Long-term succession of benthic infauna communities on constructed Spartina alterniflora marshes

Benthic infauna communities were characterized along a chronosequence (1 to 28 yr old) of 7 constructed Spartina alterniflora Loisel marshes and 7 natural (reference) marshes to identify patterns of succession following salt marsh creation. Infauna density and species richness (per 7.07 cm(2) core) achieved equivalence to comparable reference marshes within 8 yr following marsh creation. Taxa with dispersing larval stages (e.g. Streblospio benedicti, Capitella sp.) achieved equivalence within 3 yr following marsh creation. Taxa that lack a planktonic dispersal stage, such as tubificid oligochaetes and Manayunkia aesturina, were slower to develop. Densities of Manayunkia did not achieve equivalence to natural marshes until 8 yr after marsh creation. Twenty-five yr elapsed before oligochaete densities of constructed marshes were similar to natural marshes. In constructed marshes, densities of surface-deposit feeders, dominated by Streblospio and Manayunkia, achieved equivalence to natural marshes within 8 yr following marsh creation. Subsurface-deposit feeders, consisting mostly of oligochaetes, did not become equivalent to natural marshes for 25 yr. Predictable trajectories, described by an asymptotic increase over time, existed for total species, species richness, and surface- and subsurface-deposit feeders. Oligochaetes and Manayunkia, which produce non-dispersing larvae, exhibited linear trajectories over time. In constructed marshes, total density and density of dominant taxa and trophic groups were strongly correlated with soil characteristics, especially organic C, N, bulk density and macro-organic matter (MOM, the living and dead root and rhizome mat). Non-linear regressions using soil organic C and MOM explained 38 to 40% of the variation in constructed marsh infauna density. The regressions suggest that a threshold of 0.5% soil organic C (500 g m(-2), 0 to 10 cm) and 500 g MOM m(-2) (0 to 10 cm) is needed to support infauna densities comparable to densities found in natural marshes. Taxon richness of constructed marshes was more strongly related to vegetation characteristics, especially macro-organic matter quality (e.g. lignin) than soil characteristics. Development of benthic infauna communities following marsh construction requires longer than the 5 yr monitoring period required by many wetland mitigation plans, and as much as 25 yr are needed for some attributes (e.g. oligochaetes) to achieve equivalence.