Rugosity-based regional modeling of hard-bottom habitat

Systematic conservation planning is most often directed at the representation and protection of marine biodiversity. However, direct observation and sampling of marine biodiversity is extremely time consuming and expensive. Due to these constraints, marine conservation planners have sought proxies for marine biodiversity to use in their models, Hard-bottom habitats support high levels of biodiversity and are frequently used as a surrogate for it in marine spatial planning. Rugosity (i.e. the roughness of the seafloor) is an indicator of hard-bottom habitat. In the present study, we expand on previous analyses of the relationship between rugosity and hard-bottom and create the first data-driven regional rugosity model to predict hard-bottom habitat. We used logistic regression to create an empirical model and compare it to other pre-formulated definitions of rugosity with receiver operator characteristic curves. Our model performed better than all other models and was able to correctly predict the presence or absence of hard-bottom habitat with similar to 70 % accuracy. This model offers a fast and inexpensive alternative to more traditional survey methods, and should be of value to regional conservation planners and fisheries managers as an initial predictor of hard-bottom habitat. By testing this model with low-resolution (90 m) bathymetry data, we demonstrate that this type of information may be used in marine conservation plans in regions such as developing countries, where high-resolution data is not currently available. Further, our model offers a proxy for marine habitat diversity in non-coastal areas, an under-represented sector in marine conservation planning.