A spatial model to improve site selection for seagrass restoration in shallow boating environments

Abstract

Due to widespread and continuing seagrass loss, restoration attempts occur worldwide. This article presents a geospatial modeling technique that ranks the suitability of sites for restoration based on light availability and boating activity, two factors cited in global studies of seagrass loss and restoration failures. The model presented here was created for Estero Bay, Florida and is a predictive model of light availability and boating pressure to aid seagrass restoration efforts. The model is adaptive and can be parameterized for different locations and updated as additional data is collected and knowledge of how factors impact seagrass improves. Light data used for model development were collected over one year from 50 sites throughout the bay. Coupled with high resolution bathymetric data, bottom mean light availability was predicted throughout the bay. Data collection throughout the year also allowed for prediction of light variability at sites, a possible indicator of seagrass growth and survival. Additionally, survey data on boating activities were used to identify areas, outside of marked navigation channels, that receive substantial boating pressure and are likely poor candidate sites for seagrass restoration. The final map product identifies areas where the light environment was suitable for seagrasses and boating pressure was low. A composite map showing the persistence of seagrass coverage in the study area over four years, between 1999 and 2006, was used to validate the model. Eighty-nine percent of the area where seagrass persisted (had been mapped all four years) was ranked as suitable for restoration: 42% with the highest rank (7), 28% with a rank of 6, and 19% with a rank of 5. The results show that the model is a viable tool for selection of seagrass restoration sites in Florida and elsewhere. With knowledge of the light environment and boating patterns, managers will be better equipped to set seagrass restoration and water quality improvement targets and select sites for restoration. The modeling approach outlined here is broadly applicable and will be of value to a large and diverse suite of scientists and marine resource managers.