The Learning Earth for Artificial Intelligence and Physics (LEAP) Center launches in 2021. What is the core mission of the center?
LEAP’s mission is to improve the reliability, usefulness, and range of climate predictions by integrating climate and data science. Our main strategy is to improve near-term climate predictions by combining physical modeling with machine learning, from expertise in climate science and climate modeling to cutting-edge machine learning algorithms. This will really help the climate and data science community; climate scientists and modelers struggle to fully integrate the plethora of existing datasets into their models, and while machine learning algorithms such as ChatGPT have been good at simulations, they are not very good at extrapolating or predicting extremes Condition.By combining these two methods, we hope leap Will lead to major advances in data science algorithms applied to physics problems. The center is incorporating physics and causal mechanisms into machine learning algorithms to enable better generalization and extrapolation, while optimizing the use of the vast amounts of data available in climate science to better predict the future.
Can you talk about some of the key projects that LEAP has been focusing on? What breakthrough?
LEAP works on different aspects of climate science and data science, covering not only research but also knowledge transfer and education. Some recent research breakthroughs at our center have shown that by using artificial intelligence we can discover new physics (clouds or ocean turbulence) that were previously unknown. We hope to use this “new” physics in climate models to improve their accuracy, especially at extremes. Essentially, we’re using massive amounts of data, such as from satellites, so that we can improve climate models and their assessments and improve our forecasts. It is also important that we share this information with the public and private sectors in a user-friendly manner.
We also want to break down some of the historical silos with climate research – climate research is not easily translated to the public or private sector because it is highly technical and difficult to use. We’re creating a cloud platform where we can make climate data more widely available and engage with colleagues in and out of the field to see what’s really useful to them. For example, businesses will want to know how much the frequency of future floods or heat waves will change so they can adjust their operations. With LEAP, we hope to improve our models so that we can help provide more precise predictions.
AI has received a lot of attention, and Currently applying AI In your climate research and beyond. How AI is revolutionizing climate modeling, and where is it today?
The use of artificial intelligence has exploded over the past five years to better understand climate models and better represent physical processes (such as clouds, ocean and land carbon cycles, or ocean turbulence). The next big push will be how to integrate these AI algorithms into climate models that have historically used empirical equations.
The information we use to predict the future is indeed uncertain. There are many reasons for this, including the complexity of all these processes that we are trained to do when building climate models. So we’re focused on reducing and shrinking those uncertainties, providing everyone from policymakers to business leaders to educators with accurate climate forecasts to inform their own decision-making. In the agricultural sector, for example, being able to provide precise information about future climate could have a serious impact on crop productivity and yields.
The goal is to improve climate models so we can say how many days a drought is likely to last, or how likely flooding is in New York City or any other particular low-lying area. These are very critical questions. Currently, the range of estimates is so wide that actually implementing plans is challenging. We need to act now on this issue before we can actually start tackling climate change.
What excites you most about where the field is headed and how the work at LEAP will impact our future?
We are witnessing a real transformation that is centered around data—and the use of observational and simulated data to answer new hypotheses or hitherto unsolvable questions. Of course, with any new field, we still need to be cautious and make sure the results are reliable and reproducible, but I’m excited to see where the field is headed and witness the incredible pace of progress. I think we are indeed witnessing a revolution in climate science.
at this The Earth Day In 2023, what would you like to tell your children about how we invest in our planet, the importance of climate change research, and what you want to leave for them and future generations?
What I tell my three kids is to be considerate of others. We need to use fewer resources, limit our footprint, recycle and reduce emissions. In other words, we need to mitigate climate change. But we also need to adapt. Climate change is now part of our everyday lives, as evidenced by the recent surge in the number and intensity of extreme events (droughts, floods, etc.). It is clear that we need to change society as a whole. Everyone has to accept this change, so we need broad support (government, policy) so that the whole country can adapt to these changes, not just a small part of our society. Otherwise, we will fail as a society and as a nation. Social justice and climate change are closely linked—we need both to achieve our goals.
