Predição de quedas de árvores urbanas por Machine Learning na cidade de São Paulo (SP)
DOI:
https://doi.org/10.5380/revsbau.v21.102115Resumo
O aumento na incidência de quedas de árvores nas cidades brasileiras, especialmente em metrópoles e grandes capitais como São Paulo, tem se tornado uma preocupação relevante em função das mudanças climáticas aliadas à crescente urbanização sem planejamento adequado. O objetivo foi desenvolver um modelo de machine learning para prever quedas de árvores e de suas partes (galhos) no ambiente urbano, por meio da integração de dados climáticos e históricos de ocorrências. Foram consideradas variáveis como precipitação e vento, a fim de identificar padrões de comportamento que permitam antecipar medidas preventivas, mitigando danos, prejuízos e acidentes, contribuindo para a segurança pública urbana. Ambas as variáveis apresentaram correlação positiva com as quedas, sendo que a precipitação teve maior influência. O Centro foi a área com maior concentração relativa (30,4 quedas/km²), apesar de possuir apenas 4,4% do total das quedas, possivelmente devido à maior densidade urbana, infraestrutura vulnerável e melhor documentação dos eventos. Modelos como esses poderão atuar em sistemas de alerta precoce, contribuindo para uma gestão mais eficaz e proativa da arborização urbana. Além de fundamentar políticas públicas sobre manejo arbóreo, auxiliam no planejamento de plantios futuros, promovendo modelos de gestão mais seguros e sustentáveis.
Palavras-chave: Aprendizado de máquina; Mapa de calor; Modelo preditivo, Mudanças climáticas.
Referências
AHMAD, K; KHARE, M.; CHAUDHRY, K.K. Wind tunnel simulation studies on dispersion at urban street canyons and intersections — a review. Journal of Wind Engineering and Industrial Aerodynamics, v. 93, n. 9, p. 697-717, 2005.
AMORIM, M.C.C.T. Ilhas de calor urbanas: métodos e técnicas de análise. Revista Brasileira de Climatologia, v. 15, p. 22-46, 2019.
AMORIM, M.C.C.T.; DUBREUIL, V.; AMORIM, A.T. Day and night surface and atmospheric heat islands in a continental and temperate tropical environment. Urban Climate, v. 38, p. 100918, 2021.
ARAÚJO, H.C.L.; MARTINS, F.S.; CORTESE, T.T.P.; LOCOSSELLI, G. M. Artificial intelligence in urban forestry — A systematic review. Urban Forestry & Urban Greening, v. 66, p. 127410, 2021.
BURDEN, D. Urban street trees: 22 benefits specific applications. Orlando: Glatting Jackson and Walkable Communities, Inc., 2006.
CAVALARI, A.A.; VELASCO, G.N.; SILVA-LUZ, C.L.; ROSA, A.S.; WAETGE, A.A.N.; BARBOSA, E.S.; BIAZZO, F.C.M.; KAVAMURA, H.E.; SILVA FILHO, C.A.; SILVA, E.B.F.B.; LOCOSSELLI, G.M. Predicting tree failure to define roles and guidelines in risk management, a case study in São Paulo / Brazil. Urban Forestry & Urban Greening, v. 91, p. 128181, 2024.
DABRAL, P.P.; MURRY, M.Z. Modelling and forecasting of rainfall time series using SARIMA. Environmental Processes, v. 4, n. 2, p. 399-419, 2017.
FÁVERO, L.P.; BELFIORE, P. Manual de Análise de Dados: Estatística e Modelagem Multivariada com Excel®, SPSS®, Stata® e Python®. 2.ed. Rio de Janeiro: LTC, 2024, 1288p.
FERNANDES, R.; NASCIMENTO, V.; FREITAS, M.; OMETTO, J. Local Climate Zones to identify surface urban heat islands: a systematic review. Remote Sensing, v. 15, n. 4, p. 884, 2023.
GEOSAMPA. Mapa digital da cidade de São Paulo. 2025. Disponível em: <https://geosampa.prefeitura.sp.gov.br/PaginasPublicas/_SBC.aspx>. Acesso em: 05 ago. 2025.
INDRESPUTRA, F.; ZULKARNAEN, R.N.; HARIRI, M.R.; WARDANI, F.F.; HUTABARAT, P.W.; SETYANTI, D.; PRATIWI, W.A.; RAHMANINGTIYAS, L. Prediction of susceptibility for old trees (> 100 years old) to fall in Bogor botanical garden. Indonesian Journal of Forestry Research, v. 10, n. 1, p. 1-19, 2023.
INSTITUTO NACIONAL DE METEOROLOGIA - Inmet. Portal de Dados Históricos. 2025. Disponível em: <https://bdmep.inmet.gov.br/>. Acesso em: 01 maio 2025.
JANOWIAK, M.K.; BRANDT, L.A.; WOLF, K.L.; BRADY, M.; DARLING, L.; LEWIS, A.D.; FAHEY, R.T.; GIESTING, K.; HALL, E.; HENRY, M.; HUGHES, M.; MIESBAUER, J.W.; MARCINKOWSKI, K.; ONTL, T.; RUTLEDGE, A.; SCOTT, L.; SWANSTON, C.W. Climate adaptation actions for urban forests and human health. Madison: Northern Research Station, 2021, 115p.
JODAS, D.S.; BRAZOLIN, S.; VELASCO, G.N.; LIMA, R.A.; YOJO, T.; PAPA, J.P. Urban tree failure probability prediction based on dendrometric aspects and machine learning models. Computers, Environment and Urban Systems, v. 108, p. 102074, 2024.
KABIR, E.; GUIKEMA, S.; KANE, B. Statistical modeling of tree failures during storms. Reliability Engineering & System Safety, v. 177, p. 68-79, 2018.
LOCOSSELLI, G.; MIYAHARA, A.A.L.; CERQUEIRA, P., BUCKERIDGE, M.S. Climate drivers of tree fall on the streets of São Paulo, Brazil. Trees, v. 35, n. 6, p. 1807-1815, 2021.
LOPES, A.; OLIVEIRA, S.; FRAGOSO, M.; ANDRADE, J.; PEDRO, P. Wind risk assessment in urban environments: The case of falling trees during windstorm events in Lisbon. In: STŘELCOVÁ, K.; MÁTYÁS, C.; KLEIDON, A.; LAPIN, M.; MATEJKA, F.; BLAŽENEC, M.; ŠKVARENINA, J.; HOLÉCY, J. (Org.). Bioclimatology and Natural Hazards. Dordrecht: Springer, 2009, p. 55–74.
MCPHERSON, E.G.; BERRY, A.M.; VAN DOORN, N.S. Performance testing to identify climate-ready trees. Urban Forestry & Urban Greening, v. 29, p. 28-39, 2018.
MENDES, F.H.; PIEDADE, S.M.S.; MENDES, L.G.; LOPES, A.M.S.; SILVA FILHO, D.F. Probabilidade de ocorrência de altas temperaturas do ar e chuvas intensas em Piracicaba/SP. Geografia em Atos (Online), v. 1, n. 9, p. 51-65, 2019.
MENDES, F.H.; PETEAN, F.C.S.; CORREIA, E.L.T.; LOPES. A Proximity-Based Approach for the Identification of Fallen Species of Street Trees during Strong Wind Events in Lisbon. Land, v. 13, n. 5, p. 708, 2024.
ROLNICK, D.; DONTI, P.L.; KAACK, L.H.; KOCHANSKI, K.; LACOSTE, A.; SANKARAN, K.;ROSS, A.S.; MILOJEVIC-DUPONT, N.; JAQUES, N.; WALDMAN-BROWN, A.; LUCCIONI, A.S.; MAHARAJ, T.; SHERWIN, E.D.; MUKKAVILLI, S.K.; KORDING, K.P.; GOMES, C.P.; NG, A.Y.; HASSABIS, D.; PLATT, J.C.; CREUTZIG, F.; CHAYES, J.; BENGIO, Y. Tackling climate change with machine learning. ACM Computing Surveys (CSUR), v. 55, n. 2, p. 1-96, 2022.
ROODBARAKY, H.J.; BAKER, C.J.; DAWSON, A.R.; WRIGHT, C.J. Experimental observations of the aerodynamic characteristics of urban trees. Journal of Wind Engineering and Industrial Aerodynamics, v. 52, p. 171-184, 1994.
SANCHES, R.G.; NEVES, G.Z.F.; SANTOS, B.C.; SILVA, M.S.D.; PEREIRA, D.N.B.; TECH, A.R.B. Intense Rainfall in São Carlos/SP: Determination of Threshold Values Using Climate Indices and Their Spatio-Temporal Repercussion. American Journal of Climate Change, v. 7, n. 3, p. 388-401, 2018.
SCHONS, C.T.; NOTHDURFT, A.; RITTER, T.; FLEIG, F.D. Climate sensitive tree growth modelling of Myrocarpus frondosus in Southern Brazil using tree-ring time series and distributed generalized additive lag models. Forestry: An International Journal of Forest Research, v. 98, n. 2, p. 278-292, 2025.
SRIVANIT, M.; KAEWKHOW, S. A machine learning-based protocol to support visual tree assessment and risk of failure classification on a university campus. Urban Forestry & Urban Greening, v. 99, p. 128420, 2024.
YANG, Z.; REN, W.; LIU, S.; WEI, H. Artificial intelligence in forest management and tree health protection. In: ASIEGBU, F.O.; KOVALCHUK, A. (Org.). Biotechnology Applications in Forestry: Forest Microbiology: Volume 4. San Diego: Academic Press, 2025. p. 433-446.
ZHOU, Z.; FU, J.; XIAO, Y. Risk of wind destruction to urban trees: Prediction workflow and relative importance of influencing factors. Sustainable Cities and Society, v. 112, p. 105600, 2024.
Downloads
Publicado
Como Citar
Edição
Seção
Licença

Este trabalho está licenciado sob uma licença Creative Commons Attribution-NonCommercial 4.0 International License.
Direitos Autorais para artigos publicados nesta revista são do autor, com direitos de primeira publicação para a revista. Com o trabalho simultaneamente licenciado sob a Licença Creative Commons Attribution. Em virtude da aparecerem nesta revista de acesso público, os artigos são de uso gratuito, com atribuições próprias, em aplicações educacionais e não-comerciais.
