-01-22T01:30:34Z ABSTRACTThermokarst lakes are the main components of the vast Arctic and subarctic landscapes. These lakes can serve as geoindicators of permafrost degradation; therefore, proper lake distribution assessment methods are necessary. In this study, we compared four machine learning methods to improve existing lake detection systems. The northern part of Yakutia was selected as the study area owing to its complex environment. We used data from Landsat 8 and spectral indices to take into account the spectral characteristics of the lakes, and MERIT DEM data to take into account the topography. The lowest accuracy was found for the classification and regression trees (CART) method (overall accuracy = 81%). On the other hand, the random forests (RF) classification provided the best results (overall accuracy = 92%), and only this classification coped well in all problematic areas, such as shaded and humid areas, near steep slopes, burn scars, and rivers. The altitude and bands SWIR1 (Short wave infrared 1), SWIR2 (Short wave infrared 2), and Green were the most important. Spectral indices did not have significant impact on the classification results in the specific conditions of the thermokarst lakes environment. 17,700 lakes were identified with the total area of 271.43 km2. Article in Journal/Newspaper Arctic permafrost Subarctic Thermokarst Yakutia Siberia Directory of Open Access Journals: DOAJ Articles Arctic European Journal of Remote Sensing 56 1 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Thermokarst lakes subpolar areas Landsat MERIT DEM supervised classification Siberia Oceanography GC1-1581 Geology QE1-996.5 |
spellingShingle |
Thermokarst lakes subpolar areas Landsat MERIT DEM supervised classification Siberia Oceanography GC1-1581 Geology QE1-996.5 Piotr Janiec Jakub Nowosad Zbigniew Zwoliński A machine learning method for Arctic lakes detection in the permafrost areas of Siberia |
topic_facet |
Thermokarst lakes subpolar areas Landsat MERIT DEM supervised classification Siberia Oceanography GC1-1581 Geology QE1-996.5 |
description |
ABSTRACTThermokarst lakes are the main components of the vast Arctic and subarctic landscapes. These lakes can serve as geoindicators of permafrost degradation; therefore, proper lake distribution assessment methods are necessary. In this study, we compared four machine learning methods to improve existing lake detection systems. The northern part of Yakutia was selected as the study area owing to its complex environment. We used data from Landsat 8 and spectral indices to take into account the spectral characteristics of the lakes, and MERIT DEM data to take into account the topography. The lowest accuracy was found for the classification and regression trees (CART) method (overall accuracy = 81%). On the other hand, the random forests (RF) classification provided the best results (overall accuracy = 92%), and only this classification coped well in all problematic areas, such as shaded and humid areas, near steep slopes, burn scars, and rivers. The altitude and bands SWIR1 (Short wave infrared 1), SWIR2 (Short wave infrared 2), and Green were the most important. Spectral indices did not have significant impact on the classification results in the specific conditions of the thermokarst lakes environment. 17,700 lakes were identified with the total area of 271.43 km2. |
format |
Article in Journal/Newspaper |
author |
Piotr Janiec Jakub Nowosad Zbigniew Zwoliński |
author_facet |
Piotr Janiec Jakub Nowosad Zbigniew Zwoliński |
author_sort |
Piotr Janiec |
title |
A machine learning method for Arctic lakes detection in the permafrost areas of Siberia |
title_short |
A machine learning method for Arctic lakes detection in the permafrost areas of Siberia |
title_full |
A machine learning method for Arctic lakes detection in the permafrost areas of Siberia |
title_fullStr |
A machine learning method for Arctic lakes detection in the permafrost areas of Siberia |
title_full_unstemmed |
A machine learning method for Arctic lakes detection in the permafrost areas of Siberia |
title_sort |
machine learning method for arctic lakes detection in the permafrost areas of siberia |
publisher |
Taylor & Francis Group |
publishDate |
2023 |
url |
https://doi.org/10.1080/22797254.2022.2163923 https://doaj.org/article/73d3be4c3a1f4b18a512deac53b10acd |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic permafrost Subarctic Thermokarst Yakutia Siberia |
genre_facet |
Arctic permafrost Subarctic Thermokarst Yakutia Siberia |
op_source |
European Journal of Remote Sensing, Vol 56, Iss 1 (2023) |
op_relation |
https://www.tandfonline.com/doi/10.1080/22797254.2022.2163923 https://doaj.org/toc/2279-7254 doi:10.1080/22797254.2022.2163923 2279-7254 https://doaj.org/article/73d3be4c3a1f4b18a512deac53b10acd |
op_doi |
https://doi.org/10.1080/22797254.2022.2163923 |
container_title |
European Journal of Remote Sensing |
container_volume |
56 |
container_issue |
1 |
_version_ |
1766329832805761024 |