After nearly a decade of work in computer science and artificial intelligence (AI), three years ago Sasha Luccioni became deeply concerned about the climate crisis and was ready to uproot her life.
But rather than abandon her career altogether, her partner persuaded her to apply her AI knowledge to some of the challenges posed by climate change.
“You don’t have to quit your AI job to help fight 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 center Mila and became a founding member of Climate Change AI. Climate Change AI is an organization of volunteer academics who advocate using AI to solve problems related to climate change.
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 could be a useful tool for mitigating and adapting to the impacts of climate change.
Computer scientists define machine learning as a form of artificial intelligence. This allows computers to use historical data and statistical techniques to make predictions and decisions without programming.
Common applications of machine learning include predictive text, spam filters, language translation apps, streaming content recommendations, malware and fraud detection, and social media algorithms.
According to a 2019 report, applications of machine learning in climate research include climate prediction and optimization of power, transportation and energy systems.
Prepare for crop diseases
Researchers at the University of Prince Edward Island (UPEI) are using AI modeling to warn farmers about risks to their crops as weather becomes more unpredictable.
Aitazaz Farooque, interim vice-chancellor of UPEI’s Department of Climate Change Adaptation, said: “In dry years we see very few diseases, but in wet years there can be considerable disease outbreaks around plants. there is.
Researchers can plug previous year’s weather data into AI models to predict the types of diseases that could endanger crops at different times of the year, Farooque said.
“That way the growers can be a little more proactive and understand what they’re trying to do,” he said.
Watch | Visit UPEI’s School of Climate Change and Adaptation.
Farming in PEI is primarily rain-fed, and providing farmers with more accurate rainfall forecasts can increase yields, Farooque said.
“Climate change has seen various trends in cumulative precipitation totals that don’t change much, but timing is key,” he said.
“If it doesn’t happen at the right time, the sustainability of agriculture could be at risk.”
Behavioral research related to extreme weather
Another application of AI is being researched at McGill University, where researchers are using historical and recent weather data to predict the societal impact of extreme weather events impacted by climate change, such as heat waves, droughts and floods. is predicted.
According to Renee Sieber, an associate professor of geography at McGill, researchers are finding out how people have responded to destructive weather events in the past, and what that can tell us about future resilience. I would like to know how.
The team uses a form of AI called natural language processing to analyze social narratives related to weather events in newspapers and other media.
“AI is very good at organizing, synthesizing, and discovering trends and sentiment from vast amounts of unstructured text,” Sieber said.
“Basically, we throw magazine articles into a bucket and see what comes out.”
Sieber’s team says it takes findings from past articles and today’s social media and compares them to corresponding weather records to identify people’s reactions to weather events over time.
The record from the McGill Observatory is the longest and most detailed continuous record of weather patterns in Canada and contains an enormous amount of information, Sieber said. Weather recording there he began in 1863 and continued until the 1950s.
“This data is the only direct measure of climate change that we have [in Canada]said Seaver.
Optimizing energy use
Some Canadian companies are using AI to minimize waste and build more energy efficient infrastructure.
Scale AI, a Montreal-based investor group that funds projects related to supply chains, has worked with grocery chains like Loblaws and Save-on-Foods to identify buying patterns. Scale AI CEO Julien Billot says businesses can more accurately predict demand and reduce food waste through his AI.
“All the optimizations we can achieve will improve the resilience of our supply chain and contribute to the use of fewer resources,” she said.
Another Montreal company, BrainBox Al, focuses on improving energy efficiency by optimizing HVAC systems in commercial buildings.
The machine learning technology is housed in a 30 cm wide box that connects to the building’s HVAC system. Raise or lower temperatures based on data inputs such as weather forecasts, utility bills, and carbon footprint calculations.
According to BrainBox CEO Sam Ramadori, the system can reduce the energy consumed by some HVAC systems by 25%, and the company has installed the technology in 350 buildings in 18 countries in two years. Did.
“There are probably endless uses for the same kind of intelligence we bring to buildings. Pick a sector,” said Ramadri.
“The way we make cement, the way we ship goods, all of these things need to become more efficient over time as part of the fight against climate change.”
BrainBox AI is working on technology that will allow buildings to link together and communicate with the energy grid through the company’s cloud servers, Ramadori said.
This could help minimize wasted energy on a city scale by allowing the energy grid to detect more precisely where and when electricity is needed, he said.
“The power grid can say, ‘It’s going to be busy for the next two hours. Find a way to reduce consumption.'” You can say, ‘I’ve got you covered,'” said Ramadori.
Fairness restrictions on AI
Access to the kinds of AI that can help solve climate-related problems is uneven across the globe.
Wildfires in North America, for example, tend to attract more developer attention than locust infestations in East Africa, says David Rolnick, assistant professor of computer science at McGill and member of Mila.
“The impact of climate change on communities will vary greatly from region to region,” said Rolnick, who also chairs Climate Change AI.
AI technology relies on datasets, and many communities lack access to the kind of robust data they need to create machine learning algorithms, Rolnick said.
In Canada, some indigenous and remote northern communities still face a significant digital divide compared to the rest of the country, he said.
“Working on democratization is fundamentally important,” Rolnick said.
Last year Rolnick co-authored a study outlining the various limitations of implementing AI in climate change solutions in Canada. It called for increased funding for AI research and increased her AI education in primary and secondary education, as well as standards and protocols for data sharing related to climate projects.
Rapid implementation of large-scale AI literacy programs for policy makers and leaders in climate-related industries could help “demystify” AI, says the report.
“Educational programs help people understand what these tools can and cannot do, although there is often a lack of relevant knowledge,” says Rolnick.
Canadian researchers use machine learning to mitigate climate change impacts
Source link Canadian researchers use machine learning to mitigate climate change impacts