Rhythm of relationships in a social fish over the course of a full year in the wild

Abstract

Background Animals are expected to adjust their social behaviour to cope with challenges in their environment. Therefore, for fish populations in temperate regions with seasonal and daily environmental oscillations, characteristic rhythms of social relationships should be pronounced. To date, most research concerning fish social networks and biorhythms has occurred in artificial laboratory environments or over confined temporal scales of days to weeks. Little is known about the social networks of wild, freely roaming fish, including how seasonal and diurnal rhythms modulate social networks over the course of a full year. The advent of high-resolution acoustic telemetry enables us to quantify detailed social interactions in the wild over time-scales sufficient to examine seasonal rhythms at whole-ecosystems scales. Our objective was to explore the rhythms of social interactions in a social fish population at various time-scales over one full year in the wild by examining high-resolution snapshots of a dynamic social network. Methods To that end, we tracked the behaviour of 36 adult common carp, Cyprinus carpio, in a 25 ha lake and constructed temporal social networks among individuals across various time-scales, where social interactions were defined by proximity. We compared the network structure to a temporally shuffled null model to examine the importance of social attraction, and checked for persistent characteristic groups over time. Results The clustering within the carp social network tended to be more pronounced during daytime than nighttime throughout the year. Social attraction, particularly during daytime, was a key driver for interactions. Shoaling behavior substantially increased during daytime in the wintertime, whereas in summer carp interacted less frequently, but the interaction duration increased. Therefore, smaller, characteristic groups were more common in the summer months and during nighttime, where the social memory of carp lasted up to two weeks. Conclusions We conclude that social relationships of carp change diurnally and seasonally. These patterns were likely driven by predator avoidance, seasonal shifts in lake temperature, visibility, forage availability and the presence of anoxic zones. The techniques we employed can be applied generally to high-resolution biotelemetry data to reveal social structures across other fish species at ecologically realistic scales.

Developmental arcs of plasticity in whole movement repertoires of a clonal fish

“…In order to quantify behavioral plasticity in this phenotype space over time, we calculated a measure of the spread of behaviors exhibited by any particular individual as their Shannon 'behavioral entropy'. This same basic entropy measure (31) is commonly used in biology to calculate biodiversity across a species assemblage (32) or heterogeneity of physical space use in a landscape (33,34), and has also been applied to measure behavioral variability among a range of pre-defined behaviors (35) and molecular diversity across genomes (36); here we extend the application of entropy as a measure of an individual's behavioral plasticity when entropy is calculated across a multi-dimensional, cluster-delineated behavioral phenotype space as a continuous function of time. Since Shannon entropy calculations in space generally require 'space' to be discretized as a grid, gridding behavioral phenotype space was performed using unsupervised behavioral clustering via UMAP (Uniform Manifold Approximation and Projection As with CoV measures, the developmental time course of entropy was best described as an inverted U-shape (Supplementary Tables 2.3-4…”

The power and promise of interdisciplinary international research networks to advance movement ecology

“…Monk et al [ 18 ] explore social network dynamics in a social fish, common carp ( Cyprinus carpio ), at various time-scales. They examine seasonal and diurnal rhythms, the strength of social attraction, and the presence of persistent characteristic groups over time.…”