Abyssal seafloor communities cover more than 60% of Earth’s surface. Despite their great size, abyssal plains extend across modest environmental gradients compared to other marine ecosystems. However, little is known about the patterns and processes regulating biodiversity or potentially delimiting biogeographical boundaries at regional scales in the abyss. Improved macroecological understanding of remote abyssal environments is urgent as threats of widespread anthropogenic disturbance grow in the deep ocean. Here, we use a new, basin-scale dataset to show the existence of clear regional zonation in abyssal communities across the 5,000 km span of the Clarion–Clipperton Zone (northeast Pacific), an area targeted for deep-sea mining. We found two pronounced biogeographic provinces, deep and shallow-abyssal, separated by a transition zone between 4,300 and 4,800 m depth. Surprisingly, species richness was maintained across this boundary by phylum-level taxonomic replacements. These regional transitions are probably related to calcium carbonate saturation boundaries as taxa dependent on calcium carbonate structures, such as shelled molluscs, appear restricted to the shallower province. Our results suggest geochemical and climatic forcing on distributions of abyssal populations over large spatial scales and provide a potential paradigm for deep-sea macroecology, opening a new basis for regional-scale biodiversity research and conservation strategies in Earth’s largest biome.
The abyssal seabed lies between water depths of 3,000 and 6,000 m (ref. 1), representing most of the Earth’s surface and harbouring some of its most extensive but least explored ecosystems. This abyssal seabed consists of plains and hills that extend across ocean basins, interspersed with seamounts and subdivided by mid-ocean ridges, ocean trenches and fracture zones1. Lacking sunlight, energy in this environment is highly limited, with detrital particles sinking from surface waters providing the main source of food2, while temperatures are steady at 0.5–3.0 °C (ref. 3) and bottom currents are generally low (0–0.25 m s−1) (ref. 4). As a result, abyssal benthic communities typically exhibit low abundance and biomass compared with other marine environments5 but they support high species richness at regional to landscape scales6,7 and play an active role in the cycling of carbon8. Abyssal habitats are therefore considered to be reservoirs for biodiversity7,9 and sources of important ecosystem services2,10. Despite their remoteness and as a consequence of being energy-restricted and environmentally stable, abyssal habitats are also expected to be highly vulnerable to anthropogenic disturbances including climate change11,12 and emerging industrial activities such as polymetallic nodule mining13,14. To adequately conserve the biodiversity and services of abyssal seabed ecosystems, it is critical to consider and protect their full range of habitats and communities, which requires elucidating the patterns and processes controlling the distribution of abyssal populations over large, regional scales.
Macroecological studies assessing spatial variation on regional to global scales in the deep ocean15,16,17,18 have revealed that faunal body size and abundance gradually decline with decreasing energy availability along gradients such as increasing depth5,19 or latitude16,20. These relationships are thought to constrain higher-order community structure and function21, for example, reducing the diversity of larger-sized taxa with depth22. Predicted declines in species richness with depth are consistent with checklist-based analyses of large marine databases23. Overall, results from large-scale studies in the deep ocean confirm clear variation between abyssal and (shallower) bathyal seafloor ecosystems16,19,22, which encompass a variety of relatively well-known habitats, between 1,000 and 3,000 m depth, including continental slopes, submarine canyons and coral mounds. Consequently, while biogeographic provinces proposed for the deep ocean readily distinguish bathyal and abyssal provinces within oceanic basins24, potential variation within abyssal regions remains largely unresolved a decade after provincial boundaries based on surrogate environmental data were postulated24, since large-scale empirical studies (for example, bottom-up assessments) are lacking for remote abyssal ecosystems. As such, the processes regulating alpha- and beta-diversity across the scale of a seemingly connected abyssal plain remain unclear, particularly for the largest size-class of seafloor organisms, megafauna (animals >10 mm in maximum dimension). Megafauna are a conspicuous component of the abyssal benthos and a common target for investigation in deep-sea spatial ecology13,25,26,27 because occurrences of these taxa can be determined across large spatial scales from seabed imagery28 collected by deep-sea robots29.
To address these knowledge gaps, we compiled a new basin-scale standardized dataset to investigate patterns of abundance and diversity in benthic megafaunal communities across the Clarion–Clipperton Zone (CCZ, northeast Pacific). The CCZ spans ~6 million km2 of abyssal plain in areas beyond national jurisdiction between the Exclusive Economic Zones of Kiribati and Mexico (Fig. 1a) and is the largest area in the world currently in exploration phase for mineral mining30,31. We selected data from 12 expeditions using comparable seabed imaging approaches to study invertebrate benthic megafauna distribution patterns across the CCZ. We collated, reanalysed, aligned and taxonomically standardized seabed imagery data from these studies including three new sites covering a total of >150,000 m2 of seabed, which represents an area two orders of magnitude greater than that commonly assessed in studies of abyssal megafauna (for example, ref. 27,32, <5,000 m2). Our dataset spans 5,000 km across the CCZ (28 geographical locations) encompassing >50,000 megafaunal specimens from 411 morphotypes (morphologically identifiable taxonomic units33) in 13 phyla, enabling investigation of ecological gradients (for example, with latitude, depth or food supply) in seabed communities across the region to develop key understanding of macroecological patterns in Earth’s largest biome.
Results and discussion
Macroecological patterns in the northeast Pacific abyss
Our analyses reveal the existence of a clear depth zonation demarking an evident boundary between biogeographic provinces across the northeast Pacific, with distinct shallow and deep-abyssal faunas and an intermediate transition zone. Regional benthic community structure at higher taxonomic levels exhibited a more pronounced relationship to depth (Fig. 2b) than other environmental gradients (for example, Supplementary Fig. 1) while food particle supply appears to be important at more intermediate, sub-regional, spatial scales (see below). Our results show distinct communities but incomplete taxonomic replacement in the presence and relative abundance of seabed taxa across the depth range studied, both within and between phyla. The most remarkable shifts were as follows: (1) soft corals (Alcyonacea) numerically dominated at shallow-abyssal depths (mean density ~2,500 individuals ha−1) but decreased in abundance by over an order of magnitude below 4,300 m (mean density ~100 ind ha−1), becoming virtually absent below 4,800 m (mean density <10 ind ha−1); (2) brittle stars (Ophiuroidea) were most abundant at shallow-abyssal depths (mean density ~2,000 ind ha−1), numerically dominated assemblages at intermediate depths (4,300–4,800 m; mean density ~1,500 ind ha−1) but decreased to low abundances below 4,800 m (mean density <40 ind ha−1); (3) anemones (Actiniaria) consistently increased in relative abundance with depth, first replacing soft corals as the dominant Cnidaria group below 4,300 m and then replacing brittle stars as the most dominant group in assemblages below 4,800 m (Fig. 2d); (4) sea cucumbers (Holothuroidea) largely increased in relative abundance with depth, replacing brittle stars as the most abundant Echinodermata group below 4,800 m (Fig. 2d); (5) shelled molluscs (for example, Bivalvia, Gastropoda and Polyplacophora), relatively abundant above 4,400 m (mean density ~300 ind ha−1), were absent below 4,400 m, where cirrate octopuses (Cephalopoda) became the only molluscan group in megabenthic assemblages. Changes in dominant taxa at the class or phylum level (Fig. 2c) clearly delimited two distinct abyssal assemblages (Fig. 2d): a shallow community (3,800–4,300 m) and a deep community (4,800–5,300 m). These were separated by an intermediate or transitional assemblage (4,300–4,800 m), which contained elements of both communities. No major taxonomic groups were restricted to the transitional zone (Fig. 2c).
The zonation pattern was also evident at lower taxonomic levels in the 411 invertebrate morphotypes recognized in the data. For example, from the 277 morphotypes found in the deep province, 60 (41 rare, that is <5 occurrences) were found exclusively in this deep zone, while 64 morphotypes (45 rare) out of a total of 283 were found exclusively in the shallow province. However, only 25 morphotypes (23 rare) from a total of 264 were found exclusively in the transitional zone. From the total of 175 (14 rare) morphotypes that were found in both shallow and deep communities, only two were within the ten most dominant taxa in both the shallow province (first- and third-most abundant) and in the deep province (nineth- and tenth-most abundant; Supplementary Table 2): the brittle star Ophiosphalma glabrum (Fig. 1h; well-studied in the eastern CCZ34) and Thenea sp. indet. (Fig. 1g), an undescribed species of nodule-dwelling sponge.
The nature of the taxonomic shifts between provinces and the depth range at which these occur suggests that the carbonate compensation depth (CCD) plays a key role in the depth zonation we encountered. The CCD lies close to the seafloor depth in the northeast Pacific35, deepening north to south from ∼4,400 to ∼4,800 m (ref. 36). Below the CCD, calcium carbonate (CaCO3) is undersaturated, which can energetically constrain the development and distribution of species highly dependent on carbonate structures such as in calcareous Foraminifera, a ubiquitous protozoan group showing clear depth zonation at intermediate scales in the abyss37,38. It is these groups, probably depending on CaCO3 structures, that exhibit the greatest variation in relative abundance or even presence with depth in our analysis and thereby contribute most to regional beta-diversity. For example, shelled molluscs (requiring biomineralization of CaCO3 for shell development39) were absent below 4,400 m and soft corals (with scleroblast cells requiring CaCO3 to produce sclerites40) and bryozoans (most species dependent on CaCO3 exoskeletons41) were replaced by ‘soft’ anemones (Actiniaria) below 4,300 m. The echinoderm transition from ophiuroids to holothurians below 4,800 m, yielding the dominant taxon of the deep province, could be related to either differing CaCO3 requirements or metabolic efficiency, documented across echinoderms42. All of these shifts were observed within the depth range (4,400–4,800 m) of the CCD in the northeast Pacific35,36. However, our understanding of biogeochemical and metabolic cycles in abyssal ecosystems is still too limited to fully comprehend the specific role of the CCD and calcium calcite saturation, as a biogeographical driver in abyssal metazoans.
Remarkably, despite the shifts in benthic community structure between provinces, diversity rates were relatively similar across biogeographic boundaries (Fig. 3c). Richness (Hill’s q = 0) ranged between 32 and 62 morphotypes per 200 specimens in the shallow-abyssal province and between 48 and 63 in the deep one but we found no evidence of a difference in richness across the three depth ranges (overlapping 95% confidence intervals; Fig. 3c). We also found little evidence of variations between deep, transition and shallow assemblages in Shannon diversity (Hill’s q = 1; Fig. 3e), a metric more sensitive to species evenness43. However, we found very strong evidence of an increase in both taxa richness (F1,159 = 115.4, P < 0.001) and Shannon diversity (F1,159 = 94.09, P < 0.001) with increasing depth (Fig. 3d,f). This reflects a higher evenness of taxa in the deep province. For instance, the ten most abundant morphotypes in the shallow and intermediate depth ranges encompassed ~60% of the total abundance sampled in these areas, whereas below 4,800 m, the ten most abundant taxa represented only ~40% of all fauna (Supplementary Table 2). An increase in evenness across ecological communities is often related to a decrease in surface productivity44 and, in abyssal areas, greater flux of sinking food is usually found in shallower locations nearer to surface primary production2,8. This is generally the case in the northeast Pacific region45 and seems a plausible driver for the higher evenness observed at regional scale in the deep province, particularly as our sampling approach was specimen-rarefied. The higher evenness of the deep community was also evident in species accumulation patterns (that is, steeper initial accumulation; Fig. 3g). However, with increasing sampling effort, estimated total richness of the deep province remained consistently higher than the shallower communities, suggesting that both biodiversity components (richness and evenness) were not only maintained but slightly enhanced, with increasing depth.
Our results distinguish two abyssal biogeographic provinces with remarkably distinct benthic community features within a region (northeast Pacific) where a previous top-down environmental assessment identified the potential for two abyssal provinces24. We found very strong evidence of different faunal densities between the deep and the shallow provinces (Fig. 3a), with megabenthic standing stocks largely decreasing with depth (F1,82 = 51.61, P < 0.001, Fig. 3b). As was expected from local megafaunal studies25,26,27, density (range 0.06–1.46 ind m−2) was up to an order of magnitude larger in the shallow than in the deep province (Fig. 3a). Nonetheless, regional patterns of density within provinces appeared more related to variations in food supply, that is particulate organic carbon (POC) flux, to the seabed (Supplementary Fig. 3a) than depth (Fig. 3b), especially in deposit-feeding fauna (Supplementary Fig. 3b). In particular, the high variability in faunal density across the shallow province (Fig. 3a) probably reflects a latitudinal gradient in POC flux at intermediate scales, with lowest faunal densities in the northern sector (for example, range APEI-6 site: 0.3–0.4 ind m−2), increasing towards southeast locations (for example, ranges NORI-D, 0.9–1.2 ind m−2; APEI-12, 1.3–1.5 ind m−2) in closer proximity to the high primary productivity related to equatorial upwelling8, with known capacity to regulate abyssal benthic processes46. Similarly, the benthic communities surveyed in northeasternmost locations (for example, APEI-6; ref. 47)—for instance, with the largest densities we observed in scavenging megafauna (mostly decapods and isopods; Supplementary Fig. 3d) and an evidently lower total taxonomic richness (Fig. 3h)—were markedly distinct from those in the southeast locations of the shallow province (such as BGR-E, UK-1, APEI-12 and NORI-D; Fig. 2a and Supplementary Fig. 2). These differences could be driven by altered oceanographic conditions over the northeast CCZ, potentially related to the influence of the North Pacific Equatorial Current48 and the closer proximity to the continental shelf, which may transport pelagic food falls49 more commonly into this sector. Hence, while regional patterns across provinces in abundance, diversity and community composition were clearly zoned by depth over scales of thousands of kilometres, variations at intermediate scales within provinces, over hundreds of kilometres, might be more related to surface productivity gradients32 (also at the deep province; for example, Supplementary Fig. 2c) and other environmental (hydrographical) variation between sectors. Overall, this reflects a much higher ecological heterogeneity, at multiple scales, than was previously expected for benthic assemblages across the northeast Pacific abyssal seabed. This overlooked heterogeneity, stemming from geochemical and climatic forcing, has crucial implications for future ecological and macroecological research in abyssal communities and for the success of regional-scale conservation strategies implemented to protect biodiversity in the CCZ50 and probably in other abyssal areas targeted by deep-sea mining worldwide31.
The results of this study represent the largest multiphylum assessment of beta-diversity patterns to date conducted across such a vast abyssal seabed extension (>5,000 km span), provide key insights to our view of abyssal ecosystems and biodiversity and challenge some current paradigms in deep-sea macroecology. The maintenance of high taxonomic richness with increasing depth by phylum-level replacements (in presence or dominance) was surprising. Diversity, especially that of the largest-sized fauna, is commonly expected to decrease with depth and nutrient flux in the deep sea5,19,21,22, a reduction that is thought to reflect the wider geographic ranges of deep-sea species23. But our results suggest that more complex processes may control diversity in the abyss, helping maintain community richness with depth. These interpretations could be affected by the historical unbalance in the sampling conducted in abyssal ecosystems (for example, ~90% of the species living at the CCZ remain undescribed51) compared to shallower deep-sea environments52. In turn, our results add to the existing evidence53 against a source–sink dependency from bathyal to abyssal communities in the Pacific54 and stress that, given the wealth of between-phylum species replacement we encountered with increasing depth, assessments including multiple phyla might be best suited to investigate biodiversity and macroecological patterns in abyssal regions.
The presence of such a clear and unexpected boundary zone demarking the limits of provinces has not been previously documented in abyssal metazoans, opening questions for further macroecological research in the deep ocean. We suggest that the CCD may control the location of the environmental boundaries between shallow and deep-abyssal provinces for megafauna, as appears to occur in protozoans (for example, Foraminifera37,38), although more physiological research is needed to better understand how CaCO3 saturation affects different phyla. Greater understanding of CaCO3 saturation as a biodiversity driver is especially important as the CCD is expected to shoal with increased CO2 levels in the ocean associated with climate change55. In the light of our results, shoaling of the CCD could have much larger implications for abyssal biodiversity than previously expected, especially in regions like the northeast Pacific where the seafloor depths transcend the current CCD11. A CCD shoaling of tens of metres in abyssal ecosystems, given their vast area, could lead to shifts in the environmental conditions (for example, from CaCO3 saturation to undersaturation) across thousands of km2 of seabed worldwide. This could trigger large shifts in community structure and species distributions or even extinctions of highly specialized abyssal taxa, adding to the cumulative impacts of deep-sea mining and other emerging anthropogenic disturbances56 that might concur on Earth’s largest biome in the coming years.
The CCZ is an extensive abyssal plain and hill ecosystem, interspersed by seamount areas26,57, covering approximately 6 million km2 of seafloor stretching from 5° to 20° N and 115° to 160° W in the northeast Pacific basin (Fig. 1). Important abiotic factors are broadly similar across the northeast Pacific abyssal seafloor, such as low currents, constant bottom-water salinity and temperature45. In contrast, there is a gradual increase in water depth from east to west (from 3,700 to 5,500 m; Fig. 1a) resulting from the contraction of older oceanic crust to the west58 and a gradient in POC flux to the seabed, generally diminished with depth but largely enhanced towards more productive southeastern waters59 (Supplementary Methods). In turn, the CCD (the depth below which the rate of supply of calcium carbonate from the surface is equal to the rate of dissolution60) lies close to the seafloor in the northeast Pacific35, deepening north to south between 4,400 and 4,800 m (ref. 36). Surface sediments in the northeast Pacific are relatively well-oxygenated and generally dominated by either siliciclastic clay and radiolarian ooze in the northern and central parts of the CCZ, or by biogenic calcareous oozes and fine-grained sediments in southern sites45. These are supplied by extremely low sedimentation rates, regionally varying between 0.2 and 1.15 cm per 1,000 years61. These conditions are thought to promote the (extremely slow) formation of polymetallic nodules35, yielding growth rates of ∼1–12 mm per million years in the abyssal Pacific31. Nodules can vary in size, shape and abundance61 and are extensively distributed across the northeast Pacific31 but are not exclusively found there. Vast abyssal nodule field areas are present in the southeast Pacific (for example, Peru Basin), the southwest Pacific (for example, Penrhyn Basin) and in the Central Indian Ocean Basin31. Nodule fields form an unusual mosaic habitat, where the hard substratum provided by nodules acts as keystone structure increasing local seabed complexity62,63, promoting the occurrence of some of some of the most biodiverse benthic assemblages surveyed in the abyss25,27. But the biodiversity of the CCZ is poorly known, for example 90% of the species identified in the region are thought to be new to science51 and substantial levels of dark biodiversity have been predicted34. In addition, a general lack of standardization of intercalibrated taxonomic standards for the identification of benthic fauna50 has historically hampered comparability between biological studies64 and thereby the assessment of regional, macroecological patterns across the northeast Pacific basin.
Occurrences of benthic invertebrate metazoans were collated from a range of seabed image surveys collected using comparable sampling methodology across the northeast Pacific (Table 1). To ensure homogenous animal detectability and a minimum sample size per study location, we selected only available imagery obtained using deep-sea robots (remotely operated vehicles, ROVs; and autonomous underwater vehicles, AUVs; and towed-camera platforms) that fulfilled the following criteria: (1) collected between 1 and 4 m above the seabed; (2) well-lit and high resolution (that is, minimum resolution at 2 m above-seabed, 1,280 × 720 pixels); (3) total seabed survey area imaged per location >2,000 m2; (4) no overlapping frames included in image-based studies; (5) scalable stills collected vertically facing the seabed (only in data used for density analysis, see below). Imagery data were collected within 14 sites managed by the International Seabed Authority across the CCZ (Table 1 and Fig. 1a); eight mining exploration-licenced sites and six APEIs.
Megafauna specimens >10 mm were detected and counted in imagery reanalysed from the selected surveys (Table 1). Using BIIGLE 2.0 software65, animals were identified to the lowest taxonomic level possible (morphotype, typically genus or family level in undescribed species; for example, Fig. 1b–p) in accordance with an abyssal-Pacific standardized megafauna catalogue66 developed during a range of scientific workshops, in collaboration with taxonomic experts (see Acknowledgements), and by reference to existing literature67,68,69—though developed before the major higher taxonomic revision recently undergone by Octocorallia70. The catalogue follows the ref. 33 open nomenclature to report the taxonomic resolution reached in each morphotype but all taxa identified from the catalogue (411 morphotypes in this study) were deemed as sufficiently different morphologically by taxonomic experts to be confidently considered separate species. Specimens with uncertain classification at the morphotype level (~30% of all records) but identifiable to a certain higher taxonomical level (like phylum or class) were retained only for density analyses. Taxa living in a closed shell or tube (for example, most polychaetes) were excluded from analysis as it is not possible to determine whether these are alive in images. Several spatially blind reviews were conducted to the whole dataset to ensure a robust taxonomic alignment between sites, consisting in side-by side visualization of all specimens classified under the same catalogue label using the ‘Label Review Grid Overview’ tool in BIIGLE65; a process that was repeated many times by the same group of expert seabed image scientists, to minimize potential bias28. In addition, the likely feeding behaviour (suspension, deposit and scavenger or predator feeding) of each morphotype was inferred from observational knowledge and by reference to similar organisms described in the literature. With a total of 53,512 megafaunal specimens in 13 phyla, our biodiversity dataset was compiled across a wide geographical span (>5,000 km) at the abyss and is of comparable magnitude to the largest datasets that have been used to underpin our theoretical understanding in the deep-sea (for example, refs. 71,72).
We compiled invertebrate megafauna occurrences into two different subsets to investigate variations across space in different parameters. These were (1) standing stocks subset (SSdat): containing faunal count data at all taxonomical levels obtained only from scalable still images and thereby associated with a precise measure of the seabed surface area (47,087 specimens); and (2) biodiversity subset (BDdat): containing faunal count data obtained from both image and video surveys but including only specimens identified up to morphotype level (36,432 specimens). Using ArcMap v.6.10 software, we interpolated each specimen across a 10 × 10 km2 regional grid to assign and constrain occurrence data to different geographical locations (units of 100 km2) within sites. This process divided the SSdat into 23 locations and the BDdat into 28 locations (Fig. 1 and Table 1). Faunal occurrences (or image units, in SSdat) within each geographical location were then randomly resampled without replacement to generate equally sized (in terms of number of specimens encountered) replicate sample units, characterizing the community at each location while minimizing potential spatial autocorrelation biases (for example, ref. 73). In locations where surveys covered a large depth spectrum, data were further constrained to 200 m depth bins during replicate sample generation. Following recommendations of minimum sample size for accurate and precise estimation of local megabenthic community features47 and to minimize the potential effect of regional gradients in faunal abundance on diversity measures (for example, ref. 25), we generated SSdat samples containing 450–500 specimens (taxonomic resolution including specimens identified to morphotype and to higher taxonomical levels) and BDdat samples containing an exact number of 200 specimens (taxonomic resolution including only specimens identified to morphotype level). This process yielded a total of 84 samples standardized for density analyses (standing stocks) and 161 samples standardized for biodiversity analyses (richness and beta-diversity).
Depth and position (latitude and longitude) of specimens (or images) were averaged within each sample to investigate variations in communities across space and environmental gradients. Sample positions were interpolated with predictions of yearly POC flux (gCorg m−2 yr−1) to the seabed derived by applying vertical flux attenuation59 to satellite-derived export flux data (Supplementary Methods). Different ecological metrics were calculated for each sample: (1) standing stocks were assessed as the total numerical abundance of specimens per total unit (m2) of seabed surveyed in each SSdat sample; that is, for the whole assemblage, specific populations or faunal groups of interest (taxonomic or functional) and (2) to examine the range of diversity characteristics, Hill’s diversity numbers of order 0 and 1 (ref. 43) were calculated in each BDdat sample as morphotype richness (S), the exponential form of the Shannon diversity index (expH’), for the whole assemblage.
Preliminary exploration of the interaction between different environmental factors (depth, latitude, longitude and POC flux) across study locations revealed substantial correlations between depth and longitude and also between POC flux and latitude (see Supplementary Fig. 1). Consequently, we choose to investigate only potential interactions between ecological metrics (y, that is standing stock and diversity indices) and two main environmental factors (x, that is water depth and POC flux, as these were not significantly correlated; see Supplementary Fig. 1). We used the Akaike information criterion 74 to select the best fitting linear regression across a range of data transformation types (where ‘exp’ consistently outperformed ‘linear’ and ‘log’).
We followed a simple, data-driven, bottom-up approach to investigate the potential existence of biogeographic boundaries across the depth spectrum of the north Pacific abyss. First, using the whole dataset, we generated ridgeline plots depicting within-group geometric distributions of the total abundance sampled in the most dominant high taxonomic groups (>50 occurrences, minimum taxonomic rank: order). On the basis of the results of this analysis, we assessed the potential variability within and between the megabenthic communities in tentative biogeographical provinces, by grouping samples in three different depth bins: shallow <4,300 m; transition 4,300–4,800 m; deep >4,800 m). Mean values and 95% confidence intervals were calculated for different ecological parameters (faunal density based on SSdat samples and both diversity indices based on BDdat samples) within each tentative province. To contextualize patterns in morphotype richness and the representability of the sampling conducted within each depth range, morphotype accumulation curves were calculated following ref. 75, by random resampling of BDdat samples in each depth category 100 times without replacement forming increasingly larger sampling units, using EstimateS v.9.1 software76. In addition, using the same methodology, curves were also calculated for study sites encompassing a minimum of three BDdat samples (sampling effort >600 specimens identified to morphotype level).
Variations in community composition across locations, sites and potential biogeographical provinces were further explored using BDdat samples. Dissimilarity in faunal composition between all pairs of samples was calculated using the Bray–Curtis dissimilarity measure on square-root transformed (normalized77) taxon abundances. Non-metric multidimensional scaling ordinations were conducted to visualize the rate of dissimilarity (distance) between all pairs of samples; isotropic contour lines depicting broadly approximated depth ranges were fitted using generalized additive models78 in the resulting multidimensional scaling plots to aid visualization of beta-diversity patterns across environmental gradients. All data processing and analysis were implemented in R v.4.2.1 (ref. 79), using standard analysis methods and functions from the packages: ‘vegan’80, ‘AICcmodavg’81, ‘ggridges’82 and ‘ggplots2’83. We report statistical evidence of variations in ecological parameters across environmental gradients (linear regressions, that is R2 and P values) or between biogeographical provinces (95% confidence intervals; significant at P < 0.05) using the simplified language of evidence84.
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data generated for this study, invertebrate abyssal megafauna taxa occurrences across the northeast Pacific seafloor, are available at https://doi.org/10.5281/zenodo.7982461 (ref. 85). The Abyssal Pacific Seafloor Megafauna Atlas (APSMA image-based taxonomical catalogue) developed to conduct this study is available at https://doi.org/10.5281/zenodo.7765164 (ref. 66). Data handling and analyses were implemented using standard methods, software tools and code functions detailed in the Methods.
Harris, P. T., Macmillan-Lawler, M., Rupp, J. & Baker, E. K. Geomorphology of the oceans. Mar. Geol. 352, 4–24 (2014).
Smith, C. R., De Leo, F. C., Bernardino, A. F., Sweetman, A. K. & Arbizu, P. M. Abyssal food limitation, ecosystem structure and climate change. Trends Ecol. Evol. 23, 518–528 (2008).
Yasuhara, M. & Danovaro, R. Temperature impacts on deep-sea biodiversity. Biol. Rev. Camb. Philos. Soc. 91, 275–287 (2016).
Aleynik, D., Inall, M. E., Dale, A. & Vink, A. Impact of remotely generated eddies on plume dispersion at abyssal mining sites in the Pacific. Sci. Rep. 7, 16959 (2017).
Rex, M. A. et al. Global bathymetric patterns of standing stock and body size in the deep-sea benthos. Mar. Ecol. Prog. Ser. 317, 1–8 (2006).
Cordier, T. et al. Patterns of eukaryotic diversity from the surface to the deep-ocean sediment. Sci. Adv. 8, eabj9309 (2022).
Snelgrove, P. & Smith, C. A riot of species in an environmental calm: the paradox of the species-rich deep-sea floor. Oceanogr. Mar. Biol. Annu. Rev. 40, 311–342 (2002).
Jahnke, R. A. The global ocean flux of particulate organic carbon: areal distribution and magnitude. Glob. Biogeochem. Cycles 10, 71–88 (1996).
Ramirez-Llodra, E. et al. Deep, diverse and definitely different: unique attributes of the world’s largest ecosystem. Biogeosciences 7, 2851–2899 (2010).
Thurber, A. R. et al. Ecosystem function and services provided by the deep sea. Biogeosciences 11, 3941–3963 (2014).
Sweetman, A. K. et al. Major impacts of climate change on deep-sea benthic ecosystems. Elementa https://doi.org/10.1525/elementa.203 (2017).
Levin, L. A. et al. Climate change considerations are fundamental to management of deep-sea resource extraction. Glob. Change Biol. 26, 4664–4678 (2020).
Simon-Lledó, E. et al. Biological effects 26 years after simulated deep-sea mining. Sci. Rep. 9, 8040 (2019).
Jones, D. O. et al. Biological responses to disturbance from simulated deep-sea polymetallic nodule mining. PLoS ONE 12, e0171750 (2017).
Woolley, S. N. C. et al. Deep-sea diversity patterns are shaped by energy availability. Nature 533, 393–396 (2016).
Rex, M. A. et al. Global-scale latitudinal patterns of species diversity in the deep-sea benthos. Nature 365, 636–639 (1993).
McClain, C. R., Rex, M. A. & Etter, R. J. in Marine Macroecology (eds Witman, J. D. & Roy, K.) 65–100 (Univ. Chicago Press, 2009).
Jöst, A. B. et al. North Atlantic Gateway: test bed of deep‐sea macroecological patterns. J. Biogeogr. 46, 2056–2066 (2019).
Rex, M. A. Deep-sea species diversity: decreased gastropod diversity at abyssal depths. Science 181, 1051–1053 (1973).
Culver, S. J. & Buzas, M. A. Global latitudinal species diversity gradient in deep-sea benthic foraminifera. Deep Sea Res. I 47, 259–275 (2000).
McClain, C. R., Allen, A. P., Tittensor, D. P. & Rex, M. A. Energetics of life on the deep seafloor. Proc. Natl Acad. Sci. USA 109, 15366–15371 (2012).
Rex, M. A., Crame, J. A., Stuart, C. T. & Clarke, A. Large-scale biogeographic patterns in marine mollusks: a confluence of history and productivity? Ecology 86, 2288–2297 (2005).
Costello, M. J. & Chaudhary, C. Marine biodiversity, biogeography, deep-sea gradients, and conservation. Curr. Biol. 27, 511–527 (2017).
Watling, L., Guinotte, J., Clark, M. R. & Smith, C. R. A proposed biogeography of the deep ocean floor. Prog. Oceanogr. 111, 91–112 (2013).
Simon-Lledó, E. et al. Multi-scale variations in invertebrate and fish megafauna in the mid-eastern Clarion Clipperton Zone. Prog. Oceanogr. 187, 102405 (2020).
Durden, J. M. et al. Megafaunal ecology of the Western Clarion Clipperton Zone. Front. Marine Sci. https://doi.org/10.3389/fmars.2021.671062 (2021).
Amon, D. J. et al. Insights into the abundance and diversity of abyssal megafauna in a polymetallic-nodule region in the eastern Clarion–Clipperton Zone. Sci. Rep. 6, 30492 (2016).
Durden, J. M. et al. in Oceanography and Marine Biology: An Annual Review (eds Hughes, R. N. et al.) 1–72 (CRC Press, 2016).
Danovaro, R. et al. Ecological variables for developing a global deep-ocean monitoring and conservation strategy. Nat. Ecol. Evol. 4, 181–192 (2020).
Wedding, L. M. et al. OCEANS. Managing mining of the deep seabed. Science 349, 144–145 (2015).
Hein, J. R., Koschinsky, A. & Kuhn, T. Deep-ocean polymetallic nodules as a resource for critical materials. Nat. Rev. Earth Environ. 1, 158–169 (2020).
Vanreusel, A., Hilario, A., Ribeiro, P. A., Menot, L. & Arbizu, P. M. Threatened by mining, polymetallic nodules are required to preserve abyssal epifauna. Sci. Rep. 6, 26808 (2016).
Horton, T. et al. Recommendations for the standardisation of open taxonomic nomenclature for image-based identifications. Front. Mar. Sci. https://doi.org/10.3389/fmars.2021.620702 (2021).
Christodoulou, M., O’Hara, T. D., Hugall, A. F. & Arbizu, P. M. Dark ophiuroid biodiversity in a prospective abyssal mine field. Curr. Biol. 29, 3909–3912 (2019).
Mewes, K. et al. Impact of depositional and biogeochemical processes on small scale variations in nodule abundance in the Clarion–Clipperton Fracture Zone. Deep Sea Res. I 91, 125–141 (2014).
Berger, W. H., Adelseck, C. G. Jr. & Mayer, L. A. Distribution of carbonate in surface sediments of the Pacific Ocean. J. Geophys. Res. 81, 2617–2627 (1976).
Gooday, A. J. et al. The biodiversity and distribution of abyssal benthic Foraminifera and their possible ecological roles: a synthesis across the Clarion–Clipperton zone. Front. Mar. Sci. https://doi.org/10.3389/fmars.2021.634726 (2021).
Gooday, A. J. Advances in Marine Biology (Academic Press, 2003).
Feng, Q. L. et al. Polymorph and morphology of calcium carbonate crystals induced by proteins extracted from mollusk shell. J. Cryst. Growth 216, 459–465 (2000).
Watabe, N. & Kingsley, R. J. in Hard Tissue Mineralization and Demineralization (eds Suga, S. & Watabe, N.) 127–147 (Springer Japan, 1992).
Smith, A. M. Growth and calcification of marine bryozoans in a changing ocean. Biol. Bull. 226, 203–210 (2014).
Lebrato, M. et al. Global contribution of echinoderms to the marine carbon cycle: CaCO3 budget and benthic compartments. Ecol. Monogr. 80, 441–467 (2010).
Jost, L. Entropy and diversity. Oikos 113, 363–375 (2006).
Levin, L. A. & Gage, J. D. Relationships between oxygen, organic matter and the diversity of bathyal macrofauna. Deep Sea Res. II 45, 129–163 (1998).
Washburn, T. W., Jones, D. O. B., Wei, C.-L. & Smith, C. R. Environmental heterogeneity throughout the Clarion–Clipperton zone and the potential representativity of the APEI network. Front. Mar. Sci. https://doi.org/10.3389/fmars.2021.661685 (2021).
Smith, C. R. et al. Latitudinal variations in benthic processes in the abyssal equatorial Pacific: control by biogenic particle flux. Deep Sea Res. II 44, 2295–2317 (1997).
Simon-Lledó, E. et al. Megafaunal variation in the abyssal landscape of the Clarion Clipperton zone. Prog. Oceanogr. 170, 119–133 (2019).
Kawabe, M. & Fujio, S. Pacific ocean circulation based on observation. J. Oceanogr. 66, 389–403 (2010).
Simon-Lledo, E. et al. Mass falls of crustacean carcasses link surface waters and the deep seafloor. Ecology 104, e3898 (2023).
Lodge, M. et al. Seabed mining: International Seabed Authority environmental management plan for the Clarion–Clipperton Zone. A partnership approach. Mar. Policy 49, 66–72 (2014).
Rabone, M. et al. How many metazoan species live in the world’s largest mineral exploration region?. Curr. Biol. 33, 2383–2396 (2023).
Costello, M. J. et al. A census of marine biodiversity dnowledge, resources, and future challenges. PLoS ONE 5, e12110 (2010).
Hardy, S. M., Smith, C. R. & Thurnherr, A. M. Can the source–sink hypothesis explain macrofaunal abundance patterns in the abyss? A modelling test. Proc. R. Soc. B 282, 20150193 (2015).
Rex, M. A. et al. A source–sink hypothesis for abyssal biodiversity. Am. Nat. 165, 163–178 (2005).
Sulpis, O. et al. Current CaCO3 dissolution at the seafloor caused by anthropogenic CO2. Proc. Natl Acad. Sci. USA 115, 11700–11705 (2018).
Oschlies, A. A committed fourfold increase in ocean oxygen loss. Nat. Commun. 12, 2307 (2021).
Cuvelier, D. et al. Are seamounts refuge areas for fauna from polymetallic nodule fields? Biogeosciences 17, 2657–2680 (2020).
Olive, J. A. et al. Sensitivity of seafloor bathymetry to climate-driven fluctuations in mid-ocean ridge magma supply. Science 350, 310–313 (2015).
Henson, S. A., Sanders, R. & Madsen, E. Global patterns in efficiency of particulate organic carbon export and transfer to the deep ocean. Glob. Biogeochem. Cycles https://doi.org/10.1029/2011gb004099 (2012).
Morse, J. W. & Mackenzie, F. T. Geochemistry of Sedimentary Carbonates (Elsevier, 1990).
Volz, J. B. et al. Natural spatial variability of depositional conditions, biogeochemical processes and element fluxes in sediments of the eastern Clarion–Clipperton Zone, Pacific Ocean. Deep Sea Res. I 140, 159–172 (2018).
Simon-Lledó, E. et al. Ecology of a polymetallic nodule occurrence gradient: Implications for deep-sea mining. Limnol. Oceanogr. 64, 1883–1894 (2019).
Mejía-Saenz, A., Simon-Lledó, E., Partridge, L. S., Xavier, J. R. & Jones, D. O. B. Rock outcrops enhance abyssal benthic biodiversity. Deep Sea Res. I 195, 103999 (2023).
Amon, D. J. et al. Assessment of scientific gaps related to the effective environmental management of deep-seabed mining. Mar. Policy 138, 105006 (2022).
Langenkämper, D., Zurowietz, M., Schoening, T. & Nattkemper, T. W. BIIGLE 2.0—browsing and annotating large marine image collections. Front. Mar. Sci. 4, 10 (2017).
Simon-Lledó, E. et al. Abyssal Pacific seafloor megafauna atlas (version 1). Zenodo https://doi.org/10.5281/zenodo.7765164 (2023).
Bribiesca-Contreras, G. et al. Benthic megafauna of the western Clarion–Clipperton Zone, Pacific Ocean. ZooKeys https://doi.org/10.3897/zookeys.1113.82172 (2022).
Amon, D. et al. Megafauna of the UKSRL exploration contract area and eastern Clarion–Clipperton Zone in the Pacific Ocean: Annelida, Arthropoda, Bryozoa, Chordata, Ctenophora, Mollusca. Biodivers. Data J. 5, e14598 (2017).
Molodtsova, T. N. & Opresko, D. M. Black corals (Anthozoa: Antipatharia) of the Clarion–Clipperton Fracture Zone. Mar. Biodivers. 47, 349–365 (2017).
McFadden, C. S., Van Ofwegen, L. P. & Quattrini, A. M. Revisionary systematics of Octocorallia (Cnidaria: Anthozoa) guided by phylogenomics. Bull. Soc. Syst. Biol. https://doi.org/10.18061/bssb.v1i3.8735 (2022).
Sanders, H. L. & Hessler, R. R. Ecology of the deep-sea benthos. Science 163, 1419–1424 (1969).
Grassle, J. F. & Maciolek, N. J. Deep-sea species richness: regional and local diversity estimates from quantitative bottom samples. Am. Nat. 139, 313–341 (1992).
Legendre, P. Spatial autocorrelation: trouble or new paradigm? Ecology 74, 1659–1673 (1993).
Shao, J. Linear model selection by cross-validation. J. Am. Stat. Assoc. 88, 486–494 (1993).
Colwell, R. K. et al. Models and estimators linking individual-based and sample-based rarefaction, extrapolation and comparison of assemblages. J. Plant Ecol. 5, 3–21 (2012).
Colwell, R. EstimateS: Statistical Estimation of Species Richness and Shared Species from Samples (Version 9) (2013); https://www.robertkcolwell.org/pages/1407-estimates
Legendre, P. & Legendre, L. Numerical Ecology (Elsevier, 1998).
Marra, G. & Wood, S. N. Practical variable selection for generalized additive models. Comput. Stat. Data Anal. 55, 2372–2387 (2011).
R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2017).
Oksanen, J. et al. vegan: Community ecology package. R package version 2.6-2 https://CRAN.R-project.org/package=vegan (2018).
Mazerolle, M. AICcmodavg: Model selection and multimodel inference based on AIC. R package version 2.3-1 https://cran.r-project.org/package=AICcmodavg (2020).
Wilke, C. O. ggridges: Ridgeline plots in ‘ggplot2’. R package version 0.5.3 https://wilkelab.org/ggridges/ (2018).
Wickham, H. ggplot2: Elegant graphics for data analysis. R package version 3.3.6 https://ggplot2.tidyverse.org (2016).
Muff, S., Nilsen, E. B., O’Hara, R. B. & Nater, C. R. Rewriting results sections in the language of evidence. Trends Ecol. Evol. 37, 203–210 (2022).
Simon-Lledó, E. et al. Abyssal NE Pacific Seafloor Megafauna (v.1) [Data set]. Zenodo https://doi.org/10.5281/zenodo.7982462 (2023).
Uhlenkott, K., Simon-Lledó, E., Vink, A. & Martinez Arbizu, P. Investigating the benthic megafauna in the eastern Clarion Clipperton Fracture Zone (north-east Pacific) based on distribution models predicted with random forest. Sci. Rep. 12, 8229 (2022).
Jones, D. O. B. et al. Environment, ecology, and potential effectiveness of an area protected from deep-sea mining (Clarion Clipperton Zone, abyssal Pacific). Prog. Oceanogr. 197, 102653 (2021).
We thank S. Henson (National Oceanography Centre, United Kingdom) for her help in the processing and generation of POC data for the abyssal north Pacific seabed. We also thank all deep-sea researchers who helped during the generation of the standardised megafauna catalogue and other aspects of the study: D. Billett, D. Lindsay, S. Brix, S. De Grave, T. Molodtsova, H. Wiklund, A. Serpell-Stevens, D. Kersken, J. Xavier, L. Watling, A. Leitner, J. Drazen, C. Young, A. Gebruk, D. Pawson, A. Grischenko, M. Christodoulou, C. Mah, S. Laming, A. Purser, B. Bett, A. Gates and V. Huvenne. This work was part of the UK Natural Environment Research Council funded Seabed Mining and Resilience to Experimental impact (SMARTEX) project (grant reference NE/T003537/1). D.O.B.J., E.S.L., N.B., A.M.S., G.B.C. and A.G.G. also received support from TMC (The Metals Company) through its subsidiary Nauru Ocean Resources (NORI); this is contribution TMC/NORI/D/007. S.P.R. was supported by funds from FCT/MCTES in the scope of the CEEC contract (CEECIND/00758/2017) and funds attributed to CESAM (UIDP/50017/2020, UIDB/50017/2020 and LA/P/0094/2020). Samples from the BGR license area were made available by C. Rühlemann and A. Vink from the Federal Institute for Geosciences and Raw Materials (BGR) in Hannover. P.M.A. and K.U. acknowledge EU JPIO-Oceans project Ecological Aspects of Deep-Sea Mining and MinigImpact-2 (German Ministry for Science and Education BMBF contract 03F0707E and 03F0812E.). The KODOS and APEI-9 datasets had been collected through environmental baseline studies for the polymetallic manganese nodules exploration contract of Republic of Korea (Ministry of Ocean and Fisheries’ R&D no. 20160099). D.C. acknowledges funding from Mining2/0002/2017, Mining2/0005/2017, granted by FCT/MCTES and DGPM and OP AZORES 2020 (01-0145-FEDER-000140 ‘MarAZ Researchers: Consolidate a body of researchers in Marine Sciences in the Azores’ and funds attributed to Okeanos-UAç (UIDB/05634/2020, UIDP/05634/2020 and M1.1.A/REEQ.CIENTIFICOU I&D/2021/010). The funders had no role in the conceptualization, design, data collection, analysis, decision to publish or preparation of the manuscript.
The authors declare no competing interests.
Peer review information
Nature Ecology & Evolution thanks Moriaki Yasuhara and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Simon-Lledó, E., Amon, D.J., Bribiesca‐Contreras, G. et al. Carbonate compensation depth drives abyssal biogeography in the northeast Pacific. Nat Ecol Evol 7, 1388–1397 (2023). https://doi.org/10.1038/s41559-023-02122-9