Why AI Could Be Our Best Conservation Tool Yet

Sustainability AI


Since the 2nd December the 25th United Nations Climate Change Conference (COP25) has been taking place in Madrid and next week sees the 10th anniversary of the excellent Sustainable Innovation Forum, organised by Climate Action. The forum is the largest business event alongside the conference and underpins the important role business and technology has in helping accelerate efforts to mitigate climate change. At the forum this year AI will be a major talking point - both in terms of its potential to protect ecosystems, and unfortunately - to damage, unless AI is powered by genuinely green energy.

In this first of two blogs on the relationship between AI and the environment, I’m looking at some pioneering examples of how AI is helping environmental protection efforts, especially in identifying areas of concern. Without fully understanding the unique ecology and ecosystems at threat, we can't conserve them. AI is transforming the way we map and analyse the natural world, further bridging the gap between us and nature. One of the most valuable examples of these ecosystems is coral reefs.

50 Reefs is an initiative by The Ocean Agency, developed in 2017 during the filming of the critically acclaimed Netflix documentary ‘Chasing Coral’. The documentary itself showcases extraordinary underwater photography of the world’s coral reefs, uncovering their beauty and the appalling extent of their destruction. Its accessible platform and message bring powerful public awareness to the vanishing of coral reefs at an unprecedented rate. For an amazing watch, check it out here!

The 50 Reefs AI scheme that blossomed behind the scenes of 'Chasing Coral' is truly fascinating, and is accredited for technology that has elevated the efficiency of scientific reef analysis. Researchers used 360-degree cameras to take more than 56,000 images of what amounts to 3,851 kilometres worth of shallow-water reefs, according to the World Economic Forum. A fraction of these images were subsequently used to train AI to identify different coral and invertebrate types. Now, 50 Reefs has succeeded in creating a system that can catalogue reef imagery independently and, giving the project its name, identify 50 climate-change resistant reefs to protect above all else, for example, Indonesia's 'Sulawesi' reefs. You can take a closer look at their scientific study on the Society for Conservation Biology here.

The discovery of these reefs thanks to our AI analysis tools is incredibly encouraging, enabling biologists to scrutinise their genetic makeup and determine what makes them unique. It's paved the way for parallel studies too. Today, coral geneticists and researchers at the National Sea Simulator in Townsville, Australia continue to experiment with coral hybrids: crossbred corals that inherit desirable survival qualities such as heat tolerance and bleaching resistance. The hope is that the coral of the future can be engineered to survive our planet's rising temperatures, preserving these thriving ecosystems.

The core problem is obvious: our limitless global greenhouse gas emissions. It's easy to daydream about having enough fancy technology to create huge, efficient machines that extract carbon dioxide from the atmosphere… until we remember that trees already exist! Carbon negative and teeming with diverse life, our forests are environmental treasures in disastrous decline: according to the World Bank, we lost 1.3 million square kilometres of forest to deforestation between 1990-2016.

In February 2019, Microsoft partnered with The Nature Conservancy and SilviaTerra, a San Francisco based AI company using machine learning to conserve and inventory forests. Aimed at landowners, timber companies and conservation groups, Silvieterra pairs terabytes of stored satellite imagery of US forests with field data from the USFS forest inventory and analysis program, creating an algorithm that predicts the size and species of trees from studying either information: a detailed, virtual 3D forest map, effectively.

So how does this better the planet? The partnership, the Tree Potential Project, works in the interests of preserving valuable habitats, reducing forest fire risks and growing healthy trees, fully through the use of AI to measure, map and therefore manage the forest environments around us. In a promotional video for the partnership, Zac Parisa, founder of Silviaterra said that “working with Microsoft AI… allows us to scale our process globally and to democratise access to this kind of information.”

Microsoft and their other 'AI for Earth' partnerships present the new age of environmental conservation. This includes OceanMind, an initiative that prevents unregulated fishing by using AI to monitor suspicious vessel movements, and Wild Me, technology that enables the population monitoring of animals to prevent species extinction.

There is certainly hope that we can preserve our natural environment by overseeing it closely, and machine learning is making that process so much more effective already. Check out part two to this blog next week, which addresses how AI is doing its bit to solve the climate problem, as well as raise awareness, and why it’s crucial this good work is underpinned by renewable energy.


Written by Florence Grist

See Florence Grist's blog

Based in the UK, Florence Grist is a freelance writer who enjoys writing on technology and sustainability issues and especially how AI has the potential to both transform our understanding of the environment and help protect fragile ecosystems.




Related blogs

Think Before We All Jump: The Most Important AI Considerations for Humans

As I’ve recently discussed in my latest blog post, artificial intelligence operates most efficiently when it is commoditising intelligence and decision making. Simple and repetitive tasks, and later complex and repetitive tasks, will be ‘solved’ through artificial intelligence. While we are starting to see real proof of the scientific and business benefits that come from this streamlining and processing of data, there are moral and ethical discussions beginning to take place.

Read more


Iceland provides the power behind Germany's most pioneering AI start-ups

This week has seen the announcement of Analytic Engineering, a pioneering German AI engineering firm, choosing Verne Global’s data center in Iceland as the location for their intensive computing. This represents another impressive AI and Machine Learning client win for us, following DeepL joining us just before Christmas.

Read more


Explainable AI

SC18 here in Dallas is proving once again to be a fascinating melting pot of HPC insights and observations, and it's intriguing to see the continuing convergence of AI into the supercomputing ecosystem. Along these lines I started to think about the movement towards 'Explainable AI'. Being able to explain and understand how models work when making predictions about the real world is a fundamental tenet of science. Whether solving equations in a dynamic system for precise answers or using statistical analysis to examine a distribution of events, the results sought from these methods are intended to increase our clarity and knowledge of how the world works.

Read more

We use cookies to ensure we give you the best experience on our website, to analyse our website traffic, and to understand where our visitors are coming from. By browsing our website, you consent to our use of cookies and other tracking technologies. Read our Privacy Policy for more information.