Award: OPP-1840927

The goal of the work is to investigate drivers of individual behavior of krill in response to environmental conditions, particularly flow rate and direction, light, and food and predator chemical stimuli. Since krill commonly occur in large aggregations, we examined the behavior of individual krill and krill swarms. By understanding the responses of krill to environmental cues, we can better predict the distribution so as to improve our understanding of krill populations and their impacts on Antarctic ecology, and better manage this critical living marine resource. Our results will also help us understand krill schooling dynamics. Data will be used to construct an individual based model of krill behavior that can predict krill movements in oceanographic conditions informed by real data, we will and make this available to the scientific and conservation communities to help them predict krill distributions and understand their consequences. Our results showed that krill are not passive particles that move with the flow, as often has been assumed. Using cameras to record positions of krill to furnish x,y,z coordinates of swimming animals (Figure 2), allowed us to parameterize krill behavioral responses to different conditions. The reconstructed 3d paths (Figure 3) are used to determine swimming speed, turning, body orientation and other variables. Krill actively orient to horizontal flows to move upstream in speeds as low as 3 mm s-1. Food odor has complex effects, increasing swimming speed and orientation to flow at low levels, indicating krill actively seek out plankton blooms. High concentrations of food odor reduce orientation to flow and increase turning, which presumably functions to keep krill in areas of high food concentration and increases feeding. Thus, krill movements are shaped by environmental conditions, particularly flow and food. Krill responses to food are reduced in the presence of predator odor (penguin guano), which also decreases feeding rate. Thus, penguins have both direct and indirect effects that can propagate to other species. Krill respond more to flow in the absence of light, suggesting visual cues provide additional, but not, critical orientation cues. Krill dont respond as strongly to vertical flows, but food odor has the same effects. We built a specially designed circular flow chamber to examine krill swimming in groups (Figure 4). Krill form coherent aggregations only under certain conditions. Schooling is contingent upon both flow and light, and flow results in more organization when light was low/absent. Organization peaks at a density of 9 krill L-1. Lower densities might not provide sufficient cues, whereas higher densities might interfere with the ability of neighbors to see and respond to each other. When challenged with similar conditions, krill swim more quickly in groups than individuals, which supports a long standing supposition that schooling increases swimming efficiency and reduces energetic cost by reducing drag (in much the same way as cycling in a pack). Using the data on krill swimming in different conditions, we were able to identify chlorophyll and flow speed as the most important variables that determine krill behavior. Our current simulations have shown that krill form loosely structured, poorly packed balls in low flow but that in higher flows of water krill form tightly packed, elongated oval shapes that point in the direction of flow (Figure 5). Individual behaviors therefore scale up in a way that we can predict swarm structure, which helps interpret data on krill swarms collected by ships. Including chemical conditions such as food is the next step and will help us further understand how krill respond to environmental conditions and how this affects where krill are found and their swarm structure. Eventually we will model the behavior of krill in real oceanographic data to see where krill swarm (Figure 6), how long they stay in different locations and how these swarms form and disperse in realistic environmental conditions. Initial tests into this space have been conducted with preliminary results in ROMSPath showing that our work will greatly improve already existing models on krill movement in Antarctica. Last Modified: 08/28/2024 Submitted by: MarcJWeissburg