A bird’s eye view of fish schools

Gordon Swartzman
University of Washington
School of Aquatic and Fisheries Science, 355020

I begin with a model of fish school coalescence and splitting, based on Kolmogorov differential-difference equations (published in 1991) and review model assumptions from a ‘modern’ perspective. This model postulated fish schools being dynamically generated from an inherent larger local population, which later led to postulating the existence of school clusters, or aggregations of local populations. Evidence for these clusters in several fish populations is based on the distribution of next neighbor distances between schools encountered along survey transects. I then return to the original model and discuss observations of school coalescence and splitting upon predator encounter. I discuss possible extensions to model assumptions and the possibility for developing underlying population or cluster-level parameters from such a model and how these parameters might be measured.

Key words:

Clusters, populations, next-neighbor distance, sardine, anchovy, hake, Pollock
School size distribution, splitting and coalescing, Kolmogorov differential-difference model

Audio (MP3 File, Podcast Ready) Presentation Files (Zip Archive)

Back to Swarming by Nature and by Design