The Effects of Climate-Related Stressors on Social Behaviour in Fish


Various forms of social behaviour exist throughout the animal Kingdom, with benefits for predator avoidance, foraging, and reproduction. To date, however, we know almost nothing about how anthropogenic environmental disturbance affects group living in animals. Among the most critical threats to aquatic ecosystems are changes in temperature and oxygen availability due to climate change. Still, nearly everything we know about how climate-related stressors affect physiology and behaviour comes from observations of single animals in the absence of any social context. Now is the time to address this knowledge gap, given existing knowledge of environmental effects on single animals, the development of technologies for tracking of groups of animals and the urgent need to predict how populations will respond to climate change.
The most immediate effect of climate change is an increase in global surface temperatures of 1.5 to 5.8oC by 2100, with localised consequences being more severe. In ectotherms such as fish, warming increases energy demand by elevating standard metabolic rate. Supervisor Killen’s work has shown individuals with an intrinsically high SMR are less social, likely because they prioritise food acquisition and so avoid competition with groupmates. In contrast, individuals with lower energy demand prioritise protection from predators and stay closer to conspecifics. Accordingly, a greater metabolic demand stemming from temperature increase could make individuals less social and reduce cohesion of social groups, possibly affecting group foraging, predator avoidance, social learning, migration and any phenomena dependent on group cohesiveness. The maximum metabolic rate of ectotherms and aerobic scope for oxygen-consuming physiological functions are influenced by temperature, often peaking at a thermal optimum. In general, spontaneous activity in ectotherms increases with temperature but then decreases as they approach their critical thermal maximum, a benchmark for thermal sensitivity at which they experience locomotor impairment. Effects of temperature on these variables could alter which individuals become leaders because individual activity and speed are determinants of leader-follower dynamics. Social niche or placement within hierarchies can also feedback to affect traits such as boldness, which in turn can be linked to metabolism, suggesting social interactions could affect tolerance to thermal stress.
To examine these issues, the proposed project will:
(1) examine how temperature and hypoxia affect interplay between behaviour of individual
animals and their social group
(2) investigate how temperature and hypoxia influence social group functioning during group movements, foraging, and predator avoidance
(3) produce a robust movement model of group behaviour that can be applied to a range of environmental conditions and ecological contexts.
The project will use fish as a model study system. Fishes are frequently used as model taxa to study collective behaviour in animals, and the effects of environmental stressors on ectotherms. They occupy aquatic habitats across the globe, play key roles as both predators and prey, and display a variety of social behaviours. By adopting innovative empirical and theoretical approaches and combining behaviour and physiology, this project will be the first to examine how thermal shifts and hypoxia brought on by climate change will affect social behaviour.


The overall strategy for all project components will be to acclimate fish (wild minnows, Phoxinus phoxinus) to various combinations of temperature and levels of oxygen availability that reflect current day conditions as well as future projections. Fish will then be measured for minimum and maximum metabolic rates, anaerobic and aerobic capacity, swimming performance, and various metrics of social behavior. This approach will allow us to examine the independent and interactive effects of temperature and water oxygenation and individual behavior within social groups and the influence of individual physiological sensitivity to these factors. In terms of social behavior, separate experiments will be performed to examine the effects of the environment on dominance hierarchies within groups, group social networks, effects on energy savings while swimming in group formation, group foraging, and group predator avoidance. Supervisors Killen and Webster have broad expertise in studying ecophysiology and social behavior in fish and so will advise on these aspects of experiments. Importantly, data from each study will also be used in the development of groups of fish will be used to refine an existing continuous-time random walk model that incorporates interactions among individuals as a potential force affecting movement towards or away from conspecifics. This aspect of the project will be overseen by Torney, who has extensive experience in the quantitative analysis and modelling of animal social behaviour and movements. Models will be augmented with information on physiology and how ranges of interaction change with temperature and hypoxia, then re-examined for fit with data. Once we can accurately model group behaviours in various contexts, we can relate key behaviours (e.g. movement speed, leadership) to environmental sensitivities measured in individual animals, and simulate additional scenarios to form predictions for generalised responses to environmental change (e.g. different group phenotypic compositions, group sizes, or environments that match specific IPCC projections). Modelling interplay among social dynamics, food and oxygen demand, foraging, susceptibility to predation, and group formation will enable predictions of climate impacts at the population level and will be an invaluable tool for future work.

Project Timeline

Year 1

Experiments examining dominance hierarchies, social networks; initial model fitting

Year 2

Experiments examining group movements and energy savings; further model refinement

Year 3

Experiments examining group foraging and predator avoidance; model finalization

Year 3.5

Completion of analyses and thesis write-up

& Skills

The student will gain broad experience in physiological and behavioural analysis due to the multidisciplinary nature of this project. This will include measurement of animal metabolic rates using intermittent flow respirometry, the measurement of locomotor performance using various benchmark tests, and various behavioural assays. The student will also gain experience in the automated tracking and quantification of animal behavior using specialized software. The analysis of data from animals in groups will also provide training in various statistical techniques includes linear mixed-effects models. Notably, the student will be primarily placed within Killen’s lab group at the University of Glasgow. This is a large and successful group with a dynamic and cooperative working environment. More generally, the host department has 8 faculty that specialize in various facets of fish biology. The student will therefore have maximum opportunity for additional collaborations and advice as results emerge throughout the project.

References & further reading

Killen, S.S., Marras, S., Nadler, L.E., Domenici, P. 2017. The role of physiological traits in assortment among and within fish shoals. Philosophical Transactions of the Royal Society. 372: 20160236.
Cooper, B., Adriaenssens, B., Killen, S.S. 2018. Individual variation in the compromise between social group membership and exposure to preferred temperatures. Proceedings of the Royal Society B. 285: 20180884. 10.1098/rspb.2018.0884.
McLean, S., Persson, A., Norin, T., Killen, S.S. 2018. Metabolic costs of feeding predicatively alter the spatial distribution of individuals in fish schools. Current Biology. 28: 1144-1149
Killen, S.S., Fu, C., Wu, Q., Wang, Y.-W., and Fu, S-J. 2016. The relationship between metabolic rate and sociability is altered by food-deprivation. Functional Ecology. DOI: 10.1111/1365-2435.12634
Jolles, J.W., King, A.J., Killen. S.S. 2020. The role of individual heterogeneity in collective animal behaviour. Trends in Ecology and Evolution. 35: 278-291.

Further Information

Apply Now