Developing a Bayesian Network Model for Understanding River Catchment Resilience under Future Change Scenarios

The resilience of river catchments and the vital socio-ecological servicesthey provide are threatened by the cumulative impacts of future climatic and socio-economic change. Stakeholders who manage freshwatersrequire tools for increasing their understanding of catchment systemresilience when making strategic decisions. However, unravelling causes,effects and interactions in complex catchment systems is challenging,typically leading to different system components being considered inisolation. In this research, we tested a five-stage participatory method for developinga Bayesian network (BN) model to simulate the resilience of the Eden catchment in easternScotland to future pressures in a single transdisciplinary holisticframework. The five-stage participatory method involved co-developing a BNmodel structure by conceptually mapping the catchment system and identifyingplausible climatic and socio-economic future scenarios to measure catchmentsystem resilience. Causal relationships between drivers of future change andcatchment system nodes were mapped to create the BN model structure.Appropriate baseline data to define and parameterise nodes that representthe catchment system were identified with stakeholders. The BN model measured the impact of diverse future change scenarios to a2050 time horizon. We applied continuous nodes within the hybridequation-based BN model to measure the uncertain impacts of both climaticand socio-economic change. The BN model enabled interactions between futurechange factors and implications for the state of five capitals (natural,social, manufactured, financial and intellectual) in the system to beconsidered, providing stakeholders with a holistic catchment-scale approachto measure the resilience of multiple capitals and their associatedresources. We created a credible, salient and legitimate BN model tool forunderstanding the cumulative impacts of both climatic and socio-economicfactors on catchment resilience based on stakeholder evaluation. BN modeloutputs facilitated stakeholder recognition of future risks to their primarysector of interest, alongside their interaction with other sectors and thewider system. Participatory modelling methods improved the structure of theBN through collaborative learning with stakeholders while providingstakeholders with a strategic systems-thinking approach for consideringriver basin catchment resilience