Using carbon isotope ratios to verify predictions of a model simulating the interaction between coastal plant communities and their effect on ground water salinity

As sea level rises in low-lying coastal islands, lower lying salt-tolerant (halophytic) coastal vegetation communities may be able to migrate inland, replacing the freshwater vegetation, which is limited by its inability to tolerate salt stress. The pace of such shifts may be accelerated by the self-reinforcing positive feed-back between the halophytic vegetation and salinity, as well as by frequent and intensified salinity pulses associated with the increasing impact of storm surges as a consequence of sea level rise. A previously published spatially explicit individual-based model that simulates impacts on upland freshwater communities from sea-level rise and storm surge incorporated a self-reinforcing positive feedback relationship between each vegetation type and salinity. We used a modification of this model to predict the interaction between three coastal communities: mangroves, hammocks, and pinelands, which are usually arranged in well-defined bands at increasing distance from the coast. After a 20-year simulation, boundaries between communities were observed in the model landscape. An area of high pine mortality occurred at the interface between pinelands and hammocks. Model output of ground water salinity predicted fresher ground water in the pineland occupied spatial cells, while ground water salinity under both mangroves and hammocks was very high. However, the mangrove vadose zone had significantly higher salinity compared to hammocks. Model simulation predicted two qualitative characteristics regarding the interaction between these three different coastal communities: Mangroves and hammocks communities tend to have ground water with high salinities, while pineland groundwater salinity is low, and pineland located at lower elevation relative to adjacent hammock will be negatively influenced by higher ground water salinities in hammocks. We tested these predictions with foliar δ¹³C of C. erectus collected from Big Pine Key as a proxy for groundwater salinity. Measurements of groundwater salinity via this proxy confirmed the two predictions of the model. Our approach provides an approximation of the impacts of sea level rise on terrestrial vegetation communities, including threatened pineland communities, and can be used as a tool for management decisions and future modeling.