Hemelrijk, CK, DAP Reid, H Hildenbrandt, and JT Paddling. 2015. Increased swimming efficiency of fish swimming in a school. Fish and Fisheries, 16:511-521. doi: 10.1111/faf.12072
Some species of fish swim together in large social groups called schools for several reasons. Sticking together helps fish avoid predators by relying on a team of eyes that could potentially detect a predator and then signal to the rest of the group that a predator is nearby. The signal could be as simple as the fish that first sees the predator turns and suddenly swims in another direction. The other fish in the group can notice this and follow suit, even if they did not see the predator themselves. There is also likely to be hydrodynamic benefits to swimming in schools. This means that fish swimming together in a group do not have to work as hard to go the same speed as a fish swimming by itself. The concept here is similar to drafting in racing – racecar drivers and cyclists both follow each other in straight lines because the one in front breaks up the air, making it easier for the ones following to go faster.
Studying the Hydrodynamics of Schooling
Several scientists have proposed theories about the hydrodynamic benefits of swimming in a school since the 1970’s. One of the first direct theories of the hydrodynamic benefits to schooling was proposed by Daniel Weihs in an article in Nature in 1973, which predicted that swimming in a close diamond pattern is the most efficient swimming configuration for fish. However, the energetic benefits to swimming in a school has been a bit of a controversial topic with other scientists refuting the idea of hydrodynamic benefits to swimming in schools. A likely factor in the controversy is the fact that studying the way water behaves is incredibly difficult. In a new study published in Fish and Fisheries, Hemelrijk et al. took advantage of modern advances in computing power to test several of the existing theories of the hydrodynamic benefits to swimming in a school, including Weihs’ theory, using a technology called MultiParticle Collision Dynamics. MultiParticle Collision Dynamics looks at water as a collection of millions of tiny particles, simultaneously tracking the movement of each individual particle. This technology allows researchers to see the way water interacts and changes as groups of swimming fish move through it.
Thicklip grey mullets were examined in previous studies to generate computer models of the way these fish swim (Figure 1). This study utilized computer simulated fish modeled after 126mm mullets. Thicklip grey mullets were chosen in part because of the existing models of swimming performance, but also because it is considered an obligate schooler, meaning they must be in groups, and because of the way it undulates its body when swimming at cruising speed.
What They Measured
The researchers tested four different schooling configurations: A) Diamond, B) Rectangle, C) Phalanx (aka side-by-side line), and D) Line (aka front-to-back line) (Figure 2). To determine if fish gain a benefit in swimming efficiency from schooling, the researchers calculated the percentage of power exerted by the fish that was converted into forward speed, called the Froude efficiency. Measuring Froude efficiency requires measuring the useful force exerted by fish (forward thrust), separated from wasted sideways power (from moving the tail sideways), and of course the forward speed of the fish.
Is it energetically beneficial for fish to swim in schools? Can they swim faster more easily when in a group?
Hemelrijk et al. predicted that fish swimming in a single file line would be less efficient than a lone fish because they are swimming in the thrust wake of the fish in front of them, receiving the jet from that fish on their nose. Instead they found that fish can move their heads side to side to catch the wake in an advantageous position, resulting in increased efficiency! The oncoming wake can be directed to one side of the body, generating lift much like an airplane wing. This is done on alternating sides of the body, which helps to generate forward thrust for fish swimming in a single file line (see red and blue vortices in video clip below!)
Another surprising outcome from this experiment was that fish swimming close together in a diamond pattern (0.4 body lengths apart) is less efficient than when there is a bit more space between them (Figure 4), which refutes Weihs’ original theory in 1973 that a close diamond pattern would be the most efficient configuration. Instead, the diamond pattern becomes much more efficient once the fish are more than 0.4 body lengths apart from each other (Figure 4). Weihs’ theory did not include viscosity, deformable body shape, and interactions between an individual and wake, which may explain why it did not accurately predict the results from this experiment which did include these factors.
Nearly every configuration of swimming in a school is more efficient than swimming alone (all of the configurations in Figure 4 are above the Froude efficiency line of a lone fish). Only the densest phalanx configuration (side-by-side line) was less efficient than swimming alone. A phalanx with fish very close to each other is the least efficient swimming configuration because of the increased resistance of the oncoming flow over the closely grouped fish (effectively becoming one, very broad fish with a lot of drag as they push forward together through the water). Once there is enough space between the fish in a phalanx to allow water to more easily pass between the fish, swimming in this configuration is again more efficient than swimming alone. Also, as expected, the rectangular configuration is more efficient than the line and the phalanx because individuals benefit from neighbors in both directions.
Hemelrijk et al.’s experiments utilized state-of-the-art technology to conduct some of the most thorough tests of the hydrodynamic efficiency of schooling fish. The surprising findings from this study will help reshape fish biologists’ understanding of the behavior of schooling fish, as well as the energetic consequences of their behavior. Furthermore, better understanding what factors increase the efficiency with which power and thrust are converted into forward speed (which configurations are best and at what densities) may help in improving robot systems that explore the ocean floor.