International Symposium on Deep Sea Corals 7 – July 29th to August 2nd 2019
Here’s the abstract and presentation I gave at ISDSC7 in Cartagena, the Ocean Models website is here. If you’re interested in contributing, discussing or funding the project, email me at davies(at)uri.edu.
The deep ocean is one of the most challenging and expensive habitats on earth to study, and as such it remains relatively poorly explored, particularly with respects to species distributions. In recent years, the use of marine species distribution modelling (MSDM) has proliferated, in part due to (1) the urgent need to better conserve and manage deep-sea species and habitats and (2) an increase in the amount of readily available environmental and species data. However, many studies still do not adequately describe how environmental datasets were created, leading to issues in reproducibility of covariates. In addition, studies only rarely account for or explain sources of uncertainty in their environmental covariates. In this paper, several available oceanographic datasets and the approaches used to generate them are reviewed and contrasted against a new bathymetric upscaling approach. Using MSDMs of sponges and corals as examples, the utility and drawbacks of such upscaling is demonstrated at a variety of model scales. Given we are emerging into an era of open access, we present a community-driven data portal that will improve sharing of input (environmental data, species data), output (predictor and uncertainty surfaces) and computer code (R or Python) amongst MSDM practitioners, leading to improved transparency, reproducibility and confidence in deep-sea species distributions.