Automated detection of landslide scars associated with the extreme rainfall event in 2024 using Artificial Intelligence: a case study in a section of the Taquari-Antas River Basin, RS
DOI:
https://doi.org/10.5902/2236499494269Keywords:
Algorithm image classification, Mass movements, Extreme climate eventAbstract
During 2024, Rio Grande do Sul (RS) faced severe geo-hydrological disasters, such as historic floods and landslides (DT), caused by high rainfall indices (PP). In this context, remote sensing (RS) has proven essential for monitoring these extreme events. This study evaluated the accuracy of the supervised classifiers Random Forest (RF) and TempCNN in detecting landslide scars in a section of the Taquari-Antas River Basin (BHTA). For this purpose, Sentinel-2 and Digital Elevation Model (DEM) data cubes were integrated using the open-source software Satellite Image Time Series Analysis (SITS) in R Studio. The R, G, B, NIR, and ELEVATION bands were used as explanatory variables for RF, and only the optical bands for TempCNN, due to SITS limitations. Rainfall values were obtained from Google Earth Engine (GEE) using data from the Climate Hazards Center InfraRed Precipitation with Station (CHIRPS) to compare the magnitude behavior of the rainfall event (2023–2024). Results indicated an overall accuracy of 0.98% for RF and 0.97% for TempCNN. Although both techniques showed high overall accuracy, they presented low user and producer accuracies for the “Landslide” class. Adjustments are needed in sample intensification and the incorporation of additional optical variables, SAR sensors, and DEM-derived variables.
Downloads
References
ATTRI, P.; CHAUDHRY, S.; SHARMA, B. Remote Sensing & GIS based approaches for LULC change detection: a review. International Journal of Current Engineering and Technology, v. 5, n. 5, 2015.
BELGIU, M.; DRĂGUŢ, L. Random forest in remote sensing: a review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing, v. 114, p. 24–31, 2016.
CHEN, H.; H. E., Y.; ZHANG, L.; et al. A multi-input channel U-Net landslide detection method fusing SAR multisource remote sensing data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, v. 17, p. 1215-1232, 2024. Acesso em: 19 ago. 2025. DOI: https://doi.org/10.1109/JSTARS.2023.3339294.
CHEN, W.; POURGHASEMI, H. R; NAGHIBI, S. A. A comparative study of landslide susceptibility maps produced using support vector machine with different kernel functions and entropy data mining models in China. Bull Eng Geol Environ, v. 77, p. 647-664, 2018. DOI: https://doi. org/10.1007/s10064-017-1010-y.
CPRM. Serviço Geológico do Brasil. Geodiversidade do Estado do Rio Grande do Sul. P. 250, Porto Alegre, 2010. Disponível em: https://rigeo.sgb.gov.br/handle/doc/16774. Acesso em: 14 ago. 2025.
DE LIMA, F. F.; SAITO, S. M.; MORELLI, S. Rainfall thresholds and forest regeneration in post-landslide areas of the Atlantic Forest. Natural Hazards, v. 109, p. 1399–1418, 2021.
DIAS, H. C. et al. Landslide recognition using SVM, Random Forest, and Maximum Likelihood classifiers on high-resolution satellite images: a case study of Itaóca, southeastern Brazil. Brazilian Journal of Geology, v. 51, n. 4, 2021. DOI: 10.1590/2317-4889202120200105.
EGAS, H. M.; STABILE, R. A.; ANDRADE, M. R. M. et al. Inventário abrangente e avaliação inicial de deslizamentos de terra desencadeados pelas chuvas do outono de 2024 no Rio Grande do Sul, Brasil. Landslides, v. 22, p. 579–589, 2025.
EM-DAT – The International Disaster Database. Natural Disasters. Centre for Research on the Epidemiology of Disasters – CRED, Université Catholique de Louvain, Belgium, 2018.
FAGUNDES, G. C. L; AQUINO, C. M. S; SOUSA, M. A. Movimentos gravitacionais de massa: estudo de caso em Teresina, Piauí. Revista Espaço e Geografia, v. 25, n. 1, p. 295–321, 2023. Disponível em: https://periodicos.unb.br/index.php/espacoegeografia/article/view/44030. Acesso em: 1 set. 2025. DOI: 10.26512/2236-56562022e44030.
FARAH, F. Ocupação urbana e instabilidade de encostas. Coleção Habitare, São Paulo, v. 2, p. 312, 2003.
FEPAM. Fundação Estadual de Proteção Ambiental. Secretaria Estadual do Meio Ambiente. Diagnóstico ambiental da bacia do Taquari-Antas, RS: diretrizes regionais para o licenciamento ambiental das hidrelétricas – trecho médio e superior. Porto Alegre, FEPAM, 2001.
FERLA, A. K.; NARVAES, I. S.; KUPLICH, T. M. Métodos de machine learning e deep learning para mapeamento de supressão da vegetação em área de campo no bioma Mata Atlântica. In: Anais do XXI Simpósio Brasileiro de Sensoriamento Remoto, 2025, Salvador. Anais eletrônicos..., Galoá, 2025. Disponível em: https://proceedings.science/sbsr-2025/trabalhos/metodos-de-machine-learning-e-deep-learning-para-mapeamento-de-supressao-da-vege?lang=en. Acesso em: 22 out. 2025.
FERREIRA, K. R.; QUEIROZ, G. R. Medindo o Brasil a partir do espaço: tecnologias de big data e inteligência artificial. AtoZ, v. 10, n. 3, 2021, DOI: https://doi.org/10.5380/atoz.v10i3.81966.
FERREIRA, K.R.; QUEIROZ, G.R.; VINHAS, L. et al. Earth Observation Data Cubes for Brazil: Requirements, Methodology and Products. Remote Sensing, v. 12, 2020, DOI: https://doi. org/10.3390/rs12244033.
FERRI, G. A.; TOGNI, A. C. A história da bacia hidrográfica Taquari-Antas. Editora Univates. Lajeado, 2012.
FUNK, C.; PETERSON, P.; LANDSFELD, M. et al. The climate hazards infrared precipitation with stations—a new environmental record for monitoring extremes. Scientific Data v.2, n. 150066, 2015. DOI: https://doi.org/10.1038/sdata.2015.66.
HUANG, J.; JU, N. P.; LIAO, Y. J. et al. Determination of rainfall thresholds for shallow landslides by a probabilistic and empirical method. Natural Hazards and Earth System Sciences, v. 15, 2021, DOI: http://dx.doi.org/10.5194/nhess-15-2715-2015.
IBGE. Instituto Brasileiro de Geografia e Estatística. Mapa de Unidades de Relevo do Brasil. Rio de Janeiro, ed. 2, 2006.
INSTITUTO NACIONAL DE METEOROLOGIA (INMET). Eventos extremos de abril de 2024 no Brasil. Notas Técnicas, 2024. Disponível em: https://portal.inmet.gov.br/notasTecnicas. Acesso em: 11 set. 2025.
INSTITUTO NACIONAL DE METEOROLOGIA (INMET). Eventos extremos de setembro de 2023 no Brasil. Notas Técnicas, 2023. Disponível em: https://portal.inmet.gov.br/notasTecnicas. Acesso em: 27 set. 2025.
KAISER, E. A.; ROLIM, S. B. A.; GRONDONA, A. E. B. et al. Spatiotemporal Influences of LULC Changes on Land Surface Temperature in Rapid Urbanization Area by Using Landsat-TM and TIRS Images. Atmosphere, v. 13, n. 3, p. 460, 2022. DOI: http://dx.doi.org/10.3390/atmos13030460.
LE CUN, Y.; BENGIO, Y.; HINTON, G. Deep learning. Nature, v. 521 p. 436–444, 2015, DOI: https://doi.org/10.1038/nature14539.
LÓ, T. B.; CORRÊA, U. B.; ARAÚJO, R. M. et al. Temporal convolutional neural network for land use and land cover classification using satellite images time series. Arabian Journal of Geosciences, v.16, n.10, p 1-18, 2023, DOI: https://doi.org/10.1007/s12517-023-11688-4.
MEHMOOD, K.; ANEES, S. A.; MUHAMMAD, S. et al. Analyzing vegetation health dynamics across seasons and regions through NDVI and climatic variables. Scientific Reports, v. 14, n. 1, p. 1-22, 2024. DOI: http://dx.doi.org/10.1038/s41598-024-62464-7.
OLIVEIRA, G. G.; RUIZ, L. F. C.; GUASSELLI, L. A. et al. Random forest and artificial neural networks in landslide susceptibility modeling: a case study of the Fão River Basin, Southern Brazil. Natural Hazards v. 99, 1049–1073, 2019. DOI: 10.1007/s11069-019-03795-x.
OLOFSSON, P. et al. Good practices for estimating area and assessing accuracy of land change. Remote Sensing of Environment, v. 148, p. 42–57, maio 2014.
PELLETIER, C.; WEBB, G. I.; PETITJEAN, F. Temporal convolutional neural network for the classification of satellite image time series. Remote Sensing, v. 11, n. 5, 2019.
PERBET, P.; GUINDON, L.; CÔTÉ, J. F. et al. Evaluating deep learning methods applied to Landsat time series subsequences to detect and classify boreal forest disturbances events: The challenge of partial and progressive disturbances. Remote Sensing of Environment, v. 306, 2024. DOI: https://doi.org/10.1016/j.rse.2024.114107.
POSIT. RStudio: Integrated Development for R. v. 2024.05.0, Boston, 2024. Disponível em: https://posit.co/. Acesso em: 9 set. 2025.
QUEVEDO, R. P; MACIEL, D. A; REIS, M. S. et al. Land use and land cover changes without invalid transitions: A case study in a landslide-affected area. Remote Sensing Applications: Society and Environment, v. 36, p. 101314, 2024. DOI: https://doi.org/10.1016/j.rsase.2024.101314.
SALEEM, N.; HUQ, M. E; TWUMASI, N. Y. D. et al. Parameters derived from and/or used with digital elevation models (DEMs) for landslide susceptibility mapping and landslide risk assessment: a review. ISPRS International Journal of Geo-Information, v. 8, n. 12, 2019. DOI: https://doi. org/10.3390/ijgi8120545.
SANTOS, D. L.; FRAGOSO JR., C. R.; SOUZA, V. C. B. Limiar de precipitação com potencial de gerar deslizamentos nos complexos de risco em uma zona urbana. Revista de Gestão de Água da América Latina, v. 16, e12, 2019. DOI: https://dx.doi.org/10.21168/rega.v16e12.
SAUSEN, T. M.; LACRUZ, M. S. P. Sensoriamento remoto para desastres. Oficina de Textos, p. 285, São Paulo, 2015. P.
SHEYKHMOUSA, M.; MAHDIANPARI, M.; GHANBARI, M. et al. Support vector machine versus random forest for remote sensing image classification: a meta-analysis and systematic review. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, v. 13, p. 6308-6325, 2020.
SIMÕES, R. et al. Satellite image time series analysis for big earth observation data. Remote Sensing, v. 13, p. 2428, 2021.
TOAN, D. T.; DUC, D. M.; QUYNH, D. T. et al. Série de deslizamentos de terra induzidos por chuvas extremas e grandes deslizamentos de terra semelhantes a fluxos de água na comuna de Ho Bon, distrito de Mu Cang Chai, província de Yen Bai, Vietnã, em agosto de 2023. Landslides, v. 22, p. 2607–2622, 2025. DOI: https://doi.org/10.1007/s10346-025-02544-5.
TOMINAGA, L. K.; SANTORO, J.; AMARAL, R. Desastres naturais: conhecer para prevenir. São Paulo: Instituto Geológico, 2015.
TUCCI, C. E. M. Hidrologia: ciência e aplicação. 2. ed. Porto Alegre: Editora da Universidade, 1997.
XAVIER, M. O. Automatização de inventários de deslizamentos aplicados à modelagem de suscetibilidade usando métodos de aprendizado de máquina. 150 f. – Tese (Doutorado em Engenharia) – Escola de Minas, Universidade Federal de Ouro Preto, Ouro Preto, 2025.
ZÊZERE, J. L.; PEREIRA, S.; MELO, R. et al. Mapping landslide susceptibility using data-driven methods. Science of the Total Environment, v. 589, p. 347-362, 2017.
Published
Issue
Section
License
Copyright (c) 2026 Geografia Ensino & Pesquisa

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
The journal Geografia – Ensino & Pesquisa will obtain the auctorial rights for all published texts. This also implies that the text can be published anywhere in the world, including all rights on renewal, expansion and dissemination of the contribution, as well as other subsidiary rights. The author’s get permission to publish the contribution in other medias, printed or digital, may be in Portuguese or translation, since the publication is credited to Revista Geografia – Ensino & Pesquisa.The journal Geografia – Ensino & Pesquisa will obtain the auctorial rights for all published texts. This also implies that the text can be published anywhere in the world, including all rights on renewal, expansion and dissemination of the contribution, as well as other subsidiary rights. The author’s get permission to publish the contribution in other medias, printed or digital, may be in Portuguese or translation, since the publication is credited to Revista Geografia – Ensino & Pesquisa.


