“The capability to investigate massive amounts of information is critical here because we all know that it’s a rainforest that spans countries and thousands and thousands of square kilometers,” Arbeláez said. “So, to have a world view of the Amazon basin, you truly need the ability of AI and these latest models.”

The following step is critical to the biodiversity of the rainforest. Camera boxes have been placed throughout the Colombian Amazon. Anytime the cameras see movement, they take an image. A single camera can take as much as 300,000 photos. But having a researcher flip through each image would take years.

So, AI programs help speed up the method. The Department of Biological Sciences at Universidad de Los Andes contributed 110,000 images collected over the past 4 years. With the assistance of AI, researchers trained a model that detects which images contain animals and those who don’t. Now, only 10 percent of images should be manually validated, freeing up priceless hours for other research. Moreover, with the ability to quickly determine what images have been captured is vital, because if something seems misplaced, it could possibly be an indication of ecological shifts that should be addressed.

“It’s like an image in time,” Ochoa said. “Monitoring allows us to see how plants and animals live and thrive. But when we see the flawed species within the camera, that is named a bioindicator. As an illustration, if we see a bird that is generally present in the savannah and is now within the Amazon, it is a red flag. We will say to the authorities, ‘We’re getting information of a species that was previously not on this area. There’s a change within the ecosystem.’”

“This technology, and the info it generates, allows us to tell decision-makers within the policy area to assist us, especially after we are coping with invasive species,” Ochoa added.

The ultimate element of Project Guacamaya is sound. Using bioacoustics, researchers can capture sound from the Amazon and use an audio AI model to distinguish bird and non-bird sounds to categorise bird and other animal species. Guacamaya consolidates greater than 100,000 sounds into 1,000 recordings to have a baseline, after which trains a model to detect the various sounds and classify them. This helps scientists gain bird identification with reliability above 80%, which is a serious milestone considering there are greater than 2,000 bird species in Colombia.

“Using this (technology) allows us to diminish the price of the research in the sector,” Ochoa said. “For somebody to hearken to biodiversity for 3 months, is something that nobody can afford. This technology allows us to do this in a really cost-efficient way.”

Protecting the Amazon rainforest is a large-scale issue that nobody organization or entity can tackle by itself. Collaboration is a critical a part of Project Guacamaya, with quite a few groups coming together to bring their very own expertise to the table, after which hopefully create models that might be used throughout the country and the region for the greater good.

“Our goal here is to bring everyone in the identical room and work jointly towards a typical goal, because each of the partners and institutions can have a complementary role on the project,” Arbeláez said. “To guard the world’s largest rainforest from the consequences of climate change, this must be a joint effort from all of the countries which have a stake within the Amazon. We truly hope the region’s leaders and all our Latin America siblings will take this project as their very own.”

In Brazil, where the consequences of deforestation and mining within the Amazon have been most distinguished, a collaboration between Microsoft, environmental organization Imazon and nonprofit Fundo Vale Foundation called PrevisIA can be addressing the issue using technology.

The PrevisIA platform uses Microsoft’s AI to investigate data and satellite images to forecast and monitor deforestation in Brazil’s Amazon. Imazon uses satellite images after which stores them within the Azure cloud, where AI algorithms detect unofficial roads and other risk aspects of deforestation. The resulting output is visualized in an interactive map, highlighting the high-risk areas. This information is then used to support decision-making for safeguarding the rainforest.