Spatial machine learning modelling reveals that soil indicators and tree type best explain shallow landslide release
- 1Western Norway University of Applied Sciences (Bergen, Norway)
Abstract
The exploration of shallow landslide susceptibility is often impaired by biased landslide inventories, and by over-optimistic performance metrics linked to inadequate models. Here, we use a systematically mapped event inventory of 571 shallow landslides triggered in southern Norway and apply 32 gradient boosted decision tree models to rigorously test the effects of (1) a nested vs. simple cross-validation strategy, (2) spatial vs. non-spatial models, (3) four different cross-validation sampling strategies which were applied on (4) full vs. forest-only datasets. Model evaluation shows that models with random cross-validation provided the highest test performance metrics but did not account for autocorrelation and were likely over-optimistic. The spatial models with spatial cross-validation reduced autocorrelation in the model residuals at the expense of predictive power. Although no model emerged as optimal, findings across models suggest that important explanatory factors like elevation, aspect and bedrock weatherability serve as soil indicators, illustrating a need for improved datasets for soil thickness and heterogeneity. In the forest-only models, tree type was consistently an important predictor, with higher landslide susceptibility in deciduous forest, illustrating the potential of forest variables and forest-specific threshold values in shallow landslide susceptibility mapping.
Versions
Related
- Machine-learning modelling of regional landslide susceptibility in eThekwini, South Africa
- Seismic Precursors to the Blatten, Switzerland Landslide Revealed by Unsupervised Machine Learning
- Role of Topo-Climatic Factors in Landslide Susceptibility Mapping: A Machine Learning Perspective
- Landslide susceptibility mapping using machine learning models and analytical hierarchy process along NH-513, Pasighat to Yingkiong, Arunachal Pradesh, India
- Landslide susceptibility assessment in the Western Ghats adjoining Idukki township, Kerala, India, using integrated geomorphological and geotechnical parameters with AHP and machine learning
BibTeX
@article{ruther2026spatial,
title = {{Spatial machine learning modelling reveals that soil indicators and tree type best explain shallow landslide release}},
author = {Denise Christina Rüther and Kristine Flacké Haualand and Iris Louisa Johanna Peeters and Mark Andrew Kusk Gillespie},
journal = {Natural Hazards and Earth System Sciences},
year = {2026},
month = oct,
doi = {10.5194/nhess-26-4675-2026},
url = {https://nhess.copernicus.org/articles/26/4675/2026/},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Spatial machine learning modelling reveals that soil indicators and tree type best explain shallow landslide release
AU - Denise Christina Rüther
AU - Kristine Flacké Haualand
AU - Iris Louisa Johanna Peeters
AU - Mark Andrew Kusk Gillespie
PY - 2026
DA - 2026/10/01
JO - Natural Hazards and Earth System Sciences
DO - 10.5194/nhess-26-4675-2026
UR - https://nhess.copernicus.org/articles/26/4675/2026/
AB - The exploration of shallow landslide susceptibility is often impaired by biased landslide inventories, and by over-optimistic performance metrics linked to inadequate models. Here, we use a systematically mapped event inventory of 571 shallow landslides triggered in southern Norway and apply 32 gradient boosted decision tree models to rigorously test the effects of (1) a nested vs. simple cross-validation strategy, (2) spatial vs. non-spatial models, (3) four different cross-validation sampling strategies which were applied on (4) full vs. forest-only datasets. Model evaluation shows that models with random cross-validation provided the highest test performance metrics but did not account for autocorrelation and were likely over-optimistic. The spatial models with spatial cross-validation reduced autocorrelation in the model residuals at the expense of predictive power. Although no model emerged as optimal, findings across models suggest that important explanatory factors like elevation, aspect and bedrock weatherability serve as soil indicators, illustrating a need for improved datasets for soil thickness and heterogeneity. In the forest-only models, tree type was consistently an important predictor, with higher landslide susceptibility in deciduous forest, illustrating the potential of forest variables and forest-specific threshold values in shallow landslide susceptibility mapping.
SN - 1684-9981
ER -