Comparação de sistemas imageadores aerotransportados ópticos e LiDAR para quantificação da biomassa em áreas de Cerrado nativo

• Ano de defesa: 2021
• Autor(a) principal: Alves Júnior, Leomar Rufino
• Orientador(a): Ferreira, Manuel Eduardo
• Banca de defesa: Ferreira, Manuel Eduardo, Miranda, Sabrina do Couto de, Oliveira, Ivanilton José de, Rosolen, Vania Silvia, Vieira, Pedro Alves
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Goiás
• Programa de Pós-Graduação: Programa de Pós-graduação em Geografia (IESA)
• Departamento: Instituto de Estudos Socioambientais - IESA (RG)
• País: Brasil
• Palavras-chave em Português: LiDAR; RPAS; Ortomosaico; Nuvem de pontos; Métricas estruturais da vegetação
• Palavras-chave em Inglês: Orthomosaic; Point Cloud; Structural metrics of vegetation
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::GEOCIENCIAS::GEOGRAFIA FISICA::FOTOGEOGRAFIA (FISICO-ECOLOGICA)
• Link de acesso: http://repositorio.bc.ufg.br/tede/handle/tede/12421

Currently, remotely manned aerial platforms (RPAS) embedded with imaging sensors form one of the most desirable and efficient technological sets in the science of remote sensing. Among all possible applications, land use mapping is the flagship, with emphasis on the analysis of native vegetation cover and its relationship with the environment. This research aimed to analyze the woody aerial biomass contained in parcels of native vegetation of the Cerrado biome, located in the Red River basin, state of Goiás, based on advanced instruments (LiDAR and optical sensors) aboard manned aerial vehicles (airplane) and unmanned (RPAS fixedwing, eBee Plus RTK/PPK model). The research also included estimates of native vegetation cover obtained in the field, such as canopy height, number of individuals, percentage of cover and woody aerial biomass, in order to evaluate the efficiency of imaging systems and sensors. The resulting digital surface (MDS) and orthomosaic models were analyzed as a function of accuracy and precision, combining support points and the PPK (Post Processed Kinematic) system. Comparing the orthorosaics generated with support points and without the PPK system, with the orthorosaics generated only with the PPK system, both presented the same planimetric precision, but with improved altimetric precision when the support points were used. For the Cerradão, the model that best estimated the biomass was the one generated with airborne LiDAR data with manned flight, statistically significant with [(F(9.50) = 33.17; p < 0.001; R² = 0.856; SE of the estimate = 0.17 Mg/ha]. For the Dense Cerrado and Cerradão/Dense Forest, the models were statistically significant with [(F(12.27) = 121.264; p < 0.001; R² = 0.982; SE estimate = 0.049 Mg/ha] and [(F(7.42) = 23.360; p < 0.001; R² = 0.796; SE estimate = 0.329 Mg/ha], respectively, both using data obtained with LiDAR Airborne with unmanned flight It is concluded that the use of data obtained with LiDAR and by aerophotogrammetry, from manned and remotely manned aircraft, cannot replace but complement the conventional method of field survey, in the case of native vegetation of Cerrado Denso and Cerradão. As the efficiency in predicting biomass, going from the best to the worst nt, the point cloud obtained with airborne LiDAR with unmanned flight, airborne LiDAR with manned flight, and with aerophotogrammetric survey in unmanned flight. For Cerrado Denso and Cerradão, it can be concluded that all technologies used in this research showed effective results and can therefore be applied according to the availability of equipment and financial resources.