1 transfer and collective evasion in herring 2 3Many large-scale animal groups have the ability to react in a rapid and coordinated manner to 4 environmental perturbations or predators. Information transfer among organisms during such 5 events is thought to confer important antipredatory advantages. However, it remains unknown 6 whether individuals in large aggregations can change the structural properties of their 7 collective in response to higher predation risk, and if so whether such adjustments promote 8 responsiveness and information transfer. We examined the role of risk perception on the 9 schooling dynamics and collective evasions of a large herring school (~ 60 000 fish) during 10 simulated-predator encounters in a sea-cage. Using acoustics we quantified swimming 11 dynamics, collective reactions and the speed of the propagating waves of evasion induced by 12 a mobile predator model. In the higher risk condition, fish swam faster, exhibited stronger 13 circular swimming pattern, and we found an increased correlation strength indicating that the 14 school had a greater ability to collectively respond to a perturbation. When exposed to a 15 simulated threat, collective evasions were stronger and behavioural change (evasion 16 manoeuvres) propagated more quickly within the school under environmental conditions 17 perceived as being more risky. Our results demonstrate that large schools make structural and 18 behavioural adjustments in response to perceived risk in a way that improves collective 19 information transfer, and thus responsiveness, during predator attacks. 20 22Keywords: Acoustics, collective evasions, information transfer, large aggregations, predation 23 risk, schooling behaviour. 24 25

A model is developed and demonstrated for simulating echosounder and sonar observations of fish schools with specified shapes and composed of individuals having specified target strengths and behaviors. The model emulates the performances of actual multi-frequency echosounders and multi-beam echosounders and sonars and generates synthetic echograms of fish schools that can be compared with real echograms. The model enables acoustic observations of large in situ fish schools to be evaluated in terms of individual and aggregated fish behaviors. It also facilitates analyses of the sensitivity of fish biomass estimates to different target strength models and their parameterizations. To demonstrate how this tool may facilitate objective interpretations of acoustically estimated fish biomass and behavior, simulated echograms of fish with different spatial and orientation distributions are compared with real echograms of herring collected with a multibeam sonar aboard the research vessel "G.O. Sars." Results highlight the important effects of fish-backscatter directivity, particularly when sensing with small acoustic wavelengths relative to the fish length. Results also show that directivity is both a potential obstacle to estimating fish biomass accurately and a potential source of information about fish behavior.

Methods for the estimation and modeling of noise present in multi-beam sonar data, including the magnitude, probability distribution, and spatial correlation of the noise, are developed. The methods consider individual acoustic samples and facilitate compensation of highly localized noise as well as subtraction of noise estimates averaged over time. The modeled noise is included in an existing multi-beam sonar simulation model [Holmin, Handegard, Korneliussen, and Tjøstheim, J. Acoust. Soc. Am. 132, 3720-3734 (2012)], resulting in an improved model that can be used to strengthen interpretation of data collected in situ at any signal to noise ratio. Two experiments, from the former study in which multi-beam sonar data of herring schools were simulated, are repeated with inclusion of noise. These experiments demonstrate (1) the potentially large effect of changes in fish orientation on the backscatter from a school, and (2) the estimation of behavioral characteristics such as the polarization and packing density of fish schools. The latter is achieved by comparing real data with simulated data for different polarizations and packing densities.

No abstract