Cover: Multitemporal fraction images derived from Terra MODIS data for analysing land cover change over the Amazon region
| Autor: | Liana O. Anderson, Y. E. Shimabukuro, Egidio Arai |
|---|---|
| Rok vydání: | 2005 |
| Předmět: | |
| Zdroj: | International Journal of Remote Sensing. 26:2251-2257 |
| ISSN: | 1366-5901 0143-1161 |
| Popis: | L. O. ANDERSON*, Y. E. SHIMABUKURO and E. ARAIInstituto Nacional de Pesquisas Espaciais, Divisao de Sensoriamento Remoto, Av. dosAstronautas, 1758, CEP 12227-010, Sao Jose dos Campos, SP, BrazilThe Moderate Resolution Imaging Spectroradiometer (MODIS) instrtimentonboard Earth Observing System (EOS) Terra plataform has been designed toprovide improved information for monitoring land, ocean, and atmosphereconditions. The design combined characteristics oi' the Advanced Very HighResolution Radiometer (AVHRR) and the Landsat Thematic Mapper (TM),adding spectral channels in the middle and thermal infrared wavelength andproviding data in 250 m. 500 m and 1 km spatial resolutions. Spectral channels forattnospheric and cloud characterization have been included to permit both theremoval of atmospheric effects on surface observations and the provision ofatmospheric measurements (Justice et al. 1998). This work utilized the land productM0D13Q1, which is a vegetation index product with 250m spatial resolution and isa composite of 16 days of observation over Mato Grosso State (figure 1). Thisproduct contains the composite of NDVI (Normalized Difference VegetationIndex), EVI (Enhanced Vegetation Index), red reflectance band {250m spatialresolution, 620-670 nm bandwidth), near infrared reflectance band (250 m spatialresolution, 841-876 nm bandwidth), blue reflectance band (500 m spatial resolution,rearranged to 250m, 459-479nm bandwidth), medium infrared reflectance band(500m spatial resolution, rearranged to 250m, 2105-2155nm bandwidth), qualityassurance data for NDVI and EVI. view zenith angle, sun zenith angle, and relativeazimuth angle average parameters (Lozar and Balbach 2002). The composites ofNDVI and EVI were performed using the Constrained View Maximum ValueComposite (CV-MVC) algorithm.The pixel of moderate resolution sensor images, due to its spatial resolution,generally includes more than one type of terrain cover. When these sensors observethe Earth, the measured radiance is the integration of the radiance of all the objectsthat are contained within the pixel, implying the existence of the so-called mixtureproblem (Aguiar el al. 1999). The linear mixing model has been used to analyse themixture of signatures of vegetation, soil, and shade in each pixel for either high(Airborne Visible/Infrared Imaging Spectrometer (AVIRIS). TM, etc.) and coarse(AVHRR and Systeme Pour l'Observation de la Terre (SPOT-Vegetation))resolution data. The available methods estimate the proportion of each componentinside the pixel by minimizing the sum of squares of the errors. The proportionvalues must be non-negative, and they also must equal one (Shimabukuro and Smith1991). In this work, we accept that it is possible to find pure pixels of land cover typein the 250 m spatial resolution data, and the purest signatures for vegetation, soil,and shade endmembers can be seen in figure 2. For the coming work to investigate*Corresponding author. Email: liana@ltid.inpe.br |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|