Carbon dioxide (CO₂) represents about 74% of the global emissions of greenhouse gases and its atmospheric concentration has increased by 1.1% to 1.6% per year in the last three decades as a result of anthropic activities (Olivier et al. 2017). A fraction of this CO₂ is sequestered and stored by mangroves, tidal salt marshes and seagrasses within the sediments, the biomass living aboveground and belowground, and detritus, which contribute to the blue carbon pool. Carbon sequestration is among the most important ecosystem services provided by salt marshes, with an estimated global average accumulation rate in sediments of 245±26 gC m^-2 year^-1 (Chastain et al. 2018). However, factors such as climatic conditions, tidal range, vegetation type, maturity of the salt marsh, sediment type and benthos composition can modify the ecosystem functioning, and strongly affect the carbon storage capacity of a salt marsh, which may even change from sink to source (Mann 2009).

Salt marshes are among the most productive ecosystems worldwide and often have large abundances of benthic fauna (Levin and Talley 2002). On salt marshes, a low percentage of the plant primary productivity is consumed by herbivores, and most plant biomass is transformed into detritus and incorporated into the sedimentary matrix (Bergamino and Richoux 2015). Detritus is susceptible to microbial decomposition, and the associated carbon is partially incorporated into bacterial biomass, entering the benthic food web mostly through macrobenthos and meiobenthos consumption (Danovaro and Gambi 2002, Mann 2009). Salt marshes are often characterized by detritus-based food webs, in which benthic fauna is a link between detritus, decomposers, and higher trophic levels (Danovaro and Gambi 2002). Thus, benthic invertebrates are fundamental to the regulation of the ecological processes and functions of these ecosystems (Chen et al. 2009), especially those related to the transfer of carbon. Despite their importance, the macrobenthos and meiobenthos have a differential contribution to the carbon cycle and the flow of energy. A primary difference arises from the unequal amounts of carbon that these fractions consume and contain, but although the meiobenthos usually has a lower stock biomass than the macrobenthos, its higher dynamics and shorter turnover generates a high production that frequently exceeds that of the macrobenthos (Giere 2008)

Spartina alterniflora (Loisel.) is a common dominant species on salt marshes of the southwest Atlantic (Isacch et al. 2006). In the Bahía Blanca estuary, this species has increased its cover during the last few decades, forming new marshes by colonizing bare tidal flats (Pratolongo et al. 2013). The rate of expansion of S. alterniflora marshes in the Bahía Blanca estuary is similar to that found in areas where this species has been introduced as exotic (Pratolongo et al. 2013), and it has been proposed that this species is not native to South America but was introduced from North America or Europe in the early 19th century (Bortolus et al. 2015). As an ecosystem engineer, S. alterniflora is able to modify key properties of the physical and chemical environment, such as hydrology, particle movements and habitat characteristics (Levin and Talley 2002). In locations where S. alterniflora is considered an introduced species, the colonization of unvegetated tidal flats has been shown to strongly influence the structure and function of benthic fauna (Neira et al. 2005). Changes in the abiotic conditions, such as the sediment organic matter or grain size, have a major influence on the composition and distribution of invertebrates in these environments (Santos et al. 2020). In the Jiangsu coastland (Chine), Zhou et al. 2009 reported that the expansion of S. alterniflora decreased the macrobenthos diversity of the tidal flats, and forced the niche of native species composing the macrobenthos on the tidal flats to move seaward.

The effect of S. alterniflora colonization on benthic organism communities has not been studied in the study area. Therefore, the main hypothesis of this study was that the community structure and the organic carbon contained in the benthic meio- and macrobenthos change between tidal flats and salt marshes recently colonized by S. alterniflora. Accordingly, the objectives of this study were to compare tidal flats and S. alterniflora salt marshes in terms of (1) the community structure of macrobenthic and meiobenthic communities; and (2) the organic carbon contained in these fractions of the benthic fauna.

Our study demonstrated that the expansion of S. alterniflora and the replacement of bare tidal flats with salt marshes have a significant effect on the macrobenthos and meiobenthos communities. The changes involve their structure, in terms of taxonomic composition, and the amount of organic carbon contained in the benthic organisms. This finding is in agreement with several studies that have reported changes in the macrobenthos community after colonization of natural habitats by S. alterniflora (Brusati and Grosholz 2006, Su et al. 2020, Lu et al. 2022).

Regarding the macrobenthos, the presence and density of the dominant taxonomic groups in our study area were similar to those reported elsewhere on salt marshes (Chen et al. 2009, Santos et al. 2020). The density and biomass of Gastropoda were higher on the salt marsh than on the tidal flat, and Amphipoda were present only on the marsh. Gastropoda (represented by a single epifaunal species, Heleobia australis) was the characteristic taxon and accounted for 88% of the total macrobenthos on the salt marsh. Carcedo and Fiori (2011) also found a higher density of H. australis on salt marshes of S. alterniflora than on adjacent tidal flats in Bahía Blanca estuary. Molluscs were also found to be the dominant group on S. alterniflora marshes from a wide variety of geographical locations (e.g. Jiangsu coast, Chine, Ge et al. 2020; Paranaguá bay, Brazil, Netto al. 2018). Several studies (e.g. Grudemo and Bohlin 2000; De Francesco and Isla 2003) have demonstrated that the distribution of H. australis is constrained by variables of the physico-chemical (salinity, temperature and type of sediment) and biological environment (competition, parasitism and predation). The aboveground biomass of S. alterniflora may provide refuge and/or trophic support for benthic organisms (Brusati and Grosholz 2006). Canepuccia et al. (2007) suggested that temperature and dehydration may be key factors that regulate H. australis distribution, so the presence of vegetation would reduce stress, resulting in higher densities and biomass. Furthermore, the presence of macrophytes with a complex architecture increases the spatial heterogeneity, protecting the macrobenthos from predation, especially epibenthic organisms whose visibility would be reduced (Lewis and Eby 2002). Thus, the increase in the colonization of S. alterniflora on tidal flats can enhance the density and biomass of H. australis in the study area.

The density and biomass of Polychaeta were higher on the tidal flat, where it was the characteristic taxon (representing 53% of all individuals). While the aerial structures of S. alterniflora plants provide epibenthic organisms with refuge from predation and ameliorate physical conditions leading to desiccation stress, roots and rhizomes reduce habitat suitability for some infaunal taxa (Brusati and Grosholz 2006). For instance, it has been reported that dense root mats may reduce the density and biomass of infaunal invertebrates by decreasing the space available for burrowing (Neira et al. 2005, Brusati and Grosholz 2006). On the tidal flat, Laeonereis culveri was the most abundant species within the Polychaeta. This is the common dominant deposit-feeder of intertidal macrobenthic communities in estuaries of the Buenos Aires Province, Argentina (Martin and Bastida 2006). The organic matter in the sediment plays an important role as a food source of this polychaete and its biomass and secondary production could be dependent on the amount, quality and availability of organic matter in the sediment (Martin and Bastida 2006). Differences in the species composition are accompanied by differences in diet and feeding modes. The most abundant organism on tidal flats (Polychaeta) are mainly consumers of detritus, while most organisms on salt marshes (Gastropoda) consume primarily periphyton (Albertoni et al. 2001).

It is recognized that the colonization of bare tidal flats with S. alterniflora can positively or negatively alter the species richness and biomass of the meiobenthos (Wang et al. 2010, Lin et al. 2015). S. alterniflora may lead to severe physico-chemical and biological alterations of key sediment properties, such as grain size composition, benthic microalgal biomass and water and organic matter content, thus affecting the meiobenthos biodiversity (Neira et al. 2005, Lin et al. 2015). Meiobenthic organisms are especially sensitive to anoxic conditions, so they commonly inhabit the first few millimetres close to the sediment surface. However, the radial oxygen loss from S. alterniflora roots creates an oxygenated micro-environment at greater depths, favouring larger meiobenthos densities and a higher diversity in association with the rhizosphere (Wardle et al. 2001, Lin et al. 2015). In our study, meiobenthos density and total biomass were similar on the salt marsh and tidal flat, and values were within the ranges reported elsewhere on salt marshes (e. g. Gulf of Mexico, southern Brazilian coast) (Wardle et al. 2001, Netto et al. 2018). However, when separate taxa were considered, Crustacea, Turbellaria and Foraminifera showed significant differences between habitats. Crustacea and Turbellaria were only present on the salt marsh, possibly responding to the forementioned physico-chemical changes in the sediment that occurred after plant colonization.

Foraminifera, by contrast were only present on the tidal flat. The duration and frequency of tidal inundation have been pointed as the most important variables controlling the distribution of Foraminifera within the intertidal zone (Edwards et al. 2004). The close association between foraminiferal species and inundation levels has been widely used to define foraminiferal zones that provide accurate indicators of former sea levels in intertidal deposits (e.g. Horton et al. 1999). It should be noted that plant colonization may also introduce changes in the microtopography (a few centimetres) at the same elevation within the intertidal that have not been considered in this work. In this same location (Villa del Mar), Calvo-Marcilese and Pratolongo (2009) found large differences in the abundance and number of foraminiferal species at different elevations within the same salt marsh. In that study, a hundredfold increase in the abundance of Foraminifera was reported for samples collected in the lower limit of the salt marsh compared with samples obtained at mid-marsh elevations. Therefore, the differences observed between tidal flats and salt marshes may be due to local changes in the microtopography and the associated environmental variables (de Rijk and Troelstra 1997).

According to our results, the carbon stock contained in the macrobenthos (total macrobenthos biomass) was approximately 11 gC m^-2 on the tidal flat and 39 gC m^-2 on the salt marsh. The meiobenthos, in turn, represented only 0.02 and 0.04 gC m^-2 on the tidal flat and salt marsh, respectively (0.34% and 0.05% of the carbon pool associated with the macrobenthos). Despite their considerably smaller biomass, meiobenthic organisms have a higher turnover rate (P/B ratio), so they play a disproportionally large role in the trophic transfer of carbon and may be responsible for a large proportion of the secondary production (Sellanes et al. 2003). That is, there is an inverse relation between weight and production in the meiobenthos and macrobenthos. The production and energy flow are considerably higher in meiobenthos than in macrobenthos due to the more effective use of food. The mass-specific metabolic rate of meiobenthos is about five times greater than that of macrobenthos, so meiobenthos consumes five times more food (i.e. carbon) than macrobenthos per unit of biomass (Giere 2008). This is why the meiobenthos, together with the bacteria, uses the greatest portion of the energy consumed by the bottom ecosystem and forms a “small food web”. However, there is a complex interaction within this “small food web” where the meiobenthos can affect microbial communities through grazing, while spatial-temporal fluctuations of bacteria affect the distribution and abundance of meiobenthos (Leguerrier et al. 2003, Giere 2008). Also, the meiobenthos is considered to be primarily linked to the detritus/bacteria-based food chain and the energy fixed by the meiobenthos is often transmitted to higher trophic levels, such as the macrobenthos. Thus, small macrofauna and nekton (usually juvenile stages of larger fishes or crustaceans) can use meiobenthos as a food source (Heymans and Baird 1995). According to the different roles that these groups play in the energy balance and the carbon cycle, the expansion of salt marshes over tidal flats in the study area may involve a faster flow of energy and carbon transfer by the meiobenthos, along with a greater increase in food web complexity (Giere 2008).

Our results suggest that the expansion of S. alterniflora produces changes in the macrobenthic and meiobenthic communities that have an influence on the carbon cycling. This influence on the C cycle is due to the differences in the biomass of the taxa present in the two habitats, levels in the trophic web, feeding guilds (deposit feeders and herbivores), modes of life (infaunal or epifaunal) and activities such as bioturbation. It is largely recognized that the impact of plant colonization on C sequestration and storage is conditioned by plant biomass production and changes in sedimentation rates (Chastain et al. 2018), but it also depends on the indirect effects of benthic fauna. Changes in faunal composition may alter not only the amount of C that is temporarily stored in faunal biomass, but also the amount of CO₂ released by respiration and the quality of the organic matter delivered to the system through excretions. Along with changes in the species composition, there might be changes in bioturbation rates that may also alter the ventilation and irrigation of sediments. Burrowing animals, through mixing and disturbing the sediment layers, play a significant role in the release of carbon that is already sequestered in the sediment matrix. For instance, their activity increases the availability of oxygen, accelerating organic matter decomposition, and through biodeposition and excretion they enhance the localized accumulation of organic matter and promote hotspots of microbial activity (Thomson et al. 2019). Over time the expansion of S. alterniflora and the loss of bare tidal flats are expected to produce the changes in the macro- and meiobenthic assemblages described in this work. According to our results, a decreasing biomass of deposit feeders would augment the carbon sequestration capacity recognized worldwide for salt marshes, but it may also have major consequences for the trophic transfer of organic carbon to higher levels in the food webs (for example, commercially important fish). Moreover, external forces such as climate change, especially rising sea level, along with anthropic activity, can interact with marsh development to produce highly different patterns (Levin et al. 2006). Thus, more research is necessary to obtain a better understanding of the changes that the expansion of S. alterniflora introduces to the carbon cycle within the estuary.