3 Projects Using GPUs to Better Our Planet

May 31, 2019 • 5 min read


What industries come to mind when you think of GPUs? Is it gaming? Perhaps it's machine learning? Heck. Probably even Bitcoin! (Is that still a thing?)

The gradual development of GPUs today, particularly NVIDIA GPUs, have evolved way past its gaming infancy days to well-advanced and robust industries such as the ones mentioned above and now even towards the green industry!

NVIDIA GPUs and deep learning aid our understanding of ecosystems, climate patterns, managing waste, and even preserving plants and animals.

Here are three ways scientists, researchers and companies are using GPUs for a better planet.

AI Saves Trees, AI Saves Planet

Whether we are aware of this or not, our planet relies heavily on trees. Doesn't matter if we are in a tropical rainforest or jam-packed into urban green spaces, trees provide oxygen for us.

However, manually monitoring forests to track potential risks to plant health is very costly and time consuming.

A startup based in Portgual,, is using AI to monitor forests from satellite imagery in a fraction of time currently required. They use NVIDIA GPUs locally and in the cloud to process roughly around 100TB of new satellite data (daily!), helping their clients analyze tree growth, productivity and species.

Save The Whales

Today, whales are posed to a threat from commercial shipping vessels that can accidentally strike whales as they pass through shipping lanes. They are placed in this danger due to climate change as they have to adopt a new migration path.

Autonomous drone company Planck Aerosystems is working with Transport Canada, the national transportation department, to identify whales from aerial drone imagery with NVIDIA GPUs and AI. This tool is able to help scientists and biologists narrow down thousands of images to identify the few containing whales, so ships can slow down and avoid the endangered creatures.

Climate Change - It's Happening

Yep. Whether you believe it or not, climate change is happening. But at what rate?

Climate models vary in their projections of global temperature rise in the coming years, from 1.5 degrees to more than three degrees by 2100. This spreadout variation is mainly due to the difficulty of representing clouds in global climate models.

Thanks to AI, neural networks can be used to address this cloud resolution challenge, as researchers from UC Irvine, Columbia University, and the University of Munich found. Developed using an assortment of NVIDIA GPUs, their deep learning model improved performance and provided better predictions for precipitation extremes than the original climate model. This detailed view can improve scientists' ability to predict regional climate impact.



Related Content