OLI Grant: A Computational Fluid Dynamics Study of Propulsion and Swimming Energetics of Cod Larvae


Grant Funded: 2001

Proposed Research

We propose to use computational fluid dynamics (CFD) to simulate the flow field around a freely swimming cod larva. Our objective is to understand the hydrodynamics, propulsion and swimming energetics associated with different swimming techniques adopted by a freely swimming cod larva in different development stages. This interdisciplinary research will include the observed swimming kinematics of cod larvae into a CFD simulation. Such a simulation will be done for a larva in different stages (i.e., different sizes), and a novel data analysis method will be used to calculate mechanical energy consumption and efficiency by the swimming larval fish.

Regardless of whether predation or starvation is the primary cause of death in young fishes, energy consumption associated with swimming performance may be an important factor affecting survival. Directly calculating the mechanical energy consumption and efficiency for cod larvae in different developmental stages and in various swimming behaviors, such as those employed in food searching, prey attacking and predator avoiding, will be extremely helpful for identifying the cvause of mortality of these young fishes. Thus, the proposed research will benefit to improve the health of certain marine eccosystem, such as that of Georges Bank, which suffers a severe decline in cod population.

The idea of using CFD approach to study the propulsion and swimming energetics of cod larvae is original and the results should be more quantitative. Accomplishment of the proposed research will definitely provide a new tool for ocean biology, particularly for the research in fish propulsion and swimming energetics.

Progress Report

Recent observations have shown that larvae of Atlantic cod (Gadus morhua) change their swimming behavior dramatically during growth in the larval stage (Von Herbing and Gallager, 2000, 2003). Just-hatched yolk-sac larvae (about 2-3 mm long) use continuous, energetic body motions to propel themselves through the water. Their body motions take the form of a wave that travels head to tail, a swimming form termed "the anguilliform swimming" because it is like that of the common eel, Anguilla. In addition, the larvae use the "C-start" motion for take-off from rest, in which their body bends to one side around their center like the letter "C". This mode of swimming lasts for 3-4 days. At the end of this period, the larvae have absorbed their yolk sac and roughly doubled in length. The larvae then switch to a "burst-and-coast" swimming motion in which undulatory body motions are interspersed with coasting behavior, with the body held straight and rigid. During active swimming (or bursting), their body motion is "carangiform swimming" (as in mackerels, family Carangidae). In this form the body maintains a backward-moving wave with a very small or even zero amplitude of undulation in the head region and a large undulation in the immediate neighborhood of the tail. This mode of swimming is retained for the remainder of the fish's life.

The objectives of our research were to:
  • develop a Computational Fluid Dynamics (CFD) simulation framework to compute and visualize the flow field around a swimming cod larva at different developmental stages and with different swimming behaviors;
  • use the CFD simulation results to directly calculate the mechanical energy consumption and energetic efficiency for cod larvae at different developmental stages and with various swimming behaviors;
  • explain the observed differences in swimming modes as a function of age, shape, and size of the larvae and determine the salient features of behavior, morphology, and fluid dynamics that control the foraging process.
With funds from the OLI award, we have successfully developed the Computational Fluid Dynamics simulation framework. In a CFD simulation the Navier-Stokes equations, which govern fluid motions like those generated by swimming fish, are applied to a mesh or grid of points representing the flow domain around the body and then numerically solved by computer.

For the larval cod problem, the flow is generated due to the motion and deformation of the larval body. Therefore, this motion/deformation is explicitly represented in our CFD simulation framework by a moving/deforming mesh of points, internally bounded by the larval body (Figure 1). We can use different body sizes and shapes to represent different developmental stages and design different moving/deforming meshes for different swimming behaviors. The flow field around the larval body can then be calculated for different developmental stages and swimming behaviors. As an example, we show in Figures 2 and 3 the time evolutions of two flow fields corresponding to two different body sizes of a larva performing the exact same swimming behavior. It is evident that there are large differences between the two flow fields. We also have developed a method to calculate the mechanical energy consumption required to generate the flow field, and the associated energetic efficiency. The comparisons among our calculated values for mechanical energy consumption and energetic efficiency now enable us to understand how cod larvae change their swimming behavior for different developmental stages in ways that achieve a fast swimming speed, while still maintaining a low energy cost. We believe that the CFD simulation framework that we have developed will be a powerful tool for studying this type of scientific problem in many other swimming organisms.

Currently, we are putting ideas together and plan to submit a research proposal to National Sciences Foundation (NSF) Biological Oceanography Program in August 2004, in order to further pursue this interesting research.