Date Awarded


Document Type


Degree Name

Master of Science (M.Sc.)


Virginia Institute of Marine Science


David S Johnson

Committee Member

Donna M Bilkovic

Committee Member

Matthew L Kirwan

Committee Member

Christopher J Patrick


As sea-level rise converts coastal forest to salt marsh, marsh invertebrates may migrate inland; however, the resulting changes in forest and marsh invertebrate communities, including the stage of forest retreat that first supports saltmarsh species, remain unknown. Additionally, the ghost forest that forms in the wake of rapid forest retreat offers unknown quality of habitat to marsh invertebrates. In a migrating marsh on the Eastern Shore of Virginia, USA, ground-dwelling arthropod communities were assessed across the forest-to-marsh gradient, and the ecological equivalency of ghost forest and high marsh habitats was evaluated to determine if marsh invertebrates utilized expanded marsh in the same way as existing marsh. Composition and diversity patterns were evaluated across the gradient for entire arthropod communities captured by pitfall and leaf litter samples, as well as springtail, ant, and beetle communities. Ecological equivalency was assessed by comparing community structure (composition and diversity) for epifauna and infauna as well as functional metrics (diet and body condition) for two marsh species found in both high marsh and ghost forest (the detritivore amphipod, Orchestia grillus, and the hunting spider, Pardosa littoralis). Community composition differed between zones, driven largely by retreating forest taxa (e.g., Collembola), marsh taxa migrating into the forest (e.g., the saltmarsh amphipod O. grillus), and unique taxa (e.g., Hydrophilinae beetles) at the ecotone. The overlap of these groups likely contributed to an observed peak in rarefied species diversity at the ecotone for pitfall samples. The composition and diversity patterns of springtails, ants, and beetles differed from entire arthropod community patterns, with springtails showing highest diversity in the high forest, and ants exhibiting peaks in diversity in high forest and ecotone. The low forest was the most inland zone to accommodate O. grillus, a saltmarsh species that may serve as an early indicator of marsh migration into forests. Ghost forest habitat offered expanded variation in both community composition and O. grillus diet. Both forest and marsh species were present in the ghost forest, and O. grillus occupied a larger trophic niche width in the ghost forest from consuming both marsh and terrestrial material. Despite these differences, ghost forest habitat supported the majority of marsh species, and observed marsh species present in both habitats primarily consumed from the marsh grass food web with no lasting difference in body condition. Because of its capacity to support saltmarsh species, the ghost forest can be considered largely ecologically equivalent to high marsh at this site, which may inform evaluations of marsh spatial extent. Forest retreat and marsh migration thus provide an important opportunity for generalist saltmarsh invertebrates to maintain their habitat extent in the face of marsh loss due to sea-level rise.




© The Author