Complexity and scaling in the long-range collective movement of animals

In the late 1990s the dot-com financial bubble was at its frenzied peak. Shares in companies such as Qualcomm were offering traders returns of almost 2,600% in a year. Many Wall Street leaders saw how this would end. Julian Robertson, manager of the Tiger hedge fund, steadfastly refused to buy into the bubble. Because of this stance his fund lost money and investors throughout 1999, and was forced to close in March of 2000, just as the bubble burst and technology stocks began to tumble.

This example serves to highlight some of the general principles and processes that govern the behavior of complex systems. Exogenous factors invoke a response from individuals, interactions then magnify, or inhibit, this behavior and a collective response emerges. The loop is closed as the collective behavior creates pressures on the individuals, effectively altering their environment. Potentially the collective will dominate external factors and the best option for individuals (temporarily at least) is to follow the herd, regardless of their own beliefs or opinions.

Throughout the natural world this dynamic may be observed. Animal groups are a classic example of a complex system in which individuals must reconcile personal objectives with their social context. Here, as in finance, it is often better to be wrong than alone.

In these systems there is an interplay between leadership, social interaction, and environmental information that determines the accuracy of collective decisions. In the context of movement, and in particular the long-range movements associated with migration, animals have the potential to improve their navigational accuracy by traveling together. However, the use of social information may also have negative consequences. Behavior that was adaptive in one situation may be fatal when the structure of the social group changes. Further, evolution may drive systems to suboptimal states, as free-riders gain an advantage and parasitize information.

Hence, long-range animal movement is often a multi-level process in which individuals aggregate, collective behavior emerges, and conformity abounds. The purpose of this research is to explore these themes in a range of migrating animals. Four study systems will be investigated; wildebeest of the Serengeti, caribou of northern Canada, salmon in the Pacific Northwest, and bar-headed geese on their migration across the Mongolian Steppe.

By employing state-of-the-art robotics and computer vision software, the trajectories of animals will be recorded as they make their journeys. Analytical tools will be used to assess the role of sociality in movement decisions and investigate leadership roles, the effect of environmental cues, and the consequences of traveling in a group. In this way a taxonomy of collective movement will be created. This will be used to investigate the feedback between the collective and the individual, with the aim of understanding the role of social influence, and its impacts on the stability and robustness of these migrations.