How we apply the latest advances in artificial intelligence to help fight climate change and build a more sustainable, low-carbon world
Artificial intelligence is a powerful technology that will transform our future, so how can we best apply it to help fight climate change and find sustainable solutions?
Our head of climate and sustainability, Sims Witherspoon, who recently spoke about how artificial intelligence can accelerate our transition to renewable energy at TED Countdown, explains, “Climate change is a multifaceted problem with no single solution. We need to move beyond the debate what we can do and start focusing how we can do it.”
The effects of climate change on Earth’s ecosystems are incredibly complex, and as part of our effort to use AI to solve some of the world’s toughest problems, here are some of the ways we’re working to advance our understanding , to optimize existing systems. and to accelerate cutting-edge climate science and its impacts.
Understand weather, climate and their effects
A better understanding of the underlying problems and their impacts is a critical first step in addressing climate change. In collaboration with the UK Met Office, we developed a precipitation model to better understand changing weather conditions. This nowcast model is more accurate than the existing state of the art and is much preferred by the Met’s specialist meteorologists. Our climate and weather research includes short-range (less than two hours) to medium-range (ten days) forecasts, which can greatly influence how we optimize natural-based renewable energy systems resources.
From modeling the behavior of animal species across the Serengeti to supporting machine learning projects that advance conservation projects in Africa, we’ve helped scientists monitor and better understand the effects of climate change on ecosystems and biodiversity. In the future, our team also relies on artificial intelligence systems that they are used to recognize birdsong in Australia, helping to advance tools that track wildlife change at scale.
In addition, we work with non-profit Climate Change AI to fill important climate-related data gaps. Currently, this collaboration focuses on creating a comprehensive wish list of datasets whose availability would advance AI solutions to climate change. We will make this wishlist available to the general public when it is complete.
Optimize existing systems
As we transition to more sustainable infrastructure, we need to optimize the systems the world depends on today. For example, today’s computing infrastructure, including artificial intelligence itself, is energy intensive. To help solve some of these issues, we are developing artificial intelligence that can improve existing systems, including optimizing industrial cooling and more efficient computing systems.
Since our energy grids are not yet running on clean energy, it is important that we use our resources as efficiently as possible while working towards the transition to renewable energy sources. Accelerating the global transition to renewable energy sources can also significantly reduce carbon emissions.
In 2019, our climate and sustainability team worked with domain experts at a Google-owned wind farm to increase the value of wind power – ultimately, with the aim of supporting growth in the wider industry. By developing a custom AI tool to better predict wind power generation and another model to recommend commitments to deliver that anticipated power to the grid, this tool has significantly enhanced the value of wind power. Cloud is now developing a software product using this model, which is being piloted by French energy company ENGIE.
Accelerate cutting-edge science
Beyond optimizing our existing infrastructure, we need scientific breakthroughs to help us build a sustainable energy future. One particular area that holds great promise is nuclear fusion, an incredibly powerful technology with the potential to provide unlimited carbon-free energy. Fusion reactors are powered by a pressurized plasma of ionized hydrogen that is hotter than the core of the sun. The intense heat means this plasma can only be contained by a rapidly adjustable magnetic field – a notoriously difficult engineering challenge.
Mastering the magnetic control of the plasma is a fundamental part of solving the challenge of controlling the nuclear fusion process and harnessing the abundant green energy it could provide. So we worked with the Swiss Plasma Center at EPFL to develop an artificial intelligence system that learned how to successfully predict and control the plasma in a tokamak-type nuclear fusion reactor. And not just to contain the creature, but to “sculpt” it into a series of experimental shapes.
Bring us your challenges
To create effective AI solutions, researchers need a solid understanding of the challenges facing people around the world. This includes gaining access to data that represents the problems, working with domain experts to make sure we’re building reliable systems, following policy guidelines for regulatory structures, and finding real-world opportunities to test those systems. For these reasons, working with affected communities, scientists, industry professionals, regulators and governments is central to our sustainability efforts.
If you are an industry expert or climate scientist with a specific challenge to solve that could help the world understand, mitigate or adapt to climate change, our climate and sustainability team would love to to hear you.
Contact: contact-gdm-sustainability@google.com