When Less Is More

Infaunal Communities

Infauna (including macroinvertebrates, microinvertebrates and protists) were collected in replicate cores (2.5 cm diameter, 5 cm deep; volume 24.5 cm³) from each habitat type within each restoration state during the 2022 growing season (between May and October). Replication varied based on habitat presence and accessibility, with 6–15 cores per habitat type in each restoration state. Cores were rinsed through a 500 mm sieve and fixed and stained in a 10% Formalin/Rose Bengal solution. Trained observers identified infauna under a dissecting microscope to the lowest practical taxonomic level (most commonly to class). Data were square-root transformed prior to analysis. Differences in infaunal abundance were assessed with a two-way Analysis of Variance, and differences in community composition was assessed with a two-way Analysis of Similarity (ANOSIM) based on a Bray-Curtis similarity matrix. The factors were habitat type (vegetated and non-vegetated) and restoration state (U13, P13, U21, REF) (R Core Team 2020). In spring 2014, infauna were collected from the non-vegetated habitat at a subset of U13 and P13 sites; these values were qualitatively compared to 2022 values.

Across all 2022 sampling sites, the most common types of infauna were native taxa including annelid worms, ostracods, foraminiferans, insect larvae, and gastropods. Total infauna abundance in U13 sites was substantially higher in 2014 (average count per core 722 ± 341 SE) than in 2022 (35 ± 12). In P13 sites, there was little change in infaunal abundance between 2014 (36 ± 16) and 2022 (25 ± 9). In 2022, total infaunal abundance varied among habitat types and restoration state (ANOVA habitat type * restoration state df = 3, F = 15.35, p < 0.001). The highest macro/microfaunal abundances occurred in the non-vegetated areas of U13 sites (Figure 2a). In contrast, the highest protist (foraminiferan) abundances occurred in the vegetated areas of the reference sites (Figure 2b). ANOSIM indicated a high degree of overlap in infaunal community composition among habitat types and across restoration states (habitat type R = 0.149; restoration state R = 0.178). Infaunal abundance is often linked to edaphic characteristics (Craft and Sacco 2003), and previous work at an adjacent site suggested that soils in reference areas were sandier and had higher nutrient content than those in restored areas, but that all restored areas were largely similar to each other (Armitage et al. 2014). Thus, the large difference between U13 and P13 areas (Figure 2a) suggests that other factors, perhaps linked to the plant community, may have influenced infaunal abundance in these areas.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Total infaunal counts per core (24.5 cm³) in different habitat types and restoration states. a) Macro- and microinvertebrates; b) foraminiferans. See Figure 1 for description of restoration states. Box plots depict the median (bold line), first and third quartiles (box), minimum and maximum values (whiskers), and outliers (points).

Avian Communities

To assess bird use of restored and reference sites, game cameras were set up at one site in each of three restoration states: U13, P13, and REF. Cameras were installed on poles one meter from the ground, overlooking a representative mixture of vegetated and non-vegetated habitat within each site. Cameras were deployed for six two-week periods during migration and overwintering periods between October 2013 and April 2015. The photos were manually reviewed and bird abundance and identity were recorded. Counts focused on native, wetland-dependent species that forage on benthic organisms. Individual birds were not counted more than once if they appeared in a sequential series of photos. The high incidence of images with no detectable birds precluded meaningful statistical analysis, so patterns of bird use are described qualitatively. In 2022, we recorded qualitative bird observations in person during the infaunal sampling events.

In 2013–2015, birds were most frequently observed in unplanted restored areas (Figure 3). Species richness (S) at each site (cumulative across all survey dates) was substantially higher in unplanted (U13) areas (S = 17) relative to planted (P13) areas (S = 3) or reference marshes (S = 8). In 2022, all older sites (U13, P13, REF) were heavily vegetated by a monoculture of native Spartina alterniflora (Armitage 2021) and few birds were observed in those sites. Large mixed-species aggregations of shorebirds and wading birds were observed only in non-vegetated (mudflat or pond) areas, particularly in U21 sites that had little emergent vegetation coverage.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Heat map showing average bird abundance (number/2 weeks) over a series of six observation periods from 2013–2015 at one representative site in each of three restoration states as defined in Figure 1.

Potential and realized bird use of restored wetlands was linked to the availability of non-vegetated habitat such as mudflats and ponds. Shortly after restoration, bird assemblages were more diverse and abundant in areas that had not yet developed dense stands of wetland plants (Figure 3). As noted above, infaunal abundance, a proxy for trophic support for benthic foraging bird species, was highest in unplanted areas after restoration, and remained high in those areas that persisted in an unvegetated state over the following decade (Figure 2). However, dietary needs among wetland-dependent birds are not monolithic, and there are numerous smaller species (e.g., Family Scolopacidae) that consume biofilms—a slurry of protists, benthic microalgae and organic matter on the sediment surface (Kuwae et al. 2008). Based on our observed patterns of foraminiferan abundance, biofilm biomass may be higher on substrates supporting diverse stands of vegetation, such as those found in reference areas (Figure 2b).

Overall, our results indicated that an ideal restoration design should increase small-scale elevation heterogeneity to create persistent non-vegetated features to enhance shorebird use while balancing permit-mandated goals to increase vegetation cover. There is broad recognition that wetland restoration has strong potential to increase bird abundance (Ortega-Álvarez and Lindig-Cisneros 2012, Sullivan 2015), but a more intentionally designed heterogeneous elevation approach may further enhance diversity and habitat use for a wider array of avian species (Comín et al. 2001, Armitage et al. 2007, Pickens and King 2013, Boothby 2017). Planning for heterogeneity can support populations of shorebirds that forage on mudflats or in tidal creeks (Armitage et al. 2007) while also providing critical habitat for marsh birds such as rails or other marshnesting species (Pickens and King 2013). However, that heterogeneity in the elevation and vegetation landscape may not persist over time as dense monocultures of native wetland vegetation often develop rapidly in non-vegetated areas (Ward et al. 2003, Armitage 2021).

Although there is broad agreement that elevation heterogeneity is likely to increase bird use of restoration projects, challenges remain in translating that accepted ecological paradigm into quantitative elevation guidelines for the engineers designing and constructing the sites (Armitage et al. 2024). These guidelines, including specific elevation targets and ratios of vegetated, mudflat and submerged habitat, must be informed by site-specific data on bird use of habitat types. Future research should aim to produce elevation guidance that can be used by practitioners in marsh construction projects. Monitoring protocols for shorebirds, wading birds, and others are well-developed in the region (Brush et al. 2019). Yet, monitoring efforts are pervasively underfunded, yielding piecemeal data that does not completely cover the scope of bird use and health. This results in limited applicability to management in practice (Schulz 2021). Studies focused on longer-term use and life history patterns of birds could help to unravel complexities in bird habitat use; this knowledge can then be used to improve management decisions (Schulz 2021). Further, such in-depth studies would establish a robust baseline, allowing for more informed responses to future disturbances or changes in management needs.

Author Contributions

AA: Conceived the study, collected and analyzed data, wrote the paper. AGH: analyzed data, edited paper. AM, MP, JT: collected data, edited paper.