4.1. Richness, Abundance and Diversity of Seagrass Epifauna
Despite fairly stable salinities in the lagoon, species richness and diversity displayed a similar pattern of decline from the mouth as observed in typical estuaries where variation in faunal communities are largely influenced by a salinity gradient (Allanson & Baird, 1999; Barnes, 2010; Heymans & Baird, 1995). However, unlike typical estuaries, salinity levels in the lagoon increased with distance from the mouth likely due to evaporation and the lack of distinct inflow of freshwater. Variation in epifaunal community structure is not necessarily attributed directly to salinity but could be an indirect effect since salinity is often correlated with other environmental variables (Yamada et al., 2007). Factors such as current velocity, wave intensity, competition, and predation, which were not assessed in this study, could account for the unexplained variation in community structure.
In a similar enclosed bay with no freshwater input and a relatively high salinity range (22 - 35) in Arcachon, France, macrofauna in beds of another dwarf-eelgrass, Zostera noltei showed similar patterns to that observed for Langebaan Lagoon (Blanchet et al., 2004). Species richness was lowest in assemblages furthest from the mouth and at higher tidal levels while abundances were highest at these places. In that study, distance from oceanic waters (km) was used as an environmental factor and was the main attributor of differences between assemblage groups along with tidal range (Blanchet et al., 2004). Similarly, Barnes and Ellwood (2012) found richness and diversity to be highest inZ. capensis beds below mean low water levels in the marine-dominated section of the Knysna estuary, South Africa. In that large estuarine system species richness did not display the classic decline with distance from the mouth, but instead had a stable fauna separated by abrupt changes in areas where salinity levels fell below 30 (Barnes & Ellwood, 2012). In Arcachon, species occurring in areas of relatively low salinities were not distinct but rather a subset of those present at the mouth, which was not the case at Langebaan. Several species were found only at sites closer to the mouth and not at Bottelary and Geelbek. In contrast, a stable faunal composition was observed at Knysna despite the wide range in salinity (<5–35). While the observed variation in Knysna was supposedly due to changes in relative abundance of dominant species, or the lack of marine species in upstream localities and upper boundaries of seagrass beds (Barnes & Ellwood, 2012), species variation in Langebaan was more likely a result of exposure and tolerance to desiccation (this study), availability of seagrass habitat (Angel et al., 2006) as well as the ability to survive under periods of exposure to sun and wind during low tide (Day, 1959).
In Langebaan Lagoon, epifaunal community structure was dominated by grazers i.e., Siphonaria compressa , Fissurella mutabilisand Assiminea globulus . A. globulus was the main species recorded at Bottelary and Geelbek and occurred in lower abundances in high shore beds at Oesterval and Klein Oesterval, although high abundances were previously observed at Oesterval in both seagrass and saltmarsh vegetation in the high shore (Day, 1959). Regarded as a broad niche generalist and capable of adapting to a range of environmental conditions, this mud snail feeds mainly on bacteria and diatoms found on sediment surfaces and occasionally on epiphytic periphyton and is strongly averse to prolonged periods of submergence (Angel et al., 2006). Gastropods have been documented to dominate many estuarine systems (Allanson & Baird, 1999) and Barnes (2013) recorded a dominance of gastropods including 125 times greater average densities ofAssiminea in upper-estuarine and enclosed sites compared to elsewhere in the Knysna estuary. In Arcachon, seagrass beds at high shore sites and sites further from the mouth were dominated by a single gastropod, Peringia ulvae (Blanchet et al., 2004). The prevalence of gastropods in high shore areas is likely attributed to their ability to resist desiccation supported by the upward expansion of habitat provided by seagrass and saltmarsh vegetation.
High shore sites in Langebaan were also dominated by the pulmonate limpet Siphonaria compressa which has been identified as South Africa’s most threatened marine invertebrate. Classified as Critically Endangered by the IUCN (Kilburn, 1996), this limpet has an extremely narrow habitat range and grows only in Zostera capensis meadows in Langebaan Lagoon and the Knysna estuary (Allanson & Herbert, 2005). In this study, densities of 100 m-2 were recorded at Geelbek and Bottelary and <5 m-2 at Oesterval. None were found at Klein Oesterval and Centre Banks, however, Angel et al (2006) found densities between 40-80 individuals per m2 at the lower edge of seagrass beds at Klein Oesterval over a 30 year period. Analyses of faecal pellets revealed that S. compressa primarily feeds on bacteria, diatoms and filamentous algae found on seagrass leaves (Allanson & Herbert, 2005).
High abundances of the two dominant grazers Assiminea globulusand Siphonaria compressa at the same sites suggest a lack of competition and the exploitation of different niches. This hypothesis was explored by Angel et al. (2006) who assessed the interaction betweenA. globulus, S. compressa and Zostera capensis cover in relation to low tide exposure at Langebaan. They found that S. compressa thrived on the lower edge of seagrass beds subjected to shorter periods of exposure and suggested that the virtual absence of the limpet from the upper zone was largely the result of avoiding physiological stress from desiccation - a consequence of its small size and thin shell. Experimental transplants of seagrass into sandbanks saw a proliferation of S. compressa and concluded that confinement of the limpet to beds in the high shore was suboptimal and essentially due to the restriction of the seagrass beds themselves to expand further into the low shore due to sediment disturbance by the burrowing sand prawn, Kraussillichirus kraussi (Angel et al., 2006). The conclusions of that experiment do not explain the absence of S. compressa from low shore beds at Oesterval and Centre Banks in this study. Here, a likely reason is competition with grazing gastropods (Orth et al., 1984) such as Fissurella mutabilis, which is less tolerant to desiccation and occurred in high abundances at low shore sites close to the mouth and but not at high shore sites i.e., Geelbek and Bottelary.
In contrast, Angel et al. (2006) recorded high and low densities ofAssiminea globulus on the upper and lower edges of seagrass beds respectively and noted a positive correlation with exposure. Once again, this was attributed to disturbance of sediment and reduction of diatoms due to bioturbation, however these snails also appeared to prefer exposure rather than submergence (Angel et al., 2006). This species’ preference for more exposed parts of seagrass beds is likely related to the availability of food in the form of sand-dwelling diatoms that are abundant in the high shore zone (G.M. Branch, unpublished data), and therefore occupies a different niche to that of Siphonaria compressa . Given that S. compressa likely feeds essentially on periphyton on seagrass leaves (Allanson & Herbert, 2005), there was no evidence to indicate that abundance or zonation of these two species had been influenced by competition.
Environmental conditions such as a longer tidal emergence period at Bottelary and Geelbek correspond to a higher susceptibility of species to desiccation. This explains the lower species richness at these sites, and the high abundances of the desiccation resistant A. globulusand attests to the resilience of S. compressa to persist in suboptimal conditions. The cushion star Parvulastra exigua and keyhole limpet F. mutabilis were previously reported to have completely declined at Klein Oesterval (Pillay et al., 2010) but was subsequently recorded in this study. In addition, Pyura stolonifera , Sargartia ornata and Oxystele antoni not documented in that survey, were also found. These findings underscore the considerable variability inherent in seagrass ecosystems. The elevated abundances of grazers further emphasize the important trophic role of grazing within both seagrass and intertidal ecosystems more broadly (Asmus & Asmus, 1985; van Der Heijden et al., 2020).
Species richness and densities recorded in this study were comparable to the survey of macrofauna within seagrass and un-vegetated sandflats at Klein Oesterval using the traditional method of digging and sieving (Pillay et al. 2010). That study yielded a total 27 species and while infauna (n = 15) was not targeted in this study, a total 13 species were documented at Klein Oesterval - nine species were recorded in both surveys, and an additional four were recorded in this study only. These patterns allude to the strengths and weaknesses in sampling techniques, many of which fail to capture various components within faunal assemblages influenced by processes such as life history stages, succession and colonization (Moura et al., 2008) as well as temporal restrictions related to tidal regime, season and day/night sampling. While traditional methods are considered comprehensive in providing estimates of biodiversity, alternative methods including the use of a selection of indicator species in single or multimetric indices, can be useful to assess the ecological status of benthos in marine and estuarine environments (Borja et al., 2011; Dauvin et al., 2016) particularly when resources are limited. Using a subset of species has been found to provide meaningful descriptions on faunal community structure (Kuenen & Debrot, 1995; Vellend et al., 2008). Importance should therefore be placed on consistency of methods used as well as regularity of surveys to provide informative results to address research and management objectives (Magurran et al., 2010).