This blog post was originally published at NVIDIA's website. It is reprinted here with the permission of NVIDIA.
Nearly 19 million acres of forests are destroyed annually, equal to 27 football pitches a minute.
Forests serve as homes for thousands of animals, and for many people they’re a source of food, water, clothing, medicine and shelter. The so-called lungs of the Earth also mitigate climate change by acting as a carbon sink. And as cities expand and competition for space increases, managing vital green spaces is becoming increasingly important.
To deepen our understanding of these complex ecosystems, 20tree.ai, a Portugal-based startup and member of the NVIDIA Inception program, is combining AI and satellite imagery. Their work allows for the monitoring of entire forests in a fraction of the time currently required.
Bird’s-Eye View
Down on the ground, it isn’t possible to get a full understanding of the richness of resources in a given forest area or predict any potential risks to the health of plants and trees.
Comprehensive, manual surveying and data collection of forested areas can take months, with considerable strain placed on budgets and manpower.
But 20tree.ai, which was recently awarded the Copernicus Masters Planet Daily Change Challenge award, uses a combination of AI and extremely high-quality satellite imagery and radar data to speed its work.
Every day, the company uses NVIDIA GPUs to process almost 100TB of new satellite data — obtained from partners such as Airbus Defence and Space and the European Copernicus program — which is used to train a series of deep neural networks.
GPUs, running in-house and in the cloud via AWS and Google, provide the muscle power for the training, enabling it to be completed in just a few hours. The deep neural networks can then draw insights into forest health that are otherwise invisible to the human eye.
“Thanks to the power of artificial intelligence and NVIDIA’s GPUs, we are enabling faster, better decision making for our planet,” said Indra den Bakker, co-founder and deep learning engineer at 20tree.ai.
20tree.ai’s dashboard enables customers to derive deeper insights into forested areas.
Through an online user interface, 20tree.ai’s clients can check on specific factors of interest, including tree species, height and diameter, growth and productivity, as well as harvesting potential.
Customers can get these insights in a matter of minutes, allowing them to effectively monitor the progress of Mother Nature’s variables. For example, in the case of droughts, plagues or other disturbances, appropriate reactive measures can be taken earlier, with improved impact.
From Forest Economy to Saving Orangutans
The benefits of 20tree.ai’s system are manifold.
Stora Enso, a renewable materials giant headquartered in Helsinki, Finland, uses 20tree.ai to analyze tree species, wood volume, and tree height and diameter. The company can then determine whether or not they are paying the right amount for land they are purchasing and can also see exactly what they are purchasing.
The speed at which the company receives these insights means they can make faster, informed decisions on plots of land, resulting in more sustainable forest management.
But it’s not just commercial enterprises that gain from AI-powered forest intelligence. NGOs such as Rainforest Alliance are also benefiting. For example, the Sumatran Orangutan Society is working with 20tree.ai on raising awareness around the need for reforestation and restoring valuable ecosystems.
And cities, where an estimated two-thirds of the global population will live by 2050, can gain insights into green spaces, like tree canopy cover, urban forest diversity and air quality. Better monitoring and management of urban forests can help mitigate heat waves, air pollution and flooding.
“Our system enables us to gain insights into the impact of deforestation, drought, plagues and unsustainable forest management, which previously were unattainable,” said den Bakker.
Serge Lemonde
Sales Manager, Inception Program for EMEA and India Region, NVIDIA
