Introduction

Warming of the climate system is unequivocal, as is now evident from observations of increases in global average air and ocean temperatures, widespread melting of snow and ice, and rising global average sea level (IPCC 2007). At the same time, eutrophication is an important conservation problem in coastal ecosystems worldwide, altering habitat, energy fluxes, trophic networks, and community composition (deJonge et al. 2002; Bertness et al. 2007; Caffrey et al. 2007). Coastal marshes are among those ecosystems that are greatly affected by global climate change and eutrophication. For example, sea-level rise is causing extensive loss of salt marshes (Hartig et al. 2002; Morris et al. 2002; Nicholls 2004) and eutrophication may provoke dense algal growth that has been related to decline of salt marsh vegetation (Adam 1990).

Conservation and restoration of salt marshes is critical in the context of climate change because vegetation and sediments of salt marshes may act as carbon (C) sinks, reducing atmospheric carbon dioxide concentration (Caçador et al. 2004; Sousa et al. 2010a,b). Under global warming conditions, coastal wetlands are more valuable C sinks than any other wetland ecosystem because of their relatively high C sequestration rates and low methane emissions (Choi and Wang 2004). Moreover, salt marshes have been identified as strategically located nutrient sinks, mainly nitrogen (N), at the mouth of rivers that may reduce eutrophication impacts in estuaries and coastal waters (Romero et al. 1999; Wigand et al. 2007; Castro et al. 2009; Caetano et al. 2012).

Spartina species (cordgrasses) are among the most widespread halophytes in salt marshes around the world. Spartina alterniflora Loisel. (smooth cordgrass) has been widely studied showing that its prairies accumulate higher quantities of C than most grasses (Liao et al. 2007, Elsey-Quirk et al. 2011). S. alterniflora is also very efficient at N uptake, reducing eutrophication of estuaries in the Eastern North American portion of its native range (Patrick and DeLaune 1976; White and Howes 1994; McFarlin et al. 2008) and in China, where it is an invasive species (Wan et al. 2009; Zhang et al. 2009). In European salt marshes, native stands of perennial Spartina maritima (Curtis) Fernald (small cordgrass) also act as C sinks (Cartaxana and Catarino, 1997; Caçador et al. 2004; Lillebo et al. 2006). Populations of cordgrass can also reduce eutrophication through N accumulation (Caçador et al. 2007; Sousa et al. 2008; Castro et al. 2009; Sousa et al. 2010b) as the invasive Spartina densiflora Brongn. (dense-flowered cordgrass), an invasive species from South America, does in the Southwest Iberian Peninsula (Neves et al. 2010). However, the capacity of S. maritima to accumulate N in restoration plantings and expanding marshes is unknown. Furthermore, despite the fact that salt marshes are highly productive systems, there is an important lack of data about the potential of salt marshes for C storage (Chmura et al. 2003).

In this context, a specific goal of salt marsh creation and restoration projects may be to implement actions that will enhance C and N stocks to mitigate and offset negative effects of global climate change and eutrophication (Crooks et al. 2011). An innovative restoration project was carried out in the Odiel Marshes (Southwest Iberian Peninsula). Prior to our study, S. densiflora dominated this site and had been eradicated from most places. The overall objectives of the restoration project were to restore native vegetation and to phytostabilize sediments following removal of S. densiflora (Castillo and Figueroa 2009a). S. maritima was planted throughout the site to establish a native cordgrass-dominated community because S. maritima contributes effectively to sediment stabilization, while density and biomass of stands increase (Castillo et al. 2008a,b; Sousa et al. 2008; Caçador et al. 2009) at the same time that they facilitate development of ecological succession (Figueroa et al. 2003).

The primary objectives of this study were to quantify C and N stocks and sequestration rates in sediment and in vegetation dominated by restored stands of S. maritima. Along with these aims, our goal was to quantify biomass accumulation, sedimentation rate, and C and N content for sediments and plant tissue of S. maritima. Overall, our intent was to provide support for restoration and management strategies that may offset negative aspects of eutrophication and climate change in estuarine ecosystems.

Material and methods

Study site

The study was carried out in a restored salt marsh area, known locally as “Punta del Sebo”, which borders the main channel of the Huelva Estuary at the confluence of the Odiel and Tinto rivers in the European Atlantic basin (Southwest Iberian Peninsula; 37°08′–37°20′N, 6°45′–7°02′W) (Fig. 1). The Odiel Marshes are a significant wetland ecosystem recognized for their global significance with designations as a United Nations Educational, Scientific and Cultural Organization (UNESCO) Biosphere Reserve and the Ramsar Convention on Wetlands. The estuary has great historical significance for its shipping ports used by Phoenicians and ancient Greeks and as the origin of the 1492 and later expeditions led by Columbus to and from the Americas. These and later voyages are the likely source of accidental introductions of invasive South American S. densiflora that formed monospecific stands, displaced indigenous species, and transformed Odiel salt marshes (Nieva et al. 2001, Davy et al. 2009).The Odiel Marshes are now highly eutrophic due to nutrient loading from urban and agricultural runoff, and industrial processes (Elbaz-Poulichet and Dupuy 1999, Davy et al.). The plant community in our study site was restored from November 2006 to January 2007 with plantations of S. maritima and S. perennis. S. maritima clumps coming from natural populations were planted manually at a density of one clump per square meter (ca. 20 shoots per clump) after the invasive S. densiflora was eliminated manually from 2.00 ha around the site (Castillo and Figueroa 2009a).

Fig. 1
figure 1

Location of Odiel Marshes on the Atlantic coast of Southwest Iberian Peninsula (37°08′–37°20′N, 6°45′–7°02′W) and the restored area where our work was carried out (1)

Full size image

Every S. maritima population at channel edges in the Odiel Marshes have disappeared or have been degraded due to a combination of different anthropogenic impacts, erosion, and a limitation in the dispersion of S. maritima that very rarely establishes new tussocks by seeds (Castillo, Ayres et al. 2010). Thus, we decided to compare C and N stocks and sequestration for restored marshes with data from previous studies on Spartina marshes around the world.

Sediment and vegetation sampling

Sediment elevation relative to the Spanish Hydrographic Zero (SHZ) datum, redox potential, pH, electrical conductivity of interstitial water of sediments, organic matter content, and bulk dry density was recorded in S. maritima areas in May–June 2009 (n = 20). Every sediment characteristic was recorded between 0 and 10 cm deep, except for the redox potential that was sampled in surface (0–2 cm) and depth (2–10 cm). Elevation was surveyed to a resolution of 2 cm with a Leica NA 820 Theodolite (Singapore), and reference points were determined in relation to measurements of tidal extremes (Ranwell et al. 1964). Redox potential of the sediment was determined in situ with a portable meter and electrode system (pH/mV Crisonp-506). We recorded pH and electrical conductivity in the laboratory (Crison meter fitted with M-506 and 522 probes) after adding distilled water to the sediment (pH 1:1, v/v; conductivity 1:2, v/v). Sediment organic content was analyzed from triplicate subsamples by loss-on-ignition method after 4 h at 450 °C. Sediment bulk density was recorded by weighting (dry weight, DW) volume of sediments in 5 × 5-cm cylindrical cores. Sedimentation was determined by markers consisting in an iron goal (two upright poles joined by a crossbar) approximately 1.5 m tall, 0.5 m wide, and with poles of 1 cm in diameter, inserted to a depth of around 1 m in S. maritima areas. The distance from the middle of crossbar to the sediment surface was measured quarterly from March 2009 to March 2010 (n = 9). This prevents measurement errors due to the erosive vortex generated in the base of the markers (Curado et al. 2012). Sedimentation rate between consecutive measurements was calculated as the difference between the initial and final distance (in centimeters) divided by the number of years.

S. maritima plant material and sediment samples (between 0–2 cm and 2–20 cm deep, where roots were concentrated) for C and N content analysis were collected in July 2008 in S. maritima areas in two 10-m-long rows established parallel to the tidal line between +2.16 and +2.67 m over SHZ (n = 10; five equidistant sampling points per row). Superficial sediments (0–2 cm deep) were collected using nylon horizons set up 6 months earlier to ensure they were recently deposited. Previous studies have reported that S. maritima C concentration does not change seasonally (Cartaxana and Catarino 1997; Caçador et al. 2004).

S. maritima cover was recorded by contact every 2 m along nine 60- to 80-m-long stratified random transects established perpendicular to the tidal line from the lower distribution limit of S. maritima (ca. +1.5 m SHZ) to the upper border of the Spartina band (n = 9). Total area occupied by S. maritima was determined by the software ArcGIS 9 (ESRI 2008) after recording the distribution of small cordgrass in the field using a global positioning system (GPS) (model eTrex Vista Garmin). Above- and belowground biomass (AGB and BGB, respectively) were recorded in October, coinciding with the period of maximum biomass accumulation (Castellanos 1992), in 10-cm quadrant plots with monospecific cover of 100 % (n = 10). S. maritima biomass was washed carefully, separated into leaves, stems, roots, and rhizomes and dried for 48 h at 80 °C to achieve a constant DW. Then, plant material was weighed (Pinnacle P-403 balance, Denver Instrument, Denver, CO, USA) to quantify biomass. Net annual standing above- and belowground productivity (NAPP and NBPP) for S. maritima were calculated as the total AGB or BGB, respectively, divided by years since transplantation. Sampling plots for biomass were located in areas with bare sediments adjacent to clumps just after transplantation to ensure that all the standing biomass was effectively produced in situ after restoration plantings (Castillo, Leira-Doce et al. 2008).

Carbon and nitrogen content analysis

Plant samples were separated into photosynthetic organs, nonphotosynthetic stems, roots, and rhizomes. Sediment and plant samples were dried for 48 h at 80 °C and ground (Cyclotec, Foss Tecator AB, Höganäs, Sweden) to pass through an 80-μm sieve. Total C content (milligrams per gram DW) and total N content (milligrams per gram DW) were determined for undigested samples using an elemental analyzer (Leco CHNS-932, Spain). The value for each sample corresponded to the mean of three replicated measurements.

Carbon and nitrogen accumulation rates and stocks

C and N pools were calculated for every plant organ (biomass × N or C content) and for the sediment between 0–2 cm and 2–20 cm deep (sediment mass × N or C content). Sediment C and N stocks were calculated for each of two depths (0–2 and 2–20 cm) as the product of the mass of sediment (obtained as the product of the bulk density and the total volume (area × sediment depth)) and the sediment C or N content. Annual C and N accumulation in the sediments was calculated as the product of the mean annual sedimentation rate and the sediment C and N content, respectively, in the upper 2 cm deposited on nylon horizons. Soil C and N density was calculated as the product of sediment C and N content and sediment bulk density.

S. maritima C and N stocks were calculated for each organ as the product of their mean biomass density (biomass × occupied area × relative cover) and C and N content, respectively. Annual C and N accumulation in S. maritima biomass was calculated taking into consideration both NAPP or NBPP and its C and N content, respectively.

Statistical analysis

Analyses were carried out using SPSS release 12.0 (SPSS Inc., Chicago, IL, USA). Deviations were calculated as the standard error of the mean (SEM). Data were tested for normality with the Kolmogorov–Smirnov test and for homogeneity of variance with the Levene's test (P > 0.05). When homogeneity of variance between groups was not found, data were transformed using the following functions: ln(x), 1/x, and √x. Student's t test for independent samples was applied to compare two means. If homogeneity of variance or normality was not achieved by data transformation, then means were compared using Mann–Whitney U test. One-way analysis of variance (ANOVA) was applied to compare more than two means. Tukey's honestly significant difference (HSD) test between means was calculated only if F test was significant (P < 0.05). If homogeneity of variance was not achieved by data transformation, then the means were compared by Kruskal–Wallis H test nonparametric ANOVA.

Results

Carbon and nitrogen contents

Abiotic environment data for the restored S. maritima marshes are shown in Table 1. Sediment redox potential was similar at surface and at depth (t test, P > 0.05). Sediment C content was lower at surface (0–2 cm: 11 ± 1 mg C g−1) than at depth (2–20 cm: 16 ± 1 mg C g−1) (t test = −2.266, d.f. = 18, P < 0.05). Minimums were 5 mg C g−1 at the surface and 9 mg C g−1 at depth, and maximums were 19 mg C g−1 at the surface and 24 mg C g−1 at depth. In contrast, sediment N contents were similar at surface (1.6 ± 0.3 mg N g−1) and at depth (2.1 ± 0.2 mg Ng−1) (Mann–Whitney U test, P > 0.05), varying between 0.8 and 3.1 mg N g−1.

Table 1 Elevation above Spanish Hydrographic Zero (m), sediment redox potential (mV), electrical conductivity (mS cm−1), pH, organic content (%), bulk density (g cm−3), and sedimentation rate (cm year−1) in the Spartina maritima areas of restored salt marshes in the Odiel Marshes (Southwest Iberian Peninsula) (n = 9–20)
Full size table

Restoration plantings of S. maritima showed significant differences in C content between organs (ANOVA or Kruskal–Wallis, P < 0.001). The highest values were recorded in leaves and stems (ca. 420 mg C g−1) and the lowest in roots (361 ± 4 mg C g−1). S. maritima also concentrated more N in its leaves (31 ± 1 mg N g−1) than in the other organs (ANOVA, F = 19.415, P < 0.001, d.f. = 39) (Table 2).

Table 2 Carbon and nitrogen contents, biomass (dry weight), and carbon and nitrogen stock density for leaves, stems, roots, and rhizomes in restored salt marshes using Spartina maritima (8.37 ha with a relative cover of 62 %) 28 months after transplanting. Different coefficients indicate significant differences between organs (ANOVA, P < 0.05)
Full size table

Carbon and nitrogen accumulation and stocks

The total volume of sediment colonized by S. maritima was 1,674 m3 between 0 and 2 cm and 15,066 m3 between 2 and 20 cm, accumulating 14.7 and 192.8 t C, respectively, corresponding to 2,480 g C m−2 in S. maritima areas in the first 20 cm of sediment. Sediment C density was 0.009 g C cm−3 at the surface and 0.013 g C cm−3 at depth of 2–20 cm. Sediment N pool of the Spartina areas was 2.14 t N in superficial sediments and 25.31 t N at depth. Sediments (2,176 m3) were accumulated annually in areas colonized by S. maritima (8.37 ha) representing an annual accumulation of 19.1 t C and 2.8 t N.

No evidence of herbivory (that would reduce biomass and change allocation patterns) was observed during the study. Total biomass accumulated by S. maritima in monospecific stands formed after restoration was ca. 2 kg DW m−2, accumulating more biomass in aerial stems (935 ± 145 g DW m−2) than in rhizomes (424 ± 60 g DW m−2), leaves (356 ± 53 g DW m−2), or roots (192 ± 44 g DW m−2) (Kruskal–Wallis, χ 2 = 24.905, P < 0.05, d.f. = 3). Relating these biomass data with the respective C and N contents, S. maritima stems accumulated more C and N than the other organs (C: Kruskal–Wallis, χ 2 = 25.835, P < 0.01, d.f. = 3; N: Kruskal–Wallis test, χ 2 = 28.143, P < 0.001, d.f. = 3).

Twenty-eight months after transplanting, S. maritima colonized 8.37 ha with a relative cover of 62 ± 6 %, accumulating ca. 40.2 t C and ca. 2.5 t N; ca. 50 % of these C and N stocks were stored in stems. Small cordgrass prairies showed a NBPP of 264 ± 42 g DW m−2 year−1 and a NAPP of 553 ± 83 g DW m−2 year−1, which corresponded to 104 g C m−2 year−1 in BGB and 228 g C m−2 year−1 in AGB and to a total N accumulation of ca. 20 g N m−2 year−1 (Table 2).

Total C stock was 247.7 t C (29.6 t C ha−1); ca. 16 % was stored in S. maritima tissues. The C stock was increasing annually at 19.1 t C from sedimentation and 17.2 t C due to S. maritima expansion. Total N stock in 8.37 ha of the restored salt marshes colonized by S. maritima, including their sediments in the first 20 cm depth, was 30.25 t N (3.6 t N ha−1), with 8.3 % stored in the small cordgrass. N stock of 2.8 t N was being added annually by deposition of sediments and ca. 1.0 t N was sequestrated by Spartina colonization (Table 2).

Discussion

This study shows the capacity of European restored salt marshes planted with small cordgrass to sequestrate C and N in their vegetation and sediments.

Carbon and nitrogen in sediments

Sediment C content in the studied marshes was still low 28 months after transplanting in comparison with mature marshes of S. maritima and similar to the values recorded in young marshes of Spartina anglica C. E. Hubbard (common cordgrass) (Table 3). Sediment C budget in Spartina areas (2.5 kg C m−2 in the top 20 cm) was lower than those reported for S. maritima, S. alterniflora, Spartina patens (Aiton) Muhl. (salt meadow cordgrass) natural marshes (Table 3). Sediment C density (between 0.009 and 0.013 g cm3) was also lower than the average soil C density of salt marshes (0.039 ± 0.003 g cm−3) (Chmura et al. 2003).

Table 3 C and N contents and stocks in sediment, above-, and belowground biomass of S. maritima-restored marshes in the Odiel Marshes (Southwest Iberian Peninsula) in comparison with natural marshes of other Spartina species
Full size table

Sediment N content was within the range reported for bare sediments in S. maritima marshes (1.8 mg N g−1 following Castro et al. (2009)) and higher than in S. maritima Portuguese marshes (Table 3). In this sense, sediment organic content was ca. 5 %, while values ca. 11 % have been recorded for other S. maritima marshes (Lillebo et al. 2006; Castillo, Leira-Doce et al. 2008). However, our sediment N contents were comparable to those reported for S. alterniflora marshes in the USA (Table 3). Sediment N stock in S. maritima restored marshes was similar to that recorded for S. patens and S. alterniflora marshes in the USA (Table 3).

Carbon and nitrogen in S. maritima aboveground biomass

C content for S. maritima AGB was similar to that reported for S. maritima, S. alterniflora, and S. patens natural populations (Table 3). High C contents are related to fairly rigid walls that limit the amount of water influx during hypoosmotic stress, which would be beneficial in relatively stable saline environments (Touchette 2007), such as low elevations in the tidal gradient colonized by S. maritima (Castillo and Figueroa 2009b).

N content in S. maritima leaves and shoots was higher than those recorded in natural populations in the same season (Caçador et al. 2007; Sousa et al. 2008; Castro et al. 2009) and than those recorded for S. alterniflora in USA marshes (Patrick and DeLaune (1976) and McFarlinet al. (2008)) and in Chinese marshes (Liao et al. (2007)). Based on our results, it is not possible to identify the environmental factors determining the exposed interpopulation differences in N content. On one hand, N bioavailability in the sediments is a key factor determining N acquisition by halophytes (Darby and Turner 2008). On the other hand, N content depends on the age of the population, with young populations of S. maritima showing higher N contents in their AGB than mature populations (Caçador et al. 2007).

S. maritima AGB values in fully colonized patches 28 months after transplantation (1,290 ± 194 g DW m−2) were similar to those recorded for natural populations (Castillo, Rubio-Casal and Figueroa 2010). Created populations of S. maritima develop similar AGB to natural populations within ca. 2 years after transplantation (Castillo, Leira-Doce et al. 2008). AGB C stock in S. maritima biomass was also similar to those recorded in S. patens marshes and higher than in S. alterniflora prairies, while AGB N stock for restored S. maritima prairies was higher compared with other Spartina species (Table 3).

S. maritima NAPP (553 ± 83 g DW m−2 year−1) was within the wide range recorded for natural populations (120–2,800 g DW m−2 year−1 following Sousa et al. (2010a,b)) and for created marshes (131–590 g DW m−2 year−1 following Castillo et al. (2008a)). C accumulation rate of S. maritima AGB in the restored marshes (228 g C m−2 year−1) was also within the range reported for natural marshes (ca. 50–1,800 g C m−2 year−1) (Sousa et al. 2010a,b) and for the seaside alkali grass (Puccinellia maritima (Huds.) Parl.) in European marshes (Bouchard and Lefeuvre 2000). The recorded NAPP of S. maritima corresponded to a N accumulation of ca. 15 g N m−2 year−1, with relatively low sediment N content (ca. 2 mg N g−1). Sousa et al. (2008, 2010b) described N sequestration by S. maritima in the AGB that varied between ca. 2 g N m−2 year−1 for matured marshes with ca. 6 mg N g−1 in sediment and ca. 48 g N m−2 year−1 with a sediment N content of ca. 3 mg N g−1. Elsey-Quirket al. (2011) also described similar AGB N accumulation for S. patens and S. alterniflora marshes (ca. 11 g N m−2).

Carbon and nitrogen in S. maritima belowground biomass

Our C content data for S. maritima BGB coincided with that reported for S. alterniflora (Table 3). The root system of S. maritima was the organ that showed the lowest C content, coinciding with Cartaxana and Catarino (1997); however, their value was much lower than ours (179 vs. 361 mg C g−1). N content of S. maritima BGB was slightly higher than that reported for Portuguese S. maritima marshes and much higher than for Spartina foliosa Trin. (California cordgrass) and S. alterniflora (Table 3).

S. maritima BGB values (616 ± 98 g DW m−2) were much lower than those recorded in matured populations (ca. 2,500–4,800 g DW m−2) (Castillo, Rubio-Casal and Figueroa 2010). Therefore, the analyzed restored marshes had still not raised its maximum BGB even in totally colonized areas, in agreement with Castillo et al. (2008a), who reported that S. maritima transplants need between 2.5 and 4.0 years, depending on the sedimentation dynamic, to develop a similar BGB to natural marshes. The BGB C stock for S. maritima was lower than for S. patens and S. alterniflora marshes (Table 3). On the other hand, BGB N stock in the restored marshes was similar, and even higher, than in S. alterniflora and S. patens marshes in the USA (Table 3).

The NBPP of S. maritima (264 ± 42 g DW m−2 year−1) was close to the minimum values reported for created populations in the Odiel Marshes (366–3,598 g DW m−2 year−1) (Castillo, Leira-Doce et al. 2008) and lower than those recorded in natural populations (700–3,500 g DW m−2 year−1) with similar C contents, resulting in lower C accumulation rates in BGB (104 g C m−2 year−1 vs. 239–1,008 g C m−2 year−1) (Sousa et al. 2010a,b). The relatively low NBPP recorded in the restored marshes was consistent with an accretion rate of 2.2 ± 0.3 cm year−1 (Castillo, Leira-Doce et al. 2008). The NBPP of S. maritima corresponded to a N accumulation of ca. 6 g N m−2 year−1. Belowground N sequestration in Portuguese estuaries varied between 6 g N m−2 year−1 in young marshes and 45 g N m−2 year−1 in mature marshes (Sousa et al. 2008). Elsey-Quirk et al. (2011) described BGB N accumulation between 17.5 and 22.5 g N m−2 for S. patens and S. alterniflora marshes.

Total carbon and nitrogen stocks in S. maritima restored marshes

C stock in S. maritima biomass 28 months after planting (ca. 0.8 kg C m−2) was lower than that reported for S. alterniflora Chinese marshes (ca. 3.0 kg C m−2) due to higher biomass accumulation with similar C content (Liao et al. 2007). S. patens (ca. 0.9 kg C m−2) and the short-form of S. alterniflora (ca. 0.7 kg C m−2) showed similar values to S. maritima restored marshes (Elsey-Quirk et al. 2011). Thus, total C stock in S. maritima restored marshes was still low 28 months after planting (29.6 t C ha−1), when compared with natural marshes (e.g., 156–166 t C ha−1 for S. patens and S. alterniflora prairies following Elsey-Quirk et al. (2011); 209 t C ha−1 for S. alterniflora areas in the first 100 cm of sediments, following Liao et al. (2007)). On the other hand, Liao et al. (2007) reported total N stock for S. alterniflora stands of ca. 46 g N m−2, very similar to our results (47 ± 7 g N m−2). Sediment N stock (30.25 t N in the top 20 cm of 8.37 ha) was more than 10 times that stored in plant biomass (ca. 2.5 t N). Although the sediment retains N, the effect of cleaning estuarine waters is performed mostly by the biota through bacterial denitrification and N uptake by vegetation (White and Howes 1994; Dollhopf et al. 2005).

A proportion of the C and N contained in the biomass of the expanding S. maritima plantations will be accumulated gradually as dead matter in the sediments, where it is captured for the long-term, and a further proportion will be exported in the form of dead leaves and shoots, mainly following the second year after plantations (Castillo, Leira-Doce et al. 2008), since its shoots are semelparous and their mean shoot life span is about 2 years (Castellanos et al. 1998). Our results showed that the restored marshes, 2.5 years after planting, were sequestering atmospheric C, mitigating impacts of climate change, and capturing N-reducing eutrophication. Moreover, the relatively low C and N contents recorded in sediments colonized by S. maritima, at cordgrass relative cover of 62 % and low BGB, showed that these marshes can still sequester much more C and N. Craft et al. (1988) reported that sediment C content in restored marshes changed more slowly than the development of Spartina. Cornell et al. (2007) suggested that, with proper construction, most major C fluxes can be established in created salt marshes with S. alterniflora in less than 5 years.

As sea level rises, if suspended sediments and primary production are sufficient, tidal marshes will also rise or move upslope. Thus, BGB influences soil elevation rise by subsurface expansion and sediment deposit stabilization (Ford et al. 1999) and plants' belowground accumulation of organic, rather than inorganic matter, governs the maintenance of mature salt marsh ecosystems in the vertical plane (Turner et al. 2004). In North America, estuarine salt marshes have been shown to sequester C at a rate more than 10 times higher per unit area that any other wetland ecosystem due to high soil C content and constant burial due to sea-level rise (Brigham et al. 2006). Since S. maritima is effective in uptake of atmospheric CO2 and because it also counteracts salt marsh inundation caused by sea-level rise, restoration of European salt marshes using small cordgrass should be actively pursued as a method to sequester C, and environmental managers should provide conservation corridors for dispersal and migration of S. maritima with habitat shifts due to sea-level rise.

S. maritima was planted in the restored marshes after eliminating invasive S. densiflora clumps (Castillo and Figueroa 2009a). In low marsh areas as those restored, S. densiflora is not able to grow at the lowest elevations (Castillo et al. 2000) and it establishes biannual populations composed by sparse and small tussocks at higher elevations (Castillo and Figueroa 2009b). S. maritima prairies established through plantations in low-elevation marshes replace bare sediments and biannual populations of S. densiflora, raising the C and N sequestration capacity of the marshes by increasing biomass production and accumulation. Following eradication of invasive S. densiflora stands to restore native plant community diversity in an Iberian salt marsh, restoration plantings of S. maritima stabilize sediments and help retain the positive contribution of past cordgrass stands, the deep long-term storage of C in deep sediments, while the newly established S. maritima community replaces an exotic invader in the important C storage cycle.