Qualitative network model outputs and code: HIMI Patagonian toothfish fishery, stock and ecosystem responses to change

This dataset contains the supplementary data and code accompanying the publication: Ward, D., Hill, N., Melbourne-Thomas, J., Welsford, D., Arangio, R., McNeill, M., Wotherspoon, S., Ziegler, P., Corney, S. (2025) Exploring mechanisms of change in a Southern Ocean fishery with a co-produced network model, ICES Journal of Marine Science.

The study aimed to elucidate likely responses of the system to change, and mechanisms by which environmental, ecological or fishery behavioural changes could affect catch rates and toothfish stocks in the Heard and McDonald Islands (HIMI) fishing area.

In the study, we co-developed a set of 128 qualitative network models of the Heard and McDonald Islands (HIMI) Patagonian toothfish fishery–stock–ecosystem interactions system with knowledge holders from industry and science. These different model versions capture different possible configurations of interactions for which the sign of the effects (positive or negative) are unknown. For each model versions, we simulated 10,000 quantitative instances of the qualitative model. We analysed these model outputs to understand 1) mean qualitative responses of system components (nodes) in 8 test scenarios (press increases in environmental, fishery, or ecosystem nodes); 2) how ‘unknown’ effect signs affected qualitative responses of key nodes to change; and 3) relationships between responses and the strength of effects originating from nodes being perturbed. Please refer to Ward et al. (2025) for more detailed description of the methodology and findings.