Fechar
Metadados

@MastersThesis{Rodrigues::ReNoPr,
               author = "Rodrigues, Danilo Avancini",
                title = "Din{\^a}mica dos fatores de degrada{\c{c}}{\~a}o florestal em 
                         fronteira agropecu{\'a}ria da Amaz{\^o}nia: a regi{\~a}o de 
                         Novo Progresso, Par{\'a}",
               school = "Instituto Nacional de Pesquisas Espaciais (INPE)",
              address = "S{\~a}o Jos{\'e} dos Campos",
                month = "2019-08-19",
             keywords = "regress{\~a}o multivariada, degrada{\c{c}}{\~a}o florestal, 
                         fronteira agropecu{\'a}ria, multivariate regression, forest 
                         degradation, logging frontier.",
             abstract = "A degrada{\c{c}}{\~a}o florestal na Amaz{\^o}nia, causada pela 
                         explora{\c{c}}{\~a}o seletiva de madeira e pelo fogo florestal, 
                         {\'e} um dos principais impactos decorrentes da 
                         intensifica{\c{c}}{\~a}o da ocupa{\c{c}}{\~a}o dessa 
                         regi{\~a}o, com din{\^a}mica e extens{\~a}o ainda pouco 
                         conhecidas. As altera{\c{c}}{\~o}es na vegeta{\c{c}}{\~a}o, 
                         causadas por degrada{\c{c}}{\~a}o florestal, ocorrem de maneira 
                         gradual e no longo prazo, demandando longos per{\'{\i}}odos de 
                         observa{\c{c}}{\~a}o. Os fatores que influenciam a 
                         degrada{\c{c}}{\~a}o florestal variam ao longo tempo e se 
                         relacionam com os est{\'a}gios de desenvolvimento de uma 
                         regi{\~a}o. Este trabalho identificou os principais fatores, e 
                         suas respectivas contribui{\c{c}}{\~o}es (pesos), relacionados 
                         com a intensidade da degrada{\c{c}}{\~a}o florestal na 
                         regi{\~a}o de Novo Progresso, munic{\'{\i}}pio da Amaz{\^o}nia 
                         paraense, em diferentes per{\'{\i}}odos. Essa regi{\~a}o {\'e} 
                         considerada uma ativa fronteira agropecu{\'a}ria, cuja 
                         ocupa{\c{c}}{\~a}o se intensificou a partir dos anos 2000. O 
                         procedimento metodol{\'o}gico foi realizado em duas etapas: na 
                         primeira, foi realizado o mapeamento da degrada{\c{c}}{\~a}o 
                         florestal com t{\'e}cnicas de processamento de imagens e, na 
                         segunda, foram identificados os fatores explicativos que 
                         influenciam a intensidade da degrada{\c{c}}{\~a}o florestal. Na 
                         etapa de mapeamento foi realizada a classifica{\c{c}}{\~a}o 
                         espectral de imagens anuais de 2012 a 2017 para a 
                         identifica{\c{c}}{\~a}o de fei{\c{c}}{\~o}es relativas {\`a} 
                         degrada{\c{c}}{\~a}o florestal. Em seguida, foi realizada uma 
                         classifica{\c{c}}{\~a}o estrutural, para o mapeamento de 
                         padr{\~o}es de intensidade de degrada{\c{c}}{\~a}o florestal. 
                         Os dados produzidos foram incorporados a um banco de dados, 
                         formando uma s{\'e}rie hist{\'o}rica de intensidade de 
                         degrada{\c{c}}{\~a}o florestal na regi{\~a}o de Novo Progresso 
                         de 34 anos (1984 a 2017). Esta s{\'e}rie hist{\'o}rica foi 
                         dividida em tr{\^e}s per{\'{\i}}odos que refletem diferentes 
                         contextos do desenvolvimento da regi{\~a}o de estudo: 1984 a 
                         1994, 1995 a 2004 e 2005 a 2017. A obten{\c{c}}{\~a}o e 
                         an{\'a}lise dos fatores que influenciam a intensidade de 
                         degrada{\c{c}}{\~a}o florestal foram realizadas por meio de 
                         modelos de regress{\~a}o multivariados. Os modelos demonstraram 
                         que o desmatamento e as estradas, representados pelas 
                         vari{\'a}veis dist{\^a}ncia de {\'a}reas desmatadas, 
                         propor{\c{c}}{\~a}o de {\'a}rea desmatada, dist{\^a}ncia de 
                         estradas vicinais e dist{\^a}ncia da BR-163, influenciaram a 
                         degrada{\c{c}}{\~a}o florestal durante todos os 
                         per{\'{\i}}odos de estudo. A vari{\'a}vel dist{\^a}ncia da 
                         BR-163 foi a de maior contribui{\c{c}}{\~a}o para a 
                         degrada{\c{c}}{\~a}o florestal no primeiro per{\'{\i}}odo, 
                         apresentando menor contribui{\c{c}}{\~a}o no segundo 
                         per{\'{\i}}odo, n{\~a}o sendo significativa no terceiro. Esse 
                         resultado evidencia o esgotamento dos recursos florestais {\`a}s 
                         margens da BR- 163. No segundo per{\'{\i}}odo, a vari{\'a}vel 
                         dist{\^a}ncia de estradas vicinais {\'e} inclu{\'{\i}}da mas 
                         apresenta uma contribui{\c{c}}{\~a}o relativa inferior no 
                         {\'u}ltimo per{\'{\i}}odo. A dist{\^a}ncia de {\'a}rea 
                         desmatada teve sua contribui{\c{c}}{\~a}o ascendente, sendo o 
                         fator de menor contribui{\c{c}}{\~a}o no segundo 
                         per{\'{\i}}odo, e o de maior contribui{\c{c}}{\~a}o no 
                         {\'u}ltimo per{\'{\i}}odo. A vari{\'a}vel dist{\^a}ncia de 
                         assentamentos rurais foi inclu{\'{\i}}da nos dois {\'u}ltimos 
                         per{\'{\i}}odos, enquanto a vari{\'a}vel densidade de focos de 
                         calor apareceu apenas no terceiro per{\'{\i}}odo. Esses 
                         resultados mostram a heterogeneidade de fatores que influenciam a 
                         degrada{\c{c}}{\~a}o florestal ao longo do tempo e sua 
                         identifica{\c{c}}{\~a}o {\'e} fundamental para subsidiar 
                         a{\c{c}}{\~o}es e pol{\'{\i}}ticas para um controle mais 
                         eficaz da degrada{\c{c}}{\~a}o florestal na Amaz{\^o}nia. 
                         ABSTRACT: The forest degradation process in Amazonia, through 
                         selective logging and forest fires, is one of the main impacts 
                         from the intensification of the occupation in Brazilian Amazon, 
                         with dynamics not yet well known. Different from the clear cut, 
                         forest degradation changes occur gradually and in long-term, which 
                         requires long observation period. The factors related to forest 
                         degradation, such as forest fragmentation and logging activity, 
                         vary through time, related to the stages of development where it 
                         occurs. This works purpose is to identify the main factors, and 
                         its weights, related to the forest degradation intensity in Novo 
                         Progresso region, in different time periods. This region is 
                         considered an active logging frontier which occupation was 
                         intensified after 2000. The methodological procedures were 
                         separated in two steps: the first step is mapping forest 
                         degradation through digital image processing from Landsat sensors 
                         and the second step is the identification of the factors related 
                         to forest degradation using multivariate statistical analysis. 
                         Both the steps were sustained by field trips to the study area. On 
                         the mapping step, images were spectrally classified for the years 
                         2012 to 2017 to identify indicators of forest degradation. Still 
                         on the mapping phase, a second step structurally classified the 
                         forest degradation polygons into degraded forest intensity 
                         patterns. The data produced were combined with the database of 
                         forest degradation intensity from Pinheiro (2015), resulting in a 
                         33-years (1984-2017) time-series of forest degradation intensity 
                         in the region of Novo Progresso. This time series were divided in 
                         three time periods that reflects the study area context of 
                         development: 1984 to 1994, 1995 to 2004 and 2005 to 2017. The 
                         acquisition and analysis of the factors influencing forest 
                         degradation intensity were performed through multivariate 
                         regression models. The models showed that deforestation and roads, 
                         represented by the variables distance from deforested areas, 
                         deforested area proportion, distance from roads and distance from 
                         official roads influenced forest degradation during the study 
                         periods. The variable distance from rural settlements was included 
                         in the last two periods of analysis because people living in the 
                         settlements rely on wood as the primary source of money so they 
                         can invest on cattle raising and/or farming. The distance from the 
                         BR-163 was the factor with the highest contribution to forest 
                         degradation in the first period, turning out to be the variable 
                         with the lowest contribution in the second period, and the 
                         variable distance from unofficial roads becomes the most important 
                         variable. On the last period the distance from BR-163 was not 
                         included in the model, while the distance from unofficial roads 
                         presented a lower relative contribution than the previous periods, 
                         exposing the process of forest resources depletion on the margins 
                         of the BR- 163 between the two last periods. The distance from 
                         deforested areas had been ascending throughout the periods. On the 
                         last period, the kernel heat focus was the variable with the 
                         second highest influence on forest degradation intensity. It was 
                         not found evidences that measures taken by the federal government 
                         to control and reduce forest cover removal had any influence in 
                         forest degradation on Novo Progresso, evidencing the lack of 
                         governance in the region. The results can subsidize individual 
                         actions for a more efficient control of forest degradation in 
                         logging frontiers in the Amazon with context and history similar 
                         to Novo Progresso.",
            committee = "Kampel, Silvana Amaral (presidente) and Escada, Maria Isabel 
                         Sobral (orientadora) and Monteiro, Antonio Miguel Vieira and 
                         Renn{\'o}, Camilo Daleles and Silva, Edson Jos{\'e} Vidal da",
         englishtitle = "Forest degradation factors' dynamics in an Amazonian logging 
                         frontier: the region of Novo Progresso, Par{\'a}",
             language = "pt",
                pages = "99",
                  ibi = "8JMKD3MGP3W34R/3TRFQPB",
                  url = "http://urlib.net/rep/8JMKD3MGP3W34R/3TRFQPB",
           targetfile = "publicacao.pdf",
        urlaccessdate = "21 jan. 2021"
}


Fechar