Integrating Larval Dispersal, Permitting, and Logistical Factors Within a Validated Habitat Suitability Index for Oyster Restoration

Habitat suitability index (HSI) models are increasingly used to guide ecological restoration. Successful restoration is a byproduct of several factors, including physical and biological processes, as well as permitting and logistical considerations. Rarely are factors from all of these categories included in HSI models, despite their combined relevance to common restoration goals such as population persistence. We developed a Geographic Information System (GIS)-based HSI for restoring persistent high-relief subtidal oyster (Crassostrea virginica) reefs protected from harvest (i.e., sanctuaries) in Pamlico Sound, North Carolina, USA. Expert stakeholder input identified 17 factors to include in the HSI. Factors primarily represented physical (e.g., salinity) and biological (e.g., larval dispersal) processes relevant to oyster restoration, but also included several relevant permitting (e.g., presence of seagrasses) and logistical (e.g., distance to restoration material stockpile sites) considerations. We validated the model with multiple years of oyster density data from existing sanctuaries, and compared HSI output with distributions of oyster reefs from the late 1800's. Of the 17 factors included in the model, stakeholders identified four factors-salinity, larval export from existing oyster sanctuaries, larval import to existing sanctuaries, and dissolved oxygen-most critical to oyster sanctuary site selection. The HSI model provided a quantitative scale over which a vast water body (similar to 6,000 km(2)) was narrowed down by 95% to a much smaller suite of optimal (top 1% HSI) and suitable (top 5% HSI) locations for oyster restoration. Optimal and suitable restoration locations were clustered in northeast and southwest Pamlico Sound. Oyster density in existing sanctuaries, normalized for time since reef restoration, was a positive exponential function of HSI, providing validation for the model. Only a small portion (10-20%) of historical reef locations overlapped with current, model-predicted optimal and suitable restoration habitat. We contend that stronger linkages between larval connectivity, landscape ecology, stakeholder engagement and spatial planning within HSI models can provide a more holistic, unified approach to restoration.