Multi-species state-space modelling of the hen harrier (Circus cyaneus) and red grouse (Lagopus lagopus scoticus) in Scotland

摘要: State-space modelling is a powerful tool to study ecological systems. The direct inclusion of uncertainty, unification models and data, ability model unobserved, hidden states increases our knowledge about the environment provides new insights. I extend state-space framework create multi-species models, showing that ecosystem interactions limited only by data availability. are fit using both Bayesian Frequentist methods, making them independent statistical school thought. approaches can have advantage in their account for missing hierarchical structures with many parameters data; often case studies. taken fitting approach this thesis. predator-prey between hen harrier (Circus cyaneus) red grouse (Lagopus lagopus scoticus) used demonstrate modelling’s capabilities. believed be known without error, while make cyclic dynamics harder model. grouse-harrier modelled model, rather than including one species as covariate other’s Finally, included harriers’ alternate prey. singleand provide insight into species’ management. investigate aspects behaviour, from mechanisms behind cycles what motivates immigration. inferences drawn these applicable management, suggesting actions halt or mitigate conflict. Overall, suggest two popular ideas diversionary feeding habitat manipulation reduce prey densities, will not desired effect, reducing may even increase impact on chicks.