Acessibilidade / Reportar erro

Spectral and temporal behavior analysis of coffee crop in Landsat images

The definition of the spectral response of coffee crop is one of the steps to identify coffee fields in remote sensing images in order to map and estimate planted area. The objective of this work was to analyze the potential of the images acquired by the Landsat series satellites, for coffee crop mapping and forecast. A temporal analysis of the spectral behavior of coffee crop fields under development and under active production was performed through cloud free images acquired in the years of 1999 and 2001. The spectral behavior of pasture and forest was also analyzed due to their relevance in the study area. The results showed that images acquired during the dry season were more efficient to map coffee crop at early development and under production. Band 4 (near infrared) of both sensors (TM and ETM+) presented best performance for spectral differentiation between coffee crop and other scene targets. The analysis of the reflectance values for active producing coffee crop showed a high spectral variability which may be attributed to age, plants spacing, cultivar, indicating a need for field work for the identification of coffee crop in Landsat scenes.

Coffea arabica; remote sensing; geoprocessing; crop forecast


Embrapa Secretaria de Pesquisa e Desenvolvimento; Pesquisa Agropecuária Brasileira Caixa Postal 040315, 70770-901 Brasília DF Brazil, Tel. +55 61 3448-1813, Fax +55 61 3340-5483 - Brasília - DF - Brazil
E-mail: pab@embrapa.br