Using seafloor image data to build single-taxon and community distribution models for seabed fauna in New Zealand waters. Understanding the spatial distributions of seabed biodiversity is essential for effective management of the effects of human activities including fishing and mining. To improve understanding of seabed fauna distributions, we are developing a new database of benthic invertebrate occurrences in New Zealand waters by assembling quantitative data from all available seabed photographic surveys. By modelling the spatial relationships between taxon occurrences and environmental gradients across the region, we are able to predict the likelihood of individual taxa and communities being present in as-yet unsampled areas. In the first phase of the project, we concentrated on Chatham Rise; a region of high importance to commercial fisheries and with the highest density of available seabed imagery. Predictions from the models developed here are the first abundance-based models of benthic distributions in the New Zealand region and are the best-informed representations of seabed distributions on Chatham Rise to date, providing a resource that will have applications in marine environmental management and ecosystem research. All rasters are in a geotiff format at a 1000 m resolution cell size and projected to WGS 84 / Mercator 41 - EPSG:3994 coordinate system.

Using seafloor image data to build single-taxon and community distribution models for seabed fauna in New Zealand waters. Understanding the spatial distributions of seabed biodiversity is essential for effective management of the effects of human activities. However, existing knowledge of seabed faunal distributions comes overwhelmingly from records of museum specimens and fisheries and research trawl bycatch. Data from such sources have been used to build models that predict species and community distributions on the basis of correlations with environmental gradients but because these models are based on presence-only data from disparate sources and times, their predictions are considered uncertain. To improve understanding of seabed fauna distributions around New Zealand, we are developing a new database of occurrences and population densities based entirely on quantitative data from seabed photographic surveys designed to sample these fauna. By modelling the spatial relationships between taxon occurrences and environmental gradients across the region, we are able to predict the likelihood of individual taxa and communities being present in as-yet unsampled areas. In the first phase of the project, we concentrated on Chatham Rise; a region of high importance to commercial fisheries and with the highest density of existing seabed imagery. The models developed here were the first abundance-based models of benthic distributions in the New Zealand region at these spatial scales. In the second phase, we expanded the domain of the predictive models to encompass Campbell Plateau, in the south-eastern sector the EEZ. Combining data from Chatham Rise and Campbell Plateau in a single dataset of benthic invertebrate taxon occurrences and population densities enabled development of up-dated predictive distribution models for a range of individual invertebrate taxa, as well as models of the spatial variability in overall community composition. Rasters are in a geotiff format at a 1000 m resolution cell size and have their relevant projections written in their files.