New study uncovers surprising physics of ‘marine snow’

The deep ocean can often look like a real-life snow globe. As organic particles from plant and animal matter on the surface sink downward, they combine with dust and other material to create “marine snow,” a beautiful display of ocean weather that plays a crucial role in cycling carbon and other nutrients through the world’s oceans.

Now, researchers from Brown University and the University of North Carolina at Chapel Hill have found surprising new insights into how particles sink in stratified fluids like oceans, where the density of the fluid changes with depth. In a study published in Proceedings of the National Academy of Sciences, they show that the speed at which particles sink is determined not only by resistive drag forces from the fluid, but by the rate at which they can absorb salt relative to their volume.

“It basically means that smaller particles can sink faster than bigger ones,” said Robert Hunt, a postdoctoral researcher in Brown’s School of Engineering who led the work. “That’s exactly the opposite of what you’d expect in a fluid that has uniform density.”

The researchers hope the new insights could aid in understanding the ocean nutrient cycle, as well as the settling of other porous particulates including microplastics.

“We ended up with a pretty simple formula where you can plug in estimates for different parameters—the size of the particles or speed at which the liquid density changes—and get reasonable estimates of the sinking speed,” said Daniel Harris, an associate professor of engineering at Brown who oversaw the work. “There’s value in having predictive power that’s readily accessible.”

The study grew out of prior work by Hunt and Harris investigating neutrally buoyant particles—those that sink to a certain depth and then stop. Hunt noticed some strange behavior that seemed to be related to the porosity of the particles.

“We were testing a theory under the assumption that these particles would remain neutrally buoyant,” Hunt said. “But when we observed them, they kept sinking, which was actually kind of frustrating.”

That led to a new theoretical model of how porosity—specifically, the ability to absorb salt—would affect the rate at which they sunk. The model predicts that the more salt a particle can absorb relative to its size, the faster it sinks. That means, somewhat counterintuitively, that small porous particles sink faster than larger ones.

To test the model, the researchers developed a way to make a linearly stratified body of water in which the density of the liquid increased gradually with depth. To do that, they fed a large tub with water sourced from two smaller tubs, one with fresh water and the other with salt water. Controllable pumps from each tub enabled them to carefully control the density profile of the larger tub.

Using 3D-printed molds, the team then created particles of varying shapes and sizes made from agar, a gelatinous material derived from seaweed. Cameras imaged individual particles as they sank.

The experiments confirmed the predictions of the model. For spherical particles, smaller ones tended to sink faster. For thinner or flatter particles, their settling speed was primarily determined by their smallest dimension. That means that elongated particles actually sink faster than spherical ones of the same volume.

The results are surprising, the researchers said, and could provide important insights into how particles settle in more complex ecological settings—either for understanding natural carbon cycling or for engineering ways of speeding up carbon capture in large bodies of water.

“We’re not trying to replicate full oceanic conditions,” Harris said. “The approach in our lab is to boil things down to their simplest form and think about the fundamental physics involved in these complex phenomena. Then we can work back and forth with people measuring these things in the field to understand where these fundamentals are relevant.”

Harris says he hopes to connect with oceanographers and climate scientists to see what insights these new findings might provide.

Other co-authors of the research were Roberto Camassa and Richard McLaughlin from UNC Chapel Hill.