• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.8.2021.tde-16072021-150114
Document
Author
Full name
Marcos Reis Rosa
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2021
Supervisor
Committee
Barrozo, Ligia Vizeu (President)
Júnior, Laerte Guimarães Ferreira
Kawakubo, Fernando Shinji
Quintanilha, Jose Alberto
Title in Portuguese
Metodologia de classificação de uso e cobertura da terra para análise de três décadas de ganho e perda anual da cobertura florestal nativa na mata atlântica
Keywords in Portuguese
Cobertura da terra
Floresta nativa
Mata Atlântica
Abstract in Portuguese
Neste trabalho apresentamos uma análise da dinâmica anual da floresta nativa de 1985 a 2019 na Mata Atlântica. Quantificamos pela primeira vez a dinâmica da floresta nativa em larga escala e de longo prazo com base em mapas anuais de uso e cobertura da terra produzidos com uma metodologia inédita usando um algoritmo de aprendizado de máquina, amostras estáveis sorteadas em classes invariáveis e processamento em nuvem da plataforma Google Earth Engine para classificar os mosaicos anuais produzidos a partir de todas as imagens Landsat com índices de refletância, temporais e de textura. Filtros de pós-classificação espacial e temporal melhoraram a qualidade dos mapas anuais. Análises regionais demonstram que a perda continuada de florestas nativas mais antigas ameaça os ganhos obtidos com a recuperação florestal e indicam que as ações de combate ao desmatamento e preservação dos remanescentes florestais atuais devem ser realizadas paralelamente à restauração de matas nativas envolvendo a população local.
Title in English
Land use and land cover classification methodology to analyze three decades of native forest coverage loss and gain in the Atlantic Forest
Keywords in English
Atlantic Forest
Land cover
Native forest
Abstract in English
In this work, we present an analysis of the annual native forest dynamics from 1985 to 2019 in the Brazilian Atlantic Forest. We quantified for the first time, the large-scale, long-term native forest dynamics based on annual maps of land use and land cover produced with a new methodological procedure using a machine learning algorithm, stable samples located over invariant classes, and new cloud processing capabilities from the Google Earth Engine platform to classify Landsat annual mosaics produced using information from all available Landsat images with reflectance, temporal and texture indices. Spatial and temporal post-classification filters improved the accuracy of the annual maps. Regional analyses demonstrated the ongoing loss of older native forests threatens the gains obtained from forest recovery and indicates that actions to combat deforestation and preserve current forest remnants must be carried out in parallel with the restoration of native forests involving the local population.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2021-07-16
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.