EUVIMEDEUVIMED entdecken
Crossref

LANDSLIDE SUSCEPTIBILITY MAPPING BASED ON GEOMORPHOLOGICAL VARIABLES USING ARTIFICIAL NEURAL NETWORKS

Alan Lins Alves, Antonio Alex Matias Lêu, Samuel França Amorim, Analice França Lima Amorim

Veredas do Direito · 10.07.2026 · 23 (11) · e237182

DOI: 10.18623/rvd.v23.7182 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Veredas do Direito, online erschienen am 10.07.2026. Verfasst von Alan Lins Alves und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In the Metropolitan Region of Recife, mass movements, especially landslides, represent one of the main socio-environmental threats, particularly in economically vulnerable areas. The increasing frequency of these events, associated with intense rainfall and greater urban vulnerability, highlights the need for accurate and accessible methods for risk assessment. This study developed a landslide susceptibility map for the Lagoa Encantada community, located in the Barreiras Formation in Recife, Pernambuco, Brazil, using Artificial Neural Networks (ANN). Seven geomorphological variables were generated in QGIS and processed in Python for model training and validation. Model performance was evaluated using accuracy, sensitivity, F1-score, and the area under the Receiver Operating Characteristic (ROC) curve. The results showed accuracy, sensitivity, and F1-score values above 90%, as well as an area under the ROC curve greater than 0.9 for both training and test data, demonstrating the model’s high predictive capacity. Compared with traditional methods, such as SINMAP and analytical approaches, the ANN provided greater spatial continuity and a more detailed representation of susceptibility patterns, with satisfactory agreement with the landslide inventory. The results indicate that artificial neural networks represent an efficient and accessible alternative to conventional methods, providing support for landslide susceptibility assessment, risk management, and territorial planning in urban contexts.

Zitieren

APA

Alves, A. L., Lêu, A. A. M., Amorim, S. F., & Amorim, A. F. L. (2026). LANDSLIDE SUSCEPTIBILITY MAPPING BASED ON GEOMORPHOLOGICAL VARIABLES USING ARTIFICIAL NEURAL NETWORKS. Veredas do Direito, 23(11), e237182. https://doi.org/10.18623/rvd.v23.7182

Vancouver

Alves AL, Lêu AAM, Amorim SF, Amorim AFL. LANDSLIDE SUSCEPTIBILITY MAPPING BASED ON GEOMORPHOLOGICAL VARIABLES USING ARTIFICIAL NEURAL NETWORKS. Veredas do Direito. 2026;23(11):e237182. doi:10.18623/rvd.v23.7182

BibTeX
@article{Alves2026,
  author = {Alves, Alan Lins and Lêu, Antonio Alex Matias and Amorim, Samuel França and Amorim, Analice França Lima},
  title = {LANDSLIDE SUSCEPTIBILITY MAPPING BASED ON GEOMORPHOLOGICAL VARIABLES USING ARTIFICIAL NEURAL NETWORKS},
  journal = {Veredas do Direito},
  year = {2026},
  volume = {23},
  number = {11},
  pages = {e237182},
  doi = {10.18623/rvd.v23.7182}
}
Kostenlos starten →