**1. Introduction**

Transitional waters, being a continuum between continental and marine ecosystems, represent areas with high environmental heterogeneity. As such, there is a complex association between abiotic and biotic components that makes these water bodies ideal to study the distribution and dynamics of the benthic assemblages with the aim to further our understanding of the ecosystem functioning [1,2]. Lagoons have a historically relevant "social" value because they o ffer a high biological productivity [3]. For this reason, they host many human activities (i.e., fisheries, aquaculture, agriculture, industry, and tourism [4]) that, on the other side, have endangered their integrity and ecological quality status as well [5,6]. Therefore, there is general agreemen<sup>t</sup> among the scientific community, which is also recognized by legislations worldwide (e.g., the US Clean Water Act, European Water Framework Directive, Marine Strategy Framework Directive, and the National Water Act in South Africa), about the need to assess their health status and ensure proper managemen<sup>t</sup> of their resources [4,7].

Meiofauna are small benthic invertebrates that have a well-recognized role in the food webs of lagoon systems connecting microbial components to higher trophic levels that contributes to the overall carbon fluxes and organic matter mineralization [8,9]. Because of their high taxonomic diversity, rapid generation times, lack of larval stages, and various life strategies meiofaunal organisms are considered excellent bioindicators of natural or anthropogenic stressful conditions [10–12]. However, their role in ecosystems tends to be overlooked, mainly due to the lack of taxonomists and the small size of meiofauna, which require time and the appropriate techniques for their study [13].

In the Mediterranean basin, there are more than 100 coastal lagoons, half of which have available physico-chemical or ecological data in the scientific literature [4]. Among them, the largest amount of information on meiofaunal spatial pattern is available for the upper Adriatic Sea, including the Venice lagoon [14–20]. Meiofaunal diversity and assemblage structure are also well-documented in the southern part of the Adriatic Sea, including the Lesina and Varano lagoons [6,8,17,21–23]. Instead, meiofaunal studies in transitional environments along the Tyrrhenian coast, with the exception of the Stagnone of Marsala (Western Sicily), are largely lacking [17,24]. Furthermore, most of the available literature on meiofauna from coastal lagoons takes into consideration the spatial pattern of the assemblages, while only in a few cases their temporal dynamics is reported [2,8,24–26]. Finally, little is known on spatiotemporal dynamics of meiofauna in Mediterranean transitional systems characterized by di fferent physico-chemical gradients related to the riverine inflow, the connection to the sea, and the organic matter (OM) enrichment of sediments.

Within the Tyrrhenian coast, the Sardinian Island is one of the richest Italian regions in number and extension of lagoons [27], ye<sup>t</sup> knowledge on meiofaunal composition and distribution in these systems is absent. In the present study, we describe for the first time the spatiotemporal variation in meiofaunal assemblages in the Cabras Lagoon, the largest and most complex transitional system in the Sardinia Island. This lagoon is characterized by a large environmental heterogeneity, with an increasing salinity along its main longitudinal axis and varying degrees of trophic condition across the basin [28–30]. For these reasons, it represents a valuable case-study in which to test the general hypotheses on the meiofaunal dynamics in these highly variable systems. Our main objectives were to investigate the pattern of spatial variation in meiofaunal diversity and community structure in relation to the main environmental gradients, and to assess whether this pattern was consistent through time. In particular, we tested whether spatiotemporal variation could be identified in: (1) the whole meiofaunal assemblage of the three sites in terms of (i) total number of individuals, (ii) total number of taxa, and (iii) Shannon diversity (H') and Pielou's evenness (J) indices; and (2) the abundance of dominant taxa, including the ratio between nematodes and copepods. We anticipate that the response of meiofaunal assemblages to the environmental drivers (both in water and sediments) identified in the present study will provide one of the few evidences of the importance of meiofaunal studies to further our understanding of the functioning of Mediterranean lagoons.

### **2. Materials and Methods**

### *2.1. Study Area and Sampling Sites*

The Cabras Lagoon (central-western Sardinia; Figure 1) is the largest lagoon in the Sardinia island, with a surface area of 22 km<sup>2</sup> and a watershed of ~430 km<sup>2</sup> inhabited by approximately 38,000 people. Its main freshwater riverine source is the Rio Mare e Foghe located in the northern sector of the lagoon, with a minor contribution from the Rio Tanui, southward. The lagoon is connected to the adjacent Gulf of Oristano only via three narrow creeks that flow into a large channel ("scolmatore") built in the late 70's, closed in proximity of the lagoon by a 30 cm high dam. In the last two decades, the Cabras Lagoon has been extensively investigated from various perspectives and using different approaches, including physical/modeling [31,32], biogeochemical [33,34], biological [35–38], and ecological [5,39,40]. However, while several studies have been conducted in the Cabras Lagoon on the macrozoobenthos [28–30,41,42], nothing is known regarding the spatiotemporal variation in meiofaunal assemblages. In fact, no such studies are available for transitional waters in the Sardinian Island, one of the richest Italian regions in number and extension of lagoons [27], with only few examples conducted in fully marine coastal waters [43,44].

**Figure 1.** Location of the study area (Cabras lagoon, western Sardinia, Italy) and sampling sites (C1, C2, and C3). Image source: Google Earth.

For the present study, three sites (C1, C2, and C3; Figure 1) were selected along the longitudinal axis of the Cabras Lagoon, being representative of different environmental (e.g., salinity, confinement, and sediment grain-size) and trophic (e.g., sediment OM and phytopigments) conditions. Site C1 was located in the northern sector of the lagoon, connected to the main freshwater tributary the Rio Mare e Foghe. This site was characterized by sandy sediments, low OM content of sediments, and the

presence of halophytic vegetation (*Phragmites* sp.) along the shore. Site C2, was located in the satellite pond of Sali e Pauli and surrounded by halophytic vegetation (*Salicornia* sp.). This site was highly confined and characterized by a high OM content of sediments [33]. Biofilm-forming cyanobacterial strains with extremely growth rates were also found here [45,46]. Site C3 was located in the southern sector of the lagoon, at the confluence of the three creeks connecting the Cabras Lagoon to the main channel. This site was characterized by muddy-sandy sediments, limited OM enrichment of sediments, abundant submerged vegetation (e.g., *Ruppia*), and a significantly higher hydrodynamics than at the other sites [47].

### *2.2. Field Surveysand Sample Treatment*

The field surveys were carried out at sites C1, C2, and C3 on 6 July 2010 and 2 February 2011. At each site and date, water temperature, salinity, and dissolved oxygen (DO) were measured using portable probes (WTW LF 197 and WTWOxil 197, respectively). Subsequently, sediment samples for the determination of the water content (Wc) and chemical analysis (OM, chlorophyll-a, and phaeopigment content) were collected using a manual core (40 cm long, 5.5 cm diameter) gently pushed by hand into the sediments. Procedural details of sediment collection and chemical analysis are given in the companion paper by [30].

For the analyses of the meiofauna, six replicates were collected at each site by means of plexiglas corers (diameter: 3.6 cm) inserted 5 cm in the sediment. These samples were pre-filtered with magnesium chloride (MgCl2; 80 g <sup>L</sup>−1) to allow organisms to relax before fixation and facilitate subsequent taxonomic identification [48]. This treatment appears important because the "soft-bodied" taxa (e.g., Gastrotricha, Plathelminthes, and Nemertina) usually undergo the major morphological alterations after fixation and they can remain in good conditions with magnesium chloride treatment. The sediment samples were then fixed in a solution of pre-filtered seawater containing formalin bu ffered with sodium tetraborate Na2B4O7 to reach a pH of ca. 8.2 [49]. The amount of formalin to be added to the sample to obtain a final concentration of 4% was calculated based on the total volume of sediment and water present in the sample. A few drops of a Rose Bengal solution (0.5 g <sup>L</sup>−1) were added to the sample in order to facilitate the identification of organisms in the sorting phase [50].
