Burrowing crab effects across a tidal marsh successional chronosequence located along Mississippi sound

Burrowing crab engineers can affect the biological structure and sediment conditions of their environments. However, it is challenging to predict when and where burrow effects will manifest, as they are often site- and habitat-specific. We used a tidal marsh restoration chronosequence to explore crab burrow effects on plant communities (e.g., percent cover, biomass, stem heights) and sediment characteristics (e.g., bulk density, organic matter, carbon and nitrogen stocks) through early succession. In a field survey, we compared plants and sediments between plots with high and low crab burrow densities within three habitat zones: mud flat (i.e., 0-1 years old), young marsh (i.e., ~1-6 years old), and old marsh (i.e., ~6+ years old). In a manipulative experiment, we tested the physical effects of crab burrows on plants and sediments at the mud flat-young marsh ecotone using burrow mimics. In our field survey, crab burrow density did not influence plants or sediments. Rather, plant biomass and stem heights, as well as sediment bulk density, organic matter, carbon, and nitrogen, differed between habitat zones, following expected marsh successional trajectories of development over time. However, in our manipulative experiment, crab burrow mimics had a strong positive effect on plants at the ecotone, suggesting crab burrows can facilitate plant expansion into unvegetated mud flats. Thus, crab burrow effects appear to peak in early successional ecotones where burrowing mediates environmental stressors and promotes vegetative growth, with implications for the recovery of biological structure and sediment properties following restoration.

Methods
Site description

To evaluate the relationship between burrowing crabs, plant communities, and sediment conditions along a successional gradient, we used a tidal salt marsh chronosqeunce at Greenwood Island, Pascagoula, Mississippi (hereafter, GWI; 0°20'00.5"N 88°31'06.4" W). Specifically, GWI is a restored tidal marsh originally constructed through the beneficial use of dredged sediments in 2007 and was expanded with additional dredge-material applications from 2015 through 2021 (Mississippi Department of Marine Resources personal communications). This process produced a chronosequence of relative tidal marsh age since restoration, with older vegetated marsh habitat (i.e., ~6+ years old; hereafter, old marsh), younger vegetated marsh habitat (i.e., ~1-6 years old; hereafter, young marsh), and unvegetated mud flats (i.e., <1 year old; hereafter, mud flat). The Mississippi Department of Marine Resources restored this tidal marsh via beneficial use of dredged sediments, which created a unique, natural experiment that allowed us to evaluate marsh development in situ using a space-for-time approach in a model tidal marsh ecosystem. The marsh habitats at GWI are dominated by Spartina alterniflora Loisel (hereafter, Spartina), with some patches of Distichlis spicata Greene (hereafter, Distichlis) in older marsh habitats.

Fiddler crab burrows (i.e., Minuca longisignalis [Salmon and Atsaides 1968)], Minuca minax [Le Conte 1855)], Leptuca panacea [Novak and Salmon 1974], and Leptuca spinicarpa [Rathbun 1900]) are common in marsh and mud flat habitats at GWI, with densities ranging from ~0-50 burrows m-2 (Rinehart, Sharbaugh, and Dybiec personal observation). Purple marsh crabs, Sesarma reticulatum (Say, 1817), are also found in low densities in the oldest and highest elevation habitats at GWI, which were not included in this study due to the high potential for disturbance from ongoing management activities at the site. Overall, the community composition and relative densities of burrowing crabs observed at GWI are comparable to those observed throughout tidal marshes in the northern Gulf of Mexico.

Crab burrow density survey (file name-- survey)

            In November 2022, we conducted an observational survey in each habitat zone (i.e., mud flat, young marsh, old marsh) to evaluate the relationships between crab density, plant population traits, and sediment conditions. Specifically, in each habitat zone, we established five, 0.25 m2 (0.5m x 0.5 m, length x width) plots containing high crab burrow densities (i.e., ≥16 burrows m−2; hereafter, high crab plots) and five, 0.25 m2 plots with low crab burrow densities (i.e., ≤12 burrows m−2; hereafter, low crab plots; see Figure S1). All plots were at least 0.5 m apart, but they were generally clustered to reduce the natural variability in marsh structure. At the time of the survey, we documented the total number of crab burrows in each plot, and the diameter of up to five representative crab burrows. 

Plant populations: In each plot, we estimated the total percent cover of all species (i.e., Spartina and Distichlis) and the mean stem height of Spartina. We estimated the mean Spartina stem height by measuring five, randomly selected Spartina stems in each plot. We harvested the aboveground biomass in each plot by clipping stems at the air-sediment interface. Harvested plant material was then dried at 60˚ C to a constant weight. Additionally, to assess belowground biomass, we collected one sediment core (Ø = 7.9 cm, depth = 10 cm) from the center of each clipped plot. Belowground biomass cores were sub-sectioned in the field at 5 cm intervals and bagged. In the lab, belowground biomass subsections were rinsed and sorted by tissue type (i.e., roots, shoots, rhizomes) before being dried at 60°C to a constant mass.

Sediment properties: In each plot, we collected five, 24.5 cm3 soil cores to a depth of 10 cm using a Russian Peat Corer. We standardized the locations of each core within every plot to minimize bias. Specifically, we took one core in each corner and one in the center of each plot. Each core was then sub-sectioned at 2.5 cm depth intervals in the field and bagged. Sub-sections were dried at 60 °C to a constant weight to obtain bulk density. For analysis of bulk density, we calculated the mean bulk density of replicate cores collected at 1) 0-2.5 cm and 2.5-5 cm depths and 2) 5-7.5 cm and 7.5-10 cm depths to obtain plot-level mean bulk density at 0-5 cm and 5-10 cm, respectively. Once dried, samples were ground with a mortar and pestle and then ashed in a muffle furnace (six hours at 550˚ C) to estimate sediment organic matter (SOM) via loss-on-ignition. We pooled sediment from replicate cores collected at 1) 0-2.5 cm and 2.5-5 cm depths and 2) 5-7.5 cm and 7.5-10 cm depths to obtain a single plot-level sample representing 0-5 cm and 5-10 cm, respectively. These pooled samples were then sent to the Alabama Stable Isotope Laboratory for analysis of %C and %N using a Costech 4010 ECS interfaced with a Thermo Delta V isotope ratio mass spectrometer via a Thermo Conflo IV.

Crab burrow mimics at marsh ecotones experiment (file name-- experiment)

In May 2023, we established 20, 1 m2 plots along the boundary between the mud flat and young marsh habitat. Specifically, we placed one edge of the plot parallel with the vegetation line (hereafter, upper edge), so the entire plot extended into the unvegetated mud flat habitat and contained minimal vegetation at the beginning of the study (i.e., < 3% S. alterniflora cover). Plots were at least 1m apart and were placed in a line along the young marsh-mud flat ecotone (hereafter, ecotone).  In half of the plots (every other plot; n = 10; hereafter, crab burrow plots), we created crab burrow mimics by inserting a hollow, 1.5 cm diameter PVC pole vertically into the marsh sediment to a depth of 20 cm. We then removed the pole, creating a 1.5 cm diameter, 20 cm deep burrow mimic. We based the diameter of crab burrow mimics on ambient diameters observed along the ecotone at GWI (1.2 - 1.6 cm; 1.4 ± 0.04 cm; mean ± 1SE). The depth of crab burrow mimics was based on the 1) lower limits of Fiddler crab burrows observed in southeastern US tidal marshes (i.e., 16-75 cm depth; Aspey 1978) and 2) the general rooting depths of Spartina (i.e., 0-20 cm; Darby and Turner 2008). Crab burrow plots each received 20 crab burrow mimics, consistent with the density of crab burrows observed along the ecotone at GWI (range: 20-24 burrows m-2; 22.4 ± 0.98 burrows m-2; mean ± 1SE). We maintained crab burrow mimics monthly until the study concluded in November 2023. The other plots (n =10) did not receive crab burrow mimics and served as controls (hereafter, no burrow plots).

Plant populations: At the conclusion of the ecotone study, we estimated the total plant cover in each plot. We divided each plot into four quadrats (each 0.5 x 0.5 m; length x width) and estimated the total cover in each quadrant to understand the spatial variation in plant cover between the crab burrow and no burrow plots. To further understand how crab burrows influence plant growth at the ecotone, we quantified the distance that Spartina stems and rhizomes migrated into the unvegetated mud flat from the original vegetation edge. We quantified the distance that Spartina stems migrated by measuring the distance between the upper edge of the plot (i.e., the initial location of the plant ecotone) and the farthest Spartina stem at five, equidistant points (every 0.2 m) in each plot. We estimated the distance migrated by plant belowground biomass by taking three sediment cores, one at the center of the upper edge of the plot (0-10 cm from initial ecotone border, one at the middle of the plot (50-60 cm from initial ecotone border, and one at the center of the lower edge of the plot (90-100 cm from the initial ecotone border). We used a t-corer (i.d. = 10 cm) to take 5 cm deep cores at each of these points, and then searched through each core and noted the presence/absence of plant belowground biomass. If no belowground biomass was observed in any core, a zero was recorded for belowground expansion. We also estimated the mean height of Spartina in each plot by measuring the heights of five randomly selected stems.

Sediment properties: In November 2023, we collected two, 49.1 cm3 soil cores to a depth of 5 cm using a Russian Peat Corer. We standardized the locations of each core within every plot to minimize bias. Specifically, we took both cores ~0.25 m from the upper edge of each plot.  We took cores close to the original ecotone, as these areas were most likely to be colonized by Spartina over the 6-month study, meaning that these sediments should be more likely to show signs of organic matter and carbon stock development. All cores were dried at 60 °C to a constant weight to obtain bulk density and SOM, as described above.