The Effects of Plant Invasion and Ecosystem Restoration on Energy Flow through Salt Marsh Food Webs

We measured the effects of a plant invasion (Phragmites australis) on resident fish (Fundulus heteroclitus) in New England salt marshes by assessing diet quality at the food web base and by quantifying the importance of primary producers to secondary production using a recently developed Bayesian mixing model (Stable Isotope Analysis in R, "SIAR"). Spartina alterniflora, the dominant native plant, exhibited significantly greater leaf toughness and higher C/N ratios relative to P. australis. Benthic microalgae and phytoplankton (as suspended particulate matter) exhibited the lowest C/N indicating higher diet quality. We conducted a sensitivity analysis in SIAR by modeling F. heteroclitus at three separate trophic levels (1.5, 2.0, and 2.5) using species-specific discrimination factors to determine basal resource contributions. Overall, the best-fitting models include those that assume F. heteroclitus resides approximately 2.0 trophic levels above primary producers. Using discrimination factors from a range of data sources reported in the literature, our analyses revealed that consumers rely less on benthic microalgae and phytoplankton in restricted marshes (7-23 % and 11-44 %, respectively) relative to reference marshes (5-34 % and 23-48 %, respectively), resulting in a shift in diet toward invasive plant consumption (0-27 %). This is likely due to increased P. australis cover and marsh surface shading leading to decreased microalgal biomass, combined with reduced flooding of the marsh surface that favors terrestrial invertebrate assemblages. Restoration decreased the quantity of P. australis in the food web (0-15 %) and increased the importance of microalgae (1-30 %), phytoplankton (19-48 %), and native plants (23-63 %), indicating a shift in ecological recovery toward reference conditions.