Fish vs Robots

Friday 22 September, 2017
Mathews Theatre C, UNSW Kensington campus

The year is 2017. In the face of human population growth, food security requirements, and developing economies, millions of tonnes of fish from the last natural food production system on planet Earth are harvested every year. For fishers of tropical tuna species, the use of hundreds of thousands of free-floating, multi-sensor equipped buoys to locate schooling fish has become commonplace. These increasingly autonomous 'fish aggregating devices' scan the pelagic water column, collecting from tens to hundreds of tonnes of tuna, and influence the distribution of fish in ways that are not fully understood.

Until recently, the broad-scale operational details of these devices was not well known, but alongside advances in animal telemetry and modelling, the data are emerging to understand just how and why these devices work. In this talk, we will review the key factors concerning both the fish and the robots in this story, and what this means for our understanding of the sustainable exploitation of tropical tunas, changes in ideal-free distribution, and just how many fish there are left in the sea.


Dr Joe Scutt Phillips earned his PhD in Marine Ecology at the University of Southampton, spending the majority of his time working at the Secretariat of the Pacific Community in New Caledonia, providing scientific advice for a variety of Pacific fisheries stakeholders. Prior to this, Joe worked as scientist for a number of government organisations in the UK and island states in the South Pacific, focussing on climate change impacts and artisanal fisheries. He is currently a postdoctoral research fellow in the Climate Change Research Centre at UNSW, working on how tuna distribution relates to changing behaviour and habitat.

Joe’s research centres on ensuring the sustainability of exploited fish populations by examining their in situ behaviour as they interact with conspecifics, the species they prey upon, and those that would predate them, including humans. This is largely focussed on oceanic predators, and he combines the use of individual level data with simulation modelling to improve our understanding of the connectivity and vulnerability of these animals, as well as help us design better monitoring programmes for the future. Broadly, Joe is interested in complexity in ecological systems, and how it emerges from a community of interacting animals.