A new artificial intelligence tool called PrevisIA is predicting where deforestation in the Amazon will happen next. It was created by researchers at the environmental nonprofit Imazon. Instead of trying to repair damage done by deforestation after the fact, they hope to prevent it from happening at all.
Destruction across the Brazilian Amazon is close to an all-time high. Deforestation this March tripled compared to the same month last year, and the first quarter of 2023 witnessed 867 sq km of rainforest destroyed – the second largest area felled in the past 16 years.
“Existing deforestation prediction models were long-term, looking at what would happen in decades,” says Carlos Souza Jr, senior researcher at Imazon. “We needed a new tool that could get ahead of the devastation.”
The model focuses on trends in the region, looking at geostatistics and historical data from Probes, the government monitoring system for deforestation in the Amazon. Understanding what has happened in the past can help make predictions more precise.
Then the model looks at variables that put the brakes on deforestation – land protected by Indigenous communities, and areas with bodies of water, or other terrain that doesn’t lend itself to agricultural expansion.
“They are the arteries of destruction of the forest,” says Souza, referring to unofficial roads that snake through the Amazon. “These roads create the conditions for new deforestation.”
Monitoring the construction of these roads is crucial to predicting and preventing deforestation. According to Imazon, 90% of deforestation is concentrated within 5.5 km of a road. Logging is even closer, with 90% occurring within 3km and 85% of fires within 5km.
Researchers used to comb through thousands of satellite images to determining whether they could spot new roads cutting through the region. Using PrevisIA, the work is handed over to an algorithm that automates mapping, allowing for quicker analysis and more frequent updates.
Juliano Assuncao, executive director of the Climate Policy Initiative notes the obvious entities who could benefit from using PrevisIA – government agencies at all levels but also those not directly involved in monitoring the Amazon, banks, investors and those who buy products from the region. They could use this information to make better decisions, both from an economic and an environmental point of view.