Bookmarks to get started in data journalism in Brazil
In practice, that method starts in a very concrete place, the browser...
Bookmarks to get started in data journalism in Brazil

The main difficulty in data journalism today is no longer technological. It is structural, and this completely changes the way the work is done. There is more data available than ever before, but that does not mean it is usable, reliable, or even understandable.
In many cases, datasets are incomplete, poorly documented, or simply outdated, while at the same time newsrooms operate with less time and fewer people to carry out in-depth investigations.
In this context, the difference is not in access to tools, but in the ability to define a problem clearly and maintain a consistent methodology throughout the entire reporting process.
In practice, this method begins in a very concrete place: the web browser.
I usually think of my browser as a toolbox, organized through folders and saved links that allow me to quickly navigate between platforms, datasets, and references. While this happens, the reasoning process unfolds in parallel; hypotheses are tested, discarded, and reformulated until a narrative begins to emerge.
Data journalism, especially when investigative, is not linear. It is exploratory work, based on trial, error, and eventually some form of clarity. The audience only sees the final result, never the attempts behind it.
Before collecting any information, it is necessary to understand what already exists. This initial stage is often neglected, but it is where the quality of the investigation begins to be defined. Working with structured datasets allows the identification of patterns, gaps, and possible directions, especially in environmental, climate, and territorial reports.
Datasets for understanding territory
MapBiomas Water https://bacias.mapbiomas.org/ Monitors water dynamics and environmental changes over time.
Hidroweb https://www.snirh.gov.br/hidroweb/mapa Official hydrological data with historical series.
CEMADEN https://mapainterativo.cemaden.gov.br/# Real-time monitoring of extreme events and risks.
Climate TRACE https://climatetrace.org/ Global emissions data with spatial granularity.
Carbon monitoring https://carbonmonitor.org/ Near real-time emissions tracking.
Fiocruz Climate and Health Observatory https://climaesaude.icict.fiocruz.br/pagina/mapas-graficos-e-tabelas-0 Connects environmental and public health data.
Salvador Data Portal https://dados.salvador.ba.gov.br/ Example of local-level data with higher resolution.
Amazon Mining Watch https://amazonminingwatch.org/pt Monitors mining activity and territorial pressure in the Amazon.
When data is not readily available, it needs to be built. This is where the work shifts from searching to extraction, and the process becomes less about access and more about decisions. What to collect, how to structure it, and what to exclude are methodological choices that shape the result.
Tools for data collection and structuring
Colab https://colab.research.google.com/ Automates coding the extraction and collection of structured data.
Google Pinpoint https://journaliststudio.google.com/pinpoint/ Helps organize and analyze large volumes of documents.
Overpass Turbo https://overpass-turbo.eu/ Extracts geospatial data directly from OpenStreetMap.
Global Fishing Watch https://globalfishingwatch.org/ Monitors fishing activity through geospatial data.
Panjiva https://panjiva.com/ Global trade data useful for supply chain investigations.
Sociedad.info https://sociedad.info/brazil# Data on public procurement and contracts.
S2ID https://s2id.mi.gov.br/ Official disaster notification system in Brazil.
Analysis is not about identifying patterns, but about questioning them. Many platforms already offer analytical layers that can speed up the process, but they also require careful interpretation. Visualization may suggest meanings, but it does not confirm them.
Platforms for analysis and pattern detection
DataClima Panel 1.0 by AmbGeo https://dataclima.onrender.com/ Integrated visualization of hydro-climatic data for each city in Brazil.
Raven (GeoSpy) https://geospy.ai/ Cross-referenced geolocation, identifies the approximate geographic location of a photograph in a real landscape with high accuracy.
Darklight Viewer https://gee-night-lights-474517.projects.earthengine.app/view/darklight-viewer Uses nighttime light data as an indicator of human activity.
Ocean Data Platform https://www.oceandatajo.com/ Marine datasets and visual analysis tools.
Despite the growth of automation, mapping remains fundamental, not only as visualization, but as a way of structuring thought. The map is not a final result. It is part of the reasoning process.
Mapping and visualization tools
Google My Maps https://www.google.com/maps/d/ Fast and accessible mapping for early-stage work.
uMap http://umap.openstreetmap.fr/pt/ Open-source alternative based on OpenStreetMap.
StoryMapJS https://storymap.knightlab.com Builds narrative sequences through geography.
Flourish https://flourish.studio/product/data-storytelling/ Builds interactive data visualizations, dynamic graphic narratives, and creative templates for visualizing maps.
MapLibre https://maplibre.org/ Library for customized interactive maps.
MapTiler https://www.maptiler.com/ Infrastructure for tiles and geospatial rendering.
MyGeodata Cloud https://mygeodata.cloud/ Data conversion and processing.
TerraDraw https://terradraw.water-gis.com/ Vector editing directly in the browser.
When journalistic coverage involves territory, it often involves conflict. At this point, data ceases to be merely descriptive and becomes a way to understand disputes, power, and competing interests.
Platforms for analysis of conflicts and territories
JusMapp https://www.jusmapp.com/ Legal data and georeferenced court cases in Brazil.
MPF Traditional Territories Platform https://territoriostradicionais.mpf.mp.br/ Maps claims of Indigenous and traditional lands in Brazil.
Carbon Market Radiography https://infoamazonia.org/project/radiografia-do-mercado-de-carbono/ Organizes data on buyers and sellers of carbon credits in Brazil, Peru, and Colombia.
Pulitzer Center map of 20 years of reporting https://pulitzercenter.org/celebrating-20-years-pulitzer-center Interactive map gathering all the stories supported worldwide.
Some tools go beyond describing what is happening and begin to explore what could happen. This introduces a different type of narrative, less focused on documentation and more on explanation.
Modeling and simulation
VisQuill https://visquill.com/gallery/rainfall Experimental visual simulations of environmental phenomena.
River Runner Global https://river-runner-global.samlearner.com/ Geospatial simulation of the trajectory of a drop of water released anywhere in the virtual globe.
Geospatial intelligence has expanded operational capacity, especially in organizing and processing data, but it does not solve the central problem, which is interpretation. There is a real risk of automating errors if the process is not carefully controlled.
Artificial Intelligence applied to data and geoinformation
AINO https://www.aino.world/ Supports geospatial analysis and workflows.
Google Earth AI https://ai.google/earth-ai/ Experimental applications for geographic data.
PamPam https://www.pampam.city/ Creation of interactive maps with AI assistance.
Groundsource https://research.google/blog/introducing-groundsource-turning-news-reports-into-data-with-gemini/ Model for flood prediction worldwide based on thousands of news articles using Gemini.
Finally, there are methodological references that help structure investigative thinking, especially when working with open-source intelligence and digital verification.
Toolkits and references
Bellingcat online investigation toolkit https://bellingcat.gitbook.io/toolkit A consolidated guide for open-source investigations.
GIJN Resource Center https://gijn.org/resource A broad collection of guides, tools, and methodologies.
In the end, the process remains relatively stable. It begins with defining a problem, followed by the formulation of hypotheses, data collection and structuring, pattern analysis, validation of results, and finally, the construction of a narrative. Tools operate within this flow, but they do not define it. What defines the work is the ability to connect these stages with rigor, clarity, and a consistent commitment to the public interest.
메타데이터
- post_id
- 39bb1fb037ab
- slug
- bookmarks-to-get-started-in-data-journalism-in-brazil-39bb1fb037ab
- url
- https://medium.com/@CaldasHeleno/bookmarks-to-get-started-in-data-journalism-in-brazil-39bb1fb037ab
- canonical_url
- https://medium.com/@CaldasHeleno/bookmarks-to-get-started-in-data-journalism-in-brazil-39bb1fb037ab
- author_url
- https://medium.com/@CaldasHeleno
- status
- ok
- fetched_at
- 2026-06-29 01:02:39