Three ways AI Is helping fight climate change
John Harding, Regional Director UK and Ireland, Nvidia, explains how AI can help us fight against climate change and its effects
Our planet is getting warmer.
July 2021 was the hottest month ever recorded. Carbon dioxide concentration in the atmosphere was also the highest it has ever been in human history. The average global temperatures in 2020 were 1.76 degrees warmer than the 20th century average.
This rise in temperature can increase the amount of hurricanes at coastal areas, resulting in disastrous rainfall that can wash out vital agricultural fields and cause more once-in-a-thousand-year floods. Each degree Celsius contributes to 10-12% more lightning strikes, leading to more wildfires - even in places like the Arctic circle, a landscape that was thought to be fire-resistant.
Today, organizations are proving time and again that AI can help fight climate change. From wildfire detection to monitoring ocean health, they're using AI to help us better understand, monitor and react to the world we live in.
There's certainly no quick fix. But there are three key areas to focus on -- prediction, monitoring and adaptation, and mitigation.
Predicting climate change
AI can help us to predict the future. In fact, it can forecast the weather 100,000 times faster than traditional models.
Unlike weather forecasting, climate models can be multi-decade simulations that model the physics, chemistry and biology of the atmosphere, waters, ice, land and human activities.
Climate simulations today are configured at 10- to 100-kilometer resolutions. Greater resolution is needed to model variables like changes in the global water cycle, which influences storms and droughts. Meter-scale resolution is needed to simulate clouds that reflect sunlight back to space. Scientists estimate these resolutions will demand millions, or even billions of times more computing power than what's currently possible.
But, with the advances in today's AI technology, the million-x leap needed to do super-resolution climate modelling is closer than ever. Jensen Huang, Nvidia CEO, recently revealed that the company would be building Earth-2, a digital twin system that simulates the Earth's climate. This includes the compute power, training models, data pipeline and all other necessary components to predict the regional impact of mitigation and adaptation strategies.
Being able to predict extreme weather changes decades out means countries, cities and towns can get early warnings to adapt and make infrastructures more resilient.
Monitoring and adapting to climate change
AI is advancing incredibly quickly. Companies and research groups are already putting it to work on understanding, monitoring and solving climate change challenges.
Wildfires are responsible for about 6% of the planet's CO2 emissions. OroraTech, a Zurich-based startup, used their experience of AI and satellites to develop a wildfire detection and monitoring service. Despite being in a country generally unaffected by wildfires, their ambitions were to create a solution to benefit the planet. By building nano-satellites with Nvidia Jetson, a small but powerful AI computer, OroraTech's service is able to detect the first signs of wildfire anywhere in the world. Emergency teams on the ground are alerted to the visuals, where they can act immediately.
Like wildfires, serious flooding caused by monsoon season, earthquakes or just extreme weather can get deadly in a matter of minutes. A group of data scientists applied deep learning to 66,000 satellite images to create AI models for predicting flood zones. Their models can be easily replicated for application in new geographies, and work by analysing pixels for borders between things like land and water. Models like this can monitor potential flooding, alerting people to get out of harm's way before a disaster occurs, saving lives and belongings.
The Massive Graphical Processing Unit Cluster for Earth Observation (MAGEO) is also using deep learning to process Earth observation data. Based at the UK's Plymouth Marine Laboratory, over 10TB of freely available data is collected daily by sensors on more than 150 satellites orbiting the planet. Using this data, researchers at MAGEO are able to monitor fluctuations in the planet's readings. Microscopic phytoplankton, for instance, transfer about 10 gigatonnes of carbon from the atmosphere to the deep ocean. Even small changes in the growth of phytoplankton may affect the atmosphere's CO2 levels, impacting global surface temperatures.
As well as temperature, climate change affects general meteorological conditions. Research suggests both are likely to have a significant impact on bee populations worldwide. Bees pollinate 75% of leading global crops, and for commercial-sized tomato crops that's nearly half a million acres on tomato plants alone. Israel-based startup, Arugga, put AI to the task by developing an AI-powered robot that mimics a bee. It detects which flowers are ready for pollination using computer vision, then blasts air pulses at them to imitate the action of a bumblebee and initiate pollination.
Mitigating Climate Change
Taking steps now can have an impact on the trajectory of climate change. Organizations around the world, large and small, are using AI to reduce greenhouse gas emissions.
Of the 8.3 billion tonnes of virgin plastic created each year, only 9% gets recycled, despite decades of efforts to reduce the amount that ends up in landfill sites. Using AI-infused computer vision, London-based startup Recycleye is significantly increasing the capacity for recycling companies by identifying recyclable materials with more precision than ever before. Quality recyclable materials can be passed on to plastic manufacturers as part of the circular recycling process, and repurposed for another use.
Caltech's Space Solar Power Project does away with CO2 emitting power altogether. Their technology collects solar power in space, then transmits it back to Earth using microwaves. This type of transmission is unaffected by weather or time of day, meaning power could be continuously available all over the globe. The National Space Society predicts that space-based solar power has the potential to provide more energy than all other sources of energy combined, with very little negative environmental impact.
Shell, one of the world's largest energy providers, is working to become a net-zero emissions energy provider by 2050 and is using AI to accelerate its energy transition. They're combining offsets with lower-carbon energy, and aiming to reduce carbon emissions across sectors like aviation, shipping, road freight and industry.
Leveraging the AI community of passionate developers and startups, Shell is using AI hackathons to find new solutions and early-stage ideas to meet their net-zero targets. Their Shell.ai hackathon, which ran in late 2021, focused on harnessing the power of the sun. Working collaboratively with the AI community, they wanted to develop a solution to best understand the prime location for solar power -- based on variable weather conditions and hours of daylight.
Capturing and storing carbon at the source is another approach Stanford, Caltech, Purdue and Nvidia are exploring to reduce carbon dioxide levels in the atmosphere. AI can model carbon sequestration 60,000 times faster than traditional models, and it can simulate and control the pressure buildup over several years to prevent leakage of CO2. Accelerating those simulations can help to find the optimal injection rates, eventually reducing costs without exceeding safe pressure limits.
It's key initiatives like this that can make a real difference in the fight to solve climate change. By utilizing and engaging the minds of top talent around the world -- to find outside-the-box solutions.
We can work together to solve climate change
During this era of accelerated digitization, barriers to knowledge have been removed for learning communities. As the world has moved online, it's taken knowledge with it. Online conferences, webinars, workshops and communities are available to a wider audience, so people around the world can come together to learn from each other.
Through this accelerated knowledge-sharing, more companies are creating AI solutions and working together with enthusiastic developers to create a solution to climate change.
When people think of artificial intelligence, they usually think of their smartphone's voice assistant. But AI is much more than that. It may have become famous for algorithms that recommend what to watch next, navigating the fastest route on a journey or suggesting message responses, but forward-thinking teams are already creating one of its most noble applications - saving the planet.
You may also like
/news/4339576/wall-street-suffers-tesla-google-stocks-falter
Finance and Reporting
Wall Street suffers as Tesla and Google stocks falter
Nasdaq Composite and S&P 500 drop significantly
/news/4338523/tatas-uk-gigafactory-project-takes-major-step-forward
Components
Tata's UK gigafactory project takes major step forward
Sir Robert McAlpine to build multi-billion-pound factory
/news/4334709/softbank-acquires-british-ai-chipmaker-graphcore
Mergers and Acquisitions
SoftBank acquires British AI chipmaker Graphcore
The deal provides Graphcore with resources to scale its operations