This project contributed to the development of the Fisheries Research and Development Corporation 2020-2025 strategic plan (FRDC 2018-197). The CSIRO project contribution extended the use of information that was assembled on two of the possible futures that the Australian seafood sector might expect by 2030 – work completed by expert stakeholder groups. Our results show how the FRDC can explore the effect of different perturbations on elements of the future seafood systems, and to determine what indicators are most useful as predictors of trajectories of change.
Models can be used to test our assumptions about how the system-of-interest functions, and in this case, what we might expect in future.
The project team used two alternative scenarios of possible futures relevant to Australian seafood that were developed in a series of stakeholder workshops (Stage 1). We collaboratively developed qualitative models for each scenario (World A and World B), represented as signed digraphs, and then converted these to Bayes Net models (Stage 2). Our qualitative models can account for past, present and potential future dynamics of Australian aquaculture and wild-caught fisheries. The flow of activity is summarised in the schematic.
First, the qualitative models developed to describe present day dynamics were then tested to see if they could account for previously observed shocks or perturbations the Australian aquaculture and fisheries system.
Based on external input to management, employment and environmental variables, our qualitative models produced a set of predictions that were highly consistent with previously observed impacts in Australian fisheries and aquaculture.
The models were tested to see how well they compared to the dynamics described in the future scenarios (World A and B), and model predictions were found to be highly consistent with the dynamics played out in the two future scenarios – that is, both worlds are possible.
The models were then analysed to identify informative indicators for a range of possible perturbations to the system, with the intent that such indicators could inform monitoring programs and management of fishery and aquaculture systems in Australia. Selected indicators could be used by FRDC to proactively understand the emergence of a future seafood world.
- Different perturbations can be examined with the models, and the best indicators identified. For example, if regulations change the two most informative indicators for World A were (i) illegal unreported and unregulated fishing and (ii) recreational fishery production, while for World B (i) wild fish stocks and (ii) other regulations were the top two indicators. For an input to demand and preference both models had a cluster of top indicators that included (i) wild fisheries production, (ii) social acceptance and (iii) environmental quality. For an input to technology, besides technology itself, there were no other indicators that were highly informative.
This project contribution extends the use of the information from Stage 1 and allows the FRDC to test the effect of different perturbations on elements of the seafood system, and to determine what indicators are most useful as predictors of trajectories of change. These models can now be used to explore alternative perturbations, identify the informative indicators, and to determine when these models of the future (World A and B) are no longer realistic representations of real situation, and new worlds be described. If insufficient information is being gathered on these indicators, efforts could be made to collect such data, or if the information is too expensive or difficult to collect, alternative indicators can also be investigated to determine which ones provide the equivalent conformation.
Why is this model useful?
The model can be used to
- PREDICT the response of Australia’s seafood system to an input or perturbation
- To identify (DIAGNOSE) the most likely drivers of change given an observed change in the system
- Check if the way the model describes Australia’s seafood system is CONSISTENT with observed responses in the system
- Identify most informative INDICATORS for monitoring change in the system
An example scenario is provided here: FRDC_Futures_Scenario 1 – decline in the condition of the marine environment
To use the model
This FRDC Futures BBN Primer explains how to manipulate the Bayes net model to perform the above-mentioned four functions
Follow these instructions to download the files and manipulate the model.
Download Netica files
Get the Norsys Netica software
Project report – CSIRO contribution (FRDC Ocean Futures Report 2019-2020 ii)
Dambacher, J. M., D. J. Gaughan, M.-J. Rochet, P. A. Rossignol and V. M. Trenkel (2009). Qualitative modelling and indicators of exploited ecosystems. Fish and Fisheries 10: 305-322.
Dambacher, J. M., H. W. Li and P. A. Rossignol (2003). Qualitative predictions in model ecosystems. Ecological Modelling 161: 79-93.