Shifts in biodiversity and physical structure of seagrass beds across 5 decades at Carriacou, Grenadines

The primary objectives of our study are: 1. To compare changes in abundance (cover, frequency of occurrence, biomass, shoot density) and morphology of T. testudinum, considered the dominant or climax species at the onset of our study, along with changes in abundance of other native or invasive seagrass species and of associated macroalgae and sessile invertebrates (corals, sponges) across this 47-year period. 2. To use a species-level measure of abundance (frequency of occurrence) to examine tempo- ral changes in biodiversity and community composition of seagrass beds. 3. To provide concurrent measures of physical parameters of seagrass beds (extent of the sea- grass-free inshore zone, depth profiles and frequency of erosional scarps) that can be affected by changing hydrodynamic forces, and to relate these to population and commu- nity level changes documented in objectives 1 and 2. At this time of accelerating climatic change and increasing levels of global and local human impact on coastal ecosystems, the importance of establishing baselines against which ecologi- cal change can be measured becomes increasingly apparent. Longitudinal studies are particu- larly valuable in establishing relevant baselines or in providing some semblance of a pristine state, that enable comparisons across geographic regions and inform conservation and man- agement initiatives [41]. Most comparisons of seagrass systems over periods of 40 years or more are based on aerial imagery [20, 23]. Comprehensive monitoring programs that capture

ground-level observations on the structure of tropical seagrass communities before the 1980s are rare [22]. Our study provides this insight for a relatively undeveloped location in the Carib- bean that has nonetheless been impacted by broad-scale changes in biotic and abiotic factors that have shaped the local seagrass ecosystem. Materials and methods Study sites and sampling methods To provide a basic description of “turtle grass beds” (seagrass beds dominated by Thalassia tes- tudinum) around Carriacou in 1969, areas putatively dominated by T. testudinum were identi- fied as dark patches in aerial photographs of the island taken in 1966 by the Directorate of Overseas Surveys, United Kingdom (Contract 85, print scale 1:12,900). Detailed tracings of these patches provided cursory maps of seagrass distribution that were used to select study sites representative of seagrass beds around the island. For logistical reasons, most sites were on the east, windward coast. Sites were located by reference to prominent geographic features (e.g., promontories, beaches, cays, reefs) and accessed directly from shore or with a small motorboat. As a measure of ground-truthing, T. testudinum was present at all putative turtle grass sites during the initial sampling in 1969. Sampling took place in March 1969, late January to early March 1994, and January to February 2016 (except 1 transect in February 2019). Sam- pling was mainly done by snorkeling; SCUBA was used to access deeper beds in some offshore regions in 1994. Transects

Eleven sites were selected in 1969 to sample line transects that spanned fringing seagrass beds around Carriacou (i.e., located along the shoreline) [35]. Ten sites were resampled in 1994 and 2016 (Fig 1, S1 Table); one of the 1969 line transects could not be relocated. Transects were run from shore (sea level), and perpendicular to it, across all or part of a fringing bed. Transect length in 1969 was variable [35]; transects extended to at least the point where no further PLOS ONE Seagrass bed biodiversity and structure and across 5 decades PLOS ONE | https://doi.org/10.1371/journal.pone.0306897 August 1, 2024 3 / 25