Distribution of modern salt-marsh foraminifera in the Albemarle–Pamlico estuarine system of North Carolina, USA: Implications for sea-level research

We described the distributions of foraminifera from ten physiographically distinct salt marshes in the Albemarle–Pamlico estuarine system, North Carolina using 193 surface samples. We defined elevation-dependent ecological zones at individual sites using cluster analysis and detrended correspondence analysis. Additionally, seven principal biozones of salt-marsh foraminifera were identified that have distinctive spatial distributions reflecting a pattern of salinity regimes caused by the current configuration of barrier-island inlets. High salinity sites along the southern Outer Banks are associated with sub-tidal calcareous assemblages, low marshes dominated by Miliammina fusca and high marsh environments defined by Haplophragmoides wilberti, Trochammina inflata and Arenoparrella mexicana. In contrast, lower salinity marshes have Ammobaculites spp. in sub-tidal settings, Miliammina fusca-dominated low marshes and high marsh settings characterized by Jadammina macrescens. Spatial variation of foraminiferal populations and the potential for biozones to migrate in response to changing inlet configuration and salinity, suggests that datasets of modern salt-marsh foraminifera from multiple environments would be appropriate for reconstructing Holocene relative sea level in North Carolina.     

Short-term sedimentation in Tagus estuary,Portugal: the influence of salt marsh plants

Short-term sediment deposition was studied at four salt marsh areas in the Tagus estuary. In areas covered with Sarcocornia perennis, Sarcocornia fruticosa, Halimione portulacoides and Spartina maritima and also in the non-vegetated areas, sedimentation was measured as the monthly accumulation of sediments on nylon filters anchored on the soil surface, from August 2000 to May 2001. Our experiments were used also to determine the influence of the different plant species in vertical accretion rates. Short-term sedimentation rates (from 2.8 to 272.3 g m⁻² d⁻¹) did show significant differences when the four salt marshes studied in the Tagus estuary were compared to each others. Salt marshes closer to the sediment sources had higher sedimentation rates. Our results suggest that the salt marsh type and surface cover may provide small-scale variations in sedimentation and also that sediment deposition values do change according to the position of the different plant species within the salt marsh. Sedimentation is an essential factor in salt marsh vertical accretion studies and our investigation may provide support to help forecast the adaptative response of the Tagus estuary wetlands to future sea level rise.     

Feeding habits of young predatory fishes in marsh creeks situated along the salinity gradient of the Schelde estuary, Belgium and The Netherlands

Fish and macrobenthos were sampled in four different marshes along the salinity gradient of the Schelde estuary, Belgium/Netherlands, to investigate the importance of marsh creeks as foraging grounds for the dominant, larger fish species. The total density and biomass of all the main macrobenthic taxa (Corophium volutator, Nereis diversicolor, Oligochaeta, Macoma baltica and Heteromastus fliliformis) were measured. The feeding habits of the larger predatory fishes (Platichthys flesus, Dicentrarchus labrax) were investigated. Qualitative and quantitative stomach analyses included the calculation of different indices, showing the niche breadth (as diet diversity) and the niche overlap (as similarity between the predators diet) for this habitat. These analyses showed that the two most important benthic prey species for P. flesus were C. volutator and N. diversicolor. D. labrax preyed upon a wider range of species, including C. volutator, N. diversicolor, Crangon crangon, Carcinus maenas and Orchestia spp. The stomach diversity of D. labrax and P. flesus showed differences between the marshes although there was no consistent pattern in diet composition, reflecting the opportunistic nature of feeding by these large predators. The fullness indices of both flounder and sea bass did not differ significantly along the salinity gradient and the estimated minimum consumption by these predators did not indicate a top-down control of the macrobenthic community. The salt marsh creeks seem to provide excess food for the visiting fish species. The benthic prey was present in very high abundances, which may suggest that the typical nursery species such as C. crangon and C. maenas, and early juveniles of P. flesus, D. labrax and Pomatoschistus microps were not preyed upon significantly. This supports the hypothesis that salt marsh creeks provide good refuge areas for nursery species against predation by larger fish.     

Salt marsh plant-microorganism interaction in the presence of mixed contamination

In this study we investigated under laboratory conditions, the combined effect of the presence of two contaminants (Cu and PAHs) and plants (Halimione portulacoides) on a salt marsh microbial community, in terms of genetic structure, abundance and capacity to remove those contaminants. Plants changed the microbial community structure (evaluated by ARISA), but only in the treatments without PAHs. Also, in the presence of plants, Cu displayed lower toxicity (ToxScreen test) than in its absence. Nevertheless, the presence of plants interfered with the degradation (decreasing it up to 45%) of higher molecular weight PAHs by sediment microorganisms. This was in agreement with the lower microbial numbers (quantified by DAPI) observed in the presence of the plants, pointing for a competition between plants and microorganisms for nutrients. These results are new for salt marshes, and highlight the need for fertilization in order to obtain optimal effects of rhizoremediation for this type of compounds.     

Disturbance of salt marsh vegetation by wrack mats in Great Sippewissett Marsh

To quantify disturbance to salt marsh vegetation, and to test the notion that disturbance and species richness are related, we studied disturbance of vegetation by 195 wrack mats that had become stranded over Great Sippewissett Marsh. The mats varied in area, thickness, residence time, and elevation of stranding. Mats that were large, that had residence times of 3–4 months, and were stranded at higher elevations produced the most damage, ranging from total eradication to a decrease in shoot density of underlying vegatation. Thickness had no effect. About 70% of mats caused no damage, and of those that did, damage was usually 50% or less of the area beneath the mat. Plant species were differentially distributed along the tidal excursion within the marsh, and there were more species (8–9) high in the intertidal range than lower (4–6 species). Vertical distribution of species, species richness, and rate of annual disturbance were not well correlated, although highest and lowest disturbance rates corresponded to highest and lowest species richness. Most of the increases in species richness, however, occurred over a very small range of disturbance, so narrow that it seemed unlikely that disturbance affected richness. It seems more likely that factors other than disturbance rate control vertical position of species in salt marshes. It also appears that the effect of disturbance by wrack is less important than has been claimed.     

Importance of foraminifera for the formation and maintenance of a coral sand cay: Green Island, Australia

 CaCO₃ production by reef-building organisms on Green Island Reef in the Great Barrier Reef of Australia is estimated and compared with the contribution of benthic foraminifera to the sediment mass of the vegetated sand cay. Major constituents of the cay are benthic foraminifera (mainly Amphistegina lessonii, Baculogypsina sphaerulata, and Calcarina hispida), calcareous algae (Halimeda and coralline algae), hermatypic corals, and molluscs. Among these reef-building organisms, benthic foraminifera are the single most important contributor to the sediment mass of the island (ca. 30% of total sediments), although their production of CaCO₃ is smaller than other reef-building organisms. Water current measurements and sediment traps indicate that the velocity of the current around Green Island is suitable for transportation and deposition of foraminiferal tests. Abundant foraminifera presently live in association with algal turf on the shallow exposed reef flat, whose tests were accumulated by waves resulting in the formation and maintenance of the coral sand cay. Accepted: 30 June 1999     

Salt stress limitation of seedling recruitment in a salt marsh plant community

Summary Seedling recruitment in salt marsh plant communities is generally precluded in dense vegetation by competition from adults, but is also relatively rare in disturbance-generated bare space. We examined the constraints on seedling recruitment in New England salt marsh bare patches. Under typical bare patch conditions seed germination is severely limited by high substrate salinities. We examined the germination requirements of common high marsh plants and found that except for one notably patch-dependent fugitive species, the germination of high marsh plants is strongly inhibited by the high soil salinities routinely encountered in natural bare patches. Watering high marsh soil in the greenhouse to alleviate salt stress resulted in the emergence of up to 600 seedlings/225 cm². The vast majority of this seed bank consisted of Juncus gerardi, the only common high marsh plant with high seed set. We tested the hypothesis that salt stress limits seedling contributions to marsh patch secondary succession in the field. Watering bare patches with fresh water partially alleviated patch soil salinities and dramatically increased both the emergence and survival of seedlings. Our results show that seedling recruitment by high marsh perennial turfs is limited by high soil salinities and that consequently their population dynamics are determined primarily by clonal growth processes. In contrast, populations of patch-dependent fugitive marsh plants which cannot colonize vegetatively are likely governed by spatially and temporally unpredictable windows of low salinities in bare patches.     

The input of foraminiferal infaunal populations to sub-fossil assemblages along an elevational gradient in a salt marsh: application to sea-level studies in the mid-Atlantic coast of North America

The suitability of marsh sites for sea-level studies was examined based on a field study along a transect from high to low marsh. Living foraminifera at Bombay Hook (Delaware, USA) are considered to be shallow infaunal (i.e., uppermost 10 cm). Peak concentrations were found at 1–10 cm in the high marsh, 1–5 cm in the intermediate marsh, and 3–5 cm in the low marsh. However, sporadic deep infaunal inputs in the low marsh could significantly contribute to the sub-fossil assemblage. In the upper 10 cm buried (death + sub-fossil) and living assemblages showed a strong correlation, and the seasonal pattern of the buried assemblage paralleled the living one, suggesting that the buried assemblage reflected the most recent reproductive inputs. The cumulative standing crop of each dominant species was used to estimate their contribution to the buried assemblage in order to assess if the community is vertically homogeneous and, therefore, if the infaunal production causes differential sub-surface enrichment. The results showed that a “shallow” (0–5 cm) infaunal contribution is able to explain much of the sub-fossil assemblage beneath the surface in the high and intermediate marsh plots. However, in the low marsh plot, the deep infaunal contribution was greatest and significantly affected the sub-fossil assemblage. Therefore, modern analogues for sea-level studies in mid-Atlantic North American marshes should include the uppermost 5 cm. The interval proposed for the high and intermediate marsh is thin enough that epifaunal species are not underrepresented and encompasses only ~8 years (based on burial rates) providing a high temporal resolution. Handling editor: J. Saros     

Small-scale pattern: an evaluation of techniques with an application to salt marsh vegetation

This paper examines the outcome of four different procedures for quantifying small-scale association patterns in vegetation, with respect to data from Church Island salt marsh, Menai Bridge, N. Wales. A main aim is the evaluation of two plot-less techniques, proposed by Stowe & Wade (1979). The first examines frequencies of juxtaposition, the second the frequencies of target species in the regions of occurrence of other species. The results of these analyses are compared with those of two plot-based techniques. A Reciprocal averaging ordination is also employed to describe medium-scale distributional pattern within the data set.Of the two plot-less methods, the species-region method is found to be more useful, particularly because it is capable of detecting asymmetric relationships. When using the recommended collapsed data format the plot-less methods operate at an inconsistent spatial scale approximating to the mean run or clump size of species along the sampling transects. This allows detection of sociological patterns but may obscure environmental relationships. The small-scale relationships depicted by the plot-less methods are in cases in accord with the plot-based results at a range of scales, in other cases they differ: in each circumstance providing insights into the nature of interspecific relationships.Inter-specific pattern in salt marsh vegetation has rarely been studied. The methodology employed here shows that relationships do exist at the small scale, and may allow hypothesis generation with respect to their causality and their role in the dynamics of the marsh system.     

Salt marsh colonization by a rocky shore invader: Balanus glandula Darwin (1854) spreads along the Patagonian coast

Balanus glandula, an east Pacific acorn barnacle from rocky shores, was introduced to Mar del Plata, Argentina more than 40 years ago and has spread over 17 latitudinal degrees southward. Here we report the first record of this species living in a soft-bottom environment colonizing the salt marsh plant species Limonium brasiliense, Spartina densiflora, S. alterniflora and Sarcocornia perennis. In addition, we describe the size frequency distribution, density and spatial distribution of the barnacles colonizing the different plant species. The size frequency distribution of Balanus showed a bimodal pattern in all plants. Barnacles were mostly large in S. densiflora, but small in S. alterniflora, with more balanced distributions of small and large barnacles on S. perennis and L. brasiliense. The highest density of barnacles was observed on S. perennis (x = 35.8 ind/cm², SD = 40.5) and S. alterniflora (x = 33.8 ind/cm², SD = 23), while the lowest on L. brasiliense (x = 1.5 ind/cm², SD = 1.18) and S. densiflora (x = 0.17 ind/cm², SD = 0.09). More than 90% of the barnacles on any given plant were found living. While barnacles colonized only the first few centimeters above the soil surface level in S. alterniflora and L. brasiliense, they reached their highest point on S. perennis. The finding of a rocky shore species successfully colonizing soft-bottom marshes within an invaded region brings new perspectives to discussions in biological invasion ecology, and raises additional considerations for coastal environmental management.     

Living foraminifera and total populations in salt marsh peat cores: Kelsey Marsh (Clinton, CT) and the Great Marshes (Barnstable, MA)

Common species of intertidal agglutinated benthic foraminifera in salt marshes in Massachusetts and Connecticut live predominantly at the marsh surface and in the topmost sediment (0–2.5 cm), but a considerable part of the fauna lives at depths of 2.5–15 cm. Few specimens are alive at depths of 15–25 cm, with rare individuals alive between 25–50 cm in the sediments. Specimens living between the sediment surface and 25 cm deep occur in all marsh settings, whereas specimens living deeper than 25 cm are restricted to cores from the lower and middle marsh, and have an irregular distribution-with-depth. Lower and middle marsh areas are bioturbated by metazoa, suggesting that living specimens reach these depths at least in part by bioturbation. High-marsh sediments in New England consist of very dense mats of Spartina patents or Distichlis spicata roots and are not bioturbated by metazoa. In this marsh region bioturbation by plant roots and vertical fluid motion may play a role in moving the foraminifera into the sediment. The depth-distribution of living specimens varies with species: living specimens of Trochammina inflata consistently occur at the deepest levels. This suggests that species have differential rates of survival in the sediment, possibly because of differential adaptation to severe dysoxia to anoxia, or because of differing food preferences. There is no simple correlation between depth-in-core and faunal diversity, absolute abundance, and species composition of the assemblages. It is therefore possible to derive a signal of faunal changes and thus the environmental changes that may have caused them from the complex faunal signal of fossil assemblages.     

The environmental consequences of climatic change on British salt marsh vegetation

The present relationship between sea level and the zonation of salt marsh vegetation is discussed in terms of the salt marshes of the Essex and Kent coasts. These marshes are already decreasing in area as a result of a number of different environmental pressures, including the sinking of the land relative to the sea, at a rate of about 3 mm per year, the result of isostatic adjustment following the last glaciation. Because most British salt marshes are backed by a sea wall the marshes can not respond to rising sea levels by migrating landwards, thus increasing the impact of sea level change. In view of this and of the importance of salt marshes as protection for the sea walls themselves, a conceptual model has been developed, of the likely impact of climate change and the resulting sea level rise, on British salt marsh vegetation. The basis of this approach is the assumption that a rise in sea level will cause the drowning of certain existing vegetation zones and their subsequent replacement by new vegetation types appropriate to the changed sea level. Estimates have been made of the likely impact of rises in sea level of 0.5, 1.0 and 1.5 metres on the five major vegetation zones identified in East Anglia. The validity of this approach is discussed, together with the likely additive effect of present degenerative changes observed in the Essex salt marshes. It is estimated that over the next 60 years a sea level rise of only 0.5 m, when existing degeneration is taken to account, would cause a loss of over 40% of the present area of salt marsh in Essex and probably also in Kent. These losses would mainly effect the higher salt marsh vegetation zones which would be replaced by pioneer communities. These predictions would be greatly magnified by larger rises in sea level. The wider ecological implication of these changes and some possible remedial measures are considered. These predictions are discussed in relation to the situation in the rest of East Anglia and for Britain as a whole.     

Ecological engineering, design, and construction considerations for marsh restorations in Delaware Bay, USA

Public Service Electric & Gas of New Jersey is restoring approximately 4050 ha of salt marsh along Delaware Bay, USA, to offset possible effects on fish populations in the Bay from their existing once-through cooling system. Planning for this effort started with addressing three questions: Do marshes contribute significantly to fish production? How much marsh produces how much fish? Which marshes should be restored? There is ample evidence that salt marshes produce fish. The area of marsh necessary to offset potential losses was calculated from a simple aggregated food chain model and multiplied by four to provide a comfort level to the regulatory agencies. Marshes chosen for restoration were former salt marshes at appropriate tidal elevations. Planning involved experts in marsh ecology, hydrology, and engineering working with the company and regulatory agencies to establish clearly defined goals for the project. Design followed the advice of the experts and construction was overseen to follow the design. Long-term follow up is through adaptive management that is scheduled to continue for about a decade, depending upon progress of the restoration toward its goals.     

Managing Salt Marshes for Mosquito Control: Impacts of Runnelling, Open Marsh Water Management and Grid-ditching in Sub-tropical Australia

Three salt marsh sites in south-east Queensland, Australia, have been modified for mosquito control. The problem species is Ochlerotatus vigilax, a vector of Ross River virus that is an epidemic polyarthritic disease. All sites have similar vegetation and tidal influences. Each site has a different form of modification to manage the mosquitoes: runnelling, Open Marsh Water Management (OMWM) and grid-ditching. Results are compared of a treatment and control experimental design, similar for each site, but analysed separately for each site, and over the same 3-year time period. Environmental variables monitored included: for the water table, the depth, salinity and pH; for the substrate, its moisture, salinity and pH; for the vegetation, the size and density of the dominant grass (Sporobolus virginicus) and the number of mangrove (Avicennia marina) pneumatophores. Crab activity was indicated by crab hole numbers. ANOVA analyses, comparing treatment and control at each site, indicated that runnelling had the least significant impact on the environmental variables, whereas most impacts were at the grid-ditched site. Of the 10 variables included here, 5 were affected by runnelling, 7 by OMWM and 9 by grid-ditching. Statistically significant results are summarised, compared to the relevant controls, as follows: Water table: the level was significantly higher near ditches in the grid-ditched site; pH was higher at the treatment in the OMWM site but was lower in the grid-ditched site; salinity was significantly lower in the runnelled and in part of the grid-ditched marsh. Substrate: the moisture was higher at both the OMWM and grid-ditched sites; pH was higher in the OMWM treatment but lower in the grid-ditched marsh; salinity was significantly lower at the runnelled and grid-ditched sites, but was higher near the ditch at the OMWM site. Vegetation and crabs: The grass Sporobolus virginicus was less dense in the treatments in both the OMWM and grid-ditched sites, but was taller in the runnelled site and in part of the grid-ditched site. There were fewer mangrove pneumatophores in the runnelled area, but more in the OMWM one. There was more crab activity in both the runnelled and OMWM sites. It was concluded that runnelling has least impact and that grid-ditching has the most, but that none of these appears to have destroyed the marsh environment.     

Marine invasion processes: interactions between native and introduced marsh snails

We test two hypotheses to explain the success of a biological invader in marshes on the Pacific coast of North America. The first hypothesis focuses on the competitive superiority of an invading species over native species; the second hypothesis focuses on the absence of ecological equivalents in the aboriginal biota, such that no competitive interactions are necessary to facilitate successful invasion. As a model system we experimentally manipulated the introduced pulmonate snail Ovatella myosotis and the native prosobranch snails Assiminea californica and Littorina subrotundata. Manipulations included pair-wise enclosures to determine of Ovatella depressed the growth of the native;species; pair-wise associations to determine whether or not snails demonstrated negative or positive clustering; removal experiments to see if native snails invaded regions occupied by Ovatella, and reciprocal transplants on a vertical gradient to determine the physiological breath of the target species. In addition to monitoring population abundances of all three species seasonally, dietary overlap of these gastropods was also examined. No evidence of competitive superiority by Ovatella was found. Ovatella possesses the physiological capacities to exist in a semiterrestrial environment and is therefore able to take advantage of resources unavailable to native gastropods. We conclude that the successful establishment of this Atlantic snail in the Pacific Northwest did not arise at the expense of native species.     

Salt marsh vegetation of the Western Mediterranean desert of Egypt

The extensive salt marshes in the plains and depressions of the western Mediterranean desert of Egypt were classified into three habitat types: A with shallow water table and high salinity, B with relatively deep water table and high salinity, and C transitional habitats in which salinity and water table are no controlling factors. Fourteen vegetation types were distinguished, each dominated by one or two species.The dominating life forms are chamaephytes in sites of high salinity, and therophytes in sites of low salinity. Spatial and temporal variations in the standing crop biomass were pronounced. The accumulation of material started during spring and reached a maximum in autumn, when photosynthetic activity was maintained to account for transpiration losses.There was a general trend of increasing salinity and concentration of different ions from habitat type A to habitat type C through habitat type B. The periodical variation in the water table was insignificant, while a significant drop in salinity and the concentration of different ions was detected in spring, which was attributed to the diluting effect of rain water during that season.Most species exhibited clear distribution patterns and their a bundance varied significantly along gradients of different factors. Simple correlations between the compositional gradients or the distributional behaviour of species and salinity or the concentration of individual ions were generally low, while correlations with combinations of ions in the form of ratios (notably sodium and potassium adsorption ratos) were higher.During the early stages of succession the building up of soil and the decrease in salinity are the most important factors while at more advanced stages, soil texture and calcium carbonate content become more decisive.Nomenclature follows Täckholm (1974)     

Biostratigraphy and paleoceanographical significance of the Neogene planktonic foraminifera from the Pelotas Basin, southernmost Brazil

Analysis of the planktonic foraminifers recovered from 1-SCS-2 drill-hole, Florianópolis platform (Pelotas Basin, southern Brazilian Atlantic Margin), allowed recognition of the Catapsydrax dissimilis, Catapsydrax stainforthi, Globorotalia fohsi robusta, Globorotalia mayeri, Globorotalia margaritae evoluta and Globigerinoides trilobus fistulosus Miocene and Pliocene zones and four important hiatuses. Correlation with well 1-SCS-3B and other zonal schemes, as well as recognition of early diagenesis instead of a reworking process, allowed confident age assignment. The Miocene foraminifers constitute a tropical/sub-tropical assemblage, whereas in the Pliocene, the presence of scarce species associated to subantarctic water masses suggests that the Malvinas Current reaches the area but it was not a controlling factor in the paleoenvironmental conditions.     

Contribution to lignocellulose degradation and DOC formation from a salt marsh macrophyte by the ascomycete Phaeosphaeria spartinicola

Abstract Specifically radiolabeled [¹⁴C-lignin] lignocellulose and uniformly [U-¹⁴C] lignocellulose from the salt marsh grass Spartina alterniflora were incubated with the ascomycete Phaesphaeria s spartinicola . This fungus is the predominant one found on decaying standing dead S. alterniflora leaves in the salt marsh ecosystem. After 45 days of incubation at 20°C, 3.3% of the lignin moiety was mineralized to ¹⁴CO₂ and 2.7% solubilized to DO¹⁴C. Mineralization of the polysaccharides was seven times faster than that of the lignin. About 22% of the radioactivity was evolved as ¹⁴CO₂ but merely 4% was solubilized to DO¹⁴C within the incubation time. Experiments monitoring the ergosterol content of the mycelium incubated with S. alterniflora plant material were done to elucidate the carbon conversion efficiency of the fungus as well as the influence of the cinnamyl phenols p -coumaric and ferulic acid on lignocellulose degration. After 21 days of incubation, P. spartinicola showed a growth yield of 0.45 and 0.38 with and without the additional cinnamyl phenols, respectively. Grown on unextracted S. alterniflora the fungus caused a loss of organic plant material of about 50% with a corresponding growth yield of 0.38 during the incubation period. Investigation of cupric oxide oxidation products of sound and degraded lignocellulose revealed a preferential utilization of the syrinyl and cinnamly phenols compared with vanilly phenols.     

Physical Evolution of Restored Breached Levee Salt Marshes in the San Francisco Bay Estuary

Since 1972 over 940 ha (2,300 ac) of leveed former salt marsh sites around San Francisco Bay have been restored to tidal action, purposely or through natural processes. The evolution of these sites can inform predictions of rates of marshplain evolution and establishment of tidal channel systems. A review of the history of 15 re‐flooded sites ranging in size from 18 to 220 ha (45 to 550 ac) and in age from 2 to 29 years indicates that marshplain vegetation with more than 50% cover was established at nine of the sites within 4 to 20 years. The remaining six sites aged 2 to approximately 20 years continue to be less than 50% vegetated. The evolution of these sites is consistent with the following simple conceptual model of the physical evolution of restored tidal marshes in subsided breached sites. Initially, deposition of estuarine sediment builds up mudflats that allow vegetation establishment once elevations are high enough for vegetation to survive. Sites that are initially lower in the tidal frame take longer to vegetate than those that are initially higher. Three factors appear to retard the time frame for vegetation establishment: limited estuarine suspended sediment supply, erosion of deposited estuarine muds by internally generated wind waves, and restricted tidal exchange. These factors affect evolution more significantly in larger sites. The comparatively short time frame for vegetation colonization and marshplain evolution experienced in earlier, smaller, and/or less subsided breached levee restorations may not necessarily be replicable by simple levee breaching on larger subsided restoration sites now being planned. Our review of the 15 sites also indicates that the formation of tidal channels within the marshes is greatly dependent on whether and how high the site was filled before breaching. Filled sites at high intertidal elevations (above approximately 0.3 m below mean higher high water) can vegetate quickly but after several decades may show little development of tidal channels.