Simulating fluid dynamics computationally is no simple task, but clever techniques help researchers do so more efficiently.  (Video and image credit: N. Wimer et al.)

The moisture in clouds eventually condenses into droplets that grow into raindrops and fall. Some steps in this process are well understood, but others are not. In particular, scientists have struggled with the problem of how droplets grow from about 30 microns to 80 microns, where they’re big enough to start falling and merging. 

Laboratory experiments and numerical simulations (below) have shown that turbulence can help drive small water drops together. When droplets are tiny and light, they simply follow the air flow. But when they’re a little heavier, turbulent eddies (seen in orange below) act like miniature centrifuges, flinging larger water droplets (shown in cyan below) out into clusters, where they’re more likely to collide with one another.

Although this effect has been seen in experiments and simulation, it’s been difficult to capture in clouds themselves. But a new set of test flights (above) confirms that this mechanism is present in the wild as well! (Image credit: UCAR/NCAR Earth Observing Laboratory, P. Ireland et al., source; research credits: M. Larsen et al., P. Ireland et al.; via APS Physics; submitted by Kam-Yung Soh)

Scientists have long wondered whether the schooling of fish is driven by hydrodynamic benefits, but the complexity of their environment makes unraveling this complex motion difficult. A recent study uses a different tactic, combining direct numerical simulation of the fluid dynamics with techniques from artificial intelligence and machine learning to build and train autonomous, smart swimmers.

The authors use a technique called deep reinforcement learning to train the swimmers. Essentially, the swimmer being trained is able to observe a few variables, like its relative position to the lead swimmer and what its own last several actions have been -- similar to the observations a real fish could make. During training, the lead swimmer keeps a steady pace and position, and the follower, through trial and error, learns how to follow the leader in such a way that it maximizes its reward. That reward is set by the researchers; in this case, one set of fish was rewarded for keeping a set distance from their leader, one intended to keep them in a position that was usually beneficial hydrodynamically. Another set of fish was rewarded for finding the most energy-efficient method for following.