The escalating problem of global warming has become undeniable, especially highlighted by the extreme weather conditions witnessed last summer. Rising temperatures are leading to more frequent and severe climate events, and the outlook appears grim.

The integration of artificial intelligence and machine learning could play a crucial role in combating climate change. This article aims to explore the ways AI can help and what applications are currently in use.

Why is Immediate Action Necessary?

In June, Bangladesh and India experienced devastating floods, and shortly after, Pakistan faced a similar fate, with one-third of its land submerged. Concurrently, Spain and Portugal endured their worst drought in a millennium, while France and other European nations battled severe wildfires during an unprecedented heatwave. California has also seen a marked increase in destructive wildfires over the past decade.

These events underscore the direct connection between climate change and extreme weather. As global temperatures rise, the frequency and severity of such events are expected to increase. Climate models consistently indicate that without significant reductions in carbon emissions, a rise in global temperatures and extreme weather events is inevitable.

“Without immediate and deep emissions reductions across all sectors, limiting global warming to 1.5°C is beyond reach” — IPCC press release

The urgency of the situation is further emphasized by the European Union's realization of its vulnerable energy supply chain and heavy reliance on Russian gas. This has spurred a push toward transitioning from fossil fuels to renewable energy.

“We are at a crossroads. The decisions we make now can secure a livable future. We have the tools and know-how required to limit warming,” — Hoesung Lee, IPCC press release

This article will delve into the potential of machine learning and AI to facilitate energy transitions and reduce carbon emissions.

How Can Machine Learning Make a Difference?

Since its inception in 2019, Climate Change AI (CCAI) has united volunteers from academia and industry to promote the intersection of machine learning and climate change efforts.

In a recent report, CCAI outlined numerous areas where machine learning could contribute to climate solutions. The strategies were categorized as follows:

• High leverage: Areas well-suited for ML applications
• Long-term: Areas expected to see benefits after 2040
• Uncertain impact: Areas where the effects of applying ML are unpredictable due to immature technology

Electricity systems were identified as a high-leverage area. Machine learning could enhance the operation of these systems by supporting the shift to low-carbon energy sources, optimizing energy demand, and managing the grid.

Interestingly, while optimizing existing fossil fuel infrastructures could yield immediate benefits, it may also delay the transition to renewables, leading to uncertain outcomes.

The report elaborates on various AI technologies' applications. While reinforcement learning and autonomous vehicles are often discussed, virtually every AI subfield, from natural language processing to causal inference, can contribute.

“Machine learning, like any technology, does not always make the world a better place — but it can” — Climate Change AI report

A Real-World Example of ML in Action

The report offers numerous strategies and future applications. Encouragingly, various companies and researchers are already implementing these ideas.

For instance, despite the plummeting costs of wind turbines, wind patterns remain unpredictable. To address this, DeepMind has employed machine learning algorithms to forecast wind energy production. By utilizing a neural network model, they enhanced the scheduling of energy delivery to the power grid at a wind farm in central United States.

Recently, Google announced that this predictive technology will be available to wind farms via Google Cloud, enabling forecasts of wind energy output up to thirty-six hours in advance. Engie, a French company, has already signed on as the first client for this project.

Another innovative initiative is Climate TRACE, a coalition of universities leveraging computer vision to monitor greenhouse gas emissions. By employing satellite imagery and remote sensing, they identify and track emission sources, facilitating climate action. The resulting data is publicly accessible, allowing for wider application and innovation.

Satellite images are also employed to monitor sea-level rise, identify drought-prone regions, and track deforestation. Startups like Pano AI and Fion Technologies are utilizing computer vision to detect fire risks, manage wildfires, and predict their spread.

Given the variability of solar and wind energy, other projects are focusing on improving battery storage. For example, Carnegie Mellon University, in collaboration with Meta AI, has initiated the Open Catalyst Project, providing extensive datasets to enhance catalyst simulations and organize related challenges.

Agriculture, responsible for over 10% of global emissions, has seen the rise of precision agriculture, aimed at optimizing resource and fertilizer use through AI technologies.

Buildings contribute nearly one-fifth of total carbon emissions, making optimization essential. Companies are developing improved construction processes and innovative materials, while DeepMind has reported a 40% reduction in cooling costs at Google Data Centers through AI.

Some applications are less direct, yet they have garnered significant interest. Calculating a company's carbon footprint can be complex, yet various startups now offer carbon assessment services. By identifying high-emission processes, companies can take informed steps toward optimization. Watershed, for instance, evaluates and proposes emission reduction strategies for various firms.

“The world’s first trillionaire will be made in climate change.” — prediction by Chamath Palihapitiya

These initiatives demonstrate how companies are leveraging AI to combat climate change, benefiting both the environment and their business models.

Final Thoughts “We emphasize that in each application, ML is only one part of the solution; it is a tool that enables other tools across fields.” — Climate Change AI report

“There is therefore no single ‘silver bullet’ application of AI to climate change. Instead, a wide range of machine learning use cases can help in the race to decarbonize our world.” — Forbes

Global warming is an increasingly pressing issue, with the frequency of extreme events on the rise, resulting in significant damage. Predictions indicate that without urgent action, the situation will worsen.

While machine learning and artificial intelligence are not standalone solutions to global warming—especially considering their own carbon footprint—they are viewed as critical tools in the fight against climate change, particularly for energy transition and emission reduction. The report outlines which strategies and applications could benefit most from ML and AI.

Historically, academia has been engaged in researching and proposing solutions. Fortunately, numerous companies are now developing these strategies and applications. Investments in renewable energy and electric vehicles have surged in recent years, alongside a growing consumer awareness. However, greater global commitment from governments is essential.

If you are aware of other initiatives or projects utilizing AI to address climate change, please share.

If you found this article engaging: Explore my other writings, subscribe for updates on new articles, and connect with me on LinkedIn. Your support is appreciated!

Check out my GitHub repository, where I plan to compile resources related to machine learning, artificial intelligence, and more.

<div class="link-block">

<h2>GitHub - SalvatoreRa/tutorial: Tutorials on machine learning, artificial intelligence, data science…</h2>

<h3>Tutorials on machine learning, artificial intelligence, data science with math explanation and reusable code (in python…</h3>

<p>github.com</p>

</div>

Feel free to browse some of my other articles:

<div class="link-block">

<h2>How Science Contribution Has Become a Toxic Environment</h2>

<h3>How computer science has inherited the same mistakes as other disciplines</h3>

<p>towardsdatascience.com</p>

</div>

<div class="link-block">

<h2>How AI could fuel global warming</h2>

<h3>New large models are energy intensive. How much CO2 is needed for their training?</h3>

<p>towardsdatascience.com</p>

</div>

<div class="link-block">

<h2>Speaking the Language of Life: How AlphaFold2 and Co. Are Changing Biology</h2>

<h3>AI is reshaping research in biology and opening new frontiers in therapy</h3>

<p>towardsdatascience.com</p>

</div>

<div class="link-block">

<h2>A New BLOOM in AI? Why the BLOOM Model Can Be a Gamechanger</h2>

<h3>We are now used to large language models, why is this so special?</h3>

<p>pub.towardsai.net</p>

</div>

Additional resources - On global warming: here, here, here - About carbon dioxide and climate change: here, here, here, here - On investment in renewable energies and energy transition: here, here - About DeepMind AI solutions for Google data center: here, here, here - About DeepMind AI solutions for wind energy prediction: here