Publicações

Seasonally Dry Tropical Forests (SDTFs) may act as considerable carbon sinks, regulating the atmospheric and terrestrial carbon storage and fluxes with implications for local, regional and global climates. In the Caatinga, an endemic Brazilian SDTF, the research on the magnitude of the CO2 sink is still incipient. To address this gap, we conducted a comprehensive analysis of observed CO2 fluxes using the Eddy Covariance technique across the Caatinga, quantifying and assessing the seasonal and interannual variations in CO2 exchange under contrasting soil and climatic conditions. In our study we estimated whether the Caatinga functioned as a net source or sink of carbon at five sites during years with varying rates of rainfall. Results showed that the dynamics of CO2 flux components varied based on the spatio-temporal distribution and magnitude of rainfall and the corresponding variations in vegetation cover. Average annual accumulated Gross Primary Productivity ranged from 1167 g C m−2 in the Crystalline area to 2018 g C m−2 in the Agreste ecotone. Average annual Net Ecosystem Exchange was -775 g C m−2 (-7.7 t C ha−1). The Caatinga exhibited higher carbon use efficiency (CUE) compared to other dry forests in arid and semi-arid regions worldwide and to South American ecosystems, including the Amazon, as documented by FLUXNET2015 eddy covariance datasets. CUE values ranged from 0.31 in the Crystalline area to 0.58 in the Agreste ecotone. These findings provide robust, measurement-based evidence that the Caatinga is a highly efficient carbon sink, substantially contributing to atmospheric CO2 absorption and mitigating the growth rate of atmospheric CO2 concentration.

On May 2, 2024, Porto Alegre in Rio Grande do Sul, Brazil, experienced a historic flood surpassing the catastrophic 1941 event. The flood’s peak water level of 535 cm was 59 cm higher than in 1941, driven by extreme precipitation anomalies and land use changes in the Regional Metropolitan Area of Porto Alegre, with rainfall accumulations up to 500 mm in May 2024. Severe urban flooding impacted riverbanks and low-lying areas, causing significant property damage, displacement, and disruption of essential services. Agricultural lands and grasslands were extensively affected. Detailed maps from May 3 to May 8, 2024, showed floodwaters rising from 5.5 m to 8 m. Validation sample points ensured data accuracy, and flood model validation metrics (NSE = 0.63, R2 = 0.71, RMSE = 0.51 m) demonstrated the model’s predictive accuracy. Water level fluctuations at Cais Guaíba station revealed rapid rise and recession periods typical of major floods. This study highlights the urgent need for effective flood mitigation strategies, informed by accurate modeling and risk assessment, to protect vulnerable communities and infrastructure. Enhancing resilience through improved urban planning, infrastructure investment, and environmental conservation is crucial to safeguarding Porto Alegre and the broader RMPOA region from future extreme weather events. These findings offer valuable insights for other flood-prone tropical and developing regions, where the combined effects of urban expansion and intensifying climate extremes similarly increase flood risk.

Northeast Brazil (NEB) is one of the driest regions in the country, with some areas, such as the sertão, receiving less than 400 mm/year. Despite this aridity, NEB is influenced by various synoptic-scale systems that contribute to intense rainfall events throughout the year. This study aimed to identify and analyze the synoptic-scale atmospheric patterns associated with extreme precipitation events in NEB. There were selected four events that occurred in the region over the past eight years and significantly impacted metropolitan areas. Two cases occurred in eastern, one in southern, and one in northern sector of the study region. The results revealed that these extreme precipitation events were associated with different synoptic-scale systems. The easterly wave disturbance (EWD), upper tropospheric cyclonic vortex (UTCV), South Atlantic convergence zone (SACZ) and intertropical convergence zone (ITCZ). EWD and UTCV were primarily responsible for the two eastern cases. The SACZ influenced the case that occurred in the southern, whereas the Atlantic ITCZ played a role in in northern case. Our findings indicate that these events are driven primarily by intense upper-level divergence and substantial moisture transport from the Atlantic Ocean to the continent, leading to significant moisture convergence. This study provides crucial insights into the factors driving extreme precipitation events in NEB, which can aid in validating numerical weather prediction models.

 Downward longwave radiation is a crucial component of the surface radiation balance and is essential for a better understanding of the greenhouse effect. It is strongly influenced by atmospheric temperature, water vapor content, and cloudiness. This study proposes a new model to estimate downward longwave radiation under cloudy sky conditions, suitable for the application of remote sensing data. The Santa Barbara DISORT Atmospheric Radiative Transfer code was employed to develop parameterizations for cloud and atmospheric processes. The simulations indicated that the emissivity of a stratiform cloud when the cloud is thicker than 300 m (ε = 0.985) approaches that of a blackbody. It is shown that atmospheric emissivity and transmittance can be parameterized as functions of precipitable water content below cloud base. The model was validated using observational data from six different sites in Brazil. The parameterizations of primary model variables (atmospheric layer emissivity, atmospheric transmittance, and precipitable water below the cloud base) generally showed good agreement with radiosonde data. The parameterization of precipitable water in the sub-cloud atmosphere yielded good results, with a maximum RMSE of 0.44 g cm−2. The developed model presents an average overestimation of only 14 W m−2, corresponding to a percentage error of 3.5 %, and an overall average RMSE of 18.3 W m−2 (4.6 %). It shows satisfactory performance and is suitable for application in tropical and subtropical regions, as the underlying physics of this method is consistent worldwide.

About 40% of the Brazilian Amazon has been deforested or suffered changes in forest structure through degradation (selective logging, fires, and fragmentation). The impact of forest degradation on the forests sensitivity to climate extremes has not been fully explored because of a lack of data and the complex interplay of forest structure and climate drivers. Here, we combined forest structure data from 545 airborne lidar transects (375 ha each) across the Brazilian Amazon with the Ecosystem Demography Model (ED2). We explore the forests functional response to near-present (1981-2019) climate extremes under observed forest structure from lidar (Control) and two forest structure change scenarios: (1) forest recovery by excluding all future deforestation and degradation (Recovery) and (2) expansion of selective logging and deforestation (Degradation). Using the Control simulation, we found a close and positive association between local forest aboveground biomass and the predicted gross primary productivity (GPP) and evapotranspiration (ET). Moreover, both GPP and ET respond negatively to extremes in vapor pressure deficit and downwelling shortwave irradiance in degraded forests in Eastern and Southern Amazon, indicating high sensitivity to droughts. Locally high-biomass forest patches showed little or no negative response of GPP and ET to extreme drought conditions whereas low-biomass forest patches in the same locationstypically degraded forest canopiesresponded negatively to higher moisture stress. The results from the Recovery scenario showed similar results to simulations with observed structure; however, under the Degradation scenario, low-biomass forest patches became more abundant, resulting in more regions where GPP and ET are negatively impacted by hot drought conditions according to the ED2 model. Our results suggest that local forest structure is a critical determinant of an ecosystems response to climate variability, and that the loss of canopy trees in the Amazon through forest degradation could increase and expand forest vulnerability to droughts.

Climate change threatens global coffee yield stability, especially in Brazil, the largest Arabica exporter. Yield modeling is key to climate-resilient strategies and effective planning. To this end, the AquaCrop model was parameterized, calibrated, and validated for Arabica coffee in Brazil and subsequently applied to project yield under future climate scenarios. These processes were carried out across 58 municipalities in Brazil’s two largest coffee-growing regions, covering the growing years from 2014 to 2019. AquaCrop accurately simulated Arabica coffee yield during calibration and validation processes, with root mean square error values of ∼0.15 t ha−1, mean bias error of ∼0.007 t ha−1, and d-index of ∼0.76. To project yield, AquaCrop used as input daily near-surface air temperature (minimum and maximum), precipitation, and reference evapotranspiration data from 9 General Circulation Models (GCMs) from the NASA Earth Exchange Global Daily Downscaled Projections (NEX-GDDP), which are derived from the Coupled Model Intercomparison Project Phase 6 (CMIP6), under three Shared Socioeconomic Pathways scenarios (SSP2-4.5, SSP3-7.0 and SSP5-8.5). Under future climate conditions, reductions in the coffee growing cycle duration and heterogeneous yield changes are projected, with increases (between 0.25 and 0.55 t ha−1) along a belt extending from southeastern São Paulo to southern Minas Gerais, and decreases (between −0.05 and −0.15 t ha−1) in northeastern Minas Gerais. Even though yield increases are projected for much of the study region, adopting effective adaptive measures will be essential to address climate change threats.

Correction to: Naturehttps://doi.org/10.1038/s41586-024-08192-4 Published online 4 December 2024 In the version of this article initially published, there was a typographical error in the units shown on the Fig. 3b y axis, where C* (μg m3) originally read C* (μg cm3). In the article, the original text Saturation vapour pressure C* has been amended to read Saturation concentration C* in the Fig. 3b caption, in the last paragraph of the Formation of organonitrates section, and in four places in the now edited Saturation concentration Methods subsection. This correction does not influence our results and conclusions. The figure and text have been amended in the HTML and PDF versions of the article. We thank Dr. John Ness for alerting us to the need for this clarification.

The growing global emphasis on environmental, social and governance (ESG) principles has driven higher education institutions to adopt more sustainable and socially inclusive practices. However, a lack of metrics and regular reporting often limits the ability to assess the true impact of these initiatives. This study critically examined ESG practices at a public urban university in São Paulo, focusing on the Department of Agricultural Sciences (DAS). Using a case study approach, the research employed a systematic literature review, document analysis and on-site observations to assess how ESG principles are implemented. Correspondence Factor Analysis (CFA) and Discursive Textual Analysis (DTA) were used to identify patterns and insights. The findings highlight the University’s commitment to sustainability and social inclusion, particularly through initiatives such as the Accessibility Assurance Plan (PGA) and environmental conservation efforts. However, the study also identified the need for structured evaluation metrics and more transparent reporting to better monitor the impact of these initiatives. The research concludes by suggesting that future studies should explore how ESG practices influence students’ perceptions of educational quality and examine the financial costs and benefits of these initiatives to guide more strategic implementation in higher education institutions.

O governo brasileiro possui três sistemas oficiais de monitoramento por satélite da supressão da vegetação nativa dos biomas brasileiros, concebidos e executados pelo Instituto Nacional de Pesquisas Espaciais (INPE), por meio do Programa de Monitoramento dos Biomas Brasileiros (BiomasBR): 1) o Prodes Monitoramento Anual da Supressão de Vegetação Nativa; 2) o Deter – Monitoramento Diário da Supressão e Degradação de Vegetação Nativa; e 3) o TerraClass – Monitoramento Sistêmico da Cobertura e Uso da Terra, este em colaboração com a Empresa Brasileira de Agropecuária (EMBRAPA). Neste artigo são apresentadas as principais características de cada um destes sistemas, os resultados mais recentes do monitoramento da vegetação nativa brasileira, as principais políticas públicas baseadas no BiomasBR e o papel do programa no cumprimento dos compromissos climáticos internacionais assumidos pelo país.

Non-technical summary Achieving sustainability on the ground poses a challenge in decoding globally defined goals, such as sustainable development goals, and aligning them with local perspectives and realities. This decoding necessitates the understanding of the multifaceted dimensions of the sustainability challenges in a given context, including their underlying causes. In case studies from Brazilian drylands, we illustrate how an enhanced multiscale participatory method, combined with systems thinking tools, can shed light on systemic structures that currently entrench unsustainable development trajectories. This method offers insights into co-designing potential pathways toward sustainable futures and unlocking transformative capacities of the local population.Technical summary Translating United Nations global sustainable development goals (SDGs) into actions that address local realities and aspirations is an urgent challenge. It requires new thinking and approaches that foster the discussion about the main challenges to implementing the SDGs at multiple levels. This paper presents a novel multiscale participatory approach that combines the popular Three Horizons diagram with the formalism of causal loop diagrams in systems thinking. We present results from six multi-stakeholder dialogues held across drylands in Brazil with a focus on desired futures aligned with SDGs. Focusing on identifying the root causes and systemic structures of unsustainability, participants identified lock-ins, leverage points, and interventions for how these could be changed. The core lock-ins are the discontinuity of public policies, and the historical land and power concentration reinforced by the current expansion of large-scale agricultural, mining, and energy projects. The proposed interventions are structural and – if implemented – would contribute to achieving SDGs in an integrated manner. The unique approach developed in this study can provide leverage as it bridges the inclusivity of participatory visioning with the change potential of systems thinking tools to tackle root causes and unleash societal transformations.Social media summary We are not achieving SDGs. Understanding root causes of unsustainability is critical to move toward sustainable and just futures.

The Mucor circinelloides complex, a group of dung-dwelling fungi, is essential for nutrient cycling and organic matter breakdown. Despite their ecological significance, variations in germination rates and stress tolerance among isolates are not well understood. This study examined four isolatesCicarelli, Doc Beaver, Papulsa, and Zainato assess their germination on various media, including potato dextrose agar (PDA), Czapek (CZA), Sabouraud dextrose agar (SDA), Emerson (EM), and minimal media (MM). The results showed notable differences in germination rates, with CZA and EM promoting the fastest growth. The study also analysed sporangiospore tolerance to UV-B radiation and high temperatures (47 °C), with findings highlighting the role of growth media in stress resistance. Sporangiospores from Cicarelli, Doc Beaver, and Papulsa grown on PDA exhibited greater UV-B tolerance than those on CZA. Conversely, sporangiospores produced on CZA were more heat tolerant than those on PDA. The 12-h survival curve for the Zaina isolate reinforced these findings, showing similar trends in UV-B resilience and heat tolerance. This research demonstrates how growth media influence the environmental stress responses in M. circinelloides, providing insights into the adaptive potential of these fungi.

The planetary boundary layer height (zi) is a key parameter in meteorology and climatology, influencing weather prediction, cloud formation, and the vertical transport of scalars and energy near Earth’s surface. This study compares multiple machine learning (ML) models that predict zi from surface measurements at two sites in Central Amazonia-the Amazon Tall Tower Observatory (ATTO) and the Manacapuru site of the GoAmazon experiment (T3). Models were trained on ceilometer data with radiosonde measurements used for validation. We evaluated model performance by withholding approximately 10% of the data (as complete months) for testing, comparing predictions against ERA-5 reanalysis data using RMSE, nRMSE, and R2 metrics. Our results show that gradient boosted ensemble models using all available features perform best. A modified recursive feature elimination algorithm identified minimal sets of 5-7 surface measurements sufficient for accurate zi prediction, demonstrating potential for wider spatial monitoring using cost-effective sensors. The study revealed previously unrecognized variables influential in determining zi, such as deep soil temperature measurements (40 cm), suggesting new avenues for investigating land-atmosphere interactions. This study demonstrates the applicability of ML models to model zi. Plain Language Summary The height of the planetary boundary layer (PBL) is important for understanding weather, cloud formation, and the movement of pollutants and energy near the Earth’s surface. This study compares different machine learning models to predict the height of the PBL using data collected from the surface, such as temperature, energy, and meteorological measurements. The performance of these models is compared with radiosonde measurements and existing weather prediction data from the ERA-5 data set at two locations in Central Amazon. This research highlights the potential of machine learning to improve our ability to model the PBL. height accurately. This study found that you can predict the height of the PBL accurately using only 5-7 ground-based measurements. This is important because it suggests we could monitor boundary layer height over wider areas using simple, low-cost sensors combined with short-term measurement campaigns to obtain measurements of the PBL height for model training. The research also discovered some unexpected factors that influence boundary layer height such as soil temperature measured deep underground (40 cm). This opens up new directions for study and helps us better understand what controls the height of the boundary layer.