PSLSA v2.0: An automatic Python package integrating machine learning models for 
regional landslide susceptibility assessment

Guo, Zizheng; Wang, Haojie; He, Jun; Huang, Da; Song, Yixiang; Wang, Tengfei; Liu, Yuanbo; Ferrer, Joaquin Vicente

doi:10.1016/j.envsoft.2025.106367

Item

ITEM ACTIONSEXPORT

Add to Basket

Local TagsRelease HistoryDetailsSummary

Released

Journal Article

PSLSA v2.0: An automatic Python package integrating machine learning models for regional landslide susceptibility assessment

Authors

Guo, Zizheng
External Organizations;

Wang, Haojie
External Organizations;

He, Jun
External Organizations;

Huang, Da
External Organizations;

Song, Yixiang
External Organizations;

Wang, Tengfei
External Organizations;

Liu, Yuanbo
External Organizations;

/persons/resource/joaquinvicente.ferrer

Ferrer, Joaquin Vicente
Potsdam Institute for Climate Impact Research;

External Resource

No external resources are shared

Fulltext (restricted access)

There are currently no full texts shared for your IP range.

Fulltext (public)

There are no public fulltexts stored in PuRe

Supplementary Material (public)

There is no public supplementary material available

Citation

Guo, Z., Wang, H., He, J., Huang, D., Song, Y., Wang, T., Liu, Y., Ferrer, J. V. (2025): PSLSA v2.0: An automatic Python package integrating machine learning models for regional landslide susceptibility assessment. - Environmental Modelling and Software, 186, 106367.
https://doi.org/10.1016/j.envsoft.2025.106367

Cite as: https://publications.pik-potsdam.de/pubman/item/item_33123

Abstract

Accurate landslide susceptibility assessments (LSA) are crucial for civil protection and land use planning. This study introduces PSLSA v2.0 as an open-source Python package that can conduct LSA automatically. It integrates six sophisticated machine learning algorithms (C5.0, SVM, LR, RF, MLP, XGBoost), and allows arbitrary combinations of influencing factors to generate landslide susceptibility index (LSI). We demonstrate how factor contribution and hyperparameter optimization as additional outputs can enhance the model interpretability. We apply PSLSA to a case study focused from Linzhi City in the Tibetan Plateau of China, that has undergone significant engineering modifications on its slopes. The results reveal that slope and aspect are the dominant factors in determining landslide susceptibility. All the six algorithms have an accuracy of over 80%. Although the distribution patterns of LSI vary, the C5.0 model is set apart with the best performance. PSLSA provides a powerful tool for stakeholders especially the non-geohazard professionals.