NOAA Tech Memo NMFS NE 167:
Assessment and Characterization of Salt Marshes
in the Arthur Kill (New York and New Jersey)
Replanted after a Severe Oil Spill

Joseph J. Vitaliano^{17, 18} and Allen J. Bejda^{17, 19}

Postal Address: ¹⁷National Marine Fisheries Serv., 74 Magruder Rd., Highlands, NJ 07732
E-Mail Addresses: ¹⁸Joseph.Vitaliano@noaa.gov; ¹⁹Allen.Bejda@noaa.gov

INTRODUCTION

The ribbed-mussel is a dominant species in tidal salt marshes of the eastern United States. This bivalve mollusk attaches by byssal threads to the stems and roots of S. alterniflora and other substrates, and is usually most abundant within the tall S. alterniflora along the marsh edge (Bertness 1984). The ribbed-mussel enhances the survival of S. alterniflora by binding the root mat together, effectively stabilizing the substrate and strengthening the plant and the entire marsh against erosion (Bertness 1984). The ribbed-mussel also plays a major role in the food web and in the cycling of carbon, nutrients, and minerals through the saltmarsh ecosystem. This bivalve is a filter feeder on phytoplankton and suspended detritus (much of which is dead S. alterniflora), and a considerable amount of its fecal matter is deposited on the marsh surface. The ribbed-mussel is preyed upon by blue crabs (Callinectes sapidus), other crustaceans, gastropods, and birds (Bertness 1980, 1984).

Since ribbed-mussels are important in the ecology of tidal salt marshes, site-specific differences in abundance, biomass, age structure, growth rate, and the allometric relationships of ribbed-mussel populations can be explored as indicators of the success of marsh replanting. Sampling problems and restrictions, however, prevented us from quantitatively sampling the populations of ribbed-mussels at our replanted, unplanted, and reference sites in the Arthur Kill. We will therefore restrict data presentation and analysis to descriptive comparisons of several variables based on measurements of mussels taken from each site.

METHODS AND MATERIALS

A preferred method to sample ribbed-mussel populations quantitatively is to place quadrats of known dimensions randomly at similar tidal heights within the S. alterniflora zone at each site (Bertness 1980). The mussels are then destructively removed from each quadrat for enumeration and measurement. This sampling protocol could not be implemented in this study. We were not allowed to destructively sample directly in the replanted areas with the magnitude required to obtain quantitative samples of the mussel population. Mussels were also needed for determination of trace metals and hydrocarbon concentrations in the mussel meats (see "Methods" in Chapters II, "Trace Metal Contaminants in Sediments and Ribbed-Mussels," and III, "Petroleum Hydrocarbons in Sediments and Ribbed-Mussels"). To minimize damage to the replanted marsh sites and also to the unplanted and reference sites in the already impacted Arthur Kill, we attempted to make a single mussel collection at each site based on the sampling protocols needed for contaminant analysis of mussels. The mussel shells would be used for age and growth determinations, and the mussel meats would be used for contaminant analyses. This attempt failed to yield adequate numbers of mussels and it was necessary to make another collection for the age-and-growth portion of the study.

A minimum of 60 ribbed-mussels were randomly collected at each of the six sites during September 1996. At the reference sites, dense aggregations of mussels were found within a narrow area in the S. alterniflora zone. At the unplanted sites, mussels were sparse and had to be collected from both vegetated and unvegetated locations over a much larger area of the marsh and undoubtedly from different tidal levels. At the replanted sites, sampling was restricted to areas within the replanted zone, but away from permanent quadrats set up by SMRT to monitor the S. alterniflora.

In the laboratory, the sediment, epiphytes, and byssal threads were removed from each shell, the mussel was opened, and its tissue was removed. Dry weights for both shell and tissue were determined after drying to a constant weight at 60°C. After weighing, each mussel was measured with vernier calipers to determine shell length (maximum anterior-posterior dimension), width (maximum lateral dimension), and height (maximum dorsal-ventral dimension). Each mussel was aged by counting external growth rings (annuli). Length at age was determined by measuring the maximum anterior-posterior dimension at each annulus.

RESULTS

The information presented below summarizes the measurements made on mussels collected from each site. Since sampling was not quantitative, rigorous statistical comparisons among the sites were not performed.

The collections yielded 64 mussels at Old Place Creek, 60 at Saw Mill Creek North, 67 at Saw Mill Creek South, 53 at Con Ed Tower, 92 at Tufts Point, and 69 at Mill Creek. Age-frequency histograms (Figure 36) indicated that the distributions of ribbed-mussels from Old Place Creek and Saw Mill Creek North (replanted) were skewed toward younger individuals, while the distributions of mussels from Saw Mill Creek South (unplanted) and Mill Creek (reference) were more evenly distributed among age classes. Length-frequency histograms (Figure 37) indicated that Saw Mill Creek South (unplanted) and Mill Creek (reference) had larger individuals (up to 100 mm) compared to the other sites, and that the distributions of mussels from Con Ed Tower (unplanted) and Mill Creek (reference) were skewed toward larger individuals.

To reduce the variability associated with year-class differences when summarizing data on average shell dimensions, shell weight, meat weight, and allometric relationships, and when estimating growth rates, we only summarized data for individuals that were in the same age class. Based on the number of individuals in each age class at each site (Figure 36), we chose to compare separately these measures for the 2-, 3-, and 4-yr-old mussels.

For each age class, the average shell length, width, and height, shell weight, and body weight were greater at Mill Creek (reference) than at any other site (Table 13). Mussels from Mill Creek (reference) had the highest growth rates in all three age classes, while mussels from Saw Mill Creek North (replanted) had virtually the lowest growth rates in those age classes (Figure 38).

Although differences detected in the absolute growth of ribbed-mussels among sites may be due to the influence on growth of site-specific environmental factors, allometric growth measures may or may not reflect these same differences (Seed 1980). Two-, three-, and four-year-old mussels collected from Mill Creek consistently had higher body-weight-to-shell-weight ratios compared to Old Place Creek (Table 13). No large differences among sites were found in the other allometric ratios.

DISCUSSION

We will limit our discussion to a descriptive comparison of the mussel data from the two replanted sites, Old Place Creek and Saw Mill Creek North, and from the reference site, Mill Creek. Since all mussels from these three sites were collected within the S. alterniflora zone, there is a greater possibility that differences found in mussel measures indicate actual differences among these sites. There is less confidence that the data from mussels collected at Con Ed Tower and Saw Mill Creek South (both unplanted sites) accurately describe differences either between these sites or among all sites. At Con Ed Tower and Saw Mill Creek South, mussels were sampled over a wide tidal area that included both vegetated and unvegetated habitats. Tufts Point is not discussed because, although it is a reference site, it was affected to some degree by the oil spill; in certain areas, marsh grasses and mussels were destroyed, while in adjacent areas, there was no apparent effect.

The smaller, younger mussels collected at the replanted sites reflect life cycle and growth processes since the disturbance caused by the replanting procedures in 1992 and 1993. The older, larger mussels at Mill Creek (reference) represent cumulative life cycle processes over many generations at a site presumably unaffected by this particular oil spill.

Two-, three-, and four-year-old mussels from Mill Creek (reference) grew faster than same-age mussels at the two replanted sites. The slower growth rates at the replanted sites could be due to the longer-term effects of the oil spill, the disturbance caused by the replanting process, and/or the stage of maturity of the replanted marsh compared to the reference marsh. Other site-specific factors, ultimately controlled by differences in physical factors, may have also influenced growth rates. Other studies have attributed among-site variability in growth rates of ribbed-mussels to differences in shore level, temperature, salinity, current exposure, quantity and quality of food resources, and stress from contaminants (Seed 1980; Bertness 1984; Franz 1993; Franz and Tanacredi 1993). There are site-specific differences in current exposure, salinity, etc., in the Arthur Kill (C. Alderson et al., Salt Marsh Restoration Team, Natural Resources Group, New York City Parks, 200 Nevada Ave., Staten Island, NY, unpubl. data) but it is unknown to what extent, if any, these factors may have influenced growth rates of the ribbed-mussels compared to the effects of replanting.

Consistent with differences found in growth rates between the reference and replanted sites, mussels from Mill Creek (reference) were larger and weighed more than the same-age mussels at the planted sites. The lower meat-weight-to-shell-weight ratio found in mussels at Old Place Creek (replanted) compared to Mill Creek (reference) could be due to differences in stress levels among the sites that may cause mussels to differentially secrete and dissolve shell material related to weight fluctuations (Bertness 1984; Franz 1993).

Shell-shape allometric ratios do not appear to be sensitive measures of differences among marsh sites in the Arthur Kill, despite differences found in growth rates and overall size of ribbed-mussels at our sites. Similar results were reported by Bertness (1984) for three different sites along the Maryland coast, and by Franz (1993) in relationships at two shore levels in Jamaica Bay, New York.

The assessment of replanting success at sites in the Arthur Kill using measures of ribbed-mussel age structure, size-and-shape relationships, and growth rates was problematic due to sampling difficulties, and was confounded by site-specific differences in environmental factors. Depressed growth rates and sizes in mussels of the same age may be typical of the northern reaches of the Arthur Kill [Old Place Creek and Saw Mill Creek North (replanted)] compared to the southern areas [Mill Creek (reference)].

REFERENCES CITED

Bertness, M.D. 1980. Growth and mortality in the ribbed-mussel, Geukensia demissa. Veliger 23:62-69.

Bertness, M.D. 1984. Ribbed-mussels and Spartina alterniflora production in a New England salt marsh. Ecology 64:1794-1807.

Franz, D.R. 1993. Allometry of shell and body weight in relation to shore level in the intertidal bivalve Geukensia demissa (Bivalvia: Mytilidae). J. Exp. Mar. Biol. Ecol. 174:193-207.

Franz, D.R.; Tanacredi, J.T. 1993. Variability in growth and age structure among populations of ribbed-mussels, Geukensia demissa (Dillwyn) (Bivalvia: Mytilidae), in Jamaica Bay, New York (Gateway NRA). Veliger 36:220-227.

Seed, R. 1980. Shell growth and form in the Bivalvia. In: Rhoads, D.; Lutz, R., eds. Skeletal growth of aquatic organisms. New York, NY: Plenum Press; p. 23-65.

To Chapter 6