https://orcid.org/0000-0003-3880-2554
Figures
Abstract
Seagrasses are highly sensitive to human-induced disturbances and global environmental changes. Since the 1980s, Zostera marina meadows along the West Swedish coast (Skagerrak) have declined significantly, as evidenced by changes in morpho-anatomical traits, reductions in area coverage, and shifts in associated communities. However, infaunal assemblages within Z. marina meadows remain understudied compared to epifaunal communities and have not been previously used as indicators of seagrass regression. To investigate spatial variability in infaunal composition, we analysed samples from 15 coastal stations at depths of 1.5–3 m depth. Using an n-dimensional hypervolume framework, we assessed functional differences between infaunal and epifaunal communities. We examined infaunal community descriptors—such as species richness and individual abundance—biotic indices, environmental drivers (including wave exposure and Z. marina biomass), and correlations with epifauna. Variability in infaunal composition across sampling stations was primarily driven by differences in the abundance of dominant taxa, including the polychaete Capitella capitata, oligochaetes, nematodes, and chironomids. Several coastal stations, such as Marstrand and Finsbo, were classified as moderately polluted, though biotic indices, i.e., AMBI, M-AMBI and ISI, showed discrepancies. Spatial patterns in infaunal assemblages were mainly influenced by Z. marina biomass and maximum fetch, with a good representation of oligochaetes and chironomids in exposed stations. These findings suggest that infauna respond differently from epifauna but provide valuable additional insights into the ecological status, functional traits, and trophic diversity of Z. marina meadows. Integrating multiple community components is essential for a more comprehensive understanding of the processes and patterns driving seagrass ecosystem regression.
Citation: Riera R, Vasconcelos J, Baden S, Martínez A, Infantes E (2025) Towards a more integrative environmental assessment: Infauna as tool for Zostera marina conservation management. PLoS One 20(10): e0334934. https://doi.org/10.1371/journal.pone.0334934
Editor: Satheesh Sathianeson, King Abdulaziz University, SAUDI ARABIA
Received: March 20, 2025; Accepted: October 3, 2025; Published: October 21, 2025
Copyright: © 2025 Riera et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: This work was supported by the Royal Swedish Academy of Sciences (KVA) grants to Kristineberg Centre for Marine Research and Innovation, University of Gothenburg and Fundação para a Ciência e a Tecnologia I.P. (FCT) [grant numbers SFRH/BSAB/143056/2018, attributed to JV; UIDB/04292/2020, https://doi.org/10.54499/UIDB/04292/2020 and UIDP/04292/2020, https://doi.org/10.54499/UIDP/04292/2020; LA/P/0069/2020, https://doi.org/10.54499/LA/P/0069/2020]. The second author (JV) was supported by the ‘Grants for the Viera y Clavijo training program for researchers’ of the Government of the Canary Islands (VIERA Y CLAVIJO-2022-CIENCIAS1). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Human-driven disturbances are widespread across coastal areas worldwide [1,2], and their extent and variety have increased significantly in recent decades. In addition to long-established stressors such as eutrophication and chemical contamination, recently recognised or increasingly relevant pressures, such as microplastics, underwater noise, and light pollution, are intensifying their impacts on marine ecosystems [3–5]. Minimizing and mitigating these disturbances is an urgent priority at global, national, and regional levels [6]. A precise understanding of their ecological impacts is essential for advancing knowledge of coastal dynamics and processes [7]. Consequently, effective spatial management planning must consider not only the total surface area of ecosystems but also a comprehensive understanding of the biological structures and environmental drivers shaping them [8]. A thorough grasp of biodiversity responses to human-induced pressures is fundamental to ensuring that human activities remain within planetary boundaries and to accurately assessing their environmental impact in the Anthropocene [9].
Ecological baselines—reference states of habitat distribution, species composition, and abundance—are essential for informing conservation and management strategies [10]. Biodiversity monitoring programs provide crucial data on ecosystem changes, guiding research, conservation assessments, and future planning [11]. However, despite their importance, biodiversity data remain scarce and fragmented, with substantial geographic and taxonomic gaps, compounded by inconsistencies in sampling methodologies. These challenges are more pronounced in marine environments, where data collection requires extensive logistical resources, including specialized equipment, boat access, and substantial financial investment [12].
Biodiversity monitoring is particularly effective for assessing the environmental quality of marine habitats [13]. Polluted sites typically exhibit shifts in species composition, favouring taxa that thrive under altered environmental conditions. Several biotic indices have been developed to quantify ecological status based on benthic community structure. These include the Multivariate-AZTI Marine Biotic Index (M-AMBI), the Benthic Quality Index (BQI), Norwegian Sensitivity Index (NSI), Norwegian Quality Index 1 (NQI1), the Infaunal Trophic Index (ITI) and Infaunal Sensitivity Index (ISI) [14]. Most of these tools are based on the Pearson-Rosenberg model [15] which describes community responses to organic enrichment and disturbance along environmental gradients. These indices were developed to assess the ecological quality of soft-bottom benthic habitats, focusing primarily on subtidal infaunal communities [16]. As a result, they often overlook valuable ecological information provided by other components of the benthic assemblage, such as epifauna and demersal fish.
Prioritizing the monitoring of highly impacted environments is crucial for conservation and coastal management. Among marine ecosystems, seagrass meadows have been a focal point of conservation efforts due to their extensive global decline [17]. The loss of seagrass meadows extends beyond the disappearance of Zostera marina itself; it also disrupts associated communities of infauna, epifauna, and fish, leading to a sharp decline in coastal biodiversity [18,19]. Despite the recognized importance of ecological baselines for coastal management, integrative studies incorporating multiple seagrass-associated communities remain scarce (though see [8] for an exception). This is paradoxical given the well-established role of benthic invertebrates in bentho-pelagic coupling. These organisms play a fundamental role in coastal ecosystem processes, including nutrient cycling, detritus decomposition, and carbon transfer to higher trophic levels [20–23]. Incorporating multiple faunal communities into seagrass monitoring programs could enhance our understanding of ecosystem dynamics and the effects of anthropogenic disturbances, particularly when considering long-lived species with varying mobility, such as infauna and epifauna [24].
Infauna and epifauna differ significantly in trophic composition (e.g., infauna are predominantly detritivores, whereas epifauna are often herbivores), life history traits (e.g., prey-predator interactions), and niche utilization. These differences could provide deeper insights into ecological patterns and processes occurring within seagrass meadows [8]. More importantly, their contrasting responses to disturbances—due to differences in mobility—may offer additional perspectives on ecosystem changes. While infaunal species generally have limited mobility and directly exposed to environmental stressors, epifauna, with their greater mobility, can potentially evade localized disturbances within the same meadow [25]. However, these distinctions have yet to be explicitly tested.
This study aims to investigate the potential contributions of infaunal communities in Zostera marina meadows to biodiversity monitoring and environmental assessments. We focus on infaunal communities along the Skagerrak coast (Swedish west coast), spanning approximately through a coastline distance of 200 km [26]. Our main objective is structured into five specific aims: (i) to characterise the infaunal communities in terms of taxonomic composition, abundance, species richness, and functional traits; (ii) to assess the ecological health status of the sampled seagrass meadows using established biotic indices (M-AMBI, ISI, NSI, NQI1, ITI, and BENTIX); (iii) to explore the factors influencing the taxonomic and functional composition of infaunal assemblages across sampling stations, considering both Zostera-related variables (e.g., Z. marina biomass and meadow surface area) and environmental factors (e.g., hydrodynamic exposure); and (iv) to determine the relationship between infaunal and epifaunal assemblages across coastal sites, and (v) evaluate whether epifauna can serve as a proxy for long-term eutrophication effects in the region [26,27]. By integrating taxonomic and functional dimensions of biodiversity with environmental context, this study contributes to a more holistic understanding of how benthic assemblages respond to multiple pressures in temperate seagrass ecosystems. This multifaceted perspective helps interpret patterns in community structure and ecosystem functioning, while also mitigating some limitations inherent to single-season sampling.
Materials and methods
Study area
A total of 15 Zostera marina meadows were sampled along the Swedish Skagerrak coast in July 2018 (Fig 1, Table 1), coinciding with the peak abundance and biomass of epibenthic invertebrates [28]. These meadows were located in shallow waters (1–3 m depth) and spanned approximately 200 km of coastline, encompassing a range of hydrodynamic conditions.
(a) Locations of the 15 sampled seagrass meadows in Sweden. (b) Detailed view of the study sites along the Swedish West Coast: 1. N. Lindholmen; 2. Kvarnekilen; 3. S. Stridsfjorden (Sannäsfjorden); 4. Kämpersvik; 5. Valön; 6. Bottnefjord; 7. Finsbo; 8. Lindholmen; 9. Slussen; 10. Hjältön; 11. Skalhavet; 12. Björnholmen; 13. Kåkenäs; 14. Marstrand; 15. Gottskärsviken. (c) Infauna sampling in a seagrass meadow using a sediment core (QGIS, v3.2, https://qgis.org/).
To evaluate the influence of hydrodynamics on eelgrass fauna composition, we quantified wave exposure using the maximum fetch for each site (Table 1). Maximum fetch is defined as the longest horizontal distance over which wind-driven waves can develop and is considered the most reliable predictor of sediment composition. This is because storms originating from non-dominant wind directions can also affect sediment properties. Given that wave exposure influences sediment grain size, organic content, and nutrient availability in eelgrass sediments [29], we selected meadows along a gradient of wave exposures. For clarity, we categorized these into four exposure levels based on maximum fetch values: sheltered (0.5–1.5 km), semi-sheltered (1.9–2.4 km), semi-exposed (5.1–6.9 km), and exposed (8.7–13.3 km) [30]. However, data on sediment composition were not collected from the sampling sites.
Field sampling and sample processing
Infaunal samples were randomly collected using a PVC core (Ø10 cm, length 30 cm) at the selected stations. At each meadow, cores were inserted to a depth of 20 cm in vegetated habitats and immediately transferred to plastic bags. A total of six cores were taken per location, with each replicate spaced at least 2 m apart to minimize the bow-wave effect. Samples were collected by snorkelling. In the laboratory, the samples were sieved through a 0.5 mm mesh, after which the infauna was collected, preserved in 70% ethanol, and stained with Bengal rose. Specimens were identified to species level whenever possible and counted using a stereomicroscope.
For epifaunal samples, six randomly selected eelgrass specimens with associated fouling (detritus, epiphytes, and fauna) were collected per meadow using a plankton net with a 200 µm (0.2 mm) mesh size, mounted on a frame enclosing an area of 35 × 35 cm (0.123 m²). Each sample was spaced at least 5 m apart and was collected by snorkelling. The coverage of eelgrass and macroalgae was estimated in the field using a 50 × 50 cm frame placed within the meadow, with replicates spaced at least 5 m apart.
In the laboratory, Z. marina leaves were rinsed with freshwater to remove detritus, and the associated epiphytes and fauna were sieved through a 1 mm (1,000 µm) mesh (“large epifauna”) and a 200 µm (0.2 mm) mesh (“small epifauna”). Although both fractions of epifauna occupy the lowest trophic levels, they are considered distinct assemblages due to previously reported differences in species composition [28,31,32]. Therefore, they were analysed separately in this study (see Riera et al. [26] for further details on epifauna sampling and Zostera biomass).
All plant material was collected in accordance with national and international guidelines and legislation, with efforts made to minimize disturbance to the sampled meadows.
Functional space characterization
Our functional analyses followed the general protocol proposed by Mammola et al. [33], beginning with the creation of a trait matrix based on input data from the associated communities of Z. marina meadows. This matrix was then used to delineate the multidimensional trait space to describe patterns and responses to ecological questions. In this analysis, we aimed to assess whether functional space differed between infauna and small and large epifauna. We expected to observe greater polarization in functional trait distributions among these communities, potentially influenced by wave exposure.
To explore and visualize the functional differences between infaunal and epifaunal communities in Z. marina meadows, we constructed a functional trait matrix for each of the 70 recorded taxa using published literature (Table 2). This matrix was then used to represent the functional space of each community through a geometric n-dimensional hypervolume approach [34,35]. The functional characteristics of epifaunal and infaunal communities were compared to assess the extent of overlap between them, indicating whether they fulfil similar roles in ecosystem functioning. This probabilistic method employs high-dimensional kernel density estimations to delineate the shape and volume of the multidimensional space [35].
Six biological traits were selected for their relevance in distinguishing among the three studied faunal communities—namely, infauna, large epifauna, and small epifauna (Table 2, S1 Table) [36,37]. These traits represent key aspects of organism morphology (“maximum body size,” “presence of elongated body,” and “presence of eyes”), feeding strategies (“trophic guild”), behavior (“mobility”), and life history characteristics (“larval development”). Each trait reflects a specific dimension of ecological functioning in benthic communities: maximum body size is related to resource use and susceptibility to predation [36]; trophic guild captures the species’ role in energy flow (e.g., detritivory vs. suspension feeding) [36,38]; mobility influences sediment reworking and exposure to environmental stressors [37]; the presence of eyes is associated with predator avoidance and active foraging [37,39]; an elongated body facilitates burrowing and sediment penetration [38]; and larval development mode relates to dispersal capacity and recovery potential following disturbance [38]. When information on a specific trait was unavailable for a given taxon, zero values were assigned for each trait category, and the taxon was excluded from trait weighting calculations. The selected functional traits are commonly used in invertebrate studies [36–39].
Trait categories were defined for each of the 70 recorded taxa in infauna, large epifauna, and small epifauna communities in the studied Zostera marina meadows (S1 Table). Trait selection and classification were based on commonly used references in trait-based benthic studies [36–39], and specific trait sources are cited below. Definitions of each trait and their ecological relevance are provided in the main text.
As our dataset consisted predominantly of categorical traits, with only “maximum body size” as a continuous variable, we used the Gower dissimilarity measure to accommodate mixed data types and performed Principal Component Analysis (PCA) to extract orthogonal morphological axes [40]. Hypervolumes were then calculated using the package BAT v. 2.7.0 [41] in the software R [42], implementing a Gaussian kernel density estimate. The Gaussian kernel density estimation was selected because it allows for a probabilistic, rather than binary, characterization of functional space [40], a method that has been successfully applied to capture functional variation in similar marine environments [43].
To determine whether infaunal and epifaunal communities across meadows with different exposure levels were subjected to distinct filtering processes, we quantified the dispersion of functional space using the kernel.dispersion() function and the divergence method [40].
Statistical analysis
Univariate metrics, including abundance, species richness (number of taxa), Shannon–Wiener Index (H’ [44]), Hurlbert Index (ES50 [45]), Pielou’s Evenness (J’ [46]), AMBI (AZTI’s Marine Biotic Index [16]), Norwegian Sensitivity Index (NSI [47]), Indicator Species Index (ISI [47]), and BENTIX [48], were calculated for infauna, large epifauna, and small epifauna. Additionally, as multimetric or multivariate methods, M-AMBI (multivariate AMBI [16,49]) and BQI (Benthic Quality Index [13,50]) were applied. With the exception of BQI (calculated as average abundances per station), all indices were computed at the replicate level. AMBI and M-AMBI were calculated using AMBI index software v5.0 (http://ambi.azti.es), while NSI, ISI, and BENTIX were computed using the BBI R package v. 0.3.0 [51]. The remaining metrics were calculated using the vegan R package v. 2.6–6.1 [52]. The BQI formulation used in the Swedish assessment within the Water Framework Directive (WFD) was applied in this study [13].
Boundary settings for the BQI classification followed Rosenberg et al. [50] in accordance with WFD recommendations, where coastal environmental status is categorized into five levels. In the present study, BQI values ranged from 1.70 to 7.12 (reference value) for coastal stations at depths <20 m, corresponding to the five WFD ecological status classes. The classification was as follows: BQI ≥ 5.70: “High” ecological status; BQI ≥ 4.2: “Good” ecological status; BQI ≥ 2.85: “Moderate” ecological status; BQI ≥ 1.42: “Poor” ecological status; BQI < 1.42: “Bad” ecological status.
To ensure statistical independence and avoid pseudoreplication, replicate cores were first averaged within each site (n = 15 meadows). All subsequent multivariate and regression-based analyses were therefore conducted at the site level, with one set of response variables (infaunal composition and diversity indices) per meadow.
The analysis minimised overfitting from collinearity and ensured model stability by calculating site-level pairwise Spearman’s rank correlations among all candidate environmental predictors using the corrplot R package v.0.95 [53]. Where two predictors were stronlgy correlated (R² > 0.6 [54]), only the biologically most relevant predictor [55] was retained. This screening yielded eight site-level predictors: Zostera marina biomass, Chorda filum biomass, Fucus serratus biomass, Dictyota sp. Biomass, Polysiphonia sp. Biomass, Cladophora sp. Biomass, Aglaothamnion sp. Biomass, and Maximum fetch.
To assess patterns in infaunal assemblage structure among the 15 Zostera marina meadows, non-metric multidimensional scaling (n-MDS) was conducted in three dimensions (k = 3), based on the Bray-Curtis dissimilarities of relative abundances. Abundance data were standardised by row totals after adding a small constant, without transformation. Ordinations were performed using the vegan R package v2.6.6. [52]. Species potentially driving site distribution patterns were identified using the envfit function in vegan [52], retaining taxa with significant correlations (p ≤ 0.013).
Distance-based redundancy analysis (db-RDA) and multivariate generalised linear models (GLMs) were applied, using the averaged site-level community matrix (n = 15) as the response data and eight predictor variables as covariates. To test whether environmental variation (e.g., Zostera marina biomass, eelgrass coverage, and associated algal biomasses) structured infaunal assemblages across the 15 Z. marina meadows, we performed a db-RDA [56] with the dbrda() function from the vegan R package [52], computing Bray–Curtis dissimilarities on square-root-transformed, site-level relative abundances. Predictor collinearity was screened using Variance Inflation Factors (VIFs), adopting a threshold of <10. Model and axes significance were tested by permutation (999 permutations) with anova.cca() function. Ordination was visualised with a biplot showing sites grouped by wave exposure category (sheltered, semi-sheltered, semi-exposed, exposed), and environmental vectors indicating the direction and strength of association.
Complementary inference used GLMs fitted to the site-level infaunal abundance matrix with manyglm() function (mvabund R package v.4.2.1) [57], specifying a negative binomial distribution to accommodate overdispersion. Candidate predictors were selected a priori based on the db-RDA results. Significance (deviance tests and p-values) was evaluated with anova.manyglm() function using PIT-trap resampling (999 iterations). Model assumptions were evaluated by visual inspection of residual plots and quantile-quantile (QQ) plots [54]. This approach provided a robust, model-based test of the multivariate effect of environmental predictors on infaunal community composition.
Correlations between univariate variables (e.g., individual abundance, number of species, H’, J’, and ES50) for infaunal and epifaunal communities, as well as environmental variables (Z. marina and epiphyte biomass), were tested using the Pearson product-moment correlation when both variables followed a normal distribution. When normality was not met, the Spearman rank correlation was applied to assess the degree of association between variables.
The evenness of trait distributions within the total functional space was assessed using the kernel.evenness function of BAT R package v.2.9.6 [32], which quantifies the overlap between the observed functional hypervolume and a theoretical hypervolume with evenly distributed traits and abundances. The significance of observed differences in functional richness, dispersion, and evenness was tested using a Wilcoxon test.
Results
Infauna composition
A total of 18,078 individuals were collected, representing 37 taxa across eight taxonomic groups: Polychaetes, Mollusks, Oligochaetes, Insects, Amphipods, Echinoderms, Nemerteans, and Nematodes (S2 Table). The most abundant taxa included the polychaete Capitella capitata (2,885 ind.m ⁻ ², 16% of total abundance), followed by oligochaetes (2,650 ind.m ⁻ ², 15%) and chironomids (2,454 ind.m ⁻ ², 14%). In contrast, nine taxa—Aonides oxycephala, Magelona sp., Nephtys sp., Phyllodoce maculata, Sphaerosyllis histryx, Spionidae sp. 1, the mollusk Macoma balthica, the echinoderm Asterias rubens, and Nemerteans—were recorded in only a single replicate, indicating their rarity.
Species richness varied among sampling stations, with the highest diversity recorded at Kämpersvik (17 species) and Skalhavet (15 species). The lowest richness was observed at Gottskärsvikens and Kvarneviken (6 species each) (S2 Table). Mean abundance also fluctuated across stations, with Kämpersvik, Skalhavet, and Marstrand exhibiting the highest densities (>2,000 ind.m ⁻ ²), while Gottskärsvikens (177 ind.m ⁻ ²) and Kvarneviken (314 ind.m ⁻ ²) had the lowest (Fig 2, Table 1).
Mean abundances (ind.m-2 ± SE) of the most abundant infauna species in each of the 15 Zostera marina meadows sampled in 2018 along the Swedish west coast.
The n-MDS ordination plot (Fig 3) revealed no clear clustering of infaunal composition across meadows, although some site centroids were positioned farther from the ordination centre, reflecting community divergence. Importantly, Finsbo (FNSB) was distinguised by a high chironomids abundance, Lindholmen (LDHL) by elevated densities of Corophium volutator and Ericthonius difformis, Marstrand (MRST) by Scoloplos armiger, North Lindholmen (NLH) by Capitella capitata and Kvarnekilen (KKNS) by the highest nematode abundances (S2 Table). Overall, infaunal composition exhibited high spatial variability among meadows, driven primarily by differences in the relative abundances of Oligochaetes, Corophium volutator, Nematodes, Scoloplos armiger, Erichtonius difformis, Chironomids, Capitella capitata, and Gammarus locusta (Fig 3).
Db-RDA showed that maximum fetch, Z. marina biomass, and Chorda filum biomass were the most influential predictors of infaunal assemblage variation across sites. The first two canonical axes explained 61.4% of the total variation (Axis 1 = 33.7%, Axis 2 = 27.7%) (Fig 4).
The ellipses represent 95% confidence intervals around replicate cores grouped by wave exposure category (sheltered, intermediate, exposed). Black arrows represent the most influential environmental variables structuring infaunal communities across Zostera marina meadows. Only the first two canonical axes are shown.
Comparisons of nested Generalized Linear Models (GLMs) using analysis of deviance showed consistent relationships between environmental predictors and infaunal composition, with Akaike Information Criterion (AIC) supporting the same pattern. Both models achieved comparable fit across the 15 Z. marina meadows. Model 1 (Z. marina biomass, Chorda filum biomass, Fucus serratus biomass, and maximum fetch), identified Z. marina biomass (p = 0.036) and maximum fetch (p = 0.042) as significant predictors. Model 2 (Model 1 plus biomass of Polysiphonia sp., Cladophora sp., Aglaothamnion sp., and Dictyota sp.), produced similar results: Z. marina biomass remained significant (p = 0.039), and maximum fetch was marginally significant (p = 0.080). In both models, the remaining algal biomasses contributed little or no explanatory power. In both models, the remaining algal biomasses contributed little additional explanatory power. Overall, the GLMs consistently highlighted Z. marina biomass and maximum fetch as the primary predictors of infaunal community composition (Table 3).
Biotic indices
Contrasting results were obtained in the ecological status assessment of the sampling meadows. AMBI values presented the most optimistic scenario, varying widely across meadows, from 1.17 (“undisturbed”) in Gottskärsviken to 4.49 (“moderately disturbed”) in Marstrand (Table 4). The mean AMBI value was 3.15, corresponding to a “slightly disturbed” ecological status. With the exception of Gottskärsviken (“undisturbed”), all stations fell into either the “slightly disturbed” or “moderately disturbed” categories.
Due to a low number of species in Gottskärsviken and the dominance of chironomids and oligochaetes in Kåkenäs, NSI and ISI indices were not assigned to these stations (S2 Table). In contrast, the BENTIX index exhibited little variation, ranging from 2 to 2.51 (Table 4).
M-AMBI varied between 0.48 (Finsbo) and 0.91 (Skalhavet), with a mean of 0.69 (“moderately disturbed”). Three meadows were classified as “undisturbed”: Lindholmen (0.77), Kämpersvik (0.84), and Skalhavet (0.91). Two meadows were categorized as “moderately disturbed”: Finsbo (0.48) and Kvarnekilen (0.51). The remaining sampling stations were identified as “slightly disturbed”, with values ranging from 0.56 to 0.76 (Table 4).
BQI varied greatly across the sampled Z. marina meadows, with a mean value of 4.31, corresponding to a “Good” ecological status. The lowest BQI values were recorded in Kvarnekilen (1.71, “Poor”) and Finsbo (1.70, “Poor”), whereas the highest were found in Lindholmen (7.12, “High ecological status”) and Skalhavet (6.67, “High ecological status”) (Table 4).
Comparative functional approach between infauna, epifauna, and environmental variables
Analyses used the first four principal components, which together explained 91.4% of the variance, with the default per-axis bandwidth. Twelve replicates with one or no species were excluded. From the remaining data, 237 hypervolumes were reconstructed. The functional space showed only slight polarization (S1 Fig.); nevertheless, functional diversity, dispersion, and evenness differed significantly among assemblages. Specifically, functional diversity was significantly higher in large epifauna compared to infauna (W = 3531, p = 0.026) and between small and large epifauna (W = 4112, p < 0.05). Functional evenness was significantly lower in small epifauna compared to both large epifauna (W = 6067, p < 0.001) and infauna (W = 532, p < 0.001) (Fig 5f). Marginally significant differences were also found between large and small epifauna (W = 4103, p = 0.009) and between small epifauna and infauna (W = 3608, p = 0.082) (Fig 5a–5c). Within each community, no clear trends were observed regarding the degree of wave exposure across meadows (Fig 5d–5e).
Overall differences in (a) functional richness, (b) functional dispersion, and (c) functional evenness between habitats, as well as across meadows with varying degrees of wave exposure (d–f). Epifauna data extracted from Riera et al. [26].
Comparison of infaunal and epifaunal communities based on univariate descriptors (Shannon diversity, Pielou’s evenness, Hurlbert index, individual abundance, and species richness) showed no consistent patterns (Table 5). Additionally, correlations between infaunal descriptors and biomass of Z. marina and associated algae revealed no clear trends, except for two significant relationships, (i) Species richness vs. Z. marina biomass and (ii) Species richness vs. Polysiphonia sp. biomass (Table 5).
Discussion
Infaunal assemblages dynamics and environmental drivers
Infaunal assemblages varied among meadows in the study area, primarily due to shifts in the most abundant taxa, namely the capitellid polychaete Capitella capitata, Ericthonius difformis, Scoloplos armiger, oligochaetes, nematodes, and chironomids. In contrast, non-dominant taxa played a minor role in differentiating infaunal assemblages among the sampled stations. Although Zostera marina meadows are known to provide stable and favourable conditions for infaunal communities – such as sediment stability, reduced water movement, and increased sedimentation facilitated by their rhizomes and roots [58] – the studied meadows supported a low-diversity infaunal assemblage. Our site-level analysis highlights that variation in Z. marina biomass and Maximum Fetch are the primary environmental drivers structuring infaunal communities.
Studies on infauna from Scandinavian Z. marina meadows are limited and heterogeneous in terms of methodologies and spatio-temporal sampling efforts [19,59–61]. Collectively, these studies indicate that infaunal abundance in Z. marina meadows is three times higher than in adjacent non-vegetated sediments [59,61]. This difference may be attributed to variations in grain size and organic content (e.g., Dahl et al. [28]; Baden & Pihl [30,60]) as well as the increased habitat complexity provided by rhizomes.
Wave exposure can influence meadow distribution [62] and reduce seagrass survival through erosion [63,64], which in turn affects sediment grain size and habitat suitability for benthic invertebrates. Grain size is largely governed by flow velocities, particularly by the threshold of critical shear stress required to mobilise sediment particles [65,66]. Species richness within infaunal communities often increases with coarser sediments due to enhanced pore-water flow [67,68] due to enhanced pore-water flow, as observed in exposed sampling stations. Although we lacked direct grain size measurements or finding that Maximum Fetch significantly predicts community composition suggests that hydrodynamic forcing plays a key role.
Seagrass canopies attenuate wave energy by slowing the water flow beneath the leaves, promoting fine-sediment deposition and creating stable microhabitats for benthic organisms, including infauna and larvae settlement [69–72]. In our study, wave exposure – quantified by Maximum Fetch- emerged as one of the most influential environmental drivers shaping infaunal community composition, alongside with Zostera marina biomass. However, species richness did not vary consistently with exposure level: the highest value was recorded at a semi-sheltered meadow (Kämpersvik), while both sheltered (Kvarnekilen) and exposed (Gottskärsviken) meadows exhibited the lowest richness (Table 5). Other univariate descriptors, such as individual abundance and evenness, also showed no clear patterns across the exposure gradient.
Historical data are available for Norra Lindholmen (1999), Finsbo (1980–82, 1999), and Lindholmen (1982, 1999) [19,60]. Infaunal abundance at these stations exhibited significant spatial and temporal variability. However, a marked 15-fold decline has been reported when comparing data from the 1980s and 1990s to the present study. A similar downward trend was observed in epifaunal assemblages of Z. marina in the study area, largely driven by declines in small amphipods and Mytilus edulis plantigrades [19,25,60,73]. Infaunal changes over the past four decades appear closely linked to declining coastal environmental quality.
Biotic indices and constraints in ecological assessment
Ecological status was assesses using six biotic indices: AMBI, M-AMBI, BENTIX, ISI, NSI, BQI. These revealed a wide range of conditions across stations, from “Bad” to “Very Good.”. ISI, NSI, and BENTIX displayed the greatest variability among meadows, while AMBI and M-AMBI yielded more intermediate scores. Most Zostera marina meadows were neither heavily disturbed nor fully undisturbed. Some sampling stations (e.g., Marstrand and Finsbo) were consistently classified as moderately disturbed, reflecting high abundances of the opportunistic polychaete Capitella capitata (Marstrand) and chironomid larvae (Finsbo). These taxa are widely recognised as indicators of degraded water and sediment conditions [74,75] and contributed to reduced richness and evenness. Continued inter-calibration of biotic indices across water body types remains essential for ensuring consistent and reliable ecological assessments [76]. At Marstrand, long-term decline in eelgrass cover has been attributed to sediment resuspension and associated turbidity [26,77], while Finsbo proximity to a busy ferry route may contribute to ongoing environmental stress.
Caution is warranted when interpreting biotic index results, as several stations exhibited very low species richness, while others were dominated by taxa identified only to higher taxonomic levels- factor that affected the performance of indices such as BENTIX, ISI, and NSI. Additionally, a substantial proportion of taxa could not be assigned to specific ecological groups. The indices applied in this study have traditionally been used in unvegetated subtidal sediments at depths greater than 5 m and to our knowledge, this represents their first application in shallow Zostera marina meadows (1–5 m depth). Although not originally designed for intertidal or shallow habitats, recent studies have shown promising outcomes when applied to broad intertidal zones with tidal ranges exceeding >2 m (e.g., [78]). However, shallow subtidal environments (1–5 m) remain poorly represented in ecological assessments compared to deeper sandy seabeds, including those within fjords, coastal lagoons, and estuaries (see [13,78]). In this study, sampling was conducted in Z. marina meadows at depths between 1.5 and 3 m. Early work by Gislen [79] documented eelgrass beds ranging from 0.9 to 8.4 m, whereas the current mean maximum depth is approximately 4 m [26,80], reflecting a notable reduction in Z. marina habitat along the Swedish Skagerrak coast since 1926.
Epifauna-infauna interactions, functional traits and long-term trends
Changes in the composition and proliferation of epiphytic algae on Z. marina leaves may indirectly influence infaunal assemblages, given the trophic and functional interdependence between epifaunal and infaunal communities. However, no significant correlations were found between infaunal and epifaunal assemblages at the sampled meadows, suggesting that these groups respond differently to variation in Z. marina meadow structure. One possible explanation lies in the influence of seagrass roots and rhizomes on sedimentary conditions, including reduced sediment erosion [72] and microbial-driven processes such as nutrient cycling and organic matter decomposition [81]. Notably, belowground biomass production in many seagrasses can equal that of leaf tissue [82]. Several studies have highlighted the importance of buried seagrass structures in shaping infaunal communities. Although some degree of epifauna-infauna interdependence is likely, epifaunal assemblages appear to respond more directly to aboveground features, with denser Z. marina meadows typically supporting higher epifaunal diversity and abundance than sparse ones [26].
Functional hypervolume analysis of large and small epifauna, as well as infauna, revealed limited differentiation among these communities. Nevertheless, some functional metrics differed significantly across assemblages. These findings indicate that both epifauna and infauna communities contribute similarly to the overall functional space of Z. marina meadows, with no distinct segregation in trait composition. Their contributions, however, vary in quantitative terms. Large epifauna and infauna were predominantly composed of species with intermediate densities, whereas small epifauna were dominated by two to three amphipod species, resulting in lower functional diversity and evenness. A relatively high number of species were shared between epifaunal and infaunal habitats. Although typically recorded at low abundances, these species contributed to the homogenization of functional space, particularly when traits were assessed at the species level rather than the individual level. To better resolve potential functional differences, future studies should incorporate measurements of morphological traits, particularly body size, at the individual level within each environment. This could reveal stronger functional contrasts, as Z. marina leaves and sediments may serve as distinct microhabitats for different developmental stages of a given species. Similar ontogenetic habitat shifts have been documented in meiofauna within Posidonia oceanica meadows, where dominant mite species display clear life-stage-specific habitat preferences d [42,43].
From a temporal perspective, epifaunal assemblages exhibited clear trends when compared to historical data, reflecting broad regional shifts in Z. marina-associated fauna over the past two decades [26]. In contrast, the paucity of historical data on infaunal communities limits our ability to assess long-term changes in these assemblages. The regression of Z. marina meadows over recent decades appears to affect not only the seagrass itself—an important ecosystem engineer—but also associated biota, including epifauna, infauna, and fish. While epifaunal trends were relatively coherent across meadows, infaunal communities showed high spatial variability, suggesting site-specific differences among meadows. Riera et al. [26,83] demonstrated that large and small epifauna serve as effective proxies for detecting changes in Z. marina meadows in Skagerrak region. In contrast, infaunal patterns were more heterogeneous, and no consistent correlations emerged between epifaunal and infaunal communities. This aligns with findings by Reiss et al. [8], who reported significant correlations between the two groups only in terms of species richness.
Integrative ecological insights and management implications
This integrative analysis identified Zostera marina biomass and Maximum Fetch as key drivers of infaunal community structure along the Skagerrak coast. These factors influenced infaunal composition independently of biotic indices used to assess ecosystem condition and appeared more influential than the presence of other associated faunal groups, such as epifauna. The structural complexity provided by seagrass rhizomes and their stabilising role in exposed environments underscore the ecological importance of Z. marina meadows. Infaunal assemblages offer valuable insights into sediment quality and are sensitive to multiple stressors, including detritus accumulation, hydrodynamic conditions, and contamination. In parallel, the composition of epiphytic algae plays a critical role in shaping epifaunal communities, highlighting the need to consider both above- and belowground processes when assessing the integrity of seagrass ecosystems.
Infaunal communities are widely used as bioindicators of ecological condition, as shifts in their composition often reflect increasing levels of environment disturbance [15]. Given their ecological relevance, Z. marina meadows and their associated infaunal assemblages should be integrated into coastal management strategies, particularly in restoration planning and efforts to mitigate environmental stressors. The pronounced spatial variability in infauna composition observed across the study area also suggests potential differences in underlying ecosystem processes, such as secondary production and fine-scale sedimentary dynamics, which merit further investigation.
The integrative framework applied in this study may also be extended to other coastal ecosystems, including salt marshes, estuarine and lagoon environments. These habitats support diverse infaunal and epifaunal communities that, when assessed together, offer a more holistic view of ecosystem functioning. Such an approach addresses important gaps in ecological evaluation that arise when studies focus on a single biological component—such as epifauna—while overlooking other communities and key environmental drivers like maximum fetch.
Conclusions
This study highlights the complex interplay between physical drivers, habitat structure, and benthic community composition in Zostera marina meadows along the Skagerrak coast. Infaunal assemblages showed marked spatial variability, primarily driven by variation in Z. marina biomass and hydrodynamic exposure (Maximum Fetch). These factors shaped community composition regardless of biotic indices or associated faunal groups, reinforcing the need to consider structural habitat attributes and local environmental forcing in benthic assessments.
Although Z. marina meadows are typically associated with enhanced biodiversity, the observed low infaunal diversity and the dominance of opportunistic taxa in some locations indicate potential environmental stress. Biotic indices provided useful, though occasionally inconsistent, insights into ecological condition due to limitations in taxonomic resolution and the application of indices developed for deeper, unvegetated sediments.
Functional trait analysis revealed that both epifaunal and infaunal communities contribute similarly to the overall ecological functioning of seagrass meadows, despite compositional differences. Their combined assessment provides a more complete picture of benthic ecosystem integrity, especially when informed by traits at the individual level.
Long-term data revealed significant declines in infaunal abundance over recent decades, mirroring known reductions in Z. marina coverage and environmental quality. This emphasizes the importance of long-term monitoring and habitat conservation. The integrative framework applied here—linking physical habitat characteristics with taxonomic and functional metrics of multiple faunal groups—offers a promising approach for ecological assessment and coastal management. Its application to other shallow coastal habitats could improve our ability to detect and respond to environmental change at ecologically meaningful scales.
Supporting information
S1 Table. Trait assignment matrix for infaunal and epifaunal taxa [see 36–39 for details].
https://doi.org/10.1371/journal.pone.0334934.s001
(DOCX)
S2 Table. Mean abundance (ind m-2 bottom) of infauna species associated to Zostera marina meadows, per each of the 15 stations sampled along the Swedish west coast in July 2018.
1 – N. Lindholmen; 2 – Kvarnekilen; 3 – S. Stridsfjorden (Sannäsfjorden); 4 – Kämpersvik; 5 – Valön; 6 – Bottnefjord; 7 – Finsbo; 8 – Lindholmen; 9 – Slussen; 10 – Hjältön; 11 – Skallhavet; 12 – Björnholmen; 13 – Kåkenäs; 14 – Marstrand; 15 – Gottskärsviken.
https://doi.org/10.1371/journal.pone.0334934.s002
(DOCX)
S1 Fig. Functional space of 237 macrobenthic communities associated with Zostera marina, visualised in a 4-dimensional hypervolume.
Communities include large epifauna (light green), small epifauna (dark green), and infauna (brown). Each hypervolume is represented by 1,000 random points, coloured according to community identity. Larger points represent the centroids of each hypervolume. Due to the proximity of centroids, many appear visually superimposed.
https://doi.org/10.1371/journal.pone.0334934.s003
(TIF)
Acknowledgments
We are grateful to the staff from the Kristineberg Center för Marin Forskning Och Innovation of the University of Gotebörg (Lysekil), University of Gothenburg (Sweden) for all their help with logistics.
References
- 1. Halpern BS, Walbridge S, Selkoe KA, Kappel CV, Micheli F, D’Agrosa C, et al. A global map of human impact on marine ecosystems. Science. 2008;319(5865):948–52. pmid:18276889
- 2. Halpern BS, Longo C, Lowndes JSS, Best BD, Frazier M, Katona SK, et al. Patterns and emerging trends in global ocean health. PLoS One. 2015;10(3):e0117863. pmid:25774678
- 3. Depledge MH, Godard-Codding CAJ, Bowen RE. Light pollution in the sea. Mar Pollut Bull. 2010;60(9):1383–5. pmid:20736097
- 4. Amelia TSM, Khalik WMAWM, Ong MC, Shao YT, Pan H-J, Bhubalan K. Marine microplastics as vectors of major ocean pollutants and its hazards to the marine ecosystem and humans. Prog Earth Planet Sci. 2021;8(1).
- 5. Duarte CM, Chapuis L, Collin SP, Costa DP, Devassy RP, Eguiluz VM, et al. The soundscape of the Anthropocene ocean. Science. 2021;371(6529):eaba4658. pmid:33542110
- 6. Aswani S. Perspectives in coastal human ecology (CHE) for marine conservation. Biol Conserv. 2019;236:223–35. https://doi.org/10.1016/j.biocon.2019.05.047
- 7. Ehrnsten E, Sun X, Humborg C, Norkko A, Savchuk O. Slomp CP, et al. Understanding environmental changes in temperate coastal seas: Linking models of benthic fauna to carbon and nutrient fluxes. Front Mar Sci. 2020; 7, 450. https://doi.org/10.3389/fmars.2020.00450
- 8. Reiss H, Degraer S, Duineveld GCA, Kröncke I, Aldridge J, Craeymeersch JA, et al. Spatial patterns of infauna, epifauna, and demersal fish communities in the North Sea. ICES Journal of Marine Science. 2009;67(2):278–93.
- 9. Steffen W, Rockström J, Richardson K, Lenton TM, Folke C, Liverman D, et al. Trajectories of the Earth System in the Anthropocene. Proc Natl Acad Sci. 2018;33:8252–9. https://doi.org/10.1073/pnas.181014111
- 10. Rodrigues ASL, Monsarrat S, Charpentier A, Brooks TM, Hoffmann M, Reeves R, et al. Unshifting the baseline: a framework for documenting historical population changes and assessing long-term anthropogenic impacts. Philos Trans R Soc Lond B Biol Sci. 2019;374(1788):20190220. pmid:31679498
- 11. Proença V, Martin LJ, Pereira HM, Fernandez M, McRae L, Belnap J. Global biodiversity monitoring: from data sources to essential biodiversity variables. Biol Conserv. 2017;213:256–63. https://doi.org/10.1016/j.biocon.2016.07.014
- 12. Edgar GJ, Bates AE, Bird TJ, Jones AH, Kininmonth S, Stuart-Smith RD, et al. New Approaches to Marine Conservation Through the Scaling Up of Ecological Data. Ann Rev Mar Sci. 2016;8:435–61. pmid:26253270
- 13. Leonardsson K, Blomqvist M, Rosenberg R. Theoretical and practical aspects on benthic quality assessment according to the EU-Water Framework Directive--examples from Swedish waters. Mar Pollut Bull. 2009;58(9):1286–96. pmid:19520391
- 14. Puente A, Diaz RJ. Is it possible to assess the ecological status of highly stressed natural estuarine environments using macroinvertebrates indices? Mar Pollut Bull. 2008;56(11):1880–9. pmid:18790507
- 15. Pearson TH, Rosenberg R. Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Oceanogr Mar Biol Ann Rev. 1978;16:229–311.
- 16. Borja A, Franco J, Pérez V. A marine biotic index to establish the ecological quality of soft-bottom benthos within european estuarine and coastal environments. Mar Pollut Bull. 2000;40(12):1100–14. https://doi.org/10.1016/S0025-326X(00)00061-8
- 17. Waycott M, Duarte CM, Carruthers TJB, Orth RJ, Dennison WC, Olyarnik S, et al. Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proc Natl Acad Sci U S A. 2009;106(30):12377–81. pmid:19587236
- 18. Airoldi L, Balata D, Beck MW. The gray zone: relationships between habitat loss and marine diversity and their applications in conservation. J Exp Mar Bio Ecol. 2008;366(1–2):8–15. https://doi.org/10.1016/j.jembe.2008.07.034
- 19. Whippo R, Knight NS, Prentice C, Cristiani J, Siegle MR, O’Connor MI. Epifaunal diversity patterns within and among seagrass meadows suggest landscape‐scale biodiversity processes. Ecosphere. 2018;9(11).
- 20. Möller P, Pihl L, Rosenberg R. Benthic faunal energy flow and biological interaction in some shallow marine soft bottom habitats. Mar Ecol Prog Ser. 1985;27:109–21.
- 21. Mann KH. Production and use of detritus in various freshwater, estuarine, and coastal marine ecosystems. Limnol Oceanogr. 1988;33(4):910–30. https://doi.org/10.4319/lo.1988.33.4part2.0910
- 22. Boström C, Mattila J. The relative importance of food and shelter for seagrass-associated invertebrates: a latitudinal comparison of habitat choice by isopod grazers. Oecologia. 1999;120(1):162–70. pmid:28308048
- 23. Villéger S, Ferraton F, Mouillot D, De Wit R. Nutrient recycling by coastal macrofauna: intra-versus interspecific differences. Mar Ecol Prog Ser. 2012;452:297–303. https://doi.org/10.3354/meps
- 24.
Gray JS, Elliott M. Ecology of marine sediments: from science to management. Oxford: Oxford University Press. 2009.
- 25. Ramsay K, Kaiser MJ, Hughes RN. Responses of benthic scavengers to fishing disturbance by towed gears in different habitats. J Exp Mar Bio Ecol. 1998;224(1):73–89. https://doi.org/10.1016/S0022-0981(97)00170-6
- 26. Riera R, Vasconcelos J, Baden S, Gerhardt L, Sousa R, Infantes E. Severe shifts of Zostera marina epifauna: comparative study between 1997 and 2018 on the Swedish Skagerrak coast. Mar Poll Bull. 2020;158:111434. https://doi.org/10.1016/j.marpolbul.2020.111434
- 27. Baden S, Gullström M, Lundén B, Pihl L, Rosenberg R. Vanishing seagrass (Zostera marina, L.) in Swedish coastal waters. Ambio. 2003;32(5):374–7. pmid:14571969
- 28. Pihl Baden S. The cryptofauna of Zostera marina (L.): Abundance, biomass and population dynamics. Netherlands Journal of Sea Research. 1990;27(1):81–92.
- 29. Dahl M, Asplund ME, Björk M, Deyanova D, Infantes E, Isaeus M, et al. The influence of hydrodynamic exposure on carbon storage and nutrient retention in eelgrass (Zostera marina L.) meadows on the Swedish Skagerrak coast. Sci Rep. 2020;10(1):13666. pmid:32788660
- 30. Moksnes P, Röhr ME, Holmer M, Eklöf JS, Eriander L, Infantes E, et al. Major impacts and societal costs of seagrass loss on sediment carbon and nitrogen stocks. Ecosphere. 2021;12(7).
- 31. Baden SP, Pihl L. Abundance, biomass and production of mobile epibenthic fauna in Zostera marina(L.) meadows, western Sweden. Ophelia. 1984;23(1):65–90.
- 32. Jephson T, Nyström P, Moksnes PO, Baden SP. Trophic interactions in Zostera marina beds along the Swedish coast. Mar Ecol Prog Ser. 2008;369:63–76. https://doi.org/10.3354/meps07646
- 33. Mammola S, Carmona CP, Guillerme T, Cardoso P. Concepts and applications in functional diversity. Funct Ecol. 2021;35(9):1869–85. https://doi.org/10.1111/1365-2435.13882
- 34. Blonder B, Lamanna C, Violle C, Enquist BJ. The n-dimensional hypervolume. Global Ecology and Biogeography. 2014;23(5):595–609. https://doi.org/10.1111/geb.12146
- 35. Blonder B, Morrow CB, Maitner B, Harris DJ, Lamanna C, Violle C, et al. New approaches for delineating n‐dimensional hypervolumes. Methods Ecol Evol. 2017;9(2):305–19.
- 36. Nordström MC, Aarnio K, Törnroos A, Bonsdorff E. Nestedness of trophic links and biological traits in a marine food web. Ecosphere. 2015;6(9):1–14.
- 37. Gladstone-Gallagher RV, Hewitt JE, Thrush SF, Brustolin MC, Villnäs A, Valanko S, et al. Identifying “vital attributes” for assessing disturbance-recovery potential of seafloor communities. Ecol Evol. 2021;11(11):6091–103. pmid:34141205
- 38. Törnroos A, Bonsdorff E. Developing the multitrait concept for functional diversity: lessons from a system rich in functions but poor in species. Ecol Appl. 2012;22(8):2221–36. pmid:23387121
- 39. Jimenez H, Dumas P, Mouillot D, Bigot L, Ferraris J. Harvesting effects on functional structure and composition of tropical invertebrate assemblages. ICES J Mar Sci. 2015;73(2):420–8.
- 40. Mammola S, Cardoso P. Functional diversity metrics using kernel density n‐dimensional hypervolumes. Methods Ecol Evol. 2020;11(8):986–95.
- 41.
Cardoso P, Mammola S, Rigal F, Carvalho J. BAT: Biodiversity Assessment Tools. R package version 2.4.2. 2021. https://CRAN.R-project.org/package=BAT
- 42.
R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. 2021.
- 43. Martínez A, García-Gómez G, García-Herrero Á, Sánchez N, Pardos F, Izquierdo-Muñoz A, et al. Habitat differences filter functional diversity of low dispersive microscopic animals (Acari, Halacaridae). Hydrobiologia. 2021;848(11):2681–98.
- 44. Lloyd H, Zar JH, Karr JR. On the calculation of information - theoretical measures of diversity. Am Mid Nat. 1968;79:257–72.
- 45. Hurlbert SH. The Nonconcept of Species Diversity: A Critique and Alternative Parameters. Ecology. 1971;52(4):577–86. pmid:28973811
- 46. Pielou EC. The measurement of diversity in different types of biological collections. J Theor Biol. 1966;13:131–44.
- 47.
Rygg B, Norling K. Norwegian Sensitivity Index (NSI) for marine macroinvertebrates, and an update of Indicator Species Index (ISI). SNO 6475-2013. Norwegian Institute for Water Research. 2013. https://niva.brage.unit.no/niva-xmlui/handle/11250/216238
- 48. Simboura N, Zenetos A. Benthic indicators to use in ecological quality classification of Mediterranean soft bottom marine ecosystems, including a new biotic index. Mediterr Mar Sci. 2002;3(2):77–111.
- 49. Muxika I, Borja A, Bald J. Using historical data, expert judgement and multivariate analysis in assessing reference conditions and benthic ecological status, according to the European Water Framework Directive. Mar Pollut Bull. 2007;55(1–6):16–29. pmid:16844146
- 50. Rosenberg R, Blomqvist M, C Nilsson H, Cederwall H, Dimming A. Marine quality assessment by use of benthic species-abundance distributions: a proposed new protocol within the European Union Water Framework Directive. Mar Pollut Bull. 2004;49(9–10):728–39. pmid:15530516
- 51.
Cordier T. BBI: Benthic biotic indices calculation from composition data. 2018.
- 52.
Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P. vegan: Community Ecology Package. 2024.
- 53.
Taiyun W, Viliam S. R package ‘corrplot’: visualization of a correlation matrix. 2024.
- 54. Harrison XA, Donaldson L, Correa-Cano ME, Evans J, Fisher DN, Goodwin CED, et al. A brief introduction to mixed effects modelling and multi-model inference in ecology. PeerJ. 2018;6:e4794. pmid:29844961
- 55.
Bolker BM. Ecological models and data in R. Princeton University Press. 2008.
- 56. Legendre P, Anderson MJ. Distance‐based redundancy analysis: testing multispecies responses in multifactorial ecological experiments. Ecol Monogr. 1999;69(1):1–24. https://doi.org/10.1890/0012-9615(1999)069[0001:DBRATM]2.0.CO;2
- 57.
Wang Y, Naumann U, Eddelbuettel D, Wilshire J, Warton D. mvabund: Statistical Methods for Analysing Multivariate Abundance Data. 2022.
- 58. Boström C, Bonsdorff E. Community structure and spatial variation of benthic invertebrates associated with Zostera marina (L.) beds in the northern Baltic Sea. Journal of Sea Research. 1997;37(1–2):153–66.
- 59.
Berg T. Quantitative Studies of Infauna in Zostera marina Meadows on the Swedish West Coast. Biomass and Abundance Correlated to Organic Content of the Sediment (MSc Thesis in Marine Ecology). Department of Marine Ecology, Göteborg University, Sweden. 2000.
- 60. Fredriksen S, De Backer A, Boström C, Christie H. Infauna from Zostera marina L. meadows in Norway. Differences in vegetated and unvegetated areas. Mar Biol Res. 2010;6(2):189–200. https://doi.org/10.1080/17451000903042461
- 61. Infantes E, Terrados J, Orfila A, Canellas B, Alvarez-Ellacuria A. Wave energy and the upper depth limit distribution of Posidonia oceanica. Bot Mar. 2009;52:419–27. https://doi.org/10.1515/BOT.2009.050
- 62. Infantes E, Orfila A, Bouma TJ, Simarro G, Terrados J. Posidonia oceanica and Cymodocea nodosa seedling tolerance to wave exposure. Limnol Oceanogr. 2011;56(6):2223–32. https://doi.org/10.4319/lo.2011.56.6.2223
- 63. Meysick L, Infantes E, Rugiu L, Gagnon K, Boström C. Coastal ecosystem engineers and their impact on sediment dynamics: Eelgrass–bivalve interactions under wave exposure. Limnology & Oceanography. 2022;67(3):621–33.
- 64. Flemming BW. The role of grain size, water depth and flow velocity as scaling factors controlling the size of subaqueous dunes in marine sandwave dynamics, international workshop. In: 2000. 23–4.
- 65. van der Veen HH, Hulscher SJMH, Knaapen MAF. Grain size dependency in the occurrence of sand waves. Ocean Dynamics. 2006;56(3–4):228–34.
- 66. Coblentz KE, Henkel JR, Sigel BJ, Taylor CM. Technical Note: The Use of Laser Diffraction Particle Size Analyzers for Inference on Infauna-Sediment Relationships. Estuaries and Coasts. 2014;38(2):699–702.
- 67. Coblentz KE, Henkel JR, Sigel BJ, Taylor CM. Influence of sediment characteristics on the composition of soft-sediment intertidal communities in the northern Gulf of Mexico. PeerJ. 2015;3:e1014. pmid:26157603
- 68. Infantes E, Orfila A, Simarro G, Terrados J, Luhar M, Nepf H. Effect of a seagrass (Posidonia oceanica) meadow on wave propagation. Mar Ecol Prog Ser. 2012;456:63–72. https://doi.org/10.3354/meps09754
- 69. Turner SJ, Kendall MA. A comparison of vegetated and unvegetated soft sediment macrobenthic communities in the River Yealm, south western Britain. J Mar Biol Assoc UK. 1999;79:741–3. https://doi.org/10.1017/S0025315499000892
- 70. Hendriks IE, Sintes T, Bouma TJ, Duarte CM. Experimental assessment and modeling evaluation of the effects of the seagrass Posidonia oceanica on flow and particle trapping. Mar Ecol Prog Ser. 2008;356:163–73. https://doi.org/10.3354/meps
- 71. Baden S, Hernroth B, Lindahl O. Declining populations of Mytilus spp. in North Atlantic coastal waters—A Swedish perspective. J Shellfish Res. 2021;40(2):269–96. https://doi.org/10.2983/035.040.0207
- 72. Infantes E, Hoeks S, Adams M, van Katwijk MM, Bouma TJ. Seagrass roots strongly reduce cliff erosion rates in sandy sediments. Mar Ecol Prog Ser. 2022;700:1–12. >https://doi.org/10.3354/meps14196
- 73. Lencioni V, Marziali L, Rossaro B. Chironomids as bioindicators of environmental quality in mountain springs. Freshwater Science. 2012;31(2):525–41. https://doi.org/10.1899/11-038.1
- 74. Silva ACF, Tavares P, Shapouri M, Stigter TY, Monteiro JP, Machado M, et al. Estuarine biodiversity as an indicator of groundwater discharge. Estuarine, Coastal and Shelf Science. 2012;97:38–43.
- 75. Poikane S, Zampoukas N, Borja A, Davies SP, van de Bund W, Birk S. Intercalibration of aquatic ecological assessment methods in the European Union: Lessons learned and way forward. Environmental Science & Policy. 2014;44:237–46.
- 76. Song Q, Zou X, Zhang H, Yu W, Zang Z, Wang C. An approach based on M-AMBI for assessing benthic ecological status of a broad intertidal zone: A case study in the Jiangsu intertidal zone, China. Mar Pollut Bull. 2017;116(1–2):87–94. pmid:28041620
- 77. Moksnes P-O, Eriander L, Infantes E, Holmer M. Local Regime Shifts Prevent Natural Recovery and Restoration of Lost Eelgrass Beds Along the Swedish West Coast. Estuaries and Coasts. 2018;41(6):1712–31.
- 78. Borja A, Barbone E, Basset A, Borgersen G, Brkljacic M, Elliott M, et al. Response of single benthic metrics and multi-metric methods to anthropogenic pressure gradients, in five distinct European coastal and transitional ecosystems. Mar Pollut Bull. 2011;62(3):499–513. pmid:21215975
- 79.
Gislén T. Epibiosis of the Gullmar Fjord 1–2. 1877–1927. Kristinebergs Zoologiska Station. 1928.
- 80. Hansen JW, Udy JW, Perry CJ, Dennison WC, Lomstein BA. Effect of the seagrass Zostera capricorni on sediment microbial processes. Mar Ecol Prog Ser. 2000;199:83–96. https://doi.org/10.3354/meps199083
- 81. Abal EG, Loneragan N, Bowen P, Perry CJ, Udy JW, Dennison WC. Physiological and morphological responses of the seagrass Zostera capricorni Aschers, to light intensity. J Exp Mar Biol Ecol. 1994;178(1):113–29. https://doi.org/10.1016/0022-0981(94)90228-3
- 82. García-Gómez G, García-Herrero A, Sánchez N, Pardos F, Izquierdo-Muñoz A, Fontaneto D. Meiofauna is an important, yet often neglected, component of biodiversity in the ecosystem formed by Posidonia oceanica. Invertebr Biol. 2022;141(2):e12377. https://doi.org/10.1111/ivb.12377
- 83. Riera R, Bosch NE, Infantes E, Vasconcelos J. Navigating biodiversity patterns in fragmented seagrass mosaics. Sci Rep. 2025;15(1):8729. pmid:40082602