Título:	Avaliação da cobertura florestal em terra indígena na Amazônia Legal, por meio de modelagem espectral de sequestro de carbono
Autor(es):	Paiva, Yuri Roberta Yamaguchi de
Orientador(es):	Baptista, Gustavo Macedo de Mello
Assunto:	Desmatamento Amazônia Legal Sensoriamento remoto Análise espectral Processamento de imagens - técnicas digitais
Data de publicação:	14-Mar-2019
Data de defesa:	6-Ago-2018
Referência:	PAIVA, Yuri Roberta Yamaguchi de. Avaliação da cobertura florestal em terra indígena na Amazônia Legal, por meio de modelagem espectral de sequestro de carbono. 2018. 88 f., il. Dissertação (Mestrado em Geociências Aplicadas)—Universidade de Brasília, Brasília, 2018.
Abstract:	Tropical deforestation plays an important role in the global carbon cycle and in the biospheric climate. Approximately 55% of the world’s terrestrial carbon is stored on tropical forests, and deforestation contributes to the release of enormous quantities of CO2 (carbon dioxide) into the atmosphere. The establishment of protective areas, legally constituted, into large forestry areas is mandatory to contain deforestation, which also affects Indigenous Lands; these territories represent 21,2% of the Brazilian Legal Amazon. Even though Indigenous Lands present low deforestation rates, it is important to emphasize that they are susceptible to selective logging, an activity in which the most commercially valuable timber resources are removed from the forest, leaving patches in the landscape. Oppositely to clearing, forest degradation through selective logging causes subtler changes to the forest canopy; such changes are more challenging to be detected by remote sensing. Therefore, the least square linear spectral mixture model was applied in orbital images of the satellite Sentinel-2 (using 10 m spatial resolution bands, as well as 20 m spatial resolution bands resampled to 10 m), in order to estimate the separability of the pixels containing spectral mixture and to estimate the vegetation, soil and shadow fractions that constitute the Karipuna Indigenous Land, situated in the Brazilian Amazon’s state of Rondônia. It was also tested the efficiency of the spectral vegetation indice Spectral Feature Depth Vegetation Index – SFDVI in the detection of selective logging. In this study, 133.732,20 hectares, or 87,65% of photosynthetic active vegetation, have been identified in the Karipuna Indigenous Land (it was obtained through a 78% Pearson’s correlation with the field data, significant at the 1% level). The Person’s correlation goes to 79% when the spectral indice SFDVI is applied to the modeling, which represents an increase of 8.510,18 hectares in the detection of selective logging, or an efficiency 50,7% larger in the identification of the soil fraction in the modeling, resulting in the detection of 125.729 hectares of photosynthetic active vegetation in the referred Indigenous Land (82,44% of the area). The better separability of the vegetation fraction using the spectral indice SFDVI indicates a greater sensibility of the modeling integrated to this spectral indice to the detection of vegetation cover in the Legal Amazon. This is due to the presence of the bands Red Edge 1 and Green, which contributes to a better discrimination phytognomies of the vegetation, as well as the configuration of the indice’s formula itself, which aims to evaluate the depth of the spectral features. Comparing the results obtained by the modeling applied to a Landsat-8/OLI satellite image, which produced a correlation of 42%, leading to a 95% of photosynthetic active vegetation (overestimated values, if compared to the values obtained with the Sentinel-2 analysis), it is possible to conclude that the success of the processing executed in Sentinel-2 satellite images is due two factors: the 10 meters spatial resolution and the inclusion of four bands, settled for vegetation studies: three Red Edge and one band in the near infrared plateau (NIR-2). Monitoring and conservation schemes, such as REDD (Reducing Emissions from Deforestation and forest Degradation) have been demanding more accurate estimates of land cover changes. By those means, the presented results are in accordance with those guidelines, since they demonstrate better accuracy to generate more precise results for the assessment of the Amazon forest cover.
Unidade Acadêmica:	Instituto de Geociências (IG)
Informações adicionais:	Dissertação (mestrado)—Universidade de Brasília, Instituto de Geociências, Programa de Pós-Graduação em Geociências Aplicadas, 2018.
Programa de pós-graduação:	Programa de Pós-Graduação em Geociências Aplicadas e Geodinâmica
Licença:	A concessão da licença deste item refere-se ao termo de autorização impresso assinado pelo autor com as seguintes condições: Na qualidade de titular dos direitos de autor da publicação, autorizo a Universidade de Brasília e o IBICT a disponibilizar por meio dos sites www.bce.unb.br, www.ibict.br, http://hercules.vtls.com/cgi-bin/ndltd/chameleon?lng=pt&skin=ndltd sem ressarcimento dos direitos autorais, de acordo com a Lei nº 9610/98, o texto integral da obra disponibilizada, conforme permissões assinaladas, para fins de leitura, impressão e/ou download, a título de divulgação da produção científica brasileira, a partir desta data.
Aparece nas coleções:	Teses, dissertações e produtos pós-doutorado

Os itens no repositório estão protegidos por copyright, com todos os direitos reservados, salvo quando é indicado o contrário.