After spending almost a decade working in computer science and artificial intelligence (AI), Sasha Luccioni was ready to uproot her whole life three years ago after she became deeply concerned by the climate crisis.
But her partner convinced her to not give up her career completely but instead apply her knowledge of AI to some of the challenges posed by climate change.
"You don't need to quit your job in AI in order to contribute to fighting the climate crisis," she said. "There are ways that almost any AI technique can be applied to different parts of climate change."
She joined the Montreal-based AI research centre Mila and became a founding member of Climate Change AI, an organization of volunteer academics who advocate using AI to solve problems related to climate change.
Do you have a question about climate change and what is being done about it? Send an email to firstname.lastname@example.org.
Luccioni is part of a growing community of researchers in Canada who are using AI in this way.
In 2019, she co-authored a report arguing that machine learning can be a useful tool for mitigating and adapting to the effects of climate change.
Computer scientists define machine learning as a form of artificial intelligence that enables computers to use historical data and statistical methods to make predictions and decisions without having to be programmed to do so.
Common applications of machine learning include predictive text, spam filters, language translation apps, streaming content recommendations, malware and fraud detection and social media algorithms.
Applications for machine learning in climate research include climate forecasting and optimization of electricity, transportation and energy systems, according to the 2019 report.
Preparing for crop diseases
Researchers at the University of Prince Edward Island (UPEI) are using AI modelling to warn farmers about risks to their crops as weather becomes more unpredictable.
"If you have a dry year, you see very little disease, but with a wet year, you can get quite a bit of disease around plants," said Aitazaz Farooque, interim associate dean of UPEI's School of Climate Change and Adaptation.
Researchers can plug weather data from previous years into an AI model to predict the type of diseases that might jeopardize crops at different times of the year, said Farooque.
"Then the grower can be a bit proactive and have an understanding of what they're getting into," he said.
WATCH | Take a look at UPEI's School of Climate Change and Adaptation:
PEI's agriculture is mostly rain fed, and providing farmers with more accurate rainfall predictions can also help them have more successful crop yields, said Farooque.
"With climate change, we are seeing different trends where the total cumulative rainfall doesn't change much, but the timing matters," he said.
"If it doesn't happen at the right time, then the sustainability of our agriculture can be at risk."
Studying behaviour around disruptive weather
Another application of AI is being studied at McGill University, where researchers are using historical and recent weather data to predict the social impacts of extreme weather events that are being affected by climate change, such as heat waves, droughts and floods.
According to Renee Sieber, an associate professor in McGill's geography department, researchers are hoping to find out how people responded to disruptive weather events in the past and whether that can teach us anything about how resilient we will be in the future.
The team will use a form of AI called natural language processing to analyze social narratives related to weather events in newspapers and other media.
"The AI is very good for organizing, synthesizing, finding trends or some sentiment out of vast amounts of unstructured text," said Sieber.
"Basically, what you do is throw journal articles into a bucket, and you see what comes out."
Sieber said her team will take the findings from past articles and today's social media and compare them with corresponding weather records to identify people's responses to weather events over time.
Records from the McGill Observatory are the longest and most detailed uninterrupted written records of weather patterns in Canada and contain a massive amount of information, said Sieber. Weather recording there began in 1863 and continued into the 1950s.
"This data is the only direct measure of climate change that we have [in Canada]," said Sieber.
Optimizing energy use
Some Canadian companies are using AI to minimize waste and build more energy efficient infrastructure.
Scale AI, a Montreal-based investors group that funds projects related to supply chains, has worked with grocery chains such as Loblaws and Save-on-Foods to identifying purchasing patterns. Through AI, companies are able to better predict demand and less food items are going to waste, said Scale AI CEO Julien Billot.
"Every optimization we can achieve improves the resilience of supply chains and contributes to the use of less resources," she said.
Another Montreal company, BrainBox Al, is focused on improving energy efficiency by optimizing HVAC systems in commercial buildings.
The machine-learning technology is contained in a 30 cm wide box that connects to a building's HVAC system. It raises or lowers temperatures based on data inputs such as weather forecasts, utility prices and carbon-emission calculations.
The system has been able to cut energy consumed by some HVAC systems by 25 per cent, BrainBox CEO Sam Ramadori said, and over two years, the company has installed the technology in 350 buildings in 18 countries.
"The same kind of intelligence that we are bringing to buildings has probably an infinite number of applications. Just pick a sector," Ramadori said.
"How we make cement, how we ship goods — all of those need to be made more efficient over time as part of the climate change fight."
According to Ramadori, BrainBox AI is working on technology that will allow buildings to link up with each other and communicate with energy grids through the company's cloud server.
This has the potential to minimize wasted energy on a city-wide scale as energy grids more accurately detect where and when power is needed, he said.
"The utility grid can say, 'Hey, the next two hours are going to be busy. I need you to find a way we can reduce consumption.' And with the AI brain up top, it's able to say, 'OK, I can reduce a bit here and a bit there. I've got you covered,'" said Ramadori.
Equity limitations to AI
Access to the kind of AI that can help solve climate-related problems is not equal across the globe.
Forest fires in North America, for example, tend to receive more attention from developers than locust infestations in East Africa, said David Rolnick, an assistant professor of computer science at McGill and a member of Mila.
"The way in which climate change impacts a community varies greatly between different geographies," said Rolnick, who is also the chair of Climate Change AI.
AI technology relies on data sets, and many communities do not have access to enough of the kind of robust data needed to create machine-learning algorithms, Rolnick said.
In Canada, some Indigenous and remote northern communities still face significant digital divides compared with other parts of the country, he said.
"Working on democratizing that is fundamentally important," Rolnick said.
Rolnick co-authored a study last year outlining various limitations to implementing AI for climate change solutions in Canada. It called for increased funding for AI research and more AI education in primary and secondary education as well as standards and protocols for data sharing related to climate projects.
Rapidly implementing large-scale AI literacy programs for policymakers and leaders in climate-relevant industries could help "demystify" AI, the report said.
"We often see a lack of relevant knowledge, and educational programs can help people understand what these tools can and cannot do," said Rolnick.