Skip to main content

Application of a Bayesian Network Model and a Complex Systems Model to Investigate Risks of a Proposed Aquaculture Development on the Carrying Capacity of Shorebirds at the Miranda Ramsar Wetland

Report: TR 2007/04
Author: Mark Gibbs (Cawthron Institute)


This study was commissioned as a result of stakeholder’s concerns over possible effects of the proposed Western Firth aquaculture developments on the Ramsar wetland at the southern Firth of Thames. The study involved the development of a hazard assessment, and then investigating risk pathways through the use of a Bayesian network model, and a complex systems model. The results of the study may be summarised as follows:

  • The hazard assessment identified multiple pathways through which the proposed farms may interact with the wetland habitat; including through changes to primary productivity, detrital pathways and sediment dynamics.

  • Both the Bayesian network model and complex systems model suggest that the ability of the habitat to support shorebirds is non-linearly dependent upon both the habitat size, and quality.

  • Cultured mussels feed on seston (suspended particulate matter), therefore there is the potential for the proposed farms to influence the standing stock/production rates of plankton at the shorebird habitat. However, the network model suggests that the habitat quality is not strongly dependent on primary production rates in the water-column. Therefore, this result, along with the low predicted phytoplankton depletion resulting from the farms, suggests that phytoplankton drawdown will not have more than a minor influence over the ability of the habitat to support shorebirds.

  • Mussel farms can become reservoirs for numerous species of fouling organism including non-indigenous invasive species although a major vector (vessel traffic) is low in the region at present. If new mussel farm service vessels were to dock in the region of the Ramsar habitat, then biosecurity management codes of practice will be required in order to minimise risks of invasive species colonising the habitat as the introduction of pests may present risks. There is a risk that in the future some pest species may colonise the farms, then jump through natural dispersion onto hard structures in the Ramsar habitat, or increased recreational traffic may become a new vector. Once again it will be the responsibility of farmers and regulators to develop management plans to ensure that any unwanted pest species that establish on the farms are managed effectively.

  • Considerable volumes of sediment entering the southern Firth of Thames ends up on the mudflat habitats where the shorebirds forage. Hence changes to the sediment dynamics resulting from the establishment of the farms could play a role in changing both the shorebird habitat quantity, and quality. However, present best estimates of the influence of the proposed farms on sediment transport processes also suggest that this interaction will be minor.

  • By far the greatest influence on the shorebird habitat appears to be from terrestrial drivers, including the generation and delivery of sediments, organic material and nutrients. The recent dramatic expansion of the mangrove forests demonstrates the dynamic nature of this habitat, and despite the observed increase in utilisation of the habitat by Oystercatchers, these changes to a Ramsar-designated wetland are cause for concern.

The complex systems model also alluded to a possible other cause for concern that, although is beyond the scope of this study, should be highlighted. As noted above, there has been a substantial increase in the number of New Zealand migratory Pied Oystercatchers using the site. It remains to be seen whether this increase in utilisation has been at a cost to other birds, particularly the more celebrated Arctic migratory waders. The model, hints at the possibility that this may be occurring as a small change in the foraging behaviour of the Oystercatchers lead to an out-competing of the smaller bird species considered.

Application of a Bayesian Network Model and a Complex Systems Model to Investigate Risks of a Proposed Aquaculture Development on the Carrying Capacity of Shorebirds at the Miranda Ramsar Wetland
(427 kb, 61 seconds to download, 56k modem)

Table of contents

  Acknowledgements i
  Executive Summary v
1 Introduction 1
1.1 Background 1
1.2 Aims and scope 5
2 Hazard assessment 5
3 Exposure assessment 7
3.1 Bayesian network model 7
3.2 Complex systems model 11
4 Risk characterisation 12
5 Summary 16
6 References 18
Appendix A: Bayesian network model 20
Appendix B: Complex systems model 23