Ecosystem engineer morphological traits and taxon identity shape biodiversity across the euphotic–mesophotic transition

Abstract

The euphotic–mesophotic transition is characterized by dramatic changes in environmental conditions, which can significantly alter the functioning of ecosystem engineers and the structure of their associated communities. However, the drivers of biodiversity change across the euphotic–mesophotic transition remain unclear. Here, we investigated the mechanisms affecting the biodiversity-supporting potential of free-living red coralline algae—globally important habitat creators—towards mesophotic depths. Across a 73 m depth gradient, we observed a general decline in macrofaunal biodiversity (fauna abundance, taxon richness and alpha diversity), but an increase in beta-diversity (i.e. variation between assemblages) at the deepest site (86 m depth, where light levels were less than 1% surface irradiance). We identified a gradient in abundance decline rather than distinct ecological shifts, driven by a complex interaction between declining light availability, declining size of the coralline algal host individuals and a changing host taxonomy. However, despite abundance declines, high between-assemblage variability at deeper depths allowed biodiversity-supporting potential to be maintained, highlighting their importance as coastal refugia.

1. Background

Marine ecosystem engineers such as corals and macroalgae are of vast ecological significance throughout the world's oceans [1,2]. They support highly biodiverse, unique and productive ecosystems (e.g. [3,4]), which can have positive spill-over effects into neighbouring ecosystems (e.g. [5]). The ecological significance of ecosystem engineers is especially large when, for example, extreme environmental conditions do not allow alternative ecosystem engineers to persist [6].

The mesophotic oceanic zone—typically 30–150 m water depth, or the maximum depth where corals can sustain photosynthetically active symbionts (with a light limit of approximately 1% surface irradiance; [7])—is characterized by diverse and unique benthic ecosystems [8]. Mesophotic ecosystem engineers create complex habitats in an environment where few other photosynthetic organisms can survive through photosynthetic acclimation (e.g. macroalgae [7]) and/or increased reliance on heterotrophy (e.g. corals; [9]). Despite facing numerous anthropogenic disturbances at both local and global spatial scales [10], mesophotic ecosystems are comparatively sheltered from shallow-water stressors, suggesting they may be critical refugia for shallow-water communities [8,11]. However, fundamental questions about the ecological functioning of mesophotic reefs remain [7,12], especially in lesser studied macroalgal mesophotic habitats (e.g. [13,14]), including (i) the ecological community composition at mesophotic depths, (ii) how these communities differ from shallow depths and (iii) what key biotic and abiotic processes shape mesophotic communities [12].

A myriad of environmental changes occurs with increasing ocean depth, including reduced irradiance, reduced temperature and altered water flow, driving significant alterations to the physiological and ecological functioning of macroalgal ecosystem engineers [15]. Macroalgal morphological traits are commonly found to change with depth, associated with e.g. light acclimation, energy provision and environmental gradients (e.g. [16]). This can alter the quantity and quality of biogenic habitat, with cascading effect on the diversity and composition of supported communities (e.g. [17]). Depth-associated abiotic changes can also directly alter associated communities [18]. Macroalgal-associated communities may therefore be expected to significantly change through the euphotic to mesophotic (e.g. [19]).

Free-living, non-geniculate red coralline algae (Corallinaceae, Rhodophyta)—commonly known as rhodoliths or maerl—are macroalgal ecosystem engineers with a cosmopolitan global distribution and are the deepest known macroalga, with coralline algal beds recorded at 270 m depth [20]. Aggregations of free-living red coralline algae create a complex, three-dimensional substrate in otherwise dimensionless soft-bottom seafloors [21]. In turn, shallow-water free-living coralline algal beds have comparable biodiversity other marine biodiversity hotspots such as kelp forests or coral reefs [22]. Morphological traits such as their thallus size and shape are known to affect associated faunal communities through niche provision (e.g. [23]) and also change with depth (e.g. [24]). The interaction between free-living coralline algal morphology and water depth may therefore be crucial in determining the biodiversity provision of coralline algal beds at mesophotic depths.

Here, we tested the hypothesis that the interaction between the physical and environmental factors of habitat morphology and water depth drives a significant biodiversity transition through the euphotic–mesophotic zones. By comparing coralline algal-associated macrofaunal biodiversity between and within community assemblages across a 73 m depth gradient, we gained insights into the key environmental, morphological and taxonomic drivers shaping biodiversity patterns and their potential for biodiversity provision across the euphotic–mesophotic transition.

2. Methods

Free-living coralline algal thalli were collected by hand using SCUBA at five sites spanning a 73 m depth gradient (13, 40, 56, 65 and 86 m depths) around the Fernando de Noronha archipelago (electronic supplementary material, figure S1) across a 5-day period in September/October 2018. At each site, coralline algal thalli (n = 7–27 per site, 65 total) were non-intentionally targeted in intervals of five diver fin kicks, apart from the deepest site where all samples were collected within a 4 m² area due to SCUBA time restrictions. Thalli covered by large fleshy algae (e.g. Dictyotales) were avoided. Thalli were immediately transferred to individual nylon mesh bags (500 µm mesh size) at the seabed to minimize macrofaunal loss. All thalli were returned to the surface within 1 h of collection and stored in the dark at ambient water temperature before transfer to shore. Thalli were returned to the laboratory within 2 h of reaching the surface and immediately fixed in a 10% formalin/seawater solution. In situ photosynthetically active radiation (PAR) at the seabed was recorded at each site (electronic supplementary material, methods S1).

Taxonomic identification of the coralline alga from each free-living coralline algal sample was based on morphoanatomical analyses using the histological methods described in [25] and comparison to known species of the region [13,26–28].

The volume of each coralline algal thallus sampled for biodiversity was measured via liquid displacement. Sphericity (ψ) [29] was determined by measuring the x, y and z axes using digital calipers (±0.01 mm resolution, FisherScientific) [30]. Due to the destructive nature of biodiversity assessment (below), surface complexity could not be conducted on the same samples. Instead, additional thalli (n = 5–16 per site) were collected for the determination of surface complexity: air-dried thalli were three-dimensional scanned at approximately 100 µm resolution (NextEngine, Inc., USA). Surface complexity was calculated as the average of the three-dimensional surface area of three randomly chosen 1 cm² areas from each thallus.

Coralline algal-associated faunal assemblages from each thallus were determined by breaking up the thallus and removing all fauna. The washing water was also filtered using a 500 µm mesh. Vagile macrofauna were preserved in 70% ethanol and identified to Class level. Fauna were counted only when the cephalic region was preserved (Maxillopoda, Malacostraca, Polychaeta and Sipuncula); the central disc was preserved (Ophiuroidea and Echinoidea); soft body parts were present (Mollusca). Unidentifiable individuals accounted for only 0.32% of total faunal abundance.

Univariate diversity matrices of species abundance (N), number of taxa (richness, S) and Shannon–Wiener (H’) diversity were calculated using the ‘Vegan’ package in R v. 3.5.1 [31]. Differences between depths were analysed using ANOVA and a post hoc Least Square Means comparisons with Bonferroni adjustment (‘LsMeans’ package [32]). ANOVA assumptions of normality and homogeneity of variances were tested using the ‘gvlma’ package [33], and data were transformed as necessary.

Faunal–habitat relationships were investigated by constrained additive ordination [34] and reduced-rank vector generalized linear models (RR-VGLMs) [35] using the VGAM 1.1-5 R package [36]. Details on model fitting are provided in electronic supplementary material, Methods S2. All statistical analyses were performed using R v. 4.1.2.

3. Results

Light intensity decreased by two orders of magnitude across the sampling depth range, from 524 µmol photons m⁻² s⁻¹ PAR at 13 m to 6.89 µmol photons m⁻² s⁻¹ PAR at 86 m (electronic supplementary material, figure S2), equivalent to 26% (13 m) to 0.3% (86 m) of maximum surface irradiance.

Thallus complexity significantly declined with depth, with two groupings: 13–56 m and 65–86 m (figure 1a); electronic supplementary material, table S2). Biometric traits of thallus size (volume and diameter) significantly declined with depth (figure 1b,c; electronic supplementary material, table S2), with thalli significantly larger at 13 m compared to all deeper depths (except for 65 m where thallus volume did not differ). Thallus sphericity did not significantly vary with depth (figure 1d; electronic supplementary material, table S2), but was characterized by high variability at 86 m. Taxonomically, the thalli were each composed of single representatives from the Orders Corallinales, Hapalidiales, Peyssonneliales and Sporolithales (electronic supplementary material, figure S3).

Figure 1.

Changes in biogenic drivers of biodiversity across a 73 m depth range. Coralline algal morphological traits of (a) surface complexity (dimensionless), (b) thallus volume (cm³), (c) diameter (cm) and (d) sphericity (ψ). Data presented as mean ± s.e. Letters above symbols indicate statistically different groupings (electronic supplementary material, table S2). For (a): n = 10, 5, 5, 16 and 15 for sites 13, 40, 56, 65 and 86 m, respectively, for (b–d): n = 27, 11, 12, 8 and 7 for sites 13, 40, 56, 65 and 86 m, respectively, except for (c) where n = 10 at 40 m.

Eighty-eight per cent of the taxa associated with the free-living coralline algae were accounted for by Malacostraca (35% of total abundance), Gastropoda (29%), Polychaeta lato sensu (16%) and Ostracoda (8%), with limited depth specificity (figure 2). Ten other faunal classes each comprised less than 4% of the total recorded faunal abundance and demonstrated higher site specificity (figure 2). Ophiuroidea and Sipuncula were only recorded at the shallowest site. The rarest taxa (less than 0.5% total abundance)—Pycnogonida, Echinoidea, Monoplacophora and Actinopterygii—were not observed at 86 m. A single Cephalopoda was recorded at the 56 m site.

Figure 2.

Heatmap of macrofaunal abundance associated with free-living coralline algae across a 73 m depth range. Data presented as % contribution to total faunal abundance and as % abundance per site. n = 27, 11, 12, 8 and 7 thalli for sites 13, 40, 56, 65 and 86 m, respectively; total n = 2170 organisms. White colouring indicates zero observed abundance. (Online version in colour.)

There was a 10-fold and significant decline in faunal abundance with depth (figure 3a; electronic supplementary material, table S3). The highest abundance in a single thallus was 129 individuals (13 m site). A similarly significant decline was observed for taxon richness (figure 3b; electronic supplementary material, table S3). Shannon–Wiener alpha diversity (H′) also declined with depth, with a significant difference between the shallowest and deepest sites (figure 3c; electronic supplementary material, table S3). Beta-diversity (i.e. variation between assemblages) was similar at 13–56 m, highest at 86 m and lowest at 65 m (figure 3d; electronic supplementary material, table S3).

Figure 3.

Macrofaunal diversity associated with free-living coralline algae across a 73 m depth range. (a) abundance (number of individuals per thallus), (b) taxon richness (number of taxa per thallus), (c) alpha diversity measured using the Shannon–Wiener (H′) index and (d) beta-diversity (as multivariate dispersion based on a Bray–Curtis similarity index on square root transformed abundance data). Data presented as mean ± s.e. Letters above symbols indicate statistically different site groupings (electronic supplementary material, table S3). N = 27, 11, 12, 8 and 7 for depths 13 m, 40 m, 56 m, 65 m and 86 m, respectively.

Increasing mean abundance of all macrofaunal classes was characterized by a fitted gradient comprising a negative relationship with depth, PAR and some host taxa (particularly Peyssonneliaceae) and a positive relationship with thallus diameter (electronic supplementary material, figures S4 and S5, and table S4). Thallus sphericity, complexity and volume were not important explanatory variables (electronic supplementary material, figure S4 and table S4). These interacting drivers were integrated into a positive log-linear relationship along the CLO ordination gradient—indicated by consistently positive values of  (electronic supplementary material, figure S6 and table S5). Optimum conditions for an abundance maximum could not be defined for seven of the 14 classes (electronic supplementary material, table S5). The ordination curves for Bivalvia, Maxillopoda and Echinoidea suggest an abundance decline at intermediate gradient values (the composite of the abiotic and biogenic explanatory variables), while Pycnogonida, Monoplacophora and Cephalopoda exhibited an abundance maximum at intermediate gradient values (electronic supplementary material, figure S6). Low total abundances led to Ophiuroidea, Sipuncula and Actiniopterygii being most sensitive to driver change—indicated by the highest coefficients of  (electronic supplementary material, table S5).

4. Discussion

Fundamental questions remain on the ecological role and mechanistic drivers of mesophotic macroalgal habitats [12]. Here, we identified how mesophotic macrofaunal biodiversity is driven by both depth-associated changes in the environment and by biogenic variability of the underlying ecosystem engineer. Despite a marked decline in macrofaunal biodiversity with depth, beta-diversity was highest well into the mesophotic zone—indicative of an ecological ‘poise’ for community biodiversity across the euphotic–mesophotic transition.

Molluscs, crustaceans and polychaetes appear to be consistently abundant in many free-living coralline algal ecosystems worldwide (e.g. [37,38]) and sometimes dominated by Echinodermata (e.g. [39]). In terms of total fauna abundance, the maximum mean abundance (53 individuals per thallus at 13 m depth) is comparable to other shallow-water beds in Brazil and other structurally complex ecosystems such as seagrass meadows [14]. However, we observed a marked decline in faunal abundance with depth. Based on known coralline algal thallus densities at Fernando de Noronha [13], our data suggest at least approximately 15 700–1000 individuals m⁻² at 13–86 m, respectively, plus additional organisms living between thalli and/or in the underlaying sediment. This is an order of magnitude higher than other mesophotic sites (e.g. Malta, Mediterranean Sea—up to 814 macroinvertebrates m⁻² at 46 m) [38]. Given the prevalence of free-living coralline algal beds around the Fernando de Noronha archipelago to at least 100 m depth [13,14], this study demonstrates their ecological importance throughout the euphotic and mesophotic zones of the region.

Coralline algal bed biodiversity typically exhibits a negative relationship with depth [24,14,38]. However, these observations have almost all been restricted to non-mesophotic (less than 30 m) depths, often with only 2–3 depth comparisons and with a focus on fish assemblages (e.g. [40]), limiting our understanding of euphotic–mesophotic biodiversity transitions. Distinct biodiversity shifts between the euphotic and mesophotic were not evident—perhaps due to the continued capacity for photoautotrophy by the coralline algal hosts via low-light acclimation. Coralline algae in shallow waters are typically light saturated. The consequent gradual change in the ecosystem niche provision may allow for a continuous general transition in faunal communities rather than ecological ‘zones’ [41], albeit with taxon-specific sensitivities. A recurring challenge in depth-associated coralline algal biodiversity studies is the co-occurrence of algal morphological and taxonomic changes, which makes it difficult to tease apart the individual role of each biogenic trait, nor the role of biogenic habitat versus the environment (e.g. [14,38]). Our analysis indicates that host diameter and taxonomy are important in describing associated biodiversity, likely via habitat niche provision and hydrodynamic regulation. The negative effect of PAR and host taxonomy on biodiversity may be explained by species-specific production of chemical attractants and/or inhibitors (such as dimethylsulphoniopropionate), perhaps driven by high rates of coralline algal non-photochemical quenching and antioxidant cascade in shallower waters [42]. Experimental investigation using ‘discrete habitat units' (e.g. [43]) might enable a mechanistic understanding of the relative and combined role of environmental and habitat drivers on biodiversity.

Despite reduced abundance, richness and alpha diversity at 86 m, dispersion among communities (beta-diversity) was highest at this depth. The observed increase in beta-diversity combined with the observation that the pool of taxa remained largely unchanged with depth indicates that the carrying capacity of the thalli declined at depth [44]. Mesophotic community assemblages were therefore likely restricted by ecological limits rather than the availability of species [45], driven by a reduction in thallus size and a shift in host taxonomy. However, the ecological role of the habitat (as beta-diversity) was not impacted to the same extent. The limited depth specificity for all major taxa (in terms of presence/absence), combined with the limited motility of these fauna in the adult stage, suggests that deeper sites may therefore act as a long-term depth refugium (as opposed to a short-term depth refuge) [46]. This contrasts to the high depth specificity in reef fish communities [10], suggesting coralline algal biodiversity dynamics and/or invertebrate biodiversity may respond differently to the euphotic–mesophotic transition. Faunal identification to a higher taxonomic resolution would enable deeper functional insight into the refugium potential of these mesophotic communities.

Coralline algal beds have significant economic resource value for industries including carbonate extraction and commercial fisheries; mesophotic beds in particular often coincide with oil and gas activities [47]. Quantitative determination of predictive variables to identify high biodiversity regions (and therefore of high conservation priority) is therefore important for their conservation, particularly where coralline algal beds are extensively distributed. Contrasting community patterns within and between depths highlights a complexity that enables these habitats to be one of the most biodiversity marine ecosystems. Shallow-water coralline algal beds are well known to be threatened by a wide range of environmental stressors [48]. Our results indicate that their deep-water counterparts hold the potential to act as an ecological ‘buffer’. We therefore recommend consideration of the euphotic–mesophotic transition as an ecological gradient rather than as distinct ‘zones’, with formal integration of mesophotic ecosystems into marine protection policies and conservation initiatives [49].

Data accessibility

The data are provided in the electronic supplementary material [50].

Authors' contributions

S.E.V.: data curation, formal analysis, investigation, methodology, writing—original draft and writing—review and editing; B.C.M.: data curation, formal analysis, investigation and writing—review and editing; R.G.B.: formal analysis, methodology and writing—review and editing; G.H.P.-F.: conceptualization, data curation, funding acquisition, investigation, methodology, supervision, writing—original draft and writing—review and editing; T.W.Y.: methodology, software, validation, visualization and writing—review and editing; A.C.F.B.: data curation, methodology and writing—review and editing; G.M.A.-F.: funding acquisition and investigation; A.R.: formal analysis, resources, validation and writing—review and editing; G.A.T.: funding acquisition, project administration, resources, validation and writing—review and editing; I.D.W.S.: funding acquisition, project administration, resources, validation and writing—review and editing; H.L.B.: conceptualization, data curation, funding acquisition, methodology, project administration, resources, supervision, writing—original draft and writing—review and editing. All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Competing interests

We declare we have no competing interests.

Funding

Funding was provided by a Leverhulme Trust Research Project grant (no. RPG-2018-113) to H.L.B., G.A.T. and I.D.W.S., an Engineering and Physical Sciences Research Council grant (no. EP/L017008/1) to G.A.T. and I.D.W.S., and a São Paulo Research Foundation (FAPESP) individual grant (no. 2016/14017-0) to G.H.P.F.