IDENTIFICAÇÃO DE ÁREAS ALAGADAS NO BIOMA PANTANAL – BRASIL – UTILIZANDO DADOS MULTITEMPORAIS TERRA/MODIS

Autores

  • Juliana Silveira dos SANTOS UFRGS
  • Gabriel PEREIRA INPE
  • Yosio Edemir SHIMABUKURO inpe
  • Bernardo Friedrich Theodor RUDORFF inpe

Resumo

O Pantanal é a maior área úmida tropical do planeta e a principal área alagada do Brasil. As inundações sazonais são uma das principais características deste bioma sendo um fator ecológico fundamental na região. As séries temporais de imagens de sensoriamento remoto são uma importante ferramenta para auxiliar no monitoramento destas áreas. O objetivo deste trabalho foi identificar as áreas alagadas do Bioma Pantanal a partir das variações espaciais e temporais dos valores do índice de vegetação EVI (Enhanced Vegetation Index) extraídos das imagens do sensor MODIS (Moderate Resolution Imaging Spectroradiometer) empregando a análise por componentes principais. Foram utilizadas 217 imagens de EVI do período de 2000 a 2008, e dados de precipitação do Tropical Rainfall Measuring Mission (TRMM). A partir da análise por componentes principais buscou-se representar a variabilidade no tempo de cada pixel que compõe a imagem da região. As áreas que apresentaram maior variabilidade foram classificadas como áreas alagadas. Neste trabalho, foi possível identificar a dinâmica da região, ou seja, períodos de seca e alagamento e obter informações sobre a sazonalidade da vegetação bem como o perfil temporal do EVI para as áreas inundáveis na região Palavras-chave: Sensoriamento remoto. Análise de componentes principais. Índice de vegetação. Identification flooded areas of the Pantanal Biome – Brazil – using multitemporal terra/MODIS data The Pantanal is the largest tropical wetland on the planet and the major flooded area of Brazil. The seasonal floods are one of the major characteristics of this biome being a fundamental ecological factor in the region. The time series of remotely sensed images are an important tool to supervise the monitoring of these areas. The objective of this work was to identify the flooded areas of the Pantanal biome based on the spatial and temporal variation of the EVI (Enhanced Vegetation Index) values extracted from the MODIS (Moderate Resolution Imaging Spectroradiometer) images using principal components analysis. We used 217 EVI images from the period of 2000 to 2008, and precipitation data from the Tropical Rainfall Measuring Mission (TRMM). From the principal components analysis the variability in time of each pixel that composes the image was obtained for the entire region. Areas that presented the greatest variability were classified as flooded areas. In this work, it was possible to identify the dynamics of the region, i.e., periods of drought and flooding; and to obtain information about vegetation seasonality as well as the temporal profile of the EVI for flooded areas in the region. Key words: Remote sensing. Principal components analysis. Vegetation index.

Downloads

Publicado

2011-08-25

Edição

Seção

Artigos

Artigos mais lidos pelo mesmo(s) autor(es)

1 2 > >>