Introduction

The endemic Posidonia oceanica (L.) Delile, 1813 constitutes the most important meadow-forming seagrass species in the Mediterranean Sea. This marine phanerogam is actively involved in water oxygenation and has been used as bioindicator of environment well-being (Vaquer-Sunyer & Barrientos, 2021). Regarding biodiversity, P. oceanica meadows support a rich and plentiful community of marine organisms (Blanc & Grissac, 1984; Perès, 1984; Bourcier, 1989). According to these studies, the ever-increasing threat to these meadows in the Mediterranean Sea not only endangers their own survival, but also the conservation of vast diversity of living beings that find shelter and nourishment within this heterogeneous ecosystem. Three main habitats can be highlighted among these meadows: the leaves and the matte (roots and rhizomes) within the bushes and the adjacent unvegetated sediment outside them.

The most well-known fauna hosted in P. oceanica meadows include many fishes of commercial interest as well as macroinvertebrates such as crustaceans, mollusks, and sea urchins (Cetinić et al., 1999, 2011; Albano & Stockinger, 2019; Borg et al., 2006). However, knowledge is comparatively less regarding the community of marine invertebrates of small size, namely meiofauna (García-Gómez et al., 2022). This benthic community is formed by organisms capable of passing through a 500 μm mesh and being retained in a lower 63 μm mesh size sieve (Giere, 2009). Meiofauna plays an important role in the functioning of marine ecosystems (Schratzberger & Ingels, 2018) and can be a suitable model for the study of their health because of their high abundance, fast regeneration rate, and low dispersal capacity (Woodward, 2010; Zeppilli et al., 2015). Our understanding of the diversity and ecology of meiofauna inhabiting P. oceanica meadows has slowly increased in recent decades (Martínez et al., 2021; García-Gómez et al., 2022). The phylum Nematoda, on the other hand, constitutes the most abundant component of the meiofauna and has been further investigated (Novak, 1982, 1989) in these Posidonia meadows. Previous research showed a high diversity of nematodes within P. oceanica and a different community composition between its habitats. For instance, Moens & Vincx (1997) highlighted that these differences are influenced by various crucial community-shaping factors, such as food availability of habitats and feeding strategy of species, while other studies attributed this to habitat complexity (Mazzella et al., 1989; Gacia & Duarte, 2001; Boström et al., 2006).

Regarding food sources, leaves within P. oceanica ecosystem are dominated by microepiphytic organisms, such as diatoms and dinoflagellates, whose prevalence is especially clear in the summer. Moreover, during this season, the leaves reach their maximum length, which decreases hydrodynamic forces underneath (Mabrouk et al., 2011) and favors the stability of the sediment under the canopy (Gacia & Duarte, 2001; Manca et al., 2012).

As a novel contribution with respect to previous studies on nematode communities in P. oceanica meadows, our research incorporates samples from outside the canopy for comparison to leaves and matte inside bushes. This work also incorporated the analyses of sediment granulometry and concentration of organic matter and nitrogen to relate it to nematode communities. Finally, we introduced the study of functional traits that were previously unexplored in the nematode communities of P. oceanica. These traits include presence/absence of ocelli and presence/absence of denticles on the buccal cavity and tail shape, among others previously studied.

The aim of this study is to assess dissimilarity in Nematoda communities between P. oceanica’s three main habitats (leaves, matte, and adjacent unvegetated sediment) due to habitat heterogeneity. These are the main hypotheses tested in this study:

  • We expect to find higher density of nematodes inside the canopy than outside due to increased flux of food resources.

  • The diverse availability of food and the heterogeneity of the matte lead us to expect a greater richness of genera and families in this habitat in comparison to the other, more uniform habitats.

  • The resource specificity of the leaves, consisting of microepiphytes, leads to a community dominated by a group of nematodes that exploit microepiphytes as food resource, potentially resulting in a lower evenness. Conversely, the higher variety of nutrients in the matte and adjacent sediment may host more generalist feeders and foster a more even community structure.

  • We anticipate that nematodes occurring in P. oceanica show distinct functional traits as a result of their adaptation to the specific conditions of each habitat.

  • With respect to matte and adjacent sediment habitats, we expect that sediment parameters (i.e., mean grain size, nitrogen and organic matter content, and the percentages of sand, silt, and clay) will influence the genus and family composition of nematode communities.

Materials and methods

Sampling, processing, sorting and identification of meiofauna

A total of nine samples were collected by scuba diving on a P. oceanica meadow from Cala Cuartel (Santa Pola, Alicante, Mediterranean Spain; coordinates: 38° 12′ 34.04″ N, 0° 30′ 19.12″ W) on summer season, specifically on the 2nd of August 2016, when the phanerogam reaches its highest growth. A 1 km2 area within the meadow was selected, situated at depths of 4 to 7 m (Fig. 1). Within this area, we randomly designated three sites, approximately 200 m apart, where one sample was collected from each of the three primary habitats. At each site, a 20 × 20 cm quadrat was used to sample leaves, matte and the nearest unvegetated area. Leaves were first cut at the ligule level and collected in a hermetic bag; then, the underlying root sediment was shoveled into another hermetic bag (Novak, 1982, 1989; Cvitkovic et al., 2017). Lastly, the nearest adjacent sediment comprised in a quadrat was as well collected in a hermetic bag.

Fig. 1
figure 1

Location of Cala Cuartel in Santa Pola (Alicante). The rectangle with dotted outline delineates the 1 km2 meadow area where the sample sites were designated

Full size image

For the extraction of the metazoans, firstly the anesthetization by magnesium chloride technique was used to isolate the soft meiofauna (Higgins & Thiel, 1988; Schmidt-Rhaesa, 2019). Once this fraction was separated, we employed the “bubble and blot” technique to extract the hard meiofauna (Higgins & Thiel, 1988). Meiofauna was collected using a 63 μm mesh size sieve and then fixed with 7% formaldehyde. Nematodes were counted from 1/5 of each sample, a measure taken in response to the notably high abundance of metazoans within the samples. Subsequently, this count was extrapolated to represent the entire sample. Approximately 100 nematode specimens were sorted and mounted on glycerine (Bianchelli et al., 2013; Rosli et al., 2018; Semprucci et al., 2018; Rebecchi et al., 2022), encircled by a ring of paraffin, under a stereomicroscope. Nematodes were identified to genus level using an Olympus© BX51-P microscope with differential interference contrast optics equipped with an Olympus© DP-23 camera.

Granulometry and biogenic elements analysis

Sediment samples were collected under and outside the canopy using the referred 20 × 20 cm quadrat. The first 2 cm of sediment within this quadrat was carefully shoveled into a plastic bag. After air-drying the sediment, granulometry parameters were analyzed using the methods of Guitián & Carballas (1976). Particle size classes applied in this study follow the classification adopted from Blott & Pye (2001), where the following categories are included: very coarse sand, coarse sand, medium sand, fine sand, very fine sand, very coarse silt, coarse to medium silt, fine to very fine silt, and clay. The software Gradistat v.8 (Blott & Pye, 2001) was used to obtain sediment parameters following the Folk & Ward method (1957). Nitrogen content of each sediment sample was obtained following the Kjeldahl method (Bradstreet, 1954) and organic carbon following Walkley & Black (1934) adapted to microplates reader, to subsequently calculate organic matter percentage.

Statistics

All analyses were performed on R software version 4.2.2 (R Core Team, 2022).

To test the first hypothesis, primarily, density of nematodes was calculated by dividing the total number of nematodes (abundance) of a sample by the sampling area (20 cm2). Subsequently, habitat densities were compared using a one-way repeated measures ANOVA, employing the car package (Fox & Weisberg, 2019). The analysis was carried out considering the normality of data, which was verified through the shapiro.test function, and considering the site as within-subjects factor.

Richness corresponds to the total number of distinct genera and families present in a sample. For testing the second hypothesis, genera and family richness were calculated based on the identification of mounted specimens (approximately 100 specimens per sample) and compared between habitats using one-way repeated measures ANOVA with the package car (Fox & Weisberg, 2019). The analysis considered data normality, verified through the shapiro.test function, and considered the site as within-subjects factor.

Regarding the third hypothesis, genera and family composition of the nematode community was studied. First, Ruziicka index with the adespatial package (Dray et al., 2023) was used to calculate beta diversity. This index considers the abundance of each taxon included in the analysis, not only the presence of the taxon. Second, these diversity indexes were compared to test for differences between habitats using a permutational analysis of variance (PERMANOVA) using the adonis2 function of the package vegan (Oksanen et al., 2022). This PERMANOVA was carried out using the beta composition as dependent variable, the habitat type as independent variable, and to address potential non-independence of habitat samples and to account for variability between sites, it was included the site as a random factor. SIMPER routine was performed to assess which genera and families contribute with the highest dissimilarities between habitats. To explore community evenness, Pielou’s J was calculated for each sample using the function diversity of the package vegan (Oksanen et al., 2022).

To explore the fourth hypothesis, paired t-test through the function t.test was employed to examine differences between habitats with respect to the presence/absence of genera that typically possess ocelli based on the information extracted from the genus description, as they can easily get lost after fixation, and presence/absence of denticles on the buccal cavity. PERMANOVA analyses were conducted to test for variations in abundance of nematodes in each trophic group (selective, bacterial feeders (1A); non-selective deposit feeders (1B); epistrate or epigrowth (diatom) feeders (2A); predators/omnivores (2B)) (Wieser, 1953) and nematodes with different tail shape (short/rounded, clavate, conical, conico-cylindrical, elongated/filiform) between habitats. The tail classification was modified from Thistle et al. (1995) that considered clavate–conico-cylindrical as a single group. We divided this group into two, clavate and conico-cylindrical as in Maharning et al. (2023). For these PERMANOVA analyses, the Ruziicka index was used to compute beta diversity of the community, regarding the trophic group for one analysis and the tail shape for the second one. The beta diversity regarding both aspects served as the dependent variable, with habitat type as independent variable. In addition, three PERMANOVA analyses were carried out with the beta composition of the trophic group as dependent variable and mean grain size, nitrogen content, and organic matter content as independent variables. To address potential non-independence between habitat samples and to account for variability between sites, it included the site as a random factor.

In terms of the fifth hypothesis, sediment parameters (mean grain size, organic matter, and nitrogen content) were compared between the matte and the adjacent sediment using Linear Mixed-Effects Models with the function lmer in the packages lme4 (Bates et al., 2015) and Matrix (Bates et al., 2022). Linear Mixed-Effects Models were also carried out to explore nematodes density, genera richness, presence of denticles, and abundance of nematodes with each tail shape in relation to sediment parameters. Furthermore, community composition with respect to genera, families, and buccal types was examined in relation to sediment parameters through PERMANOVA analyses. In these analyses, the Ruziicka index was utilized to calculate beta diversity of the community concerning its composition in families, genera, and the abundance of nematodes within each trophic group. The beta diversity related to these aspects served as the dependent variable. Each of these parameters underwent a separate PERMANOVA against mean grain size, organic matter content, and nitrogen content as explanatory variables. To account for potential non-independence among samples and variability between sites, the site was incorporated as a random factor in each analysis.

Principal Component Analysis (PCA) from the vegan package was conducted to visualize community structure of genera composition between sites: genera composition in relation to sediment parameters; community composition according to nematodes buccal type; and community composition regarding nematode tail shape and its relationship with sediment parameters. Moreover, to visualize the differential genera community composition within habitats, data were represented with a heatmap using reshape2 (Wickham, 2007) and ggplot2 (Wickham, 2016) packages.

Results

Nematoda community composition, density, genera, and family richness between the three main habitats: leaves, matte, and adjacent sediment

The study of the community composition of nematode families and genera produced similar results. Analysis based on families showed significant differences between habitats (PERMANOVA: R2 = 0.63141; P = 0.005**), as did the analysis considering nematode genera (PERMANOVA: R2 = 0.51922; P = 0.005**, Fig. 2). In line with Novak (1989), the family Chromadoridae dominated in the leaves (81% of all nematodes in this habitat). However, this family was poorly represented in the matte and adjacent sediment (8% and 9%, respectively, Table S1). The genus Chromadora accounted for most of the presence of this family, constituting 70% of the community in the leaves and only 1% and 3% in the matte and adjacent sediment, respectively. PCA clearly illustrates the association of the genus Chromadora with leaf samples (Fig. 3A). The family Desmodoridae also showed differential relevance among the three habitats, with the highest numbers appearing in the matte (40%), less representation outside the canopy (21%), and minimal presence in the leaves (2%). In our study, Perspiria and Bolbonema were the main representatives of the latter family. Perspiria, as represented in the PCA (Fig. 3A), showed different distribution between the matte (15%) and the remaining habitats (2% outside the canopy and no presence on the leaves); similarly, Bolbonema showed the highest abundance on the sediment habitats (11% on matte and 16% on adjacent sediment), in contrast to the leaves (0.3%, see Table S2 and Fig. 3A). The highest abundance of the family Xyalidae was found in adjacent sediment samples (16%), compared to the low representation in matte (2%) and in leaves (0.3%). Moreover, the family Oncholaimidae was best represented in the adjacent sediment habitat (20%), less abundant in the leaves (12%), and poorly present in the matte (6%).

Fig. 2
figure 2

Panels showing the number of specimens per genus from leaves, matte and adjacent sediment samples. Intensity of the color increases according to the number of presences. Genera that had the greatest impact on dissimilarity were included

Full size image
Fig. 3
figure 3

Principal Component Analysis (PCA) representing A Nematoda community composition according to genera, B genera community composition in relation to sediment parameters, C community composition according to buccal type, and D community composition according to tail shape and sediment parameters; between leaves, matte, and adjacent sediment

Full size image

Nematode density did not significantly differ between habitats (ANOVA: F = 1.037, P = 0.314, Fig. 4A). However, there were differences in genera and family richness between habitats (ANOVA for genera richness: F = 10.303, P = 0.026*; ANOVA for family richness: F = 14, P = 0.016*). The richest habitat regarding genera was the matte under the canopy, although the significance was marginal when compared to the adjacent sediment (t-test: P = 0.058), and not significant when compared to the leaves (t-test: P = 0.067, Fig. 4B). Furthermore, matte family richness was significantly higher than in adjacent sediment (t-test: P = 0.0067**) and leaves (t-test: P = 0.0445*, Fig. 4C). The sediment outside the canopy showed the highest evenness of nematode genera of all the studied habitats (Pielou’s J of 0.29, 0.30 and 0.29), followed by the matte (Pielou’s J of 0.27, 0.28 and 0.25) and the leaves (Pielou’s J of 0.18, 0.19 and 0.23).

Fig. 4
figure 4

Boxplots showing A Nematoda density (ind/cm2), B genera, and C family richness between habitats. Sample 3 is highlighted to show its dissimilarity with the rest of samples of the matte habitat regarding nematodes density

Full size image

Sediment parameters as community-shaping factors

Sediment nitrogen and organic matter content were compared inside and outside the canopy habitats, showing no significant differences (%nitrogen lmer: t-value = 0.883; P = 0.43; %organic matter lmer: t-value = 1.882; P = 0.133). Nonetheless, our analyses demonstrated that the matte had a significantly smaller mean grain size than the adjacent sediment (lmer: t-value =  − 4.354; P = 0.0489*, Fig. 5).

Fig. 5
figure 5

Boxplots showing sediment A mean grain size (µm), B organic matter (%) and C nitrogen (%) of the sediment outside and inside the canopy. Sample 3 is highlighted to show its dissimilarity with the rest of samples of the matte habitat regarding organic matter and nitrogen content

Full size image

The Linear Mixed-Effects Model revealed no significant differences in nematode density associated with mean grain size (lmer: t-value =  − 0.639; P = 0.593), the percentage of organic matter (lmer: t-value =  − 0.264; P = 0.8049), or nitrogen content (lmer: t-value =  − 0.370; P = 0.7302). However, genera richness significantly decreased as mean grain size increased (lmer: t-value =  − 19.384; P = 0.0026**) but remained unaffected by the other sediment parameters included in the study (lmer: organic matter: t-value =  − 0.629; P = 0.5634; nitrogen: t-value =  − 0.661; P = 0.5448).

When comparing matte and adjacent sediment, the community composition of nematode genera was significantly affected by sediment organic matter content (PERMANOVA: R2 = 0.30658; P = 0.0403*) and nitrogen content (PERMANOVA: R2 = 0.32495; P = 0.0097**), with a marginal effect from mean grain size (PERMANOVA: R2 = 0.27326; P = 0.0569, Fig. 3B). In contrast, the community composition of nematode families was not significantly affected by any of the sediment parameters (PERMANOVA: mean grain size: R2 = 0.14623; P = 0.5514; organic matter: R2 = 0.26317; P = 0.2736; nitrogen: R2 = 0.29447; P = 0.1535).

Functional traits and habitat characteristics

Buccal types

The composition of the community was explored according to this trait, revealing significant differences between habitats (PERMANOVA: R2 = 0.73627; P = 0.005**). Epistratum feeders (2A) were the most abundant nematodes in all three habitats, constituting 82% of the community in the leaves, 67% in the matte, and 40% in the adjacent sediment. The most significant dissimilarity between leaf and matte habitats was observed in selective feeders (1A), whose abundance ranged from 5% in the leaves to 20% in the matte. The adjacent sediment community showed the highest evenness according to the buccal types, hosting 40% of epistratum feeders (2A), 30% of non-selective deposit feeders (1B), 27% of predators/omnivores (2B), and the least represented, 3% of selective feeders (1A) (Table S3; Fig. 6).

Fig. 6
figure 6

Boxplots showing percentage of nematodes with different buccal types according to Wieser (1953): selective (bacterial) feeders (1A), non-selective deposit feeders (1B), epistrate or epigrowth (diatom) feeders (2A), and predators/omnivores (2B) in the community per habitat

Full size image

When analyzing nematode buccal types in relation to sediment characteristics, none of the sediment parameters, including mean grain size (PERMANOVA: R2 = 0.24562; P = 0.3431), organic matter (PERMANOVA: R2 = 0.3618; P = 0.8), or nitrogen content (PERMANOVA: R2 = 0.42513; P = 0.0653), appeared to significantly affect the community (Fig. 3C).

Tail

Analyses show significant differences in the composition of nematode tail shapes between habitats (PERMANOVA: R2 = 0.70581; P = 0.005**). The most abundant type of tail was the conical one, contributing with the highest dissimilarities between the three habitats. Conical tail constituted the 86% of the community in the leaves, 56% in matte, and 35% in adjacent sediment. The clavate tail was most prevalent outside the canopy (31%), differentiating it from the two other habitats (leaves = 16%; matte = 16%). Likewise, the elongated/filiform tail reached its highest abundance in the adjacent sediment, comprising 13% of the community, and only 3% and 7% in the leaves and matte, respectively. Nematodes with conico-cylindrical and short/rounded tails maintained similar abundances (6, 7, and 9% for conico-cylindrical and 2, 1, and 0.3% for short/rounded) between habitats, as did nematodes with short/rounded tails (Table S4).

When analyzing the effect of sediment mean grain size, sand, silt, and clay proportions on the relative abundance of nematodes with each tail shape (Table 1), the only significant differences were found in the nematodes with elongated/filiform tails (Fig. 7). This type of tail is positively correlated to increasing mean grain size (lmer: t-value = 5.095; P = 0.0276*) as well as increasing clay percentage (lmer: t-value = 2.830; P = 0.0474*, Fig. 3D).

Table 1 lmer results relating sediment mean grain size, sand, silt, and clay proportions to the relative abundance of nematodes with each tail shape
Full size table
Fig. 7
figure 7

Linear correlation between the five types of tail shape (clavate, conical, conico-cylindrical, short/rounded, and elongated/filiform) and the sediment A mean grain size and B percentage of clay

Full size image

Ocelli

Nematodes that typically have or may present ocelli were represented in our samples by the following genera: Araeolaimus, Axonolaimus, Bolbonema, Chaetonema, Chromadora, Chromadorella, Diplopeltis, Eurystomina, Leptosomatum, Onchium, Phanoderma, and Tricoma.

The 70% of nematodes of the leaves belonged to genera mentioned above. This percentage significantly differed from the 18% of ocellated nematodes from the matte (paired t-test: t = 5.6723; P = 0.0297*) and the 30% from outside Posidonia bushes (paired t-test: t =  − 6.776; P = 0.0211*, Fig. 8A). There were no significant differences in the abundance of nematodes with ocelli between matte and adjacent sediment habitats (paired t-test: t = 3.874; P = 0.0606).

Fig. 8
figure 8

Boxplots showing the proportion of nematodes with A ocelli and B denticles per habitat

Full size image

Denticles

Nematodes that present denticles in their buccal armature are represented in our samples by the following genera: Actinonema, Chromadora, Chromadorita, Cyatholaimus, Dichromadora, Euchromadora, Eurystomina, Longicyatholaimus, Metachromadora, Paramarylynnia, Pomponema, Preacanthonchus, Rhips, and Spilophorella.

The 72% of nematodes from the leaf community had denticles in their buccal cavity. This habitat had a significant higher relative density of nematodes provided with denticles than matte (paired t-test: t = 9.3978; P = 0.0111*) and adjacent sediment (paired t-test: t =  − 9.9015; P = 0.0101*), but no significant difference was observed between the last two (paired t-test: t =  − 0.75056; P = 0.5312, Fig. 8B). This functional trait did not seem to be significantly affected by the organic matter content when comparing the two sedimentary habitats, matte, and adjacent sediment (lmer: t-value = 1.992; P = 0.1841).

Discussion

Nematode density is not significantly different between habitats

No significant differences in nematode density were observed between the studied habitats, although we expected the matte to host a higher density than the adjacent sediment. It is worth noting that one of the matte samples (sample 3) showed a significant deviation from the other two samples in terms of biogenic element concentrations. This particular sample exhibited significantly lower levels of organic matter and nitrogen compared to the other sampled mattes (Fig. 5, see the surrounded sample). Nematode density is strongly influenced by nutrient concentration (Moens & Vincx, 1997), and, as such, the atypical parameters of this sample may be a possible reason why we did not find differences between habitats. This is in contrast to the rest of the matte samples, which were characterized by higher levels of organic carbon and nitrogen than the adjacent sediment samples.

Habitat heterogeneity and food specificity shape the community

The matte within the phanerogam canopy had the highest family richness but not the highest genera richness. This habitat was dominated by nematodes from the Desmodoridae family (40%), specifically Perspiria (15%), Bolbonema (11%), and Desmodora (9%). These results align with Novak (1989), where Desmodoridae was the most abundant family in the matte habitat in September. Desmodorids are frequently reported in enriched estuarine sediments with low oxygen levels, although tolerance to anoxic conditions may be partly species specific and/or context specific (Moens et al., 2013). The abundance of this family in the matte habitat, with high organic matter concentration, would support the idea of desmodorids inhabiting enriched sediments. Most desmodorids in this study samples have conical tails, except for Spirinia, which have a conico-cylindrical shape. This is in tune with the fact that nematodes with a conical tail constitute the highest percentage in the matte (56%). In addition, the sediment under the canopy constitutes the habitat with the highest abundance of selective (bacterial) feeding nematodes (1A buccal type; Fig. 6). The constant organic matter flow from the canopy to the matte sediment likely promotes the growth of organic matter-degrading bacteria, providing an abundant food source for nematodes with such a feeding strategy. Furthermore, this habitat had the lowest abundance of nematodes with ocelli (18%). Since the matte is the darkest habitat under the leaves, photoreceptors seem a less useful adaptation in this environment and, therefore, could explain its low abundance.

The leaf habitat exhibited moderate genera richness and an uneven distribution of abundances among taxa. This habitat was dominated by the family Chromadoridae, composing the 80% of the community and mostly represented by the genus Chromadora. This habitat is also characterized by the high occurrence of specimens with denticles (72%) and conical tails (86%). Denticles are known to be used for scraping off bacteria or microalgae from substrata (Moens et al., 2013). During summertime, diatom populations and biofilms covering Posidonia leaves reach their maximum development (Mabrouk et al., 2011), providing a plentiful food source for epistratum feeding nematodes (buccal type 2A; Fig. 6). These nematodes use teeth to scrape diatoms from the surface. In addition, the leaf habitat hosted by far the highest percentage of nematodes with ocelli, likely an adaptation to the well-lighted conditions on P. oceanica leaves.

The adjacent sediment outside Posidonia’s bushes had the lowest nematode richness and the most uniform community due to the lower dominance of specific genera compared to the leaves and matte. Similar to the matte habitat, the most abundant family in the adjacent sediment was Desmodoridae (21%), mainly represented by the genus Bolbonema (16%). In contrast to the other two habitats, the adjacent sediment had a relatively high presence of the Oncholaimidae (20%) and Xyalidae (16%) families. According to Moens et al. (2013), the Xyalidae family may have developed adaptations to cope with physical disturbances (wave actions) and/or low resource availability. This fact would explain their relatively higher abundance outside the canopy where the lack of vegetation exposes the sediment to greater hydrodynamics and lower organic matter content (at least in two of our samples). On the other hand, the Oncholaimidae family is composed by nematodes that are omnivores or facultative predators (Moens et al., 2013) that could leverage whatever food resources are available and survive the unfavorable conditions of the adjacent sediment habitat. Moreover, the nematodes within this family have a good dispersal capacity (Lorenzen et al., 1987; Prein, 1988), which may also explain their prevalence on this exposed substrates (Moens et al., 2013). The abundance of nematodes with elongated/filiform tails appears to be linked to hydrodynamic and sedimentary habitats, as they are most abundant in the adjacent sediment, which combines both characteristics. Lastly, a greater number of nematodes from outside the canopy had ocelli (30%) compared to the matte. Although both are sedimentary habitats, the adjacent sediment is more exposed to direct daylight, where photoreceptors may be more useful.

Sediment granulometry and biogenic elements concentration affect the community

Comparing the two sedimentary habitats, our results show a decrease in genera richness as mean grain size increases. This contradicts the pattern highlighted in numerous investigations (see Heip and Decraemer, 1974; Tietjen, 1977; Heip et al., 1985, 1992; Steyaert et al., 1999; Vanaverbeke et al., 2002, 2011) which relate increased diversity to coarser, more permeable sediments (Vanaverbeke et al., 2011). The composition of the nematode community in our samples is influenced by nitrogen and organic matter content and tends to change with mean grain size, consistent with previous studies (Vincx et al., 1990; Schratzberger et al., 2006, 2008a, 2008b). According to our results, genera such as Bolbonema, Viscosia and Oncholaimellus seem to prefer coarser sediment grain size, while Deontolaimus and Setoplectus may be associated with high nitrogen and organic matter content (Fig. 3B). Nematodes with elongated/filiform tails are positively correlated with increasing clay percentage in our samples. On the other hand, Schratzberger et al. (2007) directly correlated nematodes with elongated/filiform tail abundance with finest sediments (but not differing between silt and clay categories). However, our samples also demonstrate the preference of nematodes with this type of tail toward coarser grain sizes. This suggests that the higher abundance of nematodes with elongated/filiform tails is not solely associated with fine sediments but may be specifically linked to the clay fraction. Riemann (1974) proposed that this tail type could be an adaptation to sand and muddy sediments where only an incomplete interstitial system exists. These elongated/filiform tails would allow the animals to retract from dead-end interstitial passageways that are too narrow to turn around and escape.

Conclusions

The habitat heterogeneity within P. oceanica meadows appears to influence the Nematoda community regarding richness and community composition, albeit not impacting its density. The richest habitat in terms of nematode genera and families was the matte, suggesting that habitat heterogeneity leads to a more diverse community. In line with the third hypothesis, the family Chromadoridae dominating the leaves are epistratum feeders (2A) that exploit the abundant food resources provided by diatoms and biofilms covering the leaves of this phanerogam during the summer season. Noteworthy differences in nematode functional traits were observed among habitats. Specifically, epistratum feeders (2A) were the most abundant type of nematodes in all habitats, but particularly in the leaves. In addition, nematodes with denticles in their buccal cavity were also more prevalent on the leaves, where this structure could be employed to scrap diatoms and biofilms attached to leaf surface. Nematodes with ocelli were notably more represented on the leaves, given the greater sun exposure in this habitat. Concerning the fifth hypothesis, our results indicate that the composition of the nematode community in our samples is influenced by nitrogen and organic matter contents and tends to vary with mean grain size. More precisely, coarser grain size negatively affects genera richness but positively influences the presence of nematodes with elongated/filiform tails. Nematodes with elongated/filiform tails were also associated with the hydrodynamic adjacent sediment habitat and a higher proportion of sediment clay. Future studies could explore seasonal variations in P. oceanica nematode communities and investigate how the life cycle of the phanerogam, and beach hydrodynamics impact the community.