CARPSIM: stochastic simulation modelling of wild carp (Cyprinus carpio L.) population dynamics, with applications to pest control

Common carp (Cyprinus carpio L.) is an important pest fish species in Australasia and North America. Carp are widely implicated in freshwater aquatic resource degradation and both the resource management and broader communities are currently seeking effective control measures. We developed CARPSIM, a simple age-based model to simulate the effects of a range of management scenarios. The model simulates change in population biomass by age and sex-specific growth and simulates change in population abundance through recruitment and sex-specific mortality. Using empirical stock–recruitment data and stochastic components derived from local hydrological data or the southern oscillation index the model simulated the population dynamics of carp populations over 200 years using biological parameters previously estimated for two Victorian populations within the Murray–Darling basin. Carp management scenarios simulated included the effects of fishing the spawning stock; of fishing the whole stock; of spawning or recruitment sabotage; and of driving the population sex ratio towards male dominance. Model predictions suggests that faster growing, shorter-lived populations may be better controlled by molecular methods inducing male-dominance, or spawning sabotage type methods whereas slower growing, long-lived populations may respond best to removal type approaches. Unselective removal, such as poisoning or trapping all age-classes is more likely to cause pseudo-extinction at levels of instantaneous fishing mortality (F>0.7); while size-selective removal at similar F levels may only be useful to reduce the biomass below 60% of virgin biomass. CARPSIM simulations show that the probability is small for any removal-based method achieving <10% of virgin biomass when F<1.4.