Machine learning to tackle climate change | by Salvatore Raieli | Sep, 2022
How AI could help against global warming and save the world from humans
The last summer showed how warming is a problem we can no longer ignore. Rising global temperatures are causing increasingly extreme events, and the future could be worse.
Machine learning and artificial intelligence could help against global warming. In this article, we will try to answer the questions: how? what are currently the applications of artificial intelligence already in the field?
Bangladesh and India were hit in June by one of the worst floods ever seen. Less than three months later Pakistan was hit by a flood that lead to a third of the country being underwater. At the same time, the worst drought in the last 1000 years hit Spain and Portugal this year. France and other European countries experience this year dire forest fires during the persistent heatwave that affected the continent this summer. California is experiencing an increase in the number of destructive wildfires in the last decade.
In summary, we know there is a link between these destructive events and climate change. The more the global temperature increase more it is expected that these events will be more frequent and more destructive. All the climate models are agreeing that without carbon dioxide reductions we will see a rise in global temperatures and extreme events.
“Without immediate and deep emissions reductions across all sectors, limiting global warming to 1.5°C is beyond reach” — IPCC press release
If this is not enough to explain the urgency, European Union has discovered how fragile its energy supply chain is and how dependent on Russian gas is. Thus, there is a call to transition from fossil fuel to renewable energy.
“We are at a crossroads. The decisions we make now can secure a liveable future. We have the tools and know-how required to limit warming,” — Hoesung Lee, IPCC press release
In this article, I will discuss how machine learning and artificial intelligence are envisioned to have a pivotal role in the energy transition and carbon dioxide emission reduction.
Since 2019 Climate Change AI (CCAI) is a volunteer-driven community (composed of members of academia and industry) that has a mission to catalyze the intersection of climate change and machine learning expertise.
They recently released a report where they presented a long list of areas and applications where machine learning could be helpful against climate change. The article is presenting divided the possible strategies as:
- high leverage, areas that are well suited for ML tools
- long-term, areas where applications are not expected to have a primary impact before 2040
- uncertain impact, areas where is difficult to interpret if applying a strategy will be helpful or leading undesirable effects (since the technology is not mature enough or the effect are unpredictable)
The authors considered electricity systems as a high-leverage area. In fact, they proposed that machine learning could contribute to the operation of electricity system technologies (helping the transition to low-carbon energy sources, energy demand optimization, management of the grid, and so on).
Interestingly, they flagged as high leverage but also the uncertain impact of the reduction of the emission by existing implants. In fact, while the transition to renewable energy is happening we could optimize existing implants (for example, use machine learning to avoid the leakage of methane from a natural gas pipeline). However, optimized fossil fuel implants could be perceived as “more green” and could delay the transition (thus, uncertain impact).
The report is investing in details of the application of many AI technologies. While it is easier to think about how reinforcement learning and autonomous vehicles could be used, they are showing as practically all the subfields of AI could be relevant (from natural language processing to causal inference and so on).
“Machine learning, like any technology, does not always make the world a better place — but it can” — Climate Change AI report
The report provides detailed suggestions for future applications and strategies. The good news is that there are already companies and researchers working to implement them.
In fact, even if the cost of the wind turbine has plummeted the wind is not constant and it is unpredictable. For this reason, DeepMind applied ML algorithms to predict wind power. Using a neural network model they optimized the delivery commitments to the power grid of the generated energy. Google and DeepMind tested the model on a wind farm in the central United States (700 MW implants).
Recently, Google decided to sell this technology through Google Cloud to wind farms. Google announced that this algorithm could predict wind power output thirty-six hours in advance. In June, Engie (a French company) has been announced as the first client of the project.
Another interesting application is Climate TRACE, a coalition of universities that are using computer vision to track greenhouse emissions. They are using satellite images and remote sensing to identify the greenhouse source, monitoring them to facilitate climate action. Moreover, the large quantity of collected data is available to the community to freely use for new potential applications. You can learn more in this video:
Satellite images are also used to monitor sea level rise or predict areas that are sensitive to drought, track deforestation, and so on. In addition, startups as Pano AI and Fion Technologies are implementing computer vision to identify areas at risk of fires, detect the wildfires, and predict how to spread.
Since solar and wind energy are variable, there are also other projects interested in improving battery storage. For example, Carnegie Mellon University in collaboration with Meta AI, build the Open Catalyst Project. They released also a large dataset to improve the catalyst simulation and they organize different related open challenges.
Moreover, agriculture is responsible for more than 10 % of global emissions, and fertilizers are not only damaging the environment but also powerful greenhouse gases. This led to the emergence of a new field: precision agriculture. In fact, different companies are using AI to optimize resources consumption, and the fertilizer usage
Furthermore, buildings are producing close to one-fifth of total carbon emission and thus it is extremely important to optimize them. There are companies focused on how to improve the building process, startups focused on improving air conditioning, finding new materials, and so on. In addition, recently DeepMind announced to use of AI to reduce the Google Data Centre cooling bill by 40%.
Other applications are less straightforward, nevertheless, they have attracted the interest of several companies. Calculating the carbon, footprint of a company is also not easy, different startups are now proposing to calculate it for large companies. Once, a company is aware of which process has the higher carbon footprint can think about actions and optimization. For example, Watershed assesses the carbon dioxide emission of other companies and propose emission reduction solution.
“The world’s first trillionaire will be made in climate change.” — prediction by Chamath Palihapitiya
All of these projects show how companies are investing in the use of artificial intelligence against climate change. And if this is good for the environment, it is also good for business.
“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, year after year, is proving to be an increasingly urgent problem. Every year the frequency of extreme events increases, resulting in extensive damage. Predictions say that if we do not take urgent and drastic action, we will face even worse and more frequent events.
Machine learning and artificial intelligence are not magic potions that can solve global warming alone (not to mention that the use of artificial intelligence and big data also produce carbon dioxide). ML and AI are, however, considered key means to combat global warming: both for energy transition and emission reduction. The report details which strategies and applications would benefit most from ML and AI.
Academia has always been engaged in studying and proposing solutions. The good news is that today there are also several companies working on developing these strategies and new applications. In addition, investment in the energy transition (renewable energy, electric vehicles, and so on) has grown significantly in recent years, and there is new consumer awareness and attention. On the other hand, there needs to be a global and stronger commitment from the side of governments.
If you know about other initiatives, other companies, and projects involved in using AI to tackle climate change, let me know.
You can look for my other articles, you can also subscribe to get notified when I publish articles, and you can also connect or reach me on LinkedIn. Thanks for your support!
Here is the link to my GitHub repository, where I am planning to collect code and many resources related to machine learning, artificial intelligence, and more.
Or feel free to check out some of my other articles on Medium:
Additional resources
- on global warming: here, here, here
- about carbon dioxide and climate change: here, here, here, here
- on investment in renewable energies and energetic transition: here, here
- about DeepMind AI solution for Google data center: here, here, here
- about DeepMind AI solution for wind energy prediction: here
How AI could help against global warming and save the world from humans
The last summer showed how warming is a problem we can no longer ignore. Rising global temperatures are causing increasingly extreme events, and the future could be worse.
Machine learning and artificial intelligence could help against global warming. In this article, we will try to answer the questions: how? what are currently the applications of artificial intelligence already in the field?
Bangladesh and India were hit in June by one of the worst floods ever seen. Less than three months later Pakistan was hit by a flood that lead to a third of the country being underwater. At the same time, the worst drought in the last 1000 years hit Spain and Portugal this year. France and other European countries experience this year dire forest fires during the persistent heatwave that affected the continent this summer. California is experiencing an increase in the number of destructive wildfires in the last decade.
In summary, we know there is a link between these destructive events and climate change. The more the global temperature increase more it is expected that these events will be more frequent and more destructive. All the climate models are agreeing that without carbon dioxide reductions we will see a rise in global temperatures and extreme events.
“Without immediate and deep emissions reductions across all sectors, limiting global warming to 1.5°C is beyond reach” — IPCC press release
If this is not enough to explain the urgency, European Union has discovered how fragile its energy supply chain is and how dependent on Russian gas is. Thus, there is a call to transition from fossil fuel to renewable energy.
“We are at a crossroads. The decisions we make now can secure a liveable future. We have the tools and know-how required to limit warming,” — Hoesung Lee, IPCC press release
In this article, I will discuss how machine learning and artificial intelligence are envisioned to have a pivotal role in the energy transition and carbon dioxide emission reduction.
Since 2019 Climate Change AI (CCAI) is a volunteer-driven community (composed of members of academia and industry) that has a mission to catalyze the intersection of climate change and machine learning expertise.
They recently released a report where they presented a long list of areas and applications where machine learning could be helpful against climate change. The article is presenting divided the possible strategies as:
- high leverage, areas that are well suited for ML tools
- long-term, areas where applications are not expected to have a primary impact before 2040
- uncertain impact, areas where is difficult to interpret if applying a strategy will be helpful or leading undesirable effects (since the technology is not mature enough or the effect are unpredictable)
The authors considered electricity systems as a high-leverage area. In fact, they proposed that machine learning could contribute to the operation of electricity system technologies (helping the transition to low-carbon energy sources, energy demand optimization, management of the grid, and so on).
Interestingly, they flagged as high leverage but also the uncertain impact of the reduction of the emission by existing implants. In fact, while the transition to renewable energy is happening we could optimize existing implants (for example, use machine learning to avoid the leakage of methane from a natural gas pipeline). However, optimized fossil fuel implants could be perceived as “more green” and could delay the transition (thus, uncertain impact).
The report is investing in details of the application of many AI technologies. While it is easier to think about how reinforcement learning and autonomous vehicles could be used, they are showing as practically all the subfields of AI could be relevant (from natural language processing to causal inference and so on).
“Machine learning, like any technology, does not always make the world a better place — but it can” — Climate Change AI report
The report provides detailed suggestions for future applications and strategies. The good news is that there are already companies and researchers working to implement them.
In fact, even if the cost of the wind turbine has plummeted the wind is not constant and it is unpredictable. For this reason, DeepMind applied ML algorithms to predict wind power. Using a neural network model they optimized the delivery commitments to the power grid of the generated energy. Google and DeepMind tested the model on a wind farm in the central United States (700 MW implants).
Recently, Google decided to sell this technology through Google Cloud to wind farms. Google announced that this algorithm could predict wind power output thirty-six hours in advance. In June, Engie (a French company) has been announced as the first client of the project.
Another interesting application is Climate TRACE, a coalition of universities that are using computer vision to track greenhouse emissions. They are using satellite images and remote sensing to identify the greenhouse source, monitoring them to facilitate climate action. Moreover, the large quantity of collected data is available to the community to freely use for new potential applications. You can learn more in this video:
Satellite images are also used to monitor sea level rise or predict areas that are sensitive to drought, track deforestation, and so on. In addition, startups as Pano AI and Fion Technologies are implementing computer vision to identify areas at risk of fires, detect the wildfires, and predict how to spread.
Since solar and wind energy are variable, there are also other projects interested in improving battery storage. For example, Carnegie Mellon University in collaboration with Meta AI, build the Open Catalyst Project. They released also a large dataset to improve the catalyst simulation and they organize different related open challenges.
Moreover, agriculture is responsible for more than 10 % of global emissions, and fertilizers are not only damaging the environment but also powerful greenhouse gases. This led to the emergence of a new field: precision agriculture. In fact, different companies are using AI to optimize resources consumption, and the fertilizer usage
Furthermore, buildings are producing close to one-fifth of total carbon emission and thus it is extremely important to optimize them. There are companies focused on how to improve the building process, startups focused on improving air conditioning, finding new materials, and so on. In addition, recently DeepMind announced to use of AI to reduce the Google Data Centre cooling bill by 40%.
Other applications are less straightforward, nevertheless, they have attracted the interest of several companies. Calculating the carbon, footprint of a company is also not easy, different startups are now proposing to calculate it for large companies. Once, a company is aware of which process has the higher carbon footprint can think about actions and optimization. For example, Watershed assesses the carbon dioxide emission of other companies and propose emission reduction solution.
“The world’s first trillionaire will be made in climate change.” — prediction by Chamath Palihapitiya
All of these projects show how companies are investing in the use of artificial intelligence against climate change. And if this is good for the environment, it is also good for business.
“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, year after year, is proving to be an increasingly urgent problem. Every year the frequency of extreme events increases, resulting in extensive damage. Predictions say that if we do not take urgent and drastic action, we will face even worse and more frequent events.
Machine learning and artificial intelligence are not magic potions that can solve global warming alone (not to mention that the use of artificial intelligence and big data also produce carbon dioxide). ML and AI are, however, considered key means to combat global warming: both for energy transition and emission reduction. The report details which strategies and applications would benefit most from ML and AI.
Academia has always been engaged in studying and proposing solutions. The good news is that today there are also several companies working on developing these strategies and new applications. In addition, investment in the energy transition (renewable energy, electric vehicles, and so on) has grown significantly in recent years, and there is new consumer awareness and attention. On the other hand, there needs to be a global and stronger commitment from the side of governments.
If you know about other initiatives, other companies, and projects involved in using AI to tackle climate change, let me know.
You can look for my other articles, you can also subscribe to get notified when I publish articles, and you can also connect or reach me on LinkedIn. Thanks for your support!
Here is the link to my GitHub repository, where I am planning to collect code and many resources related to machine learning, artificial intelligence, and more.
Or feel free to check out some of my other articles on Medium:
Additional resources
- on global warming: here, here, here
- about carbon dioxide and climate change: here, here, here, here
- on investment in renewable energies and energetic transition: here, here
- about DeepMind AI solution for Google data center: here, here, here
- about DeepMind AI solution for wind energy prediction: here
Denial of responsibility! Techno Blender is an automatic aggregator of the all world’s media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, all materials to their authors. If you are the owner of the content and do not want us to publish your materials, please contact us by email – [email protected]. The content will be deleted within 24 hours.