Gabriel Pereira

Spectral unmixing

Satellite imagery is formed by finite digital numbers representing a specific location of ground surface in which each matrix element is denominated as a picture element or pixel. The pixels represent the sensor measurements of spectral radiance. The radiance recorded in the satellite images is then an integrated sum of the radiances of all targets within the instantaneous field of view (IFOV) of the sensors. Therefore, the radiation detected is caused by a mixture of several different materials within the image pixels. For this reason, spectral unmixing has been used as a technique for analysing the mixture of components in remotely sensed images for almost 30 years. Different spectral unmixing approaches have been described in the literature. In recent years, many authors have proposed more complex models that permit obtaining a higher accuracy and use less computing time. Although the most widely used method consists of employing a single set of endmembers (typically three or four) on the whole image and using a constrained least squares method to perform the unmixing linearly, every different algorithm has its own merits and no single approach is optimal and applicable to all cases. Additionally, the number of applications using unmixing techniques is increasing. Spectral unmixing techniques are used mainly for providing information to monitor different natural resources (agricultural, forest, geological, etc.) and environmental problems (erosion, deforestation, plagues and disease, forest fires, etc.). This article is a comprehensive exploration of all of the major unmixing approaches and their applications.

Large-scale heterogeneity of Amazonian phenology revealed from 26-year long AVHRR/NDVI time-series

Depiction of phenological cycles in tropical forests is critical for an understanding of seasonal patterns in carbon and water fluxes as well as the responses of vegetation to climate variations. However, the detection of clear spatially explicit phenological patterns across Amazonia has proven difficult using data from the Moderate Resolution Imaging Spectroradiometer (MODIS). In this work, we propose an alternative approach based on a 26-year time-series of the normalized difference vegetation index (NDVI) from the Advanced Very High Resolution Radiometer (AVHRR) to identify regions with homogeneous phenological cycles in Amazonia. Specifically, we aim to use a pattern recognition technique, based on temporal signal processing concepts, to map Amazonian phenoregions and to compare the identified patterns with field-derived information. Our automated method recognized 26 phenoregions with unique intra-annual seasonality. This result highlights the fact that known vegetation types in Amazonia are not only structurally different but also phenologically distinct. Flushing of new leaves observed in the field is, in most cases, associated to a continuous increase in NDVI. The peak in leaf production is normally observed from the beginning to the middle of the wet season in 66% of the field sites analyzed. The phenoregion map presented in this work gives a new perspective on the dynamics of Amazonian canopies. It is clear that the phenology across Amazonia is more variable than previously detected using remote sensing data. An understanding of the implications of this spatial heterogeneity on the seasonality of Amazonian forest processes is a crucial step towards accurately quantifying the role of tropical forests within global biogeochemical cycles.

Monitoring the transport of biomass burning emission in South America

Abstract The main objective of this work is to use Fire Radiative Power (FRP) to estimate particulate matter with diameter less than 2.5 μm (PM2.5) and carbon monoxide (CO) emissions for the South America 2002 burning season. Sixteen small–scale combustion experiments were performed near the Laboratory of Radiometry (LARAD) at the National Institute for Space Research (DSR/INPE) to obtain the coefficient that relates the biomass consumption with the FRP released. The fire products MOD14/MYD14 from the MODIS Terra/Aqua platforms and the Wildfire Automated Biomass Burning Algorithm (WFABBA) on the Geostationary Operational Environmental Satellite (GOES) were utilized to calculate the total amount of biomass burned. This inventory is modeled in the Coupled Chemistry–Aerosol–Tracer Transport model coupled to the Brazilian developments on the Regional Atmospheric Modeling System (CCATT–BRAMS) and compared with data collected in the Large Scale Biosphere–Atmosphere (LBA) Smoke, Aerosols, Clouds, rainfall, and Climate (SMOCC) and Radiation, Cloud, and Climate Interactions (RaCCI) Experiments. The relationship between the modeled PM2.5 and CO shows a good agreement with SMOCC/RaCCI data in the general pattern of temporal evolution. The results showed high correlations, with values between 0.80 and 0.95 (significant at 0.05 level by student t–test), for the CCATT–BRAMS simulations with PM2.5 and CO. Furthermore, the slope analysis reveals an underestimation of emission values with CCATT–BRAMS modeled values, 20–30% lower than observed data with discrepancies mainly on days with large fires. However, the underestimation is similar to the uncertainties in traditional emissions methods.

Validation of MODIS MCD45A1 product to identify burned areas in Acre State - Amazon forest

Burned areas map is essential in many applications and the orbital sensors have been used to monitor fires for many years, providing a better understanding of processes at different scales and being the only practical technique to estimate fires in large areas. However, remote sensing methods have limitations that could cause errors in the final products. Therefore, the objective of this work is to evaluate the MCD45A1 burned area product derived from MODIS sensor in Amazon tropical forest by comparing this dataset with the reference data derived from the mapping of burned areas in Acre State/Brazil acquired by TM sensor aboard of Landsat 5 and with a fieldwork that took place in November 2011. The results showed that de MCD45A1 product presented 93% of omission errors in 2010 and 96% in 2011 year in relation to reference data, presenting a low confidence in identifying the burned areas in Amazon region.

O uso da energia radiativa do fogo para estimar as emissões de queimadas para a América do Sul

Every year large areas of the globe are submitted to the action of anthropogenic and natural fires. It is estimated that more than 100 million tons of aerosols from smoke are emitted into the atmosphere, which 80% occur in tropical regions of the globe. The process of biomass burning releases into the atmosphere trace gases and aerosol particles that affect significantly the air quality, the tropospheric and stratospheric chemistry, the radiation balance and the dynamics and microphysics of clouds. This study aims to use the fire radiative energy (FRE) derived from MODIS (Moderate Resolution Imaging Spectroradiometer) and GOES (Geostationary Operational Environmental Satellite) to estimate emissions of carbon monoxide (CO) and particulate matter with diameter less than 2.5 micrometer (PM2.5µm) for 2002 South America fires, and to model these emissions in CCATT-BRAMS (Coupled Chemistry-Aerosol-Tracer Transport model coupled to Brazilian Regional Atmospheric Modeling System). With a correlation greater than 86% between the aerosol emission data (in kg.s -1) and the FRE (MJ.s -1), three coefficients for GOES satellite data were originated. The use of FRE and emission coefficients to estimate the PM2.5µm m and CO emitted in biomass burning showed a correlation of approximately 91 % between modeled data and the data used as ground truth obtained from LBA (Large Scale Biosphere-Atmosphere Experiment in Amazonia) SMOCC (Smoke, Aerosols, Clouds, rainfall, and Climate) and RaCCI (Radiation, Cloud, and Climate Interactions).

AVALIAÇÃO DE IMAGENS SIMULADAS DA CÂMERA MUX DO SATÉLITE CBERS-4 APLICADAS À ANÁLISE AMBIENTAL

Metodos de simulacao de imagens orbitais sao frequentemente utilizadas na avaliacao do desempenho de determinado sistema-sensor. A partir do emprego destas tecnicas e possivel analisar e estimar o comportamento das imagens que serao geradas pelos sensores projetados, possibilitando uma estimativa da qualidade e das aplicacoes decorrentes do lancamento do satelite. Neste contexto, torna-se de fundamental importância a analise das imagens orbitais e das possiveis aplicacoes provenientes do satelite CBERS-4, que deve ser lancado ao final do ano de 2014 e tera uma politica de distribuicao gratuita dos dados. Deste modo, o objetivo deste trabalho e avaliar o potencial da câmera MUX do CBERS-4, com 20 m de resolucao espacial, para mapeamento de cobertura do solo do municipio de Apui no estado do Amazonas. Para isto, as imagens MUX sao simuladas a partir de imagens do satelite RapidEye e filtragem baseada no modelo do processo de imageamento. Para avaliar os resultados da simulacao, uma imagem da câmera TM do satelite Landsat-5 e processada para produzir um mapa de cobertura de solo, que e comparada ao mapa gerado pela imagem MUX simulada do CBERS-4. Os valores de NDVI calculados a partir das imagens MUX simulada e TM-5 tambem sao analisados. Os experimentos mostram que o processamento das imagens simuladas da câmera MUX apresentaram resultados semelhantes aos das imagens do sensor TM. Em geral, as classificacoes da cobertura do solo para os sensores MUX e TM apresentam boa concordância, com acuracia global de 87% e Kappa de 0,72. Ainda, percebe-se que os valores de NDVI estimados pela MUX sao em media 25% maiores que os valores estimados pelo TM e apresentam uma correlacao de 85% (significante a 0.05, teste t-student).

Estimation of instantaneous fire flaming and smoldering size to Amazon Rainforest

Wildfires plays a fundamental intervention in global biogeochemical cycle, by the chemical reaction occurring in the combustion process, and the organic compounds present in vegetation returns to the atmosphere and soil in a cyclical behavior. Therefore, the main objective of this work is to develop a method to estimate the instantaneous fire size in Brazil using Thematic Mapper (TM) aboard of Landsat 5 and Enhanced Thematic Mapper Plus (ETM+) aboard of Landsat 7. The results indicate that active fire to pastures/grasslands is approximately 38% higher than that found for forest areas, 31% higher than the same coefficient used to estimate the fire size in areas of herbaceous and shrub vegetation and 11% higher than the coefficient used in agricultural areas.

Spectral signature of leaves of amazon rainforest tree species

Remote sensing is based on the interaction of electromagnetic radiation with the portion of the electromagnetic radiation that interacts with Earth surface targets. Studies to analyze the spectrum of reflectance of surface targets by sensor systems could be done by terrestrial (laboratory and field), aerial and orbital acquisitions. Consequently, for vegetation studies through remote sensing observations is necessary to know the physiology of the plant studied and the reflectance spectrum, considering that the solar radiation reaches the earths surface and results in three fractions: one part is absorbed, another is reflected and a third part is transmitted. The leaves are the principal absorber of electromagnetic radiation in a canopy and represents the element that more contribute to the signal detected by orbital sensors. The present work has as its main goal the analysis of the spectral signature responses of common species of several forest functional types in a tropical forest area in the Amazon.