TED Talks Daily - How to see (and stop) deforestation from space | Tasso Azevedo

Episode Date: November 17, 2025

Nearly 20 trees are cut down every second in the Amazon rainforest, as authorities struggle to monitor millions of acres and stop illegal clear-cutting. But land reformer Tasso Azevedo and his team at MapBiomas have changed the game, transforming satellite imagery into precise, real-time maps that make every clear-cut visible — and every actor accountable. Learn how they're helping slash deforestation in the Amazon, proving that transparency is a forest's strongest defense. (This ambitious idea is part of The Audacious Project, TED’s initiative to inspire and fund global change.) Hosted on Acast. See acast.com/privacy for more information.

You're listening to TED Talks Daily, where we bring you new ideas to spark your curiosity every day. I'm your host, Elise Hugh. I don't know about you, but I'll admit that I'd be nowhere without the GPS. I pull up on my phone every time I get in my car. Maps are invaluable for so many reasons, but did you know they're also helping in the fight against deforestation? In this talk, land reformer Tasso Acevedo reveals how the collaborative mapping initiative Map Biomas has stitched together 40 years of satellite images into near real-time courtroom-ready evidence, turning Brazil's once invisible deforestation into bright, undeniable pixels. It's led to real large-scale results in saving our planet's lungs. 20, 40, 60, 800.

In every second on the last 12 months, almost 20 trees were cut down in the Amazon. See, deforestation and degradation is a kind of a disaster for the only country in the world that is named after a tree. Brazil is the fifth largest greenhouse gas emitter in the planet, but different from the top four, the bulk of our emissions do not come from burning fossil fuels. It actually comes from the way we use the land. That's 75 percent of our emissions,

and the majority of that results from forests being cutting down. You see, tropical forests are like the lungs of the Earth. You heard something like that before, right? When we cut them down, they essentially exhalate the carbon that have been stored for over many decades. And they no longer exist to absorb the carbon and to maintain the evapot transpiration that keeps the cooling system of the planet.

In 2024, the countries of the largest areas of tropical forests, like Brazil, Peru, Congo, Indonesia, lost an area of forests the size of Rwanda. And this forest has been converted to pasture, croplands, urban areas, mining, and also suffering degradation from logging, increased droughts, and wildfires. So we need to fight this degradation.

And in order to do so, we have to understand what's happening on the ground. So digital mapping can be a key role on that. Because maps, maps have power. But only if you can understand really what's going on on the ground. So global satellite technology allow us to see the world in the palm of our hands. But these are just images.

In other to understand what's happening with the land use, you need to capture the context of what we are seeing. So we can understand the transformations across the time, and then we can act. Until very recently, a detailed map like that for an entire country like Brazil was almost impossible to make because it was too costly and too slow to produce. So back in 2015, we put together a group of experts in the remote sensing, computer science, and land use to reinvent the way we produce and publish maps.

We call it Mapyomas Network. And now we can produce 40 years of maps. in six months. We do it by creating a mosaic of all the existing satellite image that we have for every year since 1985. Then we apply machine learning algorithms that allow us to classify each pixel

and make a layer for every land-use class that we want to a map, like forest, mangrove, crop land, and water. Then we put all those layers together and produce a time series with one integrated map for each year. Now we have this kind of time machine where you can see every pixel

and see the history of those pixels of 30 by 30 meters that are 9.6 billion of them that makes up Brazil. So now we can see the history of everywhere in the country in any time on the last four decades. But how this could lead to change? Have you ever crossed the red light and received, like, a penalty by email showing the license plate of your car? Yeah, so something very similar can be done with deforestation using remote sensing. Every time there is a deforestation detected in Brazil, we use high-resolution satellite imagery

to validate precisely when and where the deforestation have occurred. Then we cross this information with the land cover maps, the registry of the land properties, the catalog of protected areas, and even the authorizations for clear-cut, so we can produce very detailed reports. In 2018, all the environmental agencies in Brazil together produced less than 1,000 reports in one year.

Last year, in Mapyomas, we produced 2,000 of those reports per week. Those reports have a level of granularity and precision that can literally be used on the court of law. So in fact, between 2019 and 2024, the actions from the environmental agencies in Brazil against illegal deforestation increase from 5 to 54% of all deforestation. It's also happening on finance.

So in the last two years, the major banks in Brazil, in Brazil denied $1.5 billion on finance for 30,000 farms that have deforestation detected by mapeumas. And though this money went to a more sustainable operations. So this, together with several other actions, allows Brazil to decrease the amount of deforestation the Amazon by 54% on the last two years. And this

which would save all of us of 500 million tons of CO2 emissions. So this goes much beyond deforestation. In fact, in the last year, over 600,000 users access our data and apply on hundreds of applications, like, for example, preventing tropical disease, regulating the use of water, assessing climate change impacts, designing soil conservation practices and protecting indigenous land rights.

You know, illegal gold mining is a huge problem in the Amazon. It destroys the land, it pollutes the water, and poison indigenous people. So back in 2003, the Brazilian government decided to take action to remove 30,000 gold miners from indigenous lands. And one of the key strategies to do so was to shut down the airstrips that were close to the mines in remote areas, so the miners could not take the gold out of the forest until they have to leave.

So in order to do so, you have to find out where are those airstrips, right? So in just three weeks, our team was able to find out and map almost 3,000 air strips across the entire Amazon, which is an area of the size of Europe. So as of today, the number of gold miners in indigenous lands have dropped 19, So none of that will be possible for not applying a kind of collaborative approach.

So Mapyoma's network is a group of over 100 organizations working across South America and Indonesia. And now, with the support of that community, we aim to, through the Auditions Project, we aim to increase our reach to cover 70% of the awards tropical forests by 2013. We believe that this, you know, the ability and the power to produce locally your maps for action should be present on every tropical country. And then we can exchange the sounds of destruction by the sounds of life. That was Taso Asevedo at the TED Countdown Summit in Nairobi, Kenya in 2025.

This ambitious idea is part of the Audacious Project, TED's initiative to inspire and fund global change. Learn more at audaciousproject.org. If you're curious about Ted's curation, find out more at TED.com slash curation guidelines. And that's it for today. Ted Talks Daily is part of the TED Audio Collective. This talk was fact-checked by the TED Research Team and produced and edited by our team, Martha Estefanos, Oliver Friedman, Brian Green, Lucy Little, and Tonicaa Sung Marnivong. This episode was mixed by Christopher Faisi Bogan. Additional support from Emma Tobner and Daniela Balareso. I'm Elise Hugh. I'll be back tomorrow with a fresh idea for your feed.

Thanks for listening. Thank you.