The Congo Basin is home to the world’s second-largest tropical rainforest, spanning 2.5 million square kilometers over six countries. With over 1,000 threatened species, it is also the world’s last tropical carbon sink.
Environmental efforts receive up to €300 million per year, but biodiversity loss, deforestation, and carbon emissions continue unabated.
.png)
Project Canopy is a data-driven non-profit that works with conservation actors in the Congo Basin. We provide the analytics they need to make better decisions. How do we do it?
This allows environmental actors to save more forest cover, threatened species, and carbon stores per resource spent. In turn, funding will increase for this crucial and under-resourced ecosystem.
When you only have 30 seconds.
When you have a few minutes.
When you just need to know more.
Project Canopy is preparing several prototypes that will immediately bring value to our stakeholders.
How can local organizations use satellite imagery and AI to quickly find logging roads and monitor logging company activities?
What is the state of biodiversity in the Congo Basin rainforest? Do scientific assessments of species align with law and international regulation?
Who are the key organizations and decision-makers making conservation decisions? Who are the scientists and journalists creating & communicating essential knowledge?
Effective training data for machine learning means having clear views of the terrain, especially if you're counting on visible bands for your analysis. This can be challenging for rainforests, where cloud cover is a natural part of the ecosystem. So how do we get rid of these clouds?
View on Medium
For geospatial developers keen to use Google Earth Engine, it may not be immediately obvious how to quickly query and download Sentinel-2 satellite imagery – for example, there is no 'query' function in GEE. Here is a brief tutorial on a few methods that resolve both issues.
View on Medium
Imagery from the Sentinel constellation of satellites may be high-quality and free, but when you are running a machine learning pipeline over 2,500,000 square kilometers, you have to make a lot of decisions about how you're going to download, store and access all that data.
View on Medium
In a series of blog posts, Project Canopy data scientists Zhenya Warshavsky and David Nagy document their journey to developing a machine learning model to detect commercial logging roads, using satellite imagery of the Congo Basin. How hard could it be?
View on Medium
Jules Caron has fourteen years of campaigning, advocacy, communications, and journalism experience worldwide. He has traveled and lived in the Congo Basin since 2012, working as a campaigner for organizations such as WWF, Global Witness, and Oxfam, and collaborating with local civil-society organizations and large INGOs such as Greenpeace and Rainforest Foundation. Advocacy successes include sending elephant poachers to jail, preventing an expansion of DRC’s logging sector, and ensuring gender parity in CAR’s Truth and Reconciliation Commission. After saving the rainforest, he’s looking forward to settling in South Tyrol.
Misha Lepetic has amassed 25 years of experience and success in the information technology and services industry, including pharma (Pfizer), publishing (Hachette Book Group), non-profits (International Planned Parenthood Federation, Human Rights Watch, United Nations Democracy Fund) and educational technology (Nielsen). Skilled in IT strategy, management, and understanding organizations' technological needs, he earned a Master of Science in Technology Management from Columbia University in 2008. He lives in Brooklyn, New York. After saving the rainforest, he’s going back to DJing and studying Sanskrit.