author = "Fragal, Everton Hafemann and Silva, Thiago Sanna Freire and Novo, 
                         Evlyn M{\'a}rcia Le{\~a}o de Moraes",
          affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and 
                         {Universidade Estadual Paulista (UNESP)} and {Instituto Nacional 
                         de Pesquisas Espaciais (INPE)}",
                title = "Reconstructing historical forest cover change in the Lower Amazon 
                         floodplains using the LandTrendr algorithm",
              journal = "Acta Amazonica",
                 year = "2016",
               volume = "46",
               number = "1",
                pages = "13--24",
                month = "jan./mar.",
             keywords = "Wetlands, flooded forest, land use change, monitoring, Landsat, 
                         {\'A}reas {\'u}midas, florestas inund{\'a}veis, mudan{\c{c}}as 
                         no uso da terra, monitoramento, Landsat.",
             abstract = "The Amazon varzeas are an important component of the Amazon biome, 
                         but anthropic and climatic impacts have been leading to forest 
                         loss and interruption of essential ecosystem functions and 
                         services. The objectives of this study were to evaluate the 
                         capability of the Landsat-based Detection of Trends in Disturbance 
                         and Recovery (LandTrendr) algorithm to characterize changes in 
                         varzea forest cover in the Lower Amazon, and to analyze the 
                         potential of spectral and temporal attributes to classify forest 
                         loss as either natural or anthropogenic. We used a time series of 
                         37 Landsat TM and ETM+ images acquired between 1984 and 2009. We 
                         used the LandTrendr algorithm to detect forest cover change and 
                         the attributes of {"}start year{"}, {"}magnitude{"}, and 
                         {"}duration{"} of the changes, as well as {"}NDVI at the end of 
                         series{"}. Detection was restricted to areas identified as having 
                         forest cover at the start and/or end of the time series. We used 
                         the Support Vector Machine (SVM) algorithm to classify the 
                         extracted attributes, differentiating between anthropogenic and 
                         natural forest loss. Detection reliability was consistently high 
                         for change events along the Amazon River channel, but variable for 
                         changes within the floodplain. Spectral-temporal trajectories 
                         faithfully represented the nature of changes in floodplain forest 
                         cover, corroborating field observations. We estimated 
                         anthropogenic forest losses to be larger (1.071 ha) than natural 
                         losses (884 ha), with a global classification accuracy of 94%. We 
                         conclude that the LandTrendr algorithm is a reliable tool for 
                         studies of forest dynamics throughout the floodplain. RESUMO: As 
                         v{\'a}rzeas amaz{\^o}nicas s{\~a}o um importante componente do 
                         bioma Amaz{\^o}nico, mas impactos antr{\'o}picos e 
                         clim{\'a}ticos t{\^e}m levado {\`a} perda florestal e {\`a} 
                         interrup{\c{c}}{\~a}o de processos e servi{\c{c}}os 
                         ecossist{\^e}micos. O presente estudo teve como objetivos avaliar 
                         a aplicabilidade do algoritmo Landsat-based Detection of Trends in 
                         Disturbance and Recovery (LandTrendr) na detec{\c{c}}{\~a}o de 
                         mudan{\c{c}}as na cobertura florestal de v{\'a}rzea no Baixo 
                         Amazonas, e analisar o potencial de atributos espectrais e 
                         temporais na classifica{\c{c}}{\~a}o das perdas florestais em 
                         antr{\'o}picas ou naturais. Utilizamos uma s{\'e}rie temporal de 
                         37 imagens Landsat TM e ETM+, adquiridas entre 1984 e 2009. 
                         Aplicamos o algoritmo LandTrendr para detectar mudan{\c{c}}as na 
                         cobertura florestal e extrair os atributos de dura{\c{c}}{\~a}o, 
                         magnitude e ano de in{\'{\i}}cio das mudan{\c{c}}as, al{\'e}m 
                         de NDVI ao final da s{\'e}rie. A detec{\c{c}}{\~a}o se 
                         restringiu a {\'a}reas identificadas como cobertura florestal no 
                         in{\'{\i}}cio e/ou final da s{\'e}rie. Os atributos derivados 
                         da s{\'e}rie temporal foram classificados pelo algoritmo Support 
                         Vector Machine (SVM), diferenciando as perdas florestais 
                         antr{\'o}picas e naturais. A confiabilidade da 
                         detec{\c{c}}{\~a}o dos eventos de mudan{\c{c}}a foi 
                         consistentemente alta ao longo do rio Amazonas, e mais 
                         vari{\'a}vel no interior da v{\'a}rzea. As trajet{\'o}rias 
                         espectrais-temporais representaram fielmente os eventos de 
                         mudan{\c{c}}a na cobertura florestal, com base em 
                         averigua{\c{c}}{\~o}es em campo. A perda da cobertura florestal 
                         por causas antr{\'o}picas foi maior (1.071 ha) do que por causas 
                         naturais (884 ha), com exatid{\~a}o global de 
                         classifica{\c{c}}{\~a}o de 94%. Conclu{\'{\i}}mos que o 
                         algoritmo LandTrendr {\'e} uma ferramenta confi{\'a}vel para 
                         aplica{\c{c}}{\~a}o em estudos de din{\^a}mica da cobertura 
                         florestal de v{\'a}rzea.",
                  doi = "10.1590/1809-4392201500835",
                  url = "http://dx.doi.org/10.1590/1809-4392201500835",
                 issn = "0044-5967",
             language = "en",
           targetfile = "Fragal_reconstructing.pdf",
        urlaccessdate = "04 dez. 2020"