Close
Metadata

@MastersThesis{Amaral:1992:ReDaSi,
               author = "Amaral, Silvana",
                title = "Imagens do sistema sensor AVHRR/NOAA na dete{\c{c}}{\~a}o e 
                         avalia{\c{c}}{\~a}o de desmatamentos na Floresta Amaz{\^o}nica: 
                         rela{\c{c}}{\~o}es com dados do sistema TM/LANDSAT",
               school = "Instituto Nacional de Pesquisas Espaciais (INPE)",
                 year = "1992",
              address = "Sao Jose dos Campos",
                month = "1992-09-17",
             keywords = "vegeta{\c{c}}{\~a}o, Serra do Roncador (MT), Floresta 
                         Amaz{\^o}nica, florestas, desmatamento, sat{\'e}lites NOAA, 
                         mapeador tem{\'a}tico (Landsat), sat{\'e}lites Landsat, Sistema 
                         de Informa{\c{c}}{\~a}o Geogr{\'a}fica, calibra{\c{c}}{\~a}o, 
                         SIG, radi{\^o}metro avan{\c{c}}ado de resolu{\c{c}}{\~a}o 
                         muito alta, imagens Landsat, imagens NOAA, desflorestamento, 
                         estimativa, monitoramento, Advanced Very Resolution Radiometer 
                         (AVHRR), deforestation, geometric rectification (imagery), 
                         instrument errors, Landsat satellites, rain forest, statistical 
                         analysis, thematic mappers (Landsat), calibration, geographic 
                         information sytems.",
             abstract = "Este trabalho avalia o uso de imagens AVHRR/NOAA para 
                         detec{\c{c}}{\~a}o e avalia{\c{c}}{\~a}o de desmatamento na 
                         Floresta Amaz{\^o}nica a partir de valida{\c{c}}{\~a}o e 
                         calibra{\c{c}}{\~a}o com imagens TM/LANDSAT. Desenvolveu-se uma 
                         metodologia piloto para regi{\~a}o da Serra do Roncador-MT, onde 
                         o desmatamento e proveniente da instala{\c{c}}{\~a}o de projetos 
                         agropecu{\'a}rios de grande escala. Utilizou-se uma imagem AVHRR 
                         de 2 Km de resolu{\c{c}}{\~a}o espacial, corrigida 
                         geometricamente, para classifica{\c{c}}{\~a}o de {\'a}reas 
                         desmatadas e de floresta. A identifica{\c{c}}{\~a}o destas 
                         classes, assim como a compara{\c{c}}{\~a}o dos resultados de 
                         {\'a}rea estimada na interpreta{\c{c}}{\~a}o visual de imagens 
                         TM/LANDSAT, baseou-se no uso de Sistema Geogr{\'a}fico de 
                         Informa{\c{c}}{\~a}o. Os resultados AVHRR de {\'a}rea de 
                         floresta e desmatamento foram comparados aos dados TM em analises 
                         estat{\'{\i}}sticas, onde obteve-se forte correla{\c{c}}{\~a}o 
                         e regress{\~a}o linear entre eles (R2 = 0,93). A 
                         aplica{\c{c}}{\~a}o do modelo obtido na regi{\~a}o de S{\~a}o 
                         Jos{\'e} do Xingu, MT, de padr{\~a}o de desmatamento semelhante 
                         a primeira {\'a}rea, mostrou-se apropriada com erros m{\'e}dios 
                         de 3% para {\'a}rea total de floresta. Os resultados obtidos 
                         permitiram indicar a banda 3 AVHRR para a detec{\c{c}}{\~a}o e 
                         monitoramento de altera{\c{c}}{\~o}es em {\'a}reas florestais. 
                         Dados TM/LANDSAT s{\~a}o necess{\'a}rios para 
                         calibra{\c{c}}{\~a}o das estimativas de area. ABSTRACT: This 
                         work analyzes the use of AVHRR/NOAA images to detect and evaluate 
                         deforestation in the Amazon Forest using TM/Landsat images for 
                         validation and calibration of the results. A pilot methodology was 
                         developed for the region of {"}Serra do Roncador{"}, MT, where 
                         deforestation is caused by large agriculture and cattle projects. 
                         An AVHRR image with spatial resolution of 2 km and geometrically 
                         corrected was used to classify areas of deforestation and forest. 
                         The identification of these classes and the comparison and 
                         validation of the area estimates with the visual interpretation of 
                         the TM/Landsat images were based and on the use of Geographical 
                         Information System. The results of forest and deforestation areas 
                         obtained from AVHRR were compared to TM data using statistical 
                         analyses and strong linear correlation and regression were found 
                         between the data sets (R2 = 0.93). An application of the model for 
                         the region of S{\~a}o Jos{\'e} do Xingu, MT, with a 
                         deforestation pattern similar to that found in the first region, 
                         presented adequate results with average erros of 3% for the total 
                         forest area. The results corroborated the use of AVHRR band 3 to 
                         detect and monitor alterations in forestes areas. TM/Landsat data 
                         are necessary to calibrate estimates of areas.",
            committee = "Santos, Jo{\~a}o Roberto dos (presidente/orientador) and Setzer, 
                         Alberto Waingort (orientador) and Batista, Get{\'u}lio Teixeira 
                         and Assad, Eduardo Delgado",
           copyholder = "SID/SCD",
         englishtitle = "AVHRR/NOAA images to detect and quantify deforestation in Amazon 
                         Forest: relation with TM/LANDSAT data",
                label = "6558",
             language = "pt",
                pages = "195",
                  ibi = "6qtX3pFwXQZ3r59YD6/GP3np",
                  url = "http://urlib.net/rep/6qtX3pFwXQZ3r59YD6/GP3np",
           targetfile = "publicacao.pdf",
        urlaccessdate = "2020, May 26"
}


Close