The feeding ecology of a common saltmarsh crab, Armases cinereum (Armases), was investigated to determine how habitat (mangrove vs. ecotone, natural mangrove vs. modified mangrove fringe) influenced this species’ feeding behavior and trophic ecology in its southwestern Florida, USA, distribution. In the laboratory, Armases’ preference for mangrove material was examined using leaves of three mangrove species (Avicennia germinans, Laguncularia racemosa, and Rhizophora mangle) and leaves of different degradation levels (fresh, senescent, and partially-decomposed). Leaf material from A. germinans was preferentially consumed over the other taxa at fresh and partially-decomposed levels of leaf decay. When Armases were offered a choice between four common upland vegetation types (Iva frutescens, Borrichia frutescens, Nephrolepis biserrata, and Stenotaphrum secundatum), the preferred mangrove from the previous experiment (A. germinans, partially-decomposed), and an animal prey item (Gryllodes sigillatus, cricket), Armases displayed greatest selectivity for the animal prey item and high selectivity for both I. frutescens and A. germinans plant taxa (Manly-Chesson α selectivity and Chesson ϵ electivity). Field-based stable isotope analysis was used to determine trophic position and reconstruct dietary proportions of Armases across three pairs of natural and heavily-modified sites within Tampa Bay to determine whether Armases feeding behavior is impacted by the presence of upland forest adjacent to mangrove forest habitat. Analysis of trophic position based on δ15N signatures of Armases from each of the six populations revealed that Armases in habitats with modified connectivity had lower trophic variability and significantly higher average trophic position compared to Armases sampled from the three sites with natural connectivity. Stable isotope diet reconstruction using the Bayesian mixing model SIMMR further established Armases preference for animal-derived food material in habitats with natural and modified connectivity. This preference is likely driven by high selectivity for sources rich in Nitrogen (i.e., animal tissue, partially-decomposed A. germinans material, and I. frutescens). I determined that the use of laboratory experiments in conjunction with stable isotope mixing models is important in accurately investigating feeding preferences of Armases in mangrove intertidal regions. Together, my results show that the diet of Armases is broadly omnivorous and populations can be influenced by the heterogeneity of their habitat. Further feeding experiments, dietary analyses and a longer sampling period are needed to more definitively identify the patterns of Armases detritivory in mangrove and ecotonal upland habitats.

Recent research using stable isotope analysis has shown a dependence on migrating or breeding populations of vertebrates as vectors for the transfer of marine-derived nutrients within coastal ecosystems. Sable Island, Nova Scotia, Canada supports numerous species of plants, a variety of seabird colonies (including common [Sterna hirundo] and Arctic [Sterna paradisaea] terns), the world’s largest grey seal (Halichoerus grypus) breeding colony, and a self-sustaining population of wild (feral) horses (Equus ferus caballus). I hypothesize that nitrogen cycling within this island ecosystem is highly influenced by the input of nutrients from seals and seabirds (‘biogenic vectors’), affecting primary production and potentially stabilizing higher trophic levels (i.e., horses). To examine this relation I developed a spatially-explicit isoscape for Sable Island through stable isotope analysis of nitrogen (δ15N) in samples (n = 282) of marram grass (Ammophila breviligulata). I incorporated significant variables (i.e., distance to vector colony and distance to shoreline, r2 = 0.41) into the final parsimonious interpolation model using universal co-kriging techniques. The greatest 15N enrichment occurred within the tips and along the perimeter of the island, coinciding with greater densities of grey seals, while the lowest values occurred within the centre of the island. I then identified individual contributions of seal-, tern- and horse-mediated transfer of marine-derived nutrients inland. Marram grass exhibited higher δ15N within seal (μ = 7.5‰) and tern (μ = 5‰) colonies, while horses and biogeochemical processes (i.e., volatilization, ammonification, etc.) most likely contributed to the homogeneity within the centre of the island (μ = 3.6‰). Due to the higher densities, wider distribution, and greater 15N enrichment of marram tissues, grey seals appear to be the most important vector species while seabirds have a more localized effect. The greater availability of N within vector colonies supplemented the local vegetation community, contributing to greater vegetation cover within colony boundaries. This relation had secondary effects on the horse population, which showed correspondingly higher horse δ15N values within the tips of the island (δ15N + 1.6‰) due to consumption of 15N enriched forage. I conclude that biogenic vector species promote nutrient transfer by establishing nutrient gateways which indirectly cause cascading effects throughout the food web.