Q&A: How AI can help combat climate change

Climate change is a global emergency challenging scientists, engineers, and industry experts from a wide array of disciplines to use their knowledge and skills in pursuit of solutions to protect our planet.

Not surprisingly, some of those solutions are likely to be made possible by artificial intelligence.

“Climate data sets are enormous and take significant time to collect, analyze, and use to make informed decisions and enact actual policy change,” says Jim Bellingham, a pioneer in autonomous underwater robotics systems and executive director of the Johns Hopkins Institute for Assured Autonomy. “Using AI to factor in elements of climate change that are constantly evolving helps us make more informed predictions about changes in the environment, so that we can deploy mitigation efforts earlier.”

On Wednesday, March 15, Bellingham and Elizabeth Reilly, a senior staff scientist at JHU’s Applied Physics Laboratory, will co-lead a discussion on combatting climate change with AI at the South by Southwest Conference in Austin, Texas.

Bellingham—a professor in the Whiting School of Engineering’s Department of Mechanical Engineering and in the Asymmetric Operations Sector at APL—spoke with the Hub about how AI is being applied to the issue of climate change.

How is AI being used to address climate change?

The interesting aspect of AI is that it applies to so many things we do, including tasks that were previously activities only humans could accomplish. Climate change is one of the most difficult scientific problems that humans have ever faced. It’s a phenomenally complex system with an enormous number of variables. When people talk about climate change, they tend to focus on the physical aspects of climate, such as the amount of carbon dioxide in the atmosphere, temperatures, precipitation levels, and wind patterns. But these characteristics are all shaped by a living planet that is constantly changing. If you took life off planet Earth, it would have a very different environment.

Climate data sets are enormous and take significant time to collect, analyze, and use to make informed decisions and enact actual policy change. Using AI to factor in elements of climate change that are constantly evolving helps us make more informed predictions about changes in the environment, so that we can deploy mitigation efforts earlier.

Another use for AI is helping to reduce the carbon that is released into the atmosphere, and AI feeds into the entire chain of activities related to transitioning from a carbon-based economy to a net carbon zero economy. The ability to build larger and larger windmills depends on having materials that are exceptionally light and extraordinarily strong but that can sustain extreme weather conditions. We’re beginning to learn how to use AI to assist us in the design and creation of those materials.

AI in combination with other trends, such as the electrification of transportation, additive manufacturing, transformations in agriculture, and smart electrical grids, is a powerful enabling capability for more energy efficient (and cost saving) solutions. What is the energy benefit when we replace the delivery van visiting every house in a neighborhood as compared to a delivery van parking on the edge of the neighborhood and launching a drone for that last mile delivery?

Beginning to connect data sets from satellites and observations with model predictions is a very important part of the enterprise that ensures that we see all parts of what is going on in the environment. AI can help us be surprised less.

Why is AI technology well suited for applications related to climate change?

Artificial intelligence combines predictions based on trends and patterns with the extensive data collected. Models are at the core of prediction, but to rely on these models to make decisions, people must trust the models. Scientists are asking “what if” questions, and policymakers are weighing costs and benefits based on data collected and analyzed by researchers. AI is one tool that provides insights into where uncertainties come from related to climate change and that can help us understand what the models are telling us, which can feed back into better observation programs, improving the models, and even using AI as part of the model system. Assurance, or trust, is a key aspect of using AI.

What are some specific applications of AI to climate change?

My work, which focuses on better understanding oceans, along with the work my colleagues at APL are doing in space, have very tangible applications and outcomes. The ocean is one of the least understood parts of the Earth, as it both transfers and absorbs heat. We don’t know exactly how the ocean responds to various environmental changes, and we also don’t know specifically how climate change is impacting our oceans.

Another application of AI to climate change is how satellites orbiting in space are used to make observations and assess changes to the Earth. Satellites can help monitor forest fires and determine potential sources of carbon dioxide that are found in the environment. However, as the number of satellites in orbit increases, it’s important to make sure that space can safely hold and maintain all the satellites collecting information.

For example, the Arctic is changing rapidly and dramatically, with increasing temperatures being regularly documented. During spring, summer, and fall, ships in the Arctic collect data and other essential information that is used by scientists and policymakers. But in the winter, ice conditions in that region make it challenging for ship operations, so the ships depart, creating a significant observation and data collection gap. Using AI-powered robots in this scenario allows information to continue to be accumulated with the technology providing predictions based on trends and patterns. This would allow information to continue to be accumulated, with the technology providing predictions based on trends and patterns.

We would like robots to be able to operate on their own for up to six months in the Arctic Basin to make observations and then provide us with that data. Part of it is filling in the observation system, and another part of it is combining these various cues and allowing AI to help us see patterns we recognize in larger data sets.

What are some challenges that experts are facing when using AI to address climate change?

One of the challenges we have in really understanding climate is to begin to truly comprehend the complexities of the living part of our ecosystem, particularly our oceans. The oceans present a number of difficulties, including the fact that it is prohibitively expensive to deploy and maintain the number of ships we need to observe the ocean and collect needed data. Robots are being increasingly used for this purpose, but their autonomous capabilities need to be improved. This is where AI comes into play.

The additional oversight and prediction that AI provides to researchers is valuable, but there are expenses that need to be considered to assess the true benefits in terms of climate change work. One example is that AI relies on computers, and computers need electrical power to function, and electricity uses resources. Scientists and researchers must keep the use of electricity that is used to power AI technology in mind when assessing how beneficial the technology is in addressing climate change. On average, each new generation of processor carries out more computations for less power, but AI demands for computation are fueling an explosion of investment in computational power. The AI technology we have today is due in part to the enormous computational power we have.

What do you predict will be AI’s biggest impact on combatting climate change in the next five to 10 years?

My hope for the future of AI is that we will be able to have a meaningful impact on predicting climate change. As humans become more confident in AI, we will be able to rely on technology more to understand climate change and to make more accurate predictions and models. This will allow us to be more targeted in our strategies to mitigate the worst effects. Assurance in autonomy and AI is one area that needs to be taken very seriously. Even if we don’t think something is an AI problem today, it will be an AI problem next week or next month.