The NEXUS area covers approximately 30% of the Brazilian territory. In order to assist preservation and sustainable development policies in that region, this study proposes to replicate the work done by Yeh et al in Africa , in which a convolutional neural network estimates indicators through satellite images, each covering a region of approximately 45 km². This work compares the size and distribution of Brazil’s census tracts with those in Africa to define if the scale of images can be maintained and to define the clusters that will be used. To avoid biasing the model, special care must be taken in selecting clusters, such as keeping a balance between urban and rural sectors and, most importantly, making sure that there is little to no overlap of clusters. To do so, two approaches were proposed. The first one samples tracts in each municipality as centroids for clusters, the second merges neighboring urban tracts into a single group and fits clusters to these groups.

Amazon rainforest has been subject to intensive deforestation in the last decades, for example, illegal logging and creating pasture areas. A characteristic pattern of deforestation seen from space is the “fishbone” shape, which usually appears near roads, rivers and its tributaries. Indeed, others, more subtle, still need to be identified. These fishbone images are spatiotemporal patterns that need to be more explored with feature extraction methods. In computer vision, morphological features such as flatness, compactness, circularity, perimeter, area, and centroid are well-known to characterize the appearance of an object. In this work, we aim to characterize the shapes of deforestation in its early stages and its evolution in time, particularly in the Amazon rainforest. Thus, we propose to analyze satellite images of these regions to crop and segment by using shape features.

Research data are a vital component of the scientific record. Discovering and assessing data for possible reuse in future research is challenging. The Belmont Forum has recently awarded funds to three international teams as part of a four-year Collaborative Research Action (CRA) on Science-driven e-Infrastructure Innovation (SEI) for the Enhancement of Transnational, Interdisciplinary and Transdisciplinary Data Use to improve data management practices that will increase data reuse. One of these awardees, PARSEC, comprises two interwoven strands, one focused on improving data practices for reuse and credit, and one for synthesis science. The data specialists work alongside synthesis science researchers as they determine the influence of natural protected areas on socioeconomic outcomes for local communities. They collaborate with the researchers to better understand their motivations and work practices, and to aid them in the data-related steps that need to be taken during the research lifecycle. This will ensure their data and code are FAIR-compliant and thus enhance the likelihood of their data being reused and their analyses reproducible. The PARSEC team is working with Research Data Alliance (RDA), Earth Science Information Partners (ESIP), DataCite and ORCID to build awareness of the elements required for data creators to receive credit and automated attribution for their data contributions, and the tools that will make it easier to observe usage. Credit for data is an important incentive for researchers to make their data reusable. When data are FAIR and cited, their related publications have higher visibility. We shall discuss various ways in which we are working across the science-data interface in our multi-country and multi-disciplinary working environment to improve data (and code) reuse through better management and crediting. Make your Data FAIR, Cite your Data, Get Credit, Increase Reuse and reap the rewards!

Species Distribution Modelling (SDM) is widely used by ecologists to monitor biodiversity and manage wildlife. In the last decades, Artificial Intelligence (AI) and Machine Learning (ML) techniques became popular and were successfully applied for different tasks, including SDM. The objective of this article was to evaluate Machine Learning models for Species Distribution Modeling in the Amazon Basin region near Manaus (AM), based on meteorological and aerosol data collected by the GoAmazon 2014/15 project. The techniques were evaluated regarding their accuracy and the Decision Tree Classifier and the Maximum Entropy Model obtained good predictive performances.

With the mass adoption of data analysis in several scientific fields such as climatology, medicine, astronomy and astrophysics, the availability of an appropriate analytics infrastructure has become a necessity increasingly recognized by the scientific community. However, appropriate tools and applications are required to process the large volume of data collected and generated by researchers. One of the biggest challenges lies in the fact that these tools need to be gathered to be applied in specific domains. The area of bioclimatic data is a scientific field that still has much to improve in this matter. It is a field of study that lacks great efforts in the direction to provide methodologies and tools to facilitate the understanding of the complex phenomena involved in the influence that environmental variables have on biodiversity on the planet. Thus, the purpose of this work is to propose a big data analytics architecture that presents an ecosystem that systematizes and facilitates the task of the scientists to deal with the complexity in the bioclimatic data analysis, providing tools for storage, management, analysis using machine learning algorithms and data mining, and visualization tools. The methodological approach of this work was to make a thorough bibliographical study to verify the most used tools and the suitability of each one to the purpose of the work. In addition, the literature provided indications of software ecosystem implementations methodologies that served as a guide in the architecture design. Within the architecture, we attempted to gather a set of bioclimatic data based on a subset of data obtained from the Atmospheric Radiation Measurement (ARM) data repository for climatic data, and the Brazilian Biodiversity Portal for biodiversity data. As a result, we were able to gather a series of tools to access data such as Cassandra, distribution of processing such as Spark, programming interface represented by Jupyter Notebook, system modules for data format conversion, machine learning algorithms libraries and software for data visualization. This research discuss the importance of a domain purpose design of a data analysis architecture for bioclimatic data. We concluded that this type of ecosystem is imperative to facilitate the research process and increase the quality of the results.

This paper discusses a project aimed at detecting whether protected areas (PAs) influence the socioeconomic well-being of adjacent communities. The Belmont Forum funded PARSEC project is using satellite images and deep learning algorithms to predict socioeconomic conditions. In this paper we show our ongoing work for the selection of PAs, development of methodology using deep learning to detect socioeconomic indicators from remote sources, facilitation in data management and the approach used to handle a complex inter-disciplinary and trans-national team. We note the challenges in selecting case studies with examples from Australia, Brazil, Japan and the USA, and meshing remote sensed data with census data. We discuss the advantages of good data management for the individual and for the project and some simple steps to make this easy.

The Brazilian National Institute for Space Research (INPE) produces research that helps to understand climate and weather dynamics in Brazil and in the world, with significant impacts on national public and private strategic planning. Among the essential information for the studies are the rainfall data. In this context, ensuring the quality of this data has a direct impact on the reliability of the forecasts and analysis generated from them. Thus, this study, which is a partnership between INPE, the Laboratory of Atmospheric Physics and Polytechnic School of USP and the ARM-DoE (Atmospheric Radiation Measurement Climate Research Facility), aimed to establish computational tools that could deal with the quality of data from rain in accordance with the main international directives. Thus, it was proposed for this study the development of a specific toolkit for data from the Micro Rain Radar (MRR), disdrometers PARSIVEL2 and RD80, and rain gauge that would help researchers from INPE, USP and partners to: standardize the preparation of raw data for internationally accepted formats; processing figures to support quick analyses; analyze and process data quality and, finally, record metadata and quality analysis for publication in international data repositories.

Key measures of socioeconomic indicators are essential for making informed policy decisions, but due to the high costs and operational difficulties of traditional data collection efforts, obtaining reliable socioeconomic data remains a challenge, particularly in developing countries. This work presents a deep learning methodology to estimate socioeconomic indicators using satellite imagery. The neural network model developed was trained at the Brazilian region of Vale do Ribeira with the goal of analyzing the socioeconomic indicator of income. The preliminary results showed that models using nightlight (NL) or multispectral daytime (MS) imagery performed better than models trained only on RGB bands and that models trained exclusively on NL or MS imagery performed similar to one another and nearly as well as the combined model MS+NL. Finally, the model yielded a low performance (R2 = 0.289), but it is still promising once the dataset employed was considerably smaller than the one used in the original study that attempted to replicate.

The renewable energies have confirmed their potential to reshape the energy matrix of several countries, in the last two decades, reinforcing the set of actions to better protect the environment. For instance, the solar and wind sources of energy production can dim the production and release of pollutants associated to electricity production. Unfortunately, the atmospheric temperature rise has been linked to the increase of the gas emissions, such as Carbon Dioxide (CO2) and Methane (CH4), from manifold sources, such as the thermal power stations for electricity generation. The Amazon basin has caught special importance for Brazil and abroad, and more recently, the use of satellites, fixed instrumented stations and airborne surveys has provided data focused on studies of the environment impact. This work deals with the assessment of the solar irradiation on the Manaus city, the largest city in the Western Amazon region, using Data Science tools, in a way to help the evaluation of the renewable energy in that region.

The Amazon rainforest has a great influence on the global energy balance and carbon fluxes, responsible for the net removal of approximately 4 million tons of carbon per year, via photosynthetic activity. Climate change and deforestation have impacts on the carbon budget in Amazonia, transforming CO2 sink areas into sources. Given the complexity of the factors that govern the carbon exchange in the Amazon and its influence on biological processes, the use of Data science strategies can promote a better understanding about the main environmental factors for different scenarios, and also, assist in public policies to mitigate the global warming effects. This study aims to identify the environmental factors that determine the temporal variability of carbon exchanges between the biosphere and the atmosphere in the Tapajós National Forest, in the Amazon, applying Data Science strategies in an integrated set of environmental data from energy and carbon fluxes and remote sensing data. The specific objective is to assess the influence of a selected set of environmental variables on the variability of carbon exchanges, with the use of an artificial neural networks classification model to identify the variables with great impact on source, sink and neutrality scenarios in Tapajós National Forest. Data Science strategies were applied to an integrated dataset of ground-based carbon flux measurements and remote sensing data, considering the period between 2002 and 2006. An artificial neural network (ANN) classification model was developed to identify the environmental variables with great impact on carbon source, sink and neutrality conditions. The average global score of ANN model was 65%. It was possible to identify the predictor variables with greatest impact to the carbon sink condition: radiation at the top of the atmosphere, sensible and latent energy fluxes and leaf area index. Thus, the ANN model with an ensemble of Data Science strategies can improve a better understanding of variability CO2 fluxes and be a powerful tool to promote new knowledge.