North West Shelf Joint Environmental Management Study:InVitro Inputs - Parameter estimation

For continental shelf reef habitat the base-case parameters for the habitat equations were determined by least squares optimization. The Simplex method of minimising the sum of squares was used to fit the model to the benthos observations, with some parameters constrained to a biologically meaningful range. For the seagrass, macroalgae and mangrove habitats there was considerably less information available from which to estimate madel parameters. For these habitats the base-case parameters for the habitat equations were determined from expert information and heuristic fitting to available data on historical cover and distributions. The pessimistic and optimistic parameterisations were determined by considering the extremes of the relevant parameters in the literature and by exploring the dynamics of the system in the parameter space around the base case results. The pessimistic parameters were selected so that the impact of disturbances on the habitat groups was stronger than in the base case and the rate of recovery was slower. Conversely, in the optimistic specification, the parameters were selected so that impacts were smaller and the rate of recovery faster. However, the parameter selection was constrained such that the resulting habitat cover predictions were plausible given the available data sets and expert opinion on historical and present habitat cover. Where statistical methods could be applied the 80% confidence interval was used to identify the optimistic and pessimistic bounds. The optimistic and pessimistic bounds were much wider for the seagrass, macroaglae and mangrove habitats than for the continental shelf habitats, because of the different quantity and quality of the information available.