Using complex systems vulnerability models to synthesise environmental windows for dredging: a seagrass case study

Environmental windows, periods during which dredging is allowed, are a management tool used to help reduce the impact of dredging and ensure that critical environmental and ecological thresholds are not exceeded. However,... [ view full abstract ]

Environmental windows, periods during which dredging is allowed, are a management tool used to help reduce the impact of dredging and ensure that critical environmental and ecological thresholds are not exceeded. However, evaluating the long-term impact of dredging on ecosystem health and resilience is challenging due to uncertainty in data, knowledge, and the emergence of system dynamics from cumulative interactions and their interdependencies. A complex systems model can be used to integrate information to capture the connections between different factors of the system, their interactions, and ultimately the impact of disturbances such as dredging.

This paper presents a vulnerability based framework and case study application of a complex systems model to the analysis of environmental windows for seagrass meadows. Risk, the confluence between uncertainty (some probability) and consequence (a loss), is a powerful approach to modelling such problems. However, it is also necessary to incorporate recovery in order to capture resilience and thus the trajectory of the ecosystem over time following a disturbance. Thus, by considering the net effects of loss and recovery and their associated probabilities, it is possible to derive a quantitative framework to capture vulnerability. Specifically, the vulnerability of the ecosystem at the current time is a function of the predicted cumulative effect of loss and recovery at some future time under a range of scenarios.

Although there has been substantial work on ecosystem resilience and vulnerability and its constituent factors, a quantitative framework linking the ecological and mathematical domains of the problem remains a challenge. Here, a proposed vulnerability framework is presented for seagrass meadows under dredging disturbances using a Dynamic Bayesian Network (DBN) model. The DBN provides a visual and mathematical representation of the factors and their relationships, linking biological, environmental and ecological factors in a whole-of-systems model whose boundaries are determined by the specific vulnerability based management objectives of maintenance dredging. It integrates available information including expert knowledge, domain knowledge and data. Due to the dynamic and probabilistic nature of the model, it can then be used to evaluate cumulative impacts to ascertain the distribution of potential outcomes and their probability.