One of the key challenges in facing a rapidly-changing climate is accurately predicting how global warming will affect individual regions around the planet: one area may suddenly be stricken with prolonged drought, while another may be inundated with catastrophic flooding. Keeping ahead of potentially disastrous conditions that can lead to situations such as these will be required for policy makers and emergency planners if they are to save lives and livelihoods, but our current climate models still appear to be inadequate in providing the fine details needed. The solution: according to a professor at Columbia University, the secret to our survival lies in the branch of artificial intelligence called machine learning.
Pierre Gentine, an associate professor of earth and environmental engineering at Columbia Engineering, has published a paper that demonstrates that machine learning can be used to improve the resolution of models of cloud formation down to roughly 100 kilometers (62 miles), with the potential to improve the results even further as the process is refined.
"This could be a real game-changer for climate prediction," explains Gentine. "We have large uncertainties in our prediction of the response of the Earth’s climate to rising greenhouse gas concentrations. The primary reason is the representation of clouds and how they respond to a change in those gases. Our study shows that machine-learning techniques help us better represent clouds and thus better predict global and regional climate’s response to rising greenhouse gas concentrations."
Using a deep neural network called the Cloud Brain (CBRAIN), they trained the AI to learn from an existing simulation that specifically focused on cloud formation. CBRAIN was able to skillfully predict many of the aspects of cloud mechanics essential to climate simulation, including cloud heating, moistening, and radiative features.
"Our approach may open up a new possibility for a future of model representation in climate models, which are data driven and are built ‘top-down,’ that is, by learning the salient features of the processes we are trying to represent," according to Gentine. The paper also notes that given the intrinsic link between global temperature sensitivity to CO2 and cloud formation, even CBRAIN’s demonstrative model itself may be able to improve future temperature predictions.
In regards to machine intelligence, the Master of the Key said that we "cannot survive without it. An intelligent machine will be an essential tool when rapid climate fluctuation sets in. Your survival will depend on predictive modeling more accurate than your intelligence, given the damage it has sustained, can achieve." You can read Whitley’s account of his encounter with this insightful and enigmatic individual in The Key, available through Amazon, Barnes & Noble, Kobo, and Audible.