ABSTRACT The Amazon region is one of the most pristine continental areas whose concentrations of atmospheric trace gases and aerosol particles are very low, mainly in the wet season. This study provides observational results of aerosol optical and radiative characteristics in situ as well as atmospheric columnar at a pristine forest in Central Amazonia. Spectral variation of the aerosol properties (aerosol optical depth [AOD], single scattering albedo [SSA], and asymmetry parameter [AP]) was evaluated using the AErosol RObotic NETwork (AERONET) data. The SSA values under natural atmospheric column conditions (AERONET) were compared to the SSA values calculated with in situ measurements. The values of shortwave aerosol radiative forcing (SWARF) on top of the atmosphere (TOA) and the surface (SUR) were estimated using the SBDART model and were validated with the AERONET values with regression analysis. SWARF had a high correlation to TOA (0.97) and SUR (0.92), including dry and wet seasons. Monthly, seasonal and annual mean values of SWARFTOA and SWARFSUR were negative while SWARFATM values were positive. SWARFTOA was -9.18 ± 2.80 W m-2 and SWARFSUR was -20.77 ± 5.04 W m-2 in the dry season, inducing a heating rate (HR) of 0.37 ± 0.13 K.day-1. This study showed that, for a long series of measurements, the effects caused by aerosols on the radiative flux in the pristine forest of Central Amazonia were of the order of SWARFTOA of -3.66 ± 1.59 W m-2 and SWARFSUR of -11.86 ± 2.35 W m-2 during the wet season.

RESUMEN La región del Amazonas es una de las áreas continentales más impolutas, cuyas concentraciones de gases traza atmosféricos y partículas de aerosol son muy bajas, sobre todo en la estación húmeda. El presente estudio proporciona resultados observacionales in situ tanto de las características ópticas y radiativas como de la columna atmosférica de los aerosoles en un bosque virgen de la Amazonia Central. Se evaluó la variación espectral de las propiedades de los aerosoles (profundidad óptica de aerosoles [AOD, por sus siglas en inglés], albedo de dispersión simple [SSA] y parámetro de asimetría [AP]) utilizando para ello datos de la Red Robótica de Aerosoles (AERONET). Los valores de SSA en condiciones naturales de la columna atmosférica (obtenidos de AERONET) se compararon con valores de SSA obtenidos de mediciones in situ. Se estimaron los valores del forzamiento radiativo de onda corta (SWARF) en el techo de la atmósfera (TOA) y en la superficie (SUR) con datos del modelo SBDART y se validaron con valores de AERONET utilizando análisis de regresión. Los valores de SWARF tuvieron una alta correlación con los de TOA (0.97) y SUR (0.92), incluyendo las temporadas húmeda y seca. Los valores promedio mensuales, estacionales y anuales de SWARFTOA y SWARFSUR fueron negativos, en tanto que los de SWARFATM fueron positivos. Los valores de SWARFTOA fueron de -9.18 ± 2.80 W m-2 y los de SWARFSUR de -20.77 ± 5.04 W m-2 en la estación seca, induciendo una tasa de calentamiento (HR) de 0.37 ± 0.13 K día-1. El presente estudio muestra que, para series largas de medición, los efectos ocasionados por los aerosoles en el flujo radiativo en un bosque virgen de la Amazonia Central fueron del orden de -3.66 ± 1.59 W m-2 para SWARFTOA y de -11.86 ± 2.35 W m-2 para SWARFSUR durante la temporada húmeda.

Resumo A emissão de metano é um fator importante na gestão de áreas alagadas como o Pantanal. Zonas úmidas em todo o mundo contribuem para o ciclo global do CH4, emitindo cerca de um terço do CH4 global. No entanto, a dinâmica de CH4 na atmosfera acima do Pantanal, continua mal compreendida. Neste trabalho estudou-se a variabilidade de CH4 na atmosfera na região do Pantanal com base na série temporal de concentração de CH4 (2003-2013), a partir de dados do sensor AIRS a bordo do satélite AQUA, e se analisou sua relação com a precipitação. A metodologia utilizada baseou-se em estatísticas descritivas que incluem médias semanais, anomalias baseadas na variância, correlação cruzada entre precipitação e concentração de CH4, e transformada wavelet. Os resultados mostraram que existe uma concentração média sazonal e a variabilidade da anormalidade de CH4 com uma precipitação é negativa. Variações sub-anuais de CH4 foram inversamente proporcionais às de precipitação, devido possivelmente ao aumento da concentração de hidroxila [OH-] durante a estação chuvosa, o que diminui o saldo de emissão de CH4. A análise wavelet permitiu observar um ciclo anual e interanual da concentração de CH4 entre 2009 a 2013.

Abstract Methane emission is an important factor in the management of wet areas such as the Pantanal region. The temporal variability of CH4 concentrations on the Pantanal was therefore studied using remote sensing data from 2003 to 2013 from an AIRS sensor on board of AQUA satellite. Wetlands worldwide contribute to the overall CH4 cycle, emitting around one-third of the overall CH4. However, the dynamics of CH4 remain poorly understood in the Pantanal. This work studied the variability of CH4 in the atmosphere over the Pantanal of Mato Grosso, based on the time series of CH4 concentration (2002-2013). The methodology used was based on descriptive statistics that included weekly averages, variance anomalies, cross-correlation between precipitation and concentration of CH4, and the transformed wavelet technique. The results showed a seasonal average concentration and the variability of CH4 abnormality in a negative cross-correlation with precipitation. Sub-annual variations of CH4 were inversely related to precipitation, and seemed to be influenced by the concentration of hydroxyl [OH-] during the rainy season, which reduces the balance of CH4 emissions. The wavelet analysis found an annual and inter-annual cycle of CH4 concentration between 2009-2013.

Brazil plays an important role with regard to the emission of aerosols into the atmosphere. Ecosystems such as the Amazonian tropical forest and Brazilian Cerrado are undergoing sweeping changes. These human activities promote an increase in the levels of gases and particles emitted into the atmosphere. To determine how these activities have affected the emission of coarse and fine particulate matter (PM10 and PM2.5), atmospheric aerosol samples were collected continuously from June 2004 to April 2005 in the Cuiaba Basin of south-central Mato Grosso, Brazil. Aerosols were sampled using stacked filter units. Filters were analysed for particulate mass, black carbon and for concentrations of seventeen elements by Particle Induced X-Ray Emission. The results revealed that Mg, Al, Si, S, K, Ca, Ti and Fe were the major components detected in PM2.5 and PM10. The high enrichment factors observed in PM2.5 and PM10 suggest that anthropogenic activities were the predominant source for elements such as Cu, Zn and Pb. These data indicate that most of the aerosols measured in the Cuiaba Basin originate from land conversion and biomass burning. These patterns are similar to those observed in other Cerrado and Amazonian forest areas that are undergoing rapid and spatially extensive land-cover change.

Resumo As variações das propriedades óticas dos aerossóis podem interferir nos processos de transferência de energia entre a atmosfera e a superfície terrestre. A AOD (Aerosol Optical Depth) é a variável mais usada para este propósito, a qual possibilita obtenção das principais tendências e periodicidades das variações. Este trabalho objetivou verificar a habilidade das técnicas SSA (Singular Spectrum Analysis) e Wavelet em descrever as flutuações, e respectivas periodicidades, presentes em uma série temporal de 12 anos (2003-2014) de AOD na faixa do espectro de 550 nm (AOD 550 nm) sobre a região do Pantanal brasileiro. Os dados de AOD 550 nm foram adquiridos dos produtos do sensor MODIS, a bordo dos satélites TERRA e AQUA, e foram submetidos a análise de SSA, permitindo a descrição das principais periodicidades da AOD 550 nm, além de constatar dois períodos dominantes, 1 e 0,5 ano, tendo pequenas variações. A Wavelet de Morlet inferiu as periodicidades estatisticamente significativas, entre as quais também foram detectadas periodicidades anual e semestral. Ambas as técnicas reproduziram as principais oscilações dos valores da AOD 550 nm, bem como a gradativa redução dos valores de AOD 550 nm a partir de 2011. A Wavelet possibilitou a interpretação e visualização das principais periodicidades ao longo do tempo, sendo possível verificar em quais instantes as periodicidades ocorreram.

Abstract Variations of the optical properties of aerosols can interfere in the energy transfer processes between the atmosphere and the Earth's surface. Aerosol Optical Depth (AOD) has been most used variable to investigate such variations, which enabling the achievement of main trends and periodicities from AOD time series. This study aimed to verify the sensitivity of Singular Spectrum Analysis (SSA) and Wavelet techniques on a time series of 12 years (2003-2014) AOD at wavelengths 550 nm (AOD 550 nm) over the Brazilian Pantanal region. AOD 550 nm data set were acquired from MODIS sensor and were subjected by SSA, allowing description of the main frequencies of the AOD 550 nm and of two dominant periods of 1 and 0.5 year. From Morlet Wavelet model were inferred major statistically significant periodicity, among which were also detected an annual and a biannual frequencies. Both techniques explained the main variations of the values of AOD 550 nm on the area of study, moreover, detected a gradual reduction in AOD 550 nm values from 2011. The Wavelet analysis allowed the interpretation and visualization of the main periodicities over time, being possible to identify at what instant such periodicities series occurred.

The Earth's atmosphere has a large number of small solid and liquid particles which differ in size, shape and composition. One of the most obvious characteristics of the phenomena of air pollution is linked to the occurrence of atmospheric aerosol. The research was conducted in a protected nature reserve, Private Natural Heritage Reserve (RPPN) Social Service of Commerce (SESC) Pantanal , at 111 m of altitude (16º39' S, 56º47' W) and is referred to here as SESC - Advanced Research Center of Wetland , located about 160 km from Cuiabá - MT. The sampling device consists of samplers fine and coarse particles, separating the fine fraction of the aerosol (dp <2.5 µm) from the coarse fraction (2.5 <dp <10 µm). In August 2012, concentrations of fine particles were up to nine times higher (up to 18 µg m-3) than that recorded in the rainy months of 2012 and 2013 (amounts around 2 µg m-3). The concentrations of coarse particles were up to three times higher in the dry season, compared to the rainy season.