Skip to main content
Log in

An evolutionary approach for spatial prediction of landslide susceptibility using LiDAR and symbolic classification with genetic programming

  • Original Paper
  • Published:
Natural Hazards Aims and scope Submit manuscript

Abstract

This research examines the potential of spatial prediction of landslide susceptibility by implementing an evolutionary approach using symbolic classification with genetic programming (GP). Specifically, the light detection and ranging (LiDAR)-based digital elevation model was used to generate topographic prediction attributes and to digitize the location of shallow landslides by derivatives such as hillshade maps and contours. The presented approach tested a total of 72 runs with different parameter configurations for producing a good outcome among a number of possible solutions by varying population size, tournament group size and mutation probability. The final solution depicted a total of three important variables including slope, wetness index and solar insulation that were used in the prediction. The GP methodology used symbolic expression trees for the development of the predictive models that were tested and validated in the northern portion of the Cuyahoga Valley National Park located in northeast Ohio. The selected solution from the implemented approach showed that the area under the curve from the receiver operating characteristic curves had a high discrimination power in separating the areas with high susceptibility. The presented model yielded an accuracy of 85.0 % classifying a total of 13.4 % as high susceptibility area with an overall quantitative index of accuracy corresponding to 0.9082. Based on obtained results, the potential of the presented GP approach for mapping landslide susceptibility is promising and further exploration of its capabilities is suggested for finding new avenues of possible landslide research and practical implementations.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

Data availability

All datasets are derived from LiDAR-based DEMs.

Code availability

Not applicable, “Free and open-source software” (FOSS) software used and referenced in the manuscript.

References

Download references

Acknowledgements

The author would like to thank the two anonymous reviewers, who provided helpful suggestions and excellent additions to the manuscript.

Funding

This research received no external funding.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Pece V. Gorsevski.

Ethics declarations

Conflict of interest

The author declares no conflicts of interest

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gorsevski, P.V. An evolutionary approach for spatial prediction of landslide susceptibility using LiDAR and symbolic classification with genetic programming. Nat Hazards 108, 2283–2307 (2021). https://doi.org/10.1007/s11069-021-04780-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11069-021-04780-z

Keywords

Navigation