DEFORESTATION PATTERN ALGORITHMS

World Resources Institute (WRI) and Orbital Insight apply experimental algorithms to high resolution satellite images to detect patterns that indicate the risk of deforestation before it happens.

Orbital Insight and World Resources Institute (WRI) announced a partnership to apply experimental algorithms to high resolution satellite images. Orbital Insight’s algorithms will be used as part of Global Forest Watch to detect patterns that indicate the risk of deforestation before it happens.

Using a technique called deep learning, Orbital Insight will analyze high resolution satellite images of tens of millions of acres of forest provided by Global Forest Watch to discover patterns that identify indicators of deforestation risk. Deep learning is a form of artificial intelligence that involves processing tremendous amounts of data to solve problems in a way that roughly mimics the human brain.

Global Forest Watch researchers will be able to use the deep learning algorithms to recognize patterns, such as road-building in previously undisturbed areas, and forecast deforestation before the destruction occurs. The technology will also increase transparency around the impact of global commodity supply chains on forest loss. So-called “forest-risk commodities” such as palm oil, beef, soy, pulp, and paper has led to over 70% of deforestation in tropical forests, often illegally.

“We currently know a lot about forest clearing that has happened in the past, but soon we will have the power to look ahead and identify the forest areas at greatest risk,” said Aaron Steele, Chief Technology Officer, WRI. “This partnership makes cutting-edge technology available to everyone and demonstrates how deep learning can have clear benefits for people and the planet.”

Launched in 2014, Global Forest Watch is a dynamic online forest monitoring and alert system that empowers people everywhere to better manage forests. Global Forest Watch partners, led by WRI, use data from University of Maryland, Google, NASA, and others to monitor tree cover loss and other global forest trends. While existing data on Global Forest Watch detects recent past deforestation, Orbital Insight’s deep learning algorithms that analyze satellite images will help WRI move towards a predictive and preventive approach to global forest monitoring.

“WRI has already made great strides toward global transparency in forest data,” said Dr. James Crawford, founder and CEO, Orbital Insight. “This partnership will have a tangible role in curbing deforestation by harnessing the power of deep learning to provide timely actionable insights.”

“Satellites already enable us to monitor many stages of commodities supply chains, from land use change and production to the transportation of raw materials.” said Steele. “Orbital Insight’s algorithms will help us analyze and understand these complex networks to the point where we could forecast deforestation risk based on indicators like the number of logging trucks seen in an area.”

More than one billion people depend on forests for their livelihoods and forests support much of the world’s biodiversity, but these ecosystems are under significant pressure. Between 2000 and 2012, the world lost 2.3 million square kilometers of tree cover—the equivalent of losing 50 soccer fields’ worth of forests every minute of every day.

WRI and Orbital Insight will release analysis as it is developed. WRI and Orbital Insight are committed to using cutting-edge technology to turn information into action to protect resources that are the foundation of economic opportunity and human well-being.

Source: World Resources Institute website

Author: Daniel Melling

More: Global Forest Watch