Oyster Habitat Suitability in the Northern Gulf of Mexico

Oysters provide a wide variety of ecosystem services including furnishing habitat, supporting an economic industry, and enhancing water quality. To identify suitable areas for oyster restoration and aquaculture, areas of oystersuitable habitat must first be identified. A habitat suitability model is an ideal tool for identifying sites for species restoration. Because it relies on presence-only data, MaxEnt is a particularly useful habitat-suitabilitymodel for identifying restoration and species introduction sites. Habitat suitability models rely on the selection of environmental factors, which are assumed to be important for the target species; however, this selection of environmental factors is often arbitrary and there are few existing guidelines. This work applies an oyster habitat suitability model to the St. Louis Bay in the northern Gulf of Mexico. Six environmental factors, namely average salinity, maximum temperature, minimum temperature, water depth, minimum dissolved oxygen, and average total suspended solids were chosen to simulate habitat suitability. The environmental factors were obtained from a calibrated hydrodynamic and water quality model of the estuary. The habitat suitability model was run with every possible combination of environmental factors, including one, two, three, four, five, and six inputs. Model results showed that at this location, salinity is the most important environmental input. Furthermore, model results showed that increasing the number of inputs optimizesmodel results. There is a diminishing return on the addition of environmental factors and there is a point at which the continued addition of environmental factors will not continue to notably improve model optimization. The ranges of simulated suitable values for the environmental factors were contextualized within measured values. This study shows how a statistical model can be used to identify restoration locations. It is a particularly compelling methodology because (1) it does not require prior information regarding suitable ranges of environmental factors; (2) it does not require information regarding the level of importance for those environmental factors; (3) it requires presence-only data; (4) results are based on spatiotemporal data at a finer scale than is generally measured in the field; and (5) model results can provide an additional line of evidence that complements knowledge by biologists and oystermen.