Exploring Clouds' Perspectives Dually Through AI in Climate Research
We're dealing with a intricate issue regarding climate change, and it requires an exceptionally potent computer to truly model the Earth System and its future with fine detail. I recently came across Mike Pritchard, a research manager at Nvidia and a professor at UC Irvine, discussing this topic.
He pointed out that "The physics span 10 orders of magnitude in space and time." For instance, figuring out if a cloud particle tends to attract water vapor is one of the complexities involved.
If you want to simulate the planet numerous times, to explore various 'what if' scenarios of the future, you unfortunately, despite having the most powerful super computers, can't fully account for the complexity involved," he stated. "Our questions about the future climate are too broad for simulation technology."
As an illustration, he spoke about observing a certain kind of cloud on his commute between San Diego and Irvine.
"It looks like a gray band on the horizon," he said. "We call it the marine layer. If it comes close on a beach day, it's disappointing because it makes you feel cold. But what matters is that it's the edge of a big sheet of low clouds you'll see out of the window from an airplane, halfway through a flight from San Diego to Hawaii, and that cloud reflects a lot of energy from the planet, keeping it cooler than it would be otherwise. So if it disappears, that will intensify global warming...but if it thickens up, that will mitigate it. And that's a multi-trillion-dollar uncertainty. And it's a simulation problem. We know these clouds take high resolution to simulate, but we can't afford that in climate simulation yet."
A Flurry of Systems
Pritchard also mentioned the term 'ensemble,' which has a unique meaning in weather prediction.
"You don’t predict one hurricane," he said, "you predict hundreds of hurricanes. You hope for the best, and plan for the worst...card-carrying atmosphere scientists at the University of Washington are using AI weather models, which were trained on the chaotic and complex real atmosphere, and then after the fact, testing them to see if they've learned physics."
He discussed how this works with technology and building an archive of evidence for AI's ability to aid in weather prediction.
Development of New Software and Technology
As an example, Pritchard referred to the capabilities of Nvidia AI tools such as Modulus, Earth2Studio which enable the research, development, and validation of AI Forecast Models.
The company, which has become a leader on the U.S. stock market, has many research projects and collaborations with the atmospheric science community. These are released in the open source domain and include prominent models like:
StormCast research - demonstrating a generative AI model that mimics atmospheric dynamics, examining mesoscale weather phenomena and making predictions.
CorrDiff - another generative AI model that produces high-resolution weather forecasts. You can learn more about AI-powered downscaling using pretrained Corrdiff here.
FourCastNet - this model achieves 25-km -resolution weather forecasting for various parts of the world using Spherical Fourier Neural Operators, recently calibrated for huge ensembles. You can learn more about medium range global forecasting using pretrained Fourcastnet here.
Earth-2 Platform is a digital twin cloud platform that helps enterprises leverage these AI advances and speed up traditional numerical simulations to reduce the computational bottleneck of climate and weather simulations. Combining these advances with advances in computer graphics like RTX rendering technology, we can create digital twins of Earth's climate and weather to help scientists explore, analyze, and explain complex weather patterns, especially in the context of changing climate.
More on Climate Works
Pritchard also discussed promoting optimal dispersion in large ensemble AI weather predictions and referred to new papers that provide more detail about the emerging science of using AI to simulate low-likelihood, high-impact climate extremes. This, he said, will provide climate risk modelers with new tools to help us understand and protect against extreme weather events.
Moving Forward and Backward
Another aspect of Pritchard's discussion involved the usefulness of AI models. He described traditional climate informatics processes as like "going to an Oracle" – large simulators create large data sets, users then have to mine them to help inform 'what if' scenarios and questions about future climate. AI predictions, he added, can run forwards and backwards, which should help users more easily figure out what could have changed, given a different initial input.
"We might be entering a future where we can understand our influence on the future more easily, without having to experience all the bottlenecks of conventional simulation," he said.
The Power of Twins
In conclusion, Pritchard also talked about the idea of digital twinning applied to the biggest single item we have in our world – the world itself.
"I think that the really important paradigms of interactivity are chains and cascades of AI digital twins, ... So you can imagine a future that's evolving towards AI digital twins of the climate, coupled to AI digital twins of extreme weather events."
With research trends and everyone's actions serving as our guide, Pritchard provides some thought-provoking ideas on tackling our current climate predicament using technology that surpasses simple big data sets. Keep an eye out for more insights gained from recent AI advancements and their impact on our Earth, right here in Boston.
Utilizing advanced computer technologies, researchers can harness the power of AI to create digital twins of Earth's climate and weather. For instance, Nvidia's tools like Modulus and Earth2Studio enable the development and validation of AI forecast models, such as StormCast research and FourCastNet, which contribute to improving weather prediction and climate simulation.
Furthermore, the media can play a crucial role in promoting an understanding of these technological advancements and their implications on climate change. By reporting on groundbreaking research, scientists like Mike Pritchard can ensure that their findings reach a broader audience, helping to inspire action and drive progress in the fight against climate change.
