Determining the likely future distribution of interacting fish species due to climate change: A combined diffusion and multivariate autoregressive state space model

Background/Question/Methods

Spatial distributions of fish species are expected to change due to climate change and the resultant shifts in bioclimatic envelopes.  Proper management of these species and the fisheries that depend on them relies on anticipating and predicting the likely future distributions and population sizes and structures of these species as they move. In this study, I develop a diffusion model describing movement of species combined with a modified multispecies Gompertz population model (a biotic interaction model) in order to predict the potential future distributions of several theoretical groups of interacting species due to climate change. The theoretical model is run on three theoretical, interacting species under scenarios of differing strengths of biotic interactions among species in order to demonstrate a range of effects that biotic interactions can have on a distribution model. In addition, I examine a set of three commercially important, interacting species from George’s Bank by incorporating shifting bioclimatic envelopes and biotic interactions in a multivariate autoregressive state space model to determine the likely future distribution of the species due to climate change on a time horizon of approximately 50 years. 

Results/Conclusions

In the theoretical model, depending on the scenario being examined, we can see changes in distribution and abundances over time. These future distributions and abundances of the species being examined are different when biotic interactions are included in the bioclimatic envelope model than when biotic interactions are not explicitly accounted for in the model. The general results are a pole-ward shift in distributions, but abundances of fish depend upon, not only their preference for habitat and temperature, but also the overlap of the distributions of predator and prey species. For example, if a predator species is forced out of a region because of temperature requirements, a prey species in that region may increase in abundance. Including biotic interactions in distribution models can provide improved predictions of future distributions and abundances due to climate change.