Modelling current and future abundances of benthic invertebrates using bathymetric LiDAR and oceanographic variables

A key requirement for managing commercial fisheries is understanding the geographic footprint of the resource, the level of exploitation and the potential impacts of changing climate or habitat conditions. The development of spatially explicit predictive models of species distributions combined with predictions of changing oceanographic conditions provides the opportunity to obtain new insights of species-habitat associations. Here, generalized linear models (GLMs) were used to model the abundance of two commercially important marine macro-invertebrates, blacklip abalone Haliotis rubra and long-spined sea urchin Centrostephanus rodgersii, along the coast of Victoria, Australia. We combined abundance data from fisheries independent diver surveys with environmental variables derived from bathymetric light detection and ranging (LiDAR) and oceanographic parameters derived from satellite imagery. The GLM was used to predict species responses to environmental gradients where reef complexity, sea surface temperature (SST) and depth were strongly associated with species distributions. The abundance of H. rubra declined with increasing summer SST. In comparison, the abundance of C. rodgersii increased with increasing winter SST. The GLM showed that the projected increase in ocean temperatures will likely lead to a decline in abundance across the H. rubra fishery. Conversely, a range expansion of C. rodgersii is likely due to the strengthening of the East Australian Current. For species that exhibit a high affinity to specific seascape features, this research demonstrated how recent advances in seabed mapping can allow the identification of areas with high conservation or fisheries value at a fine-scale relevant to resource exploitation across large geographic regions.