A Labeling Intercomparison of Retrogressive Thaw Slumps by a Diverse Group of Domain Experts

ABSTRACT Deep‐learning (DL) models have become increasingly beneficial for the detection of retrogressive thaw slumps (RTS) in the permafrost domain. However, comparing accuracy metrics is challenging due to unstandardized labeling guidelines. To address this, we conducted an experiment with 12 international domain experts from a broad range of scientific backgrounds. Using 3 m PlanetScope multispectral imagery, they digitized RTS footprints in two sites. We evaluated label uncertainty by comparing manually outlined RTS labels using Intersection‐over‐Union (IoU) and F1 metrics. At the Canadian Peel Plateau site, we see good agreement, particularly in the active parts of RTS. Differences were observed in the interpretation of the debris tongue and the stable vegetated sections of RTS. At the Russian Bykovsky site, we observed a larger mismatch. Here, the same differences were documented, but several participants mistakenly identified non‐RTS features. This emphasizes the importance of site‐specific knowledge for reliable label creation. The experiment highlights the need for standardized labeling procedures and definition of their scientific purpose. The most similar expert labels outperformed the accuracy metrics reported in the literature, highlighting human labeling capabilities with proper training, site knowledge, and clear guidelines. These findings lay the groundwork for DL‐based RTS monitoring in the pan‐Arctic.

Authors

Institutions

Publication Details

Journal
Permafrost and Periglacial Processes
Published
2024-10-20
DOI
https://doi.org/10.1002/ppp.2249
Citations
18
Primary Topic
Landslides and related hazards
Type
article
Field-Weighted Citation Impact
12.78

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A Labeling Intercomparison of Retrogressive Thaw Slumps by a Diverse Group of Domain Experts

Zhuoxuan Xia, Nina Nesterova, Chandi Witharana, Alexander Kizyakov et al.
18 citations
Permafrost and Periglacial Processes
Landslides and related hazards
12.78
article

A Labeling Intercomparison of Retrogressive Thaw Slumps by a Diverse Group of Domain Experts

Zhuoxuan Xia, Nina Nesterova, Chandi Witharana, Alexander Kizyakov, Philipp Bernhard, Melissa Ward Jones, Alexandra Runge, Lingcao Huang, Mark J. Lara, Sophia Barth, Ingmar Nitze, Alexandra Veremeeva, Anna Liljedahl, Jurjen van der Sluijs
article en
18 citations

Abstract

ABSTRACT Deep‐learning (DL) models have become increasingly beneficial for the detection of retrogressive thaw slumps (RTS) in the permafrost domain. However, comparing accuracy metrics is challenging due to unstandardized labeling guidelines. To address this, we conducted an experiment with 12 international domain experts from a broad range of scientific backgrounds. Using 3 m PlanetScope multispectral imagery, they digitized RTS footprints in two sites. We evaluated label uncertainty by comparing manually outlined RTS labels using Intersection‐over‐Union (IoU) and F1 metrics. At the Canadian Peel Plateau site, we see good agreement, particularly in the active parts of RTS. Differences were observed in the interpretation of the debris tongue and the stable vegetated sections of RTS. At the Russian Bykovsky site, we observed a larger mismatch. Here, the same differences were documented, but several participants mistakenly identified non‐RTS features. This emphasizes the importance of site‐specific knowledge for reliable label creation. The experiment highlights the need for standardized labeling procedures and definition of their scientific purpose. The most similar expert labels outperformed the accuracy metrics reported in the literature, highlighting human labeling capabilities with proper training, site knowledge, and clear guidelines. These findings lay the groundwork for DL‐based RTS monitoring in the pan‐Arctic.

Permafrost and Periglacial ProcessesVol. 36(1)
Alfred-Wegener-Institut Helmholtz-Zentrum für Polar- und Meeresforschung (DE), Woodwell Climate Research Center (US), University of Connecticut (US), University of Alaska Fairbanks (US), University of Illinois Urbana-Champaign (US), University of Potsdam (DE), Chinese University of Hong Kong (HK), Lomonosov Moscow State University (RU), ETH Zurich (CH), GFZ Helmholtz Centre for Geosciences (DE), Gamma Remote Sensing (Switzerland) (CH), Government of Northwest Territories (CA)
National Science Foundation, National Aeronautics and Space Administration, European Space Agency, Deutscher Akademischer Austauschdienst, Nuclear Safety and Security Commission, Helmholtz Association, International Permafrost Association, Lomonosov Moscow State University
Openalex Percentile: Top 1%
Landslides and related hazards
12.78
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.