Week 1 | What Can Machine Learning Do for Climate Change?
Climate change is one of the biggest challenges facing our world today. Rising global temperatures, shifting weather patterns, and more extreme weather events are already impacting communities worldwide. Machine learning, a type of artificial intelligence, offers new hope in the fight against climate change.
Published on September 11, 2023 by Bronte Sihan Li
climate change machine learning AI
1 min READ
Climate change is one of the biggest challenges facing our world today. Rising global temperatures, shifting weather patterns, and more extreme weather events are already impacting communities worldwide. Machine learning, a type of artificial intelligence, offers new hope in the fight against climate change.
Machine learning algorithms can analyze massive datasets quickly to uncover hidden insights. This makes them uniquely suited to tackling the complexity of climate science, incorporating many different factors in making better predictions for future changes. In the exploration of how machine learning can be applied to tackle a specific problem, there are three main questions to consider:
- What is the problem we want to solve / needs to be solved?
- What is the data available / what kinds of data should be applied?
- What is the solution?
One way machine learning is being applied is in predicting and managing wildfires. Wildfires are increasing in frequency and severity due to hotter, drier conditions. Computer vision algorithms can analyze satellite imagery to identify areas at highest risk of wildfires. This allows fire officials to proactively allocate resources and better plan controlled burns. For example, we can estimate the spread of the fire in the short term, which helps authorities safely evacuate residents in the fire’s path. After fires, machine learning can help assess damage to man-made structures and environmental assets. This information guides recovery and restoration efforts. Machine learning can even identify ways to improve infrastructure design to be more resilient to future fires.
My current project focuses on developing machine learning models for wildfire modeling here in North America. By training algorithms on satellite imagery and data like vegetation, wind direction, etc., we hope to better understand wildfire behavior in the short term, and its impact on marginalized communities in the long term. This will arm officials with science-based, insights to help our community adapt to our hotter, drier future. Machine learning offers data-driven solutions to curb the impacts of climate change and help mitigate risks to the most vulnerable populations.