Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria
Rock glaciers (RG) are landforms that occur in high latitudes or elevations and — in their active state — consist of a mixture of rock debris and ice. Despite serving as a form of groundwater storage, they are an indicator for the occurrence of (former) permafrost and therefore carry significance in...
Published in: | Applied Computing and Geosciences |
---|---|
Main Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Elsevier
2022
|
Subjects: | |
Online Access: | https://doi.org/10.1016/j.acags.2022.100093 https://doaj.org/article/5de8f1e93f24495bb14ef83418c01084 |
id |
ftdoajarticles:oai:doaj.org/article:5de8f1e93f24495bb14ef83418c01084 |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:5de8f1e93f24495bb14ef83418c01084 2023-05-15T16:37:38+02:00 Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria Georg H. Erharter Thomas Wagner Gerfried Winkler Thomas Marcher 2022-12-01T00:00:00Z https://doi.org/10.1016/j.acags.2022.100093 https://doaj.org/article/5de8f1e93f24495bb14ef83418c01084 EN eng Elsevier http://www.sciencedirect.com/science/article/pii/S2590197422000155 https://doaj.org/toc/2590-1974 2590-1974 doi:10.1016/j.acags.2022.100093 https://doaj.org/article/5de8f1e93f24495bb14ef83418c01084 Applied Computing and Geosciences, Vol 16, Iss , Pp 100093- (2022) Rock glacier inventory Permafrost Hydrological catchment Digital mapping Machine learning Image segmentation Geography. Anthropology. Recreation G Geology QE1-996.5 Electronic computers. Computer science QA75.5-76.95 article 2022 ftdoajarticles https://doi.org/10.1016/j.acags.2022.100093 2022-12-30T19:40:11Z Rock glaciers (RG) are landforms that occur in high latitudes or elevations and — in their active state — consist of a mixture of rock debris and ice. Despite serving as a form of groundwater storage, they are an indicator for the occurrence of (former) permafrost and therefore carry significance in the research for the ongoing climate change. For these reasons, the past years have shown rising interest in the establishment of RG inventories to investigate the extent of permafrost and quantify water storages. Creating these inventories, however, usually involves manual, laborious, and subjective mapping of the landforms based on aerial image - and digital elevation model analysis. We propose an approach for RG mapping based on supervised machine learning which can help to increase the mapping efficiency and permits rapid RG mapping in vast and not yet covered areas. We found deep convolutional artificial neural networks (ANN) that are specifically designed for image segmentation (U-Net architecture) to be well suited for this classification problem. The general workflow consists of training the ANNs with orthophotos and slope maps of digital elevation models as input. The output (RG label-maps) is derived from a recently published RG inventory of the Austrian Alps that features 5769 individual RGs and was compiled manually by several scientists. To increase the generalization capabilities, we use live data augmentation during training. Based on this inventory, the ANNs have learned the average expert opinion and the RG map generated by the ANN can be used to increase the consistency and completeness of already existing RG inventories. Moreover, this ANN approach might be valuable for other landform mapping tasks beyond rock glaciers (e.g., other mass movements). Article in Journal/Newspaper Ice permafrost Directory of Open Access Journals: DOAJ Articles Applied Computing and Geosciences 16 100093 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
Rock glacier inventory Permafrost Hydrological catchment Digital mapping Machine learning Image segmentation Geography. Anthropology. Recreation G Geology QE1-996.5 Electronic computers. Computer science QA75.5-76.95 |
spellingShingle |
Rock glacier inventory Permafrost Hydrological catchment Digital mapping Machine learning Image segmentation Geography. Anthropology. Recreation G Geology QE1-996.5 Electronic computers. Computer science QA75.5-76.95 Georg H. Erharter Thomas Wagner Gerfried Winkler Thomas Marcher Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria |
topic_facet |
Rock glacier inventory Permafrost Hydrological catchment Digital mapping Machine learning Image segmentation Geography. Anthropology. Recreation G Geology QE1-996.5 Electronic computers. Computer science QA75.5-76.95 |
description |
Rock glaciers (RG) are landforms that occur in high latitudes or elevations and — in their active state — consist of a mixture of rock debris and ice. Despite serving as a form of groundwater storage, they are an indicator for the occurrence of (former) permafrost and therefore carry significance in the research for the ongoing climate change. For these reasons, the past years have shown rising interest in the establishment of RG inventories to investigate the extent of permafrost and quantify water storages. Creating these inventories, however, usually involves manual, laborious, and subjective mapping of the landforms based on aerial image - and digital elevation model analysis. We propose an approach for RG mapping based on supervised machine learning which can help to increase the mapping efficiency and permits rapid RG mapping in vast and not yet covered areas. We found deep convolutional artificial neural networks (ANN) that are specifically designed for image segmentation (U-Net architecture) to be well suited for this classification problem. The general workflow consists of training the ANNs with orthophotos and slope maps of digital elevation models as input. The output (RG label-maps) is derived from a recently published RG inventory of the Austrian Alps that features 5769 individual RGs and was compiled manually by several scientists. To increase the generalization capabilities, we use live data augmentation during training. Based on this inventory, the ANNs have learned the average expert opinion and the RG map generated by the ANN can be used to increase the consistency and completeness of already existing RG inventories. Moreover, this ANN approach might be valuable for other landform mapping tasks beyond rock glaciers (e.g., other mass movements). |
format |
Article in Journal/Newspaper |
author |
Georg H. Erharter Thomas Wagner Gerfried Winkler Thomas Marcher |
author_facet |
Georg H. Erharter Thomas Wagner Gerfried Winkler Thomas Marcher |
author_sort |
Georg H. Erharter |
title |
Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria |
title_short |
Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria |
title_full |
Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria |
title_fullStr |
Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria |
title_full_unstemmed |
Machine learning – An approach for consistent rock glacier mapping and inventorying – Example of Austria |
title_sort |
machine learning – an approach for consistent rock glacier mapping and inventorying – example of austria |
publisher |
Elsevier |
publishDate |
2022 |
url |
https://doi.org/10.1016/j.acags.2022.100093 https://doaj.org/article/5de8f1e93f24495bb14ef83418c01084 |
genre |
Ice permafrost |
genre_facet |
Ice permafrost |
op_source |
Applied Computing and Geosciences, Vol 16, Iss , Pp 100093- (2022) |
op_relation |
http://www.sciencedirect.com/science/article/pii/S2590197422000155 https://doaj.org/toc/2590-1974 2590-1974 doi:10.1016/j.acags.2022.100093 https://doaj.org/article/5de8f1e93f24495bb14ef83418c01084 |
op_doi |
https://doi.org/10.1016/j.acags.2022.100093 |
container_title |
Applied Computing and Geosciences |
container_volume |
16 |
container_start_page |
100093 |
_version_ |
1766027928721686528 |