Show simple item record

dc.contributor.authorMaxwell, Paul S
dc.contributor.authorPitt, Kylie A
dc.contributor.authorOlds, Andrew D
dc.contributor.authorRissik, David
dc.contributor.authorConnolly, Rod M
dc.description.abstractThe relationship between ecological impact and ecosystem structure is often strongly nonlinear, so that small increases in impact levels can cause a disproportionately large response in ecosystem structure. Nonlinear ecosystem responses can be difficult to predict because locally relevant data sets can be difficult or impossible to obtain. Bayesian networks (BN) are an emerging tool that can help managers to define ecosystem relationships using a range of data types from comprehensive quantitative data sets to expert opinion. We show how a simple BN can reveal nonlinear dynamics in seagrass ecosystems using ecological relationships sourced from the literature. We first developed a conceptual diagram by cataloguing the ecological responses of seagrasses to a range of drivers and impacts. We used the conceptual diagram to develop a BN populated with values sourced from published studies. We then applied the BN to show that the amount of initial seagrass biomass has a mitigating effect on the level of impact a meadow can withstand without loss, and that meadow recovery can often require disproportionately large improvements in impact levels. This mitigating effect resulted in the middle ranges of impact levels having a wide likelihood of seagrass presence, a situation known as bistability. Finally, we applied the model in a case study to identify the risk of loss and the likelihood of recovery for the conservation and management of seagrass meadows in Moreton Bay, Queensland, Australia. We used the model to predict the likelihood of bistability in 23 locations in the Bay. The model predicted bistability in seven locations, most of which have experienced seagrass loss at some stage in the past 25 years providing essential information for potential future restoration efforts. Our results demonstrate the capacity of simple, flexible modeling tools to facilitate collation and synthesis of disparate information. This approach can be adopted in the initial stages of conservation programs as a low-cost and relatively straightforward way to provide preliminary assessments of nonlinear dynamics in ecosystems.
dc.description.sponsorshipDepartment of Climate Change and Energy Efficiency
dc.description.sponsorshipDepartment of Climate Change
dc.description.sponsorshipDepartment of Climate Change and Energy Efficiency
dc.publisherEcological Society of America
dc.relation.ispartofjournalEcological Applications
dc.subject.fieldofresearchEnvironmental Impact Assessment
dc.subject.fieldofresearchEnvironmental Sciences
dc.subject.fieldofresearchBiological Sciences
dc.subject.fieldofresearchAgricultural and Veterinary Sciences
dc.titleIdentifying habitats at risk: simple models can reveal complex ecosystem dynamics
dc.typeJournal article
dc.type.descriptionC1 - Articles
dc.type.codeC - Journal Articles
gro.facultyGriffith Sciences, Griffith School of Environment
gro.hasfulltextNo Full Text
gro.griffith.authorConnolly, Rod M.
gro.griffith.authorPitt, Kylie A.
gro.griffith.authorRissik, David
gro.griffith.authorOlds, Andrew D.
gro.griffith.authorMaxwell, Paul S.

Files in this item


There are no files associated with this item.

This item appears in the following Collection(s)

  • Journal articles
    Contains articles published by Griffith authors in scholarly journals.

Show simple item record