Air pollution in Brazilian agricultural-urban interfaces is strongly driven by land-use change and seasonal biomass burning. In southeastern Brazil's Atlantic Forest biome, these processes interact with local meteorology to produce complex air-quality dynamics. This study provides one of the first long-term (2017-2025), observation-based assessments of air pollution variability and machine-learning diagnostic prediction of particulate matter in an agricultural-urban interface of the Atlantic Forest biome affected by recurrent biomass burning. In situ measurements of carbon monoxide (CO), nitrogen oxides (NOx), ozone (O-3), fine particulate matter 2.5 mu m (PM2.5) and coarse particulate matter 10 mu m (PM10) from the local air-quality monitoring station revealed pronounced seasonality, with particulate concentrations peaking during the winter (dry season) under intense solar radiation, low humidity, strong biomass burning, and weak winds that favor pollutant accumulation. Exceedances of World Health Organization (WHO) air-quality guidelines occurred on more than 30% of monitored days for PM2.5, with 2024 being the most polluted year. Multiple statistical and machine learning models, including multiple linear regression, generalized additive models (GAM), random forest (RF), extreme gradient boosting (XGBoost), deep neural network (DNN), and long-short term memory (LSTM), were evaluated for same-day particulate matter (PM) prediction. Multiple linear regression, GAM, and tree-based models captured the mean behavior of PM2.5 and PM10 but systematically underestimated high-concentration regimes, while the LSTM model did not improve the representation of daily PM variability. Incorporating feature-engineered temporal predictors improved performance, particularly for XGBoost, which identified CO, NOx, relative humidity, and dew-point temperature as dominant predictors. The feature-engineered DNN, obtained with time-lagged features, achieved the best results (PM2.5 and PM10: R-2 = 0.96; and MAE = 2.02 and 5.17 mu g m(-3), respectively), better reproducing both low-and high-concentration regimes. A simple persistence baseline was used as a reference benchmark. RF, XGBoost, and DNN consistently outperformed it, whereas the LSTM did not outperform the persistence baseline, even with feature engineering, highlighting the limitations of recurrent architectures for this dataset and modeling configuration. These results show that particulate matter variability in this agricultural-urban interface is controlled by the combined influence of combustion tracers, atmospheric moisture, and temporal persistence. They also highlight the value of combining long-term observations with diagnostic machine-learning benchmarks to support PM estimation, gap filling, and the future development of site-specific air-quality assessment frameworks in biomass-burning regions.

Urban areas in tropical regions are increasingly exposed to thermal extremes because of rapid urbanization, land-use changes, and air pollution. Atmospheric aerosols can modulate urban energy balance through radiative interactions; however, their relationship with air temperature in tropical cities remains insufficiently quantified. In this study, we investigated the relationship between atmospheric aerosols and maximum air temperature (T-MAX) in the Metropolitan Region of Bel & eacute;m (MRB), Eastern Amazon, using a combination of long-term in situ meteorological observations (2003-2024) and aerosol estimates from MERRA-2 reanalysis. The results indicate a positive and statistically significant, although moderate, relationship between aerosol optical depth (AOD) and T-MAX during the dry season under clear-sky conditions (r = 0.41, p < 0.001). Surface concentrations of fine particulate matter (PM2.5) and black carbon (BC) also showed statistically significant relationships with T-MAX (r = 0.20 and r = 0.27, respectively; p < 0.001), supporting the hypothesis that absorbing aerosols may influence thermal conditions near the surface. Despite predominantly negative radiative forcing at the surface, atmospheric absorption associated with BC may contribute to increased atmospheric heating rates, favoring higher T-MAX. Trend analysis revealed a simultaneous increase in BC concentrations (0.18 mu gm(-3)) and T-MAX (1.2 degrees C) over the last two decades. These findings reinforce that aerosol-radiation interactions can contribute to maximum air temperature variability in tropical urban environments, although this influence is modulated by multiple atmospheric processes; this underscores the need to incorporate the effects of air pollution into urban heat risk assessments for urban areas in the Amazon.

Air pollution remains a major global environmental risk, and exposure to fine particulate matter (PM2.5) is associated with adverse health outcomes even at low concentrations. Meteorological conditions influence PM2.5 variability, and precipitation is often expected to reduce particle loads through wet removal. However, humid and wet conditions may coincide with elevated PM2.5 under specific atmospheric and compositional conditions. Here, we investigate long-term relationships between precipitation regimes and PM2.5 concentrations in the Metropolitan Region of Bel & eacute;m (Eastern Amazonia) over the period 1980-2024. We combined PM2.5 from the MERRA-2 reanalysis (including a bias-corrected product) with in situ precipitation records, and classified precipitation conditions using the Standardized Precipitation Index (SPI). We find statistically significant positive long-term tendencies in both precipitation and PM2.5. Stratified analyses show that PM2.5 concentrations are significantly higher under wet conditions, with a weak but significant positive relationship between SPI and PM2.5 (r = 0.23 for the full period; r = 0.24 for the wet class, p-value < 0.01). These findings indicate that increased precipitation in a strong humid tropical urban environment does not necessarily lead to improved air quality. Instead, wet conditions may favor processes such as hygroscopic growth and secondary aerosol formation, contributing to higher PM2.5 concentrations on a monthly scale. Overall, this study highlights the importance of considering precipitation regimes and associated atmospheric processes when assessing air quality in tropical urban environments.

Atmospheric aerosols play a crucial role in modulating the energy available to the Earth's surface, influencing the hydrological cycle, ecosystems, and climate. In the Amazon, previous studies have mainly examined how aerosols scatter and absorb radiation. However, little is known about their interactions with energy partitioning (i.e., sensible and latent heat fluxes). Here, we investigate how regimes of high (AOD >0.40) and low (AOD <0.13) aerosol optical depth (AOD) affect surface energy and carbon dioxide (CO2) fluxes in an undisturbed Amazon rainforest. For this, we used long-term meteorological measurements from the Amazon Tall Tower Observatory (ATTO) collected between 2016 and 2022. We find that enhanced aerosol presence reduces both sensible heat flux and energy available for evapotranspiration by approximately 13.5 % and 2.1 % respectively, while increasing CO2 uptake (i.e., CO2 flux becoming more negative) by about 39.5 %. The impact of aerosols on turbulent surface fluxes is reflected in a cooling of approximately 0.9 degrees C at the canopy top, caused by a 2.8% reduction in incoming shortwave radiation. These results demonstrate that aerosols modify turbulent energy exchange, with consequences for the forest microclimate and the coupled carbon and water cycles.

Aerosol particles formed by new particle formation (NPF) are essential for cloud condensation nuclei and can strongly influence cloud properties and climate. However, the mechanisms behind NPF in the Amazon boundary layer have remained elusive. Classical "banana" NPF events, common in other continental regions, are rarely observed in the Amazon, while most detected sub-50 nm particles have been linked to precipitation- and downdraft-related episodes, often called Amazonian banana events. Here, we analyse a decade of particle number size distributions (10-420 nm) from the Amazon Tall Tower Observatory (ATTO) during the wet season and demonstrate the presence of a distinct phenomenon called Quiet NPF. This process represents a subtle but persistent background particle formation, occurring on days without clear banana-type growth signatures. Using a statistical approach, we show that Quiet NPF links freshly formed 10 nm particles to their subsequent growth into the Aitken mode. This mechanism is characterized by a growth rate of 2.4 +/- 0.1 nm h-1, about half that of Amazonian banana events, but occurs much more frequently. Quiet NPF accounts for similar to 45 % of 10-25 nm particle production during the wet season, revealing an overlooked but important source of nanoparticles that contributes to sustaining Amazonian aerosol populations.

Cities account for over half of global greenhouse gas (GHG) emissions, yet monitoring is very limited, especially in the Global South. A GHG monitoring network was established in 2019 in S & atilde;o Paulo, the largest metropolitan area in the Southern Hemisphere. Five sites were strategically selected, including four measuring surface mole fraction of carbon monoxide (CO), carbon dioxide (CO2), and methane (CH4): Pico do Jaragua (PDJ; -23.46 degrees, -46.77 degrees), IAG (-23.56 degrees, -46.73 degrees), UNICID (-23.54 degrees, -46.56 degrees), and ICESP (-23.56 degrees, -46.67 degrees). The fifth site, CIENTEC (-23.65 degrees, -46.62 degrees), was dedicated to measuring CO2 fluxes within a vegetated area. At IAG, delta C-13-CO2 was also measured. The results revealed the influence of local sources, vegetation, and surface characteristics on GHG patterns. The elevated, forested site of PDJ provided urban background GHG levels, while urban core sites were heavily affected by traffic and biomass burning. CO2 levels increased across the city, with annual growth rates of 4.86 +/- 2.84 ppm at UNICID (from 2021 to 2024), 4.92 +/- 0.56 ppm at IAG (from 2021 to 2024), and 1.67 +/- 0.38 ppm at PDJ (2019-2024). CH4 levels were more stable, exhibiting minor seasonal variability and no clear trend at PDJ. At UNICID, strong CO2-CO correlations (r > 0.9) during traffic peaks suggested emissions dominated by combustion, while weaker correlations during outside peak hours (r < 0.7) indicated the uptake of CO2 by the vegetation. Diurnal CO2 patterns across stations reflected the balance between human activity and biogenic processes, driven by photosynthesis and respiration, emphasizing the role of vegetation in modulating urban CO2 levels. These results highlight the importance of urban GHG measurements for understanding emission dynamics, assessing the influence of urban structure and meteorology, and supporting mitigation strategies at the city scale. This work clearly highlights the need to establish similar networks across Latin America to fill this critical data gap, which is key for developing effective climate mitigation strategies.

This study analyzes spatial-temporal deforestation patterns in Amazonas using 36 years of land use and land cover changes . We identified contiguous deforestation patches for each year and characterized their evolution using two geometric metrics: compactness, related to the shape of the patch and equivalent radius, proportional to the deforested area. These metrics enabled the aggregation of deforestation patches into four distinct regions within the Amazon, each exhibiting unique yet consistent characteristics with different temporal evolution. Typical distributions were found for these two metrics that allow to characterize space and time evolution of the deforestation for different land-use. Pasture patches showed a gamma distribution, while agricultural lands followed a lognormal distribution. Over time, pastures exhibited a trend towards lower compactness values, whereas agriculture and silviculture demonstrated shifts towards higher compactness. The equivalent radius distribution showed increased frequency of larger deforested areas over time. These findings underscore the utility of simple geometric metrics in understanding deforestation's spatial and temporal evolution, offering valuable insights into land-use dynamics estimation in the Amazon and providing a foundation for more effective monitoring and conservation strategies.

In central Amazonia, aerosol sources, weather, and chemical processes create a highly variable aerosol population. The aerosols' optical properties, shaped by composition and size, determine sunlight interaction and the regional radiation budget. Previous studies observed differences in the particles' physical properties during smoke events and described their vertical gradients during clean periods. However, a complete characterization of these properties at two height levels considering both seasons is still missing. This study connects aerosol optical measurements from the Amazon Tall Tower Observatory (ATTO), at 60 and 325 m heights, to particle composition and sources, characterizing different aerosol populations, assessing their vertical gradients, and associating them with the influence of various emission sources and atmospheric processes. A seasonally segregated clustering method was applied to five years of optical data (2018–2023), allowing for the identification of periods with low biomass-burning impact, long-range transport (LRT) events, and regional pollution episodes. Aerosols from Saharan dust events showed the highest real and imaginary refractive index, along with a large inorganic mass fraction (around 26 %), which differs from typical Amazonian conditions. Furthermore, regional biomass-burning emissions during the dry season promoted elevated fine-mode particle concentrations (median 2250 cm−3), dominated by absorbing carbonaceous material. These particles also showed the maximum mass scattering efficiency, which was consistently higher at the 60 m height, underscoring the importance of vertical transport and aerosol aging processes. These results indicate that the clustering method can discriminate between aerosol populations and elucidate differences between particles of different sources and processes influencing the Amazonian atmosphere.

Air pollution has significant implications for the climate and poses irreversible risks to human health. The Amazon region of Brazil is severely affected by biomass burning (BB) emissions, yet air quality monitoring remains highly inadequate. Given the scarcity of surface-based observations, reanalysis models have become essential tools for assessing air pollution. Although MERRA-2 and CAMS PM2.5 products are widely utilized, their validation and comprehensive evaluation for the Amazon Basin remain limited. This study assesses the performance of these products in a semi-urbanized region in the southern Amazon. The calibrated time series was employed to analyze PM2.5 concentrations from 2003 to 2023. Our results showed satisfactory performance of both products for the 24-h averages of PM2.5, with linear correlations above 0.76. However, it was found that both products overestimate surface concentrations. MERRA-2 performed better, with approximately 30

Air quality monitoring stations are unequally distributed worldwide despite the relevance of the air pollution impacts. After validation, atmospheric composition reanalysis can fill information gaps in locations where air quality observations are absent. Reanalysis datasets are based on global emission inventories, often overlooking regional characteristics. This underscores the need for regional and local evaluation studies, which remain scarce in South America. This study presents the first evaluation of atmospheric composition reanalysis products in the Metropolitan Area of São Paulo (MASP), Brazil. Two reanalysis products were evaluated: MERRA-2 (Modern-Era Retrospective Analysis for Research and Applications) and CAMS (Copernicus Atmospheric Monitoring Service). Four air pollutants were assessed, considering monthly concentration time series between 2015 and 2019: ozone (O3), nitrogen dioxide (NO2), inhalable (PM10) and fine particulate matter (PM2.5). Near-surface reanalysis was compared with air quality monitoring stations in the MASP. CAMS correctly reproduced the seasonal and interannual variability of concentrations, with significant Pearson correlation coefficients in the range of 0.75–0.89. However, CAMS overestimated O3, PM2.5 and PM10 by 136, 50 and 16% respectively. By contrast, MERRA-2 failed to reproduce the main features of air pollutant seasonal variability in the MASP, especially for PM2.5 and PM10. Based on these findings, we conclude that CAMS adequately represents near-surface air quality conditions in the MASP, although bias corrections are required. This means that the CAMS reanalysis data may be used to obtain information about air quality conditions in cities where local monitoring is absent, at least in Brazilian cities near the MASP. Further studies are necessary to investigate the adequacy of CAMS in other Brazilian regions.

The Amazon rainforest is one of Earth's most diverse ecosystems, playing a key role in maintaining regional and global climate stability. However, recent changes in land use, vegetation, and the climate have disrupted biosphere-atmosphere interactions, leading to significant alterations in the water, energy, and carbon cycles. These disturbances have far-reaching consequences for the entire Earth system. Here, we quantify the relative contributions of deforestation and global climate change to observed shifts in key Amazonian climate parameters. We analyzed long-term atmospheric and land cover change data across 29 areas in the Brazilian Legal Amazon from 1985 to 2020, using parametric statistical models to disentangle the effects of forest loss and alterations of temperature, precipitation, and greenhouse gas mixing ratios. While the rise in atmospheric methane (CH4) and carbon dioxide (CO2) mixing ratios is primarily driven by global emissions (>99%), deforestation has significantly increased surface air temperatures and reduced precipitation during the Amazonian dry season. Over the past 35 years, deforestation has accounted for approximately 74% of the  ~ 21 mm dry season-1 decline and 16.5% of the 2°C rise in maximum surface air temperature. Understanding the interplay between global climate change and deforestation is essential for developing effective mitigation and adaptation strategies to preserve this vital ecosystem.

Scientific research in Amazonia plays a fundamental role in identifying pathways to sustainable development for the region, addressing the challenges posed by climate change, preserving its unique ecosystems, and aligning with societal challenges and rights advocated by its diverse populations. This paper encompasses a broad range of scientific publications, spanning from 1977 to 2024, and highlights key research areas, analyzing their results and trends to inform future developments. It also identifies areas that require deeper investigation. The results emphasize a focus on agricultural, biological, and environmental sciences. On the other hand, there is a need for more extensive research within the social sciences. As shown, research on indigenous land rights, cultural heritage, and the socio-economic impacts of environmental disruptions is essential for developing comprehensive conservation strategies. Furthermore, research on governance, policy, and socio-political dynamics in Amazonia can provide innovative approaches to addressing the challenges and opportunities for its people, biodiversity, and role in climate regulation, as demonstrated by the findings. The strategic research fields identified in this paper provide a guide for future studies and policy development aimed at protecting the forest and its inhabitants. This study emphasizes the need for approaches that integrate both natural and social sciences as essential for addressing the complex ecological and socio-economic challenges that continue to shape the contemporary research landscape. Furthermore, this paper highlights the importance of unity and cooperation among Amazonian countries and research institutions in achieving these goals. In this context, reinforcing long-term, large-scale research programs such as the LBA (Large-Scale Biosphere–Atmosphere Experiment in Amazonia) and the Scientific Panel for the Amazon (SPA) are crucial to advancing integrated, policy-relevant science for the sustainable future of the region.

As the impacts of climate change become increasingly evident, understanding the role of atmospheric aerosols in regulating Earth’s climate is crucial. Despite their significance, the optical properties of aerosols remain unclear to the general public and, particularly, to students in Physical and Earth sciences. This paper presents a didactic approach to exploring variations in aerosol optical properties, focusing on refractive indices and utilizing data from the National Aeronautics and Space Administration’s (NASA) AErosol RObotic NETwork (AERONET) under typical atmospheric conditions in central Amazon. The discussion presented here can be integrated into undergraduate and graduate university courses, linking Physics and environmental sciences and enhancing theoretical discussions in electromagnetism courses with observational data. The Amazon rainforest, with its dynamic atmospheric conditions influenced by seasonal changes, biomass-burning events, and long-range aerosol transport, provides a rich context for this analysis. Our study identifies seasonal patterns in aerosol optical depth, with biomass burning affecting light scattering at shorter wavelengths. African dust and smoke show strong absorption across a broader spectrum. We characterize the refractive index’s wavelength-dependent behavior for the different aerosol conditions, highlighting their unique optical properties. This interdisciplinary analysis contributes to a deeper understanding of atmospheric sciences and its implications for climate modeling and environmental assessment.

We present results from ATTO-Campina, a new permanent observational site in central Amazon, about 4 km from the ATTO towers. Operational since 2020, ATTO-Campina characterizes atmospheric, cloud and rainfall properties through remote sensing. The goal is to provide continuous, complementary measurements to the ATTO towers, addressing the rainforest’s complex gas-aerosol-cloud-precipitation dynamics. Using a 3.5-year dataset, we classified convective clouds into three types: shallow cumulus ( ShCu ), congestus ( Con ) or ( Deep ) clouds. The shallow-to-deep transition takes about three hours, starting with ShCu formation at 11:00 local time. The accumulated rainfall peak follows at about 16:00. Only weak downdrafts are present in the upper troposphere where previous studies indicate new particle formation (NPF) occurrence. Strong downdrafts are mostly limited to heights below 5 km. Con and Deep convective days have higher concentrations of ultrafine aerosol and lower concentrations of accumulation-mode particles compared to ShCu . Convective clouds also significantly modify gas mixing ratios. Deep convective clouds are associated with high near-surface O 3 , consistent with downward transport from the midtroposphere. Our results showcase the added detail achieved by integrating data from the ATTO towers and ATTO-Campina sites. Together, these sites support better understanding of interconnected gas-aerosol-cloud-precipitation processes in the Amazon and their evolution under climate change.

Although fungal spores are of major importance to the Amazon Forest and exhibit a direct and indirect role in local and global climate regulation, there is still much unknown about their diversity, ecology and underpinning factors. Here, we show the impacts of two distinct forest types (“Campinarana” and “Terra Firme”) on the diversity and concentrations of airborne fungal spores collected within the Amazon Tall Tower Observatory (ATTO) site during the 2024 wet season. A total of 3,176,880 spores were counted and 20 morphotype-species were identified. Whilst “Terra Firme” exhibited a 34 This graphical abstract depicts the comparative dynamics of airborne fungal spores across two distinct types of forest with the Amazon biome, “Terra Firme” and “Campinarana”. The central panel shows the area of interest, with the relative position between the two sampling areas, and the sampling strategy. The left panel represents the “Campinarana” ecosystem, noted for its sandy, oligotrophic soils, lower canopy height, and sparser vegetation structure. In contrast, the right panel depicts the “Terra Firme” forest, characterised by a dense canopy, clay-dominated soil, higher plant biomass, and significant litter productivity. Over both landscapes, schematic spore icons indicate the presence of primary biological aerosol particles (PBAPs) that play crucial roles in cloud formation, ice nucleation, and overall climate regulation, with the major contribution of Cladosporium sp. (ca. 40

The Amazon region is an excellent laboratory for analysis of natural aerosols in the wet season; however, in the dry season the biomass burning emissions highly influence it, which considerably alters the physical and chemical properties of the atmosphere. We analyzed long-term time series (2000–2017) of optical and radiative properties of aerosols during dry and rain seasons from nine AERONET stations located in the Amazon Basin (Western Brazil). Aerosols have been classified into two groups: organic carbon (OC) and elemental carbon (EC), which allowed quantification of their effects on the radiative forcing for these sites. It was possible to conclude that the optical depth values of aerosols (AOD 500 nm), which remained in a downward inclination beginning in 2010 to 2012, returned to rise since 2013. The analysis showed that the fraction of biogenic particles varied from approximately 38

Numerous studies globally have centered on atmospheric air pollution due to its profound health and climate effects. NASA’s AERONET (National Aeronautics and Space Administration - AErosol RObotic NETwork) network has been one of the world’s leading tools for accessing the physical properties of atmospheric aerosols from various sources, mainly anthropogenic ones. This study proposes a new approach to evaluate the Aerosol Optical Depth (AOD) and precipitable water vapor (PWV) seasonality and the influence of short-term perturbations, such as the presence of local and regional aerosol sources or meteorological events, based on the temporal autocorrelation function (ACF). We introduce the adimensional seasonal assessment autocorrelation function, Δ _ACF,k , as a parameter to quantify the influence of the short-term perturbation, and we use its average, ⟨Δ _ACF,k⟩ , as a proxy for seasonality loss. The smaller ⟨Δ _ACF,k⟩ , the lower the influence of high-frequency perturbations on seasonality. Nine AERONET network sites in South America with different environmental characteristics were evaluated. The selected sites were São Paulo, Rio Branco, Manaus, ATTO (Amazon Tall Tower Observatory), Alta Floresta, Ji-Paraná, Cuiabá, Arica, and La Paz. The results showed that sites with less local anthropogenic aerosol sources acting as short-term perturbations had pronounced AOD seasonality and a linear relationship between the ACF functions of AOD, PWV, and the simulated direct solar radiation. As local anthropogenic sources become more prominent, the AOD ACF is attenuated and has less amplitude in seasonal oscillations. In addition, the relationship between AOD and PWV ACF becomes more attenuated. Buenos Aires has shown to be the most affected site, with ⟨Δ _ACF,AOD⟩ of 0.47, followed by São Paulo and La Paz. The areas in the Amazonian deforestation arc had relatively close average Δ _ACF,AOD , with Alta Floresta representing the most influenced by short-term perturbations. Central Amazonian sites had the lowest Δ _ACF,AOD averages, of about 0.25, which means that constant local anthropogenic sources do not dominate the AOD seasonality and that the wet deposition still plays an essential role in regulating the aerosol sources in the atmosphere. In contrast, the behavior of ⟨Δ _ACF,PWV⟩ in the Amazon region varies mainly due to meteorological influences, with the highest values observed in the central region, likely related to the high amount of water vapor in the atmosphere, and more pronounced seasonality near deforestation arcs and major cities. The proposed method eliminates the need for a reference site when comparing seasonalities of different time series, enabling valid comparisons across different areas without a comparative reference point. The method can be further applied to other atmospheric time series, including greenhouse gases.

This study investigates the rain-initiated mixing and variability in the mixing ratio of selected trace gases in the atmosphere over the central Amazon rain forest. It builds on comprehensive data from the Amazon Tall Tower Observatory (ATTO), spanning from 2013 to 2020 and comprising the greenhouse gases (GHGs) carbon dioxide (CO2) and methane (CH4); the reactive trace gases carbon monoxide (CO), ozone (O3), nitric oxide (NO), and nitrogen dioxide (NO2); and selected volatile organic compounds (VOCs). Based on more than 1000 analyzed rainfall events, the study resolves the trace gas mixing ratio patterns before, during, and after the rain events, along with vertical mixing ratio gradients across the forest canopy. The assessment of the rainfall events was conducted independently for daytime and nighttime periods, which allows us to elucidate the influence of solar radiation. The mixing ratios of CO2, CO, and CH4 clearly declined during rainfall, which can be attributed to the downdraft-related entrainment of pristine air from higher altitudes into the boundary layer, a reduction of the photosynthetic activity under increased cloud cover, and changes in the surface fluxes. Notably, CO showed a faster reduction than CO2, and the vertical gradient of CO2 and CO is steeper than for CH4. Conversely, the O3 mixing ratio increased across all measurement heights in the course of the rain-related downdrafts. Following the O3 enhancement by up to a factor of 2, NO, NO2, and isoprene mixing ratios decreased. The temporal and vertical variability of the trace gases is intricately linked to the diverse sink and source processes, surface fluxes, and free-troposphere transport. Within the canopy, several interactions unfold among soil, atmosphere, and plants, shaping the overall dynamics. Also, the mixing ratio of biogenic VOCs (BVOCs) clearly varied with rainfall, driven by factors such as light, temperature, physical transport, and soil processes. Our results disentangle the patterns in the trace gas mixing ratio in the course of sudden and vigorous atmospheric mixing during rainfall events. By selectively uncovering processes that are not clearly detectable under undisturbed conditions, our results contribute to a better understanding of the trace gas life cycle and its interplay with meteorology, cloud dynamics, and rainfall in the Amazon.

The wet-season atmosphere in the central Amazon resembles natural conditions with minimal anthropogenic influence, making it one of the rare preindustrial-like continental areas worldwide. Previous long-term studies have analyzed the properties and sources of the natural Amazonian background aerosol. However, the vertical profile of the planetary boundary layer (PBL) has not been assessed systematically. Since 2017, such a profile assessment has been possible with the 325 m high tower at the Amazon Tall Tower Observatory (ATTO), located in a largely untouched primary forest in the central Amazon. This study investigates the variability of submicrometer aerosol concentration, size distribution, and optical properties at 60 and 325 m in the Amazonian PBL. The results show significant differences in aerosol volumes and scattering coefficients in the vertical gradient. The aerosol population was well-mixed throughout the boundary layer during the daytime but became separated upon stratification during the nighttime. We also found a significant difference in the spectral dependence of the scattering coefficients between the two heights. The analysis of downdrafts and the related rainfall revealed changes in the aerosol populations before and after rain events, with absorption and scattering coefficients decreasing as optically active particles are removed by wet deposition. The recovery of absorption and scattering coefficients is faster at 325 m than at 60 m. Convective events were concomitant with rapid increases in the concentrations of sub-50 nm particles, which were likely associated with downdrafts. We found that the aerosol population near the canopy had a significantly higher mass scattering efficiency than at 325 m. There was also a clear spectral dependence, with values for λ=450, 525, and 635 nm of 7.74±0.12, 5.49±0.11, and 4.15±0.11 m2 g−1, respectively, at 60 m, while at 325 m the values were 5.26±0.06, 3.76±0.05, and 2.46±0.04 m2 g−1, respectively. The equivalent aerosol refractive index results, which were obtained for the first time for the wet season in the central Amazon, show slightly higher scattering (real) components at 60 m compared to 325 m of 1.33 and 1.27, respectively. In contrast, the refractive index's absorptive (imaginary) component was identical for both heights, at 0.006. This study shows that the aerosol physical properties at 60 and 325 m are different, likely due to aging processes, and strongly depend on the photochemistry, PBL dynamics, and aerosol sources. These findings provide valuable insights into the impact of aerosols on climate and radiative balance and can be used to improve the representation of aerosols in global climate models.