Deep midwater Annelida from the NORI-D exploration area of the Clarion-Clipperton Zone, Pacific Ocean Open Access

Baseline species occurrence data, supported by reliable taxonomy, are essential for biodiversity research and environmental monitoring. In the eastern equatorial Pacific Ocean, the abyssal seafloor of the Clarion-Clipperton Zone (CCZ) holds polymetallic nodules of commercial interest. Exploration of mineral resources and recent testing of a pilot collection system have been conducted at the Nauru Ocean Resources Inc. exploration contract area D (NORI-D) at approximately 4,200 m depth in the southeastern CCZ (International Seabed Authority, 2024). These activities necessitated the submission of an environmental impact statement to the International Seabed Authority (Nauru Ocean Resources Inc., 2022a), to assess the effects on marine life (International Seabed Authority Council, 2013).

In addition to directly perturbing the seafloor, deep-sea mining is expected to impact the midwater by generating a discharge plume as unwanted sediment is expelled from a hydraulic riser system (Muñoz-Royo et al., 2021; Ouillon et al., 2022). Based on field measurements and modeling, the midwater discharge plume is expected to act in two phases: an initial dynamic phase upon release, operating on the scale of tens to hundreds of meters; and an ambient phase of advection, settling, and diffusion, operating over larger spatial and temporal scales (Muñoz-Royo et al., 2021). The impact of the benthic collector plume into the overlying water column, in contrast, is expected to be minimal. During pre-prototype nodule collector trials in the eastern CCZ, 92–98% of the suspended sediment mass remained within 2 m of the seafloor (Muñoz-Royo et al., 2022), and the highest suspended particle concentrations occurred at only 1 m altitude, reaching background concentrations by 50 m altitude (Gazis et al., 2025). The release of the discharge plume near the seabed, rather than in midwater, has therefore been suggested as a means to reduce the total environmental footprint of mining operations (Gazis et al., 2025).

The choice of discharge depth is expected to have major implications for pelagic communities; the higher in the water column, the greater the potential for cascading impacts on suspension-feeding and light-sensitive organisms (Christiansen et al., 2020; Drazen et al., 2020). To circumvent the biologically productive epipelagic zone (0–200 m depth) and mesopelagic zone (200–1,000 m), the depth of the NORI-D discharge outlet has been set at 1,200 m (Nauru Ocean Resources Inc., 2022a). Increased suspended sediment concentrations are an established stressor of shallow-water aquatic animals, and deep-sea animals are expected to respond with similar or greater sensitivity (van der Grient and Drazen, 2022). In a case study on the deep-sea jellyfish Periphylla periphylla (Péron & Lesueur, 1810), exposure to simulated sediment plumes elicited production of excess mucus, which represents an energetically expensive stress response, as well as decreased health scores and changes in gene expression suggesting additional physiological impairment (Stenvers et al., 2023). Starvation is another anticipated risk to the midwater food web, given the depleted amino acid content of sediment plume particles compared to the naturally occurring diet of midwater animals (MH Dowd and colleagues, personal communication, 03/03/2025). Sensitivity metrics for deep-sea taxa will be important to inform environmentally acceptable threshold levels (Muñoz-Royo et al., 2021; Gazis et al., 2025).

Numerous studies have inventoried the benthic fauna of the CCZ (Blake, 2019; Bribiesca-Contreras et al., 2022; Rabone et al., 2023; Wiklund et al., 2023), but the pelagic taxa have only scarcely been investigated. In the CCZ and globally, biodiversity data are extremely limited for the mesopelagic, bathypelagic (>1,000 m), and abyssopelagic zones (>3,000–4,000 m; Webb et al., 2010; Sutton et al., 2017; Christiansen et al., 2020; Drazen et al., 2020; Kennedy and Rotjan, 2023; Perelman et al., 2025). At NORI-D and its vicinity in the eastern tropical Pacific, the midwater is notably heterogeneous, as most of the mesopelagic zone intersects an oxygen minimum zone spanning approximately 100–900 m depth (Sutton et al., 2017; Nauru Ocean Resources Inc., 2022a; Vasiliu et al., 2022; Perelman et al., 2023; Perelman et al., 2025). The dissolved oxygen concentrations of <0.5 ml L⁻¹ in the core of the oxygen minimum zone are considered severely hypoxic, flanked by regions of intermediate hypoxia at 0.5–1.4 ml L⁻¹ (Hofmann et al., 2011; Levin, 2018). Oxygen levels are expected to constrain the vertical and biogeographic distributions of midwater taxa (Perelman et al., 2021; 2023).

Annelids are abundant and ecologically important in deep midwater ecosystems (Haddock and Choy, 2024) and are considered energy-rich prey items for deep-sea fishes and large zooplankton (Thuesen and Childress, 1993). In characterizing midwater communities, many annelids and other soft-bodied zooplankton are not adequately sampled by conventional trawls or acoustic methods. For example, in recent surveys of micronekton and macroplankton at NORI-D using a Multiple Opening and Closing Net and Environmental Sensing System, annelids comprised a substantial fraction of the numerical abundance and biomass below 3,000 m depth but were difficult to sort and identify due to damage (JC Drazen and colleagues, personal communication, 03/03/2025). Remotely operated vehicles (ROVs) are the ideal method to survey these fragile pelagic taxa (Hetherington et al., 2022; International Seabed Authority Legal and Technical Commission, 2023; Perelman et al., 2025). In this work, we identify and describe annelids documented from NORI-D during the first midwater ROV surveys of the CCZ, emphasizing the linkage of voucher specimens, images, and reference DNA sequences.

ROV-based midwater biodiversity surveys were conducted as part of the Environmental and Social Impact Assessment performed by The Metals Company during Environmental Campaigns 5b (March–April 2021) and 5e (November–December 2021) aboard the offshore supply ship Maersk Launcher. Within NORI-D (Figure 1a), survey locations included the collector test area (CTA), designated as a representative target site for seafloor polymetallic nodule mining, and the preservation reference zone (PRZ), designated as a non-target area and similar to CTA in terms of zooplankton species diversity (Nauru Ocean Resources Inc., 2022b).

Specimens were collected to provide genetic and morphological vouchers for taxa identified from imagery. Target taxa were collected opportunistically at 200–4,000 m depth in conjunction with horizontal transects. The seafloor depth was approximately 4,270 m and the maximum transect depth occurred 5 m above bottom. Video annotations of annelids were reviewed to select representative images of additional taxa that were not collected. These occurrences do not convey the abundance or full depth range of each taxon; community composition will be addressed in a separate work. Additional specimens for comparison were collected in the Pescadero Basin, Gulf of California, Mexico, under CONAPESCA permit PPFE/DGOPA-090/21, with ROV SuBastian on R/V Falkor cruise FK210922 (October–November 2021).

ROV Odysseus (Pelagic Research Services) was equipped with a Mini Zeus 4K video camera (Insite Pacific Inc.) with a 1/2.5 inch Exmor R CMOS sensor and a 20X Optical Zoom lens (4.4 mm to 88.4 mm), and five LED Sealite units (LSL-2000 Flood, DeepSea Power & Light). Video was streamed to the shipboard lab and recorded as Quicktime .mov files by a video ingest server (Cinedeck ZX85) using the Apple ProRes 4:2:2 codec. Time was synchronized using a modular NTP time server (IMS-LANTIME-M1000, Meinberg Funkuhren GmbH & Co. KG). Environmental conditions were measured using a CTD (SBE 19plus V2 SeaCAT, Sea-Bird Scientific) and a dissolved oxygen sensor (SBE 43, Sea-Bird Scientific). ROV navigational data and environmental data were ingested, timestamped, and logged using a Python script. Using an image annotation system (GreyBitsBox, GreyBits Engineering), frames were captured from the streaming 4K video and linked to environmental and navigation data in a database via the Squidle+ web interface with SqCapture and SquidJam plugins (Sangekar et al., 2023). Size indications are not available for some images because the ROV was not equipped with a laser scaler.

Specimens were collected using either a six-canister suction sampler with a variable-speed hydraulic pump or one of four detritus samplers (D-samplers), in which the target enters a vertically oriented open cylinder and is enclosed by hydraulically activated horizontally sliding doors (Youngbluth, 1984). Animals were maintained in chilled seawater at 4°C during shipboard processing, following the principles of sample and data management in Glover et al. (2015).

Following guidelines in Rouse et al. (2022), annelid specimens were relaxed with a 7% magnesium chloride freshwater solution and photographed alive with a Canon EOS Rebel T3i digital camera or under a Leica S8Apo stereomicroscope with a Canon EOS Rebel T6i digital camera. Gulf of California FK210922 specimens were photographed under a Leica MZ12.5 stereomicroscope with a Canon Rebel T7i digital camera. Tissue from the posterior of each specimen, or an entire small specimen, was subsampled into 95–100% ethanol and maintained at −20°C for genetic analysis. Most voucher specimens were fixed in 10% formalin (4% formaldehyde buffered with sodium borate, in seawater) for at least 24 hours, rinsed with distilled water, and preserved in 50% ethanol for archival. Selected Swima voucher specimens were fixed and maintained in 4% paraformaldehyde in Sorenson’s phosphate buffer (pH 7.4, 0.128 M sodium phosphate dibasic, 0.037 M sodium phosphate monobasic) with 8% sucrose. Two Flota specimens (DL100, DL106) were subsampled into ethanol and the remainder fixed and stored in 10% formalin (4% formaldehyde buffered with sodium borate, in seawater). These specimens were deposited with the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) and the Scripps Institution of Oceanography Benthic Invertebrate Collection (SIO-BIC).

Specimens were identified morphologically using descriptions, reviews, and keys for pelagic annelids (Dales, 1957; Salazar-Vallejo and Zhadan, 2007; Osborn et al., 2011; Osborn and Rouse, 2011; Fernández-Álamo, 2021a; 2021b; Kolbasova and Neretina, 2021; Rouse et al., 2022; Kolbasova et al., 2023). Parapodia and chaetae were imaged using a Leica DMR interference contrast microscope and a Canon EOS Rebel T6s camera.

DNA was extracted using the Zymo Research Quick-DNA Miniprep Plus Kit or Microprep Plus Kit (Irvine, California, USA). Mitochondrial cytochrome c oxidase subunit I (COI) was amplified using the polymerase chain reaction (PCR) primers LCO1490 and HCO2198 (Folmer et al., 1994) in the following reaction: 12.5 µl Apex 2.0x Taq Red DNA Polymerase Master Mix (Genesee Scientific, El Cajon, California, USA), 1 µl forward and reverse primer (10 μM), 8.5 µl ddH₂O, and 2 µl of eluted DNA. PCR was performed on a thermal cycler with the following temperature profiles: for SIO-BIC specimens: 94°C/180 s – (94°C/30 s – 47°C/45 s – 72°C/60 s) * 5 cycles – (94°C/30 s – 52°C/45 s – 72°C/60 s) * 30 cycles – 72°C/300 s; for JAMSTEC specimens: 94°C/300 s – (94°C/60 s – 45°C/120 s – 72°C/180 s) * 40 cycles – 72°C/300 s. PCR products were purified with ExoSAP-IT (USB Corporation, Cleveland, Ohio, USA), and Sanger sequencing was performed by Eurofins Genomics (Louisville, Kentucky, USA). Consensus sequences were assembled using Geneious Prime (v. 2022.0 through v. 2024.0) and submitted to GenBank.

Genetic identification was facilitated by comparing sequences to the National Center for Biotechnology Information GenBank database (Clark et al., 2016) using the nucleotide BLAST suite (Zhang et al., 2000; Boratyn et al., 2013). Sequences of interest were aligned using the MAFFT online service v7.511, option L-INS-I (Katoh et al., 2018). COI haplotype networks were created with PopART v1.7 (Leigh and Bryant, 2015) using the TCS algorithm (Clement et al., 2002). Uncorrected pairwise distances (p-distances) among aligned sequences were calculated using PAUP* v.4.0a169 (Swofford, 2003).

In presenting our results, we have followed the annelid classification in Rouse et al. (2022). Midwater species are presented in alphabetical order by family, followed by one benthopelagic observation. Taxonomic uncertainty is expressed using Open Nomenclature qualifiers (Sigovini et al., 2016; Horton et al., 2021), for example, “gen. indet.” (genus indeterminabilis) to indicate that identification to the genus level is not possible based on the lack of diagnostic characters in the available material or images. Records from NORI-D are summarized in Table 1. Additional details, including records from the Gulf of California, are provided in Table S1.

Acrocirridae Banse, 1969

Swima Osborn, Haddock, Pleijel, Madin & Rouse, 2009 

Swima bombiviridis Osborn, Haddock, Pleijel, Madin & Rouse, 2009 

Material examined: SIO-BIC A13750 from NORI-D (one specimen; voucher fixed in 4% paraformaldehyde, posterior tissue sample in 95% ethanol); SIO-BIC A14014 from the Gulf of California, Mexico (one specimen; voucher fixed in 10% seawater-buffered formalin, posterior tissue sample in 95% ethanol).

Morphology: The post-collection live body length of the NORI-D specimen was relatively small at approximately 10 mm (Figure 2a and 2b), compared to the previously reported size range of 18–93 mm for Swima bombiviridis (Osborn et al., 2009). The anterior gut appeared full of ingested matter but not darkly pigmented (Figure 2a), consistent with S. bombiviridis as opposed to S. fulgida Osborn, Haddock & Rouse, 2011 or S. tawitawiensis Osborn, Haddock & Rouse, 2011. As in S. bombiviridis (Osborn et al., 2009) and S. fulgida (Osborn et al., 2011), the chaetae were longer than the body width and uniform along the length of the body, as well as uniform dorsally and ventrally (Figure 2a and 2b). Palps were observed on the NORI-D specimen in situ (Figure 2c) but were lost upon recovery, nor were elliptical “bomb” branchiae observed post-collection. These features are easily detached from Swima specimens (Osborn et al., 2009; Osborn et al., 2011). The single median subulate branchia, not easily lost in this species (Osborn et al., 2009), was retained (Figure 2d). The presence of only one subulate branchia as opposed to three excluded identification as S. tawitawiensis. Digitiform branchiae were not observed in this specimen, although their presence is variable among individuals of S. bombiviridis and S. fulgida (Osborn et al., 2009; Osborn et al., 2011). As in all described species of Swima, the specimen showed clavate papillae visible as opaque spots throughout the gelatinous sheath, one pair of nephridiopore papillae (Figure 2e), and lollipop-shaped inter-ramal papillae projecting beyond the gelatinous sheath (Figure 2f).

Genetics: The three currently described species of Swima are separated by a minimum uncorrected COI distance of 14%, with intraspecific COI distances less than 3% (Osborn et al., 2011). Specimen SIO-BIC A14014, with a COI sequence identical to that of the holotype of S. bombiviridis (Figure 3), confirmed the occurrence of this species in the Gulf of California, along with an existing GenBank sequence (OK209527.1). COI from the NORI-D specimen SIO-BIC A13750 was 99.83% identical to that of the holotype (Figure 3), establishing this specimen as a southern range extension for S. bombiviridis.

Remarks: Although in retrospect the morphology was consistent with Swima bombiviridis, COI sequencing provided important verification to rule out S. fulgida or an undescribed species of Swima.

Distribution: Swima bombiviridis was originally described near Monterey Bay, off central California, USA, at depths of 2,732–3,744 m (Figure 1a and 1b; Osborn et al., 2009). This work extends its distribution from Astoria Canyon, off Oregon, USA (GenBank MZ848603.1), to NORI-D (Figure 1a), representing a total latitudinal range of approximately 4,000 km and an extension of approximately 1,600 km from the previous southernmost record in the Gulf of California.

Swima fulgida Osborn, Haddock & Rouse, 2011 

Material examined: SIO-BIC A13752 from NORI-D (one specimen; voucher fixed in 4% paraformaldehyde, posterior tissue sample in 95% ethanol); SIO-BIC A14013 from the Gulf of California, Mexico (one specimen; voucher fixed in 10% seawater-buffered formalin, posterior tissue sample in 95% ethanol).

Morphology: The morphology was diagnostic of Swima fulgida. The post-collection live body length of the NORI-D specimen was approximately 25 mm (Figure 4a), consistent with previously reported lengths of 29 mm and >30 mm (Osborn et al., 2011). The chaetae were longer than the body width and uniform along the length of the body, as well as uniform dorsally and ventrally (Figure 4a). Palps remained attached to the specimen through recovery (Figure 4a–4c), but elliptical “bomb” branchiae were not observed post-collection. A single median subulate branchia and a row of digitiform branchiae were observed (Figure 4b), as well as dark pigmentation of the anterior gut and purple-brown pigmentation of the lateral lips (Figure 4c). The pigmentation of these features distinguishes S. fulgida from other Swima species (Osborn et al., 2011), although the pigment was not easily visible in situ due to the small size and rapid motion of the animal (Figure 4d). As in all described species of Swima, the specimen showed clavate papillae visible as opaque spots throughout the gelatinous sheath and lollipop-shaped inter-ramal papillae projecting beyond the gelatinous sheath (Figure 4e). Gonads were visible ventrally as opaque masses at the posterior margins of chaetigers 4–6 (Figure 4f) and resemble those documented in two female paratypes of S. fulgida (Osborn et al., 2011). Although the male anatomy of S. fulgida is unknown, we infer the NORI-D specimen to be a mature or near-mature female based on the close match to the described females.

Genetics: Specimen SIO-BIC A14013, with a COI sequence 99.21% identical to that of the Swima fulgida holotype (Figure 5), confirmed the occurrence of this species in the Gulf of California. COI from the NORI-D specimen was 99.35% identical to that of the holotype (Figure 5).

Remarks: Species-level identification of Swima specimens may be impractical under typical imaging conditions of ROV surveys (Figures 2c and 4d) and genetic identification of Swima specimens is often prudent given the extent of undescribed diversity in this group (Osborn et al., 2011).

Distribution: Swima fulgida has been recorded only from its type locality near Monterey Bay at depths of 3,267–3,625 m (Figure 1a and 1b), and its distribution is thought to extend to Oregon and the Gulf of California based on unconfirmed observations (Osborn et al., 2011). This work establishes its southern range to NORI-D, representing an extension of approximately 2,900 km (Figure 1a).

Flabelligeridae de Saint-Joseph, 1894

Flota Hartman, 1967 

Flota sp. KJO-2008 sensu Osborn and Rouse (2008) 

Material examined: SIO-BIC A13753 (one specimen; voucher fixed in 10% seawater-buffered formalin and stored in 50% ethanol, posterior tissue sample in 95% ethanol); SIO-BIC A13754 (one specimen; preparation same as SIO-BIC A13753); JAMSTEC DL100 (one specimen; voucher fixed and stored in 10% seawater-buffered formalin, tissue subsample fixed in 95% ethanol); JAMSTEC DL106 (one specimen; preparation same as JAMSTEC DL100).

Morphology: The NORI-D specimens (Figures 6 and 7a–7e) appeared morphologically consistent with Flota sp. KJO-2008, a possibly novel species of gelatinous flabelligerid recorded from 2,816–3,452 m depth near Monterey Bay (Figure 1a and 1b; Osborn and Rouse, 2008). Flota sp. KJO-2008 is depicted in figure 1 of Osborn and Rouse (2008) and figure 3 of Osborn and Rouse (2011), although it has not been formally described. Gelatinous worms such as Flota are susceptible to dramatic contraction of the body during collection and fixation (Salazar-Vallejo and Zhadan, 2007; Osborn and Rouse, 2011). The in situ habitus of Flota sp. KJO-2008 is shown in Figure 7a–7e. Approximate in situ body lengths were 43 mm (DL100) and 60 mm (DL106), measured from images when the specimens collided with a part of the D-sampler for which the dimensions were known. Approximate post-collection body lengths were 31 mm (SIO-BIC A13753, Figure 6a) and 27 mm (SIO-BIC A13754, Figure 6e), although SIO-BIC A13754 experienced more severe contraction and may have been larger than SIO-BIC A13753 in situ. In Flota sp. KJO-2008, male gonads can be observed as opaque small white masses in chaetigers 5 and 6 (Osborn and Rouse, 2008). Similar features appear present in SIO-BIC A13754 (Figure 6h) and an individual that was not collected (Figure 7e) but not in SIO-BIC A13753 (Figure 6a and 6b), suggesting that the latter specimen may be reproductively immature.

Flota Hartman, 1967, was synonymized with Buskiella by Salazar-Vallejo and Zhadan based on morphology (Salazar-Vallejo and Zhadan, 2007). The genus presently contains three abyssopelagic species: Buskiella abyssorum McIntosh, 1885 with a type locality off Sierra Leone, western Africa; Buskiella flabelligera (Hartman, 1967) with a type locality off southern Chile; and Buskiella vitjasi (Buzhinskaja, 1977) with a type locality near the Kuril-Kamchatka Trench, northwestern Pacific Ocean (Salazar-Vallejo and Zhadan, 2007). The described species have nine to eleven chaetigers and multiarticulated chaetae (Salazar-Vallejo and Zhadan, 2007), as does Flota sp. KJO-2008 from Monterey Bay with nine chaetigers (Osborn and Rouse, 2008) and the NORI-D specimens, all with nine chaetigers (Figures 6a, 6d, and 6f and 7b–7e).

The only morphological characters that distinguish the three described species are the number of gonopodial lobes (only one pair on chaetiger 4, or an additional pair on chaetiger 3) and the shape of these lobes (Salazar-Vallejo and Zhadan, 2007). Two pairs of gonopodial lobes were observed on chaetigers 3 and 4 of NORI-D specimen SIO-BIC A13754 (Figure 6h), but none were observed on SIO-BIC A13753 (Figure 6b), possibly reflecting undescribed ontogenetic variation or sexual dimorphism given the absence of obvious gonads in the latter specimen. The gonopodial lobes of SIO-BIC A13754 (Figure 6h) most closely resembled those of Buskiella flabelligera in their tapering cirriform, as opposed to blunt, appearance (Hartman, 1967; Salazar-Vallejo and Zhadan, 2007).

Genetics: The NORI-D COI sequences differed from each other by only two base pairs and from reference sequences of Flota sp. KJO-2008 (GenBank EU694128.1, EU694129.1) by 14–17 base pairs (sequences 97.21–97.70% identical; Figure 8). Considering that up to 6% COI divergence has been accepted as intraspecific variation in another deep-sea holopelagic flabelligerid, Poeobius meseres Heath, 1930 (Jimi et al., 2019), the NORI-D Flota specimens can reasonably be interpreted as the same species as Flota sp. KJO-2008.

Remarks: Osborn and Rouse (2008; 2011) maintained the genus name Flota for the Monterey species KJO-2008 because its morphology was not an exact match to those of the three described species and may warrant further revision of this group. We retain this name for the NORI-D specimens for consistency. DNA sequences are not currently available from the three described species of Buskiella to evaluate their genetic distances to Flota sp. KJO-2008. At NORI-D, Flota sp. KJO-2008 was frequently encountered motionless and neutrally buoyant in a head-down position (Figure 7e). We observed the swimming ability of these annelids to be slow and poorly coordinated, corroborating previous remarks (Salazar-Vallejo and Zhadan, 2007). In towed net surveys of micronekton and macroplankton at NORI-D, Flota sp. KJO-2008 was the most conspicuous component of the annelid fraction, which accounted for approximately 25% of the numerical density below 3,000 m depth (JC Drazen and colleagues, personal communication, 03/03/2025).

Distribution: This work extends the known range of Flota sp. KJO-2008 from Monterey Bay to NORI-D, an extension of approximately 2,900 km (Figure 1a). Specimen JAMSTEC DL106 increases the maximum depth reported for this species by 611 m (Osborn and Rouse, 2008), for a total depth range of 2,816–4,063 m (Figure 1b). If the morphological resemblance between Flota sp. KJO-2008 and Buskiella flabelligera can be confirmed with new genetic data from the type locality, the range of B. flabelligera would extend from Monterey Bay to southern Chile at depths of 2,816–4,063 m (Hartman, 1967; Salazar-Vallejo and Zhadan, 2007).

Poeobius meseres Heath, 1930

Material examined: A representative individual was photographed on ROV Odysseus Dive 30 at 749 m depth (Figure 7f and Table 1).

Distribution: Poeobius meseres occurs in the northern and eastern Pacific from central Japan to northern Chile (25.4°S), with genetically confirmed records from 433–1,801 m depth across this geographical range (Figure 1a) and additional literature records from 25–3,975 m depth (Figure 1b; Seid et al., 2020). The distribution of P. meseres was once thought to track with the North Pacific subarctic water mass and the California Current (McGowan, 1960), but evidently is not so simple as this species also occurs well into the southeastern Pacific (Seid et al., 2020).

Remarks: The photographed representative of P. meseres at NORI-D was observed motionless and neutrally buoyant with its tentacles extended for feeding (Figure 7f). This individual occurred in severely hypoxic conditions (Table 1), consistent with the known occurrence of P. meseres in the oxygen minimum zone off California (Gilly et al., 2013) and the low oxygen consumption rate and relatively inactive lifestyle of this species (Thuesen and Childress, 1993).

Lopadorrhynchidae Claparède, 1870

Pelagobia Greeff, 1879

Pelagobia sp. indet.

Material examined: SIO-BIC A13759 (one mid-body fragment; subsample fixed in 95% ethanol and entirely used for DNA extraction, remainder frozen at −80°C and thawed into 95% ethanol for further work).

Morphology: These small orange swimming worms were repeatedly targeted for collection, but only one mid-body fragment was recovered (Figure 9a and 9b). Based on the elongated digitiform dorsal and ventral cirri extending past the parapodial lobes (Figure 9c–e), this specimen belongs to Pelagobia (Rouse et al., 2022; Kolbasova et al., 2023). Also consistent with Pelagobia, the chaetae of SIO-BIC A13759 were compound spinigers with a smooth shaft and serrated blade (Figure 9f). The in situ images were not taxonomically informative (Figure 9g).

Genetics: COI sequencing attempts were unsuccessful.

Remarks: Pelagobia contains extensive undescribed diversity and warrants revision, requiring both genetic and morphological approaches (Kolbasova et al., 2023). Without DNA sequences or images of taxonomically diagnostic anterior features, further identification of the NORI-D specimen is not possible. We notate this specimen as Pelagobia sp. indet. Based on the smooth (as opposed to serrated) chaetal shafts in specimen SIO-BIC A13759 (Figure 9f), Pelagobia serrata Southern, 1909 can be excluded (Kolbasova and Neretina, 2021). Pelagobia rubromaculata Kolbasova & Neretina, 2021 can likely also be excluded, as the live coloration of specimen SIO-BIC A13759 did not show rows of brownish-red spots (Kolbasova and Neretina, 2021). A possibly undescribed species of Pelagobia (“sp. A”), with orange pigmentation and reaching 20 mm length, has been reported as “very abundant” at 2,400 m depth off southern California (Thuesen and Childress, 1993), but no images or DNA sequences are available. If Pelagobia sp. A could be collected again for vouchering, morphology, and genetics, it would offer an interesting comparison to the NORI-D morphospecies.

Pelagobia sp. indet. occurred within the severely hypoxic oxygen minimum zone (Table 1). Its presence in low-oxygen waters was somewhat surprising, because this morphospecies and others (see below) swam rapidly to evade collection. Active swimming behavior and a high oxygen consumption rate have been reported for Pelagobia sp. A (Thuesen and Childress, 1993), although the evasive swimming may not have represented typical behavior for the NORI-D species. As with other lopadorrhynchids (Jumars et al., 2015), whether the NORI-D Pelagobia species captures food primarily by cruising or a sit-and-wait strategy is unclear.

Lopadorrhynchidae gen. indet.

Material examined: One individual was imaged in situ, but not collected, on ROV Odysseus Dive 32 at 810 m depth (Figure 9h and Table 1).

Morphology: This yellow-orange morphospecies showed a more fusiform appearance than the collected Pelagobia morphospecies. This specimen likely belongs to Maupasia Viguier, 1886 and the two pairs of cirri on segment 2 (Figure 9h) appear to exclude Lopadorrhynchus Grube, 1855 (Rouse et al., 2022).

Remarks: This morphospecies also occurred within the severely hypoxic oxygen minimum zone (Table 1). Additional lopadorrhynchid species may be present at NORI-D, based on the abundance of individuals observed but not collected or linked to high-quality images. In ROV surveys of the upper 1,000 m at NORI-D, Lopadorrhynchidae accounted for at least 3% of relative abundance of pelagic taxa in the lower oxycline (700–1,000 m stratum; Perelman et al., 2025). The ecological role of these annelids in the CCZ is unknown.

Tomopteridae Grube, 1850

Tomopteris Eschscholtz, 1825

Tomopteris sp. USNM_IZ_1181815

Material examined: SIO-BIC A13760 (one adult specimen photographed and four juvenile specimens collected: one specimen fixed in 10% seawater-buffered formalin and stored in 50% ethanol; two specimens fixed in 95% ethanol, of which one was entirely used for DNA extraction; one specimen fixed in RNAlater).

Morphology: An adult Tomopteris, interpreted as a parent demonstrating brooding behavior, was observed encircling a clutch of juveniles against its ventral surface by maintaining a tightly curved posture with head-over-tail tumbling motions (Figure 10a–10c and File S1). Capture of the adult was attempted using the D-sampler, but only four juveniles were recovered. We interpret the parent worm as the mother, based on indications of internal fertilization in other tomopterid species (Rouse et al., 2022).

The body of the mother appeared completely transparent, with no obvious gut pigmentation. The mother had approximately 15 parapodia-bearing body segments, followed by a thin, smooth tail-shaped posterior projection lacking parapodia and of approximately uniform width, somewhat abruptly demarked from the rest of the body (Figure 10a–10c). The mother had two eyes, and no notch was apparent between the laterally extended prostomial horns, which appeared to span at least the width of the trunk but not the full width of the extended parapodia (Figure 10b). We estimated the length of the elongated aciculae of the second segment (“acicular streamers”) to be approximately 50–75% of the main body length (Figure 10a–10c).

The post-collection live body length of a representative juvenile specimen was 4.5 mm (Figure 10d). Based on the juvenile, we estimated the main body length of the mother to be 53 mm, plus 21 mm of tail (Figure 10c). From these estimates, the tail constituted approximately 42% of the main body length and the acicular streamers approximately 56% of the main body length.

The juveniles had eight parapodia-bearing body segments, on which the last two pairs of parapodia appeared quite incompletely developed, and no tail-shaped projection was evident (Figure 10d). Yellow spots were visible within many of the parapodial pinnules in live specimens (Figure 10e). In the juveniles, the eyes were visible, the prostomium was notched in apparent contrast to the mother, and no aciculae or cirri were visible at the anterior (Figure 10d).

A cross-section of the second pair of parapodia (i.e., relatively advanced in development) in an unstained formalin-fixed juvenile (Figure 10f) showed the yellow spots represented in Figure 10e to consist of clusters of yellow tubes, loosely radiating from the center of each pinnule to the margins. We interpret these structures to be incompletely developed glands, most likely chromophile glands, which in many Tomopteris species are conspicuous as bundles of yellowish tubes converging to a common opening on the ventral margin of the pinnule of the ventral ramus (Dales, 1957; Fernández-Álamo, 2021b). In our juvenile specimen, the tubes were present in both the dorsal and ventral rami and did not appear to converge on a single opening, likely reflecting incomplete development or multiple gland types.

Genetics: The COI sequence of the NORI-D specimen was 99.85–100.00% identical to five Tomopteris sequences from the Gulf of California, 235–1,238 m depth (GenBank OL504928.1, OL504929.1, OL504930.1, OL504936.1, OL504943.1), and >98% identical to eight additional Tomopteris sequences from Monterey Bay, 350–959 m depth (Table 2). These divergences represent differences of 13 base pairs or fewer (Figure 11). We confidently interpret the >99% identical sequences as belonging to the same species, based on a recent phylogenetic study of unnamed molecularly delimited species of northern Pacific Tomopteris (Kin et al., 2022). Kin et al. (2022) reported a minimum interspecific COI p-distance of 8.2%, and we computed a maximum intraspecific distance of 1.67% for the same set of species. The intraspecific COI distances are reflected in the branch lengths in figure 2 of Kin et al. (2022) but were not directly stated in the text. Too few Tomopteris COI sequences on GenBank are currently annotated with species names to enable further intraspecific distance comparisons. We notate the NORI-D specimen as Tomopteris sp. USNM_IZ_1181815 to acknowledge the >99% identical COI matches.

The Monterey Bay sequences in Table 2 with >98% identity can also be reasonably interpreted as a species-level match to the sequences from NORI-D and the Gulf of California. COI divergences of 2–3% have been accepted as intraspecific variation for recent examples in Phyllodocidae (Pearson and Rouse, 2022; Teixeira et al., 2023). Although there is no global species delimitation threshold for Annelida (Kvist, 2016), extensive multi-species analyses have reported average maximum intraspecific divergences of 4.9% (p-distance; Kvist, 2016) and 5.9% (K2P; Carr et al., 2011).

Supporting 5–6% COI divergence as a reasonable benchmark for species delimitation in Tomopteris, the next closest available COI match, Tomopteris sp. USNM_IZ_1450332 (GenBank OK340129.1, Monterey Bay, depth not provided), showed only 93.92% identity to the NORI-D specimen, and a representative live image shows these species to be morphologically distinct (Burns et al., 2024). Tomopteris sp. USNM_IZ_1450332 has short parapodial segments visible along the length of the tail, whereas the tail of Tomopteris sp. USNM_IZ_1181815 at NORI-D appeared smooth (Figure 10a, 10c, and 10g).

Remarks: Tomopteridae is a taxonomically challenging group due to the morphological similarity across species and the wide or cosmopolitan geographical distributions reported for many species. The key characters for identification are the presence or absence of a tail region, the presence or absence of a small cirrus between the prostomial horns and the acicular streamers, and the occurrence of four types of parapodial glands (Dales, 1957; Fernández-Álamo, 2021b). The latter two characters could not be discerned from the in situ images of the mother. Given the major morphological disparities between the mother and juveniles due to their stage of development, we used only the morphology of the mother to assess the identification of Tomopteris sp. USNM_IZ_1181815.

Based on taxonomic literature for tomopterids occurring in the Pacific, we identified several described species as potential morphological matches to Tomopteris sp. USNM_IZ_1181815 (listed in alphabetical order and then discussed in descending order of similarity): Tomopteris aloysii sabaudiae Rosa, 1908, originally described from the Pacific off Mexico and also occurring in the Indian Ocean (Rosa, 1908; Dales, 1957; Dales and Peter, 1972); T. apsteini Rosa, 1908, originally described off Sicily and recorded in the Atlantic and Pacific, from Japan to Mexico and at least as far south as 4°S (Dales, 1957; Berkeley and Berkeley, 1964; Day, 1967; Dales and Peter, 1972); T. duccii Rosa, 1908, originally described from the Pacific off Mexico and also occurring in the Atlantic, Red Sea, and Indian Ocean (Rosa, 1908; Dales, 1957; Day, 1967; Dales and Peter, 1972); T. nationalis Apstein, 1900, originally described off Naples and occurring in the Atlantic, Indian, and Pacific Oceans including off Mexico and in the Gulf of California (Dales, 1957; Dales and Peter, 1972; Fernández-Álamo, 2000; Fernández-Álamo, 2006; 2021b); and T. pacifica Izuka, 1914, originally described off Japan and occurring in the northern Pacific including Kamchatka, Vancouver Island, and Monterey Bay (Izuka, 1914; Dales, 1955; 1957; Day, 1967; Dales and Peter, 1972).

Tomopteris pacifica can reach 40 mm in length (Dales, 1957), with 20 segments followed by a slender tail approximately one-third of the total length (i.e., half the main body length) and acicular streamers (“second tentacular cirri”) approximately 75% of the main body length (Izuka, 1914). The original drawing shows a soft indentation between the prostomial horns (Izuka, 1914), although the prostomium has been reported as not notched (Day, 1967). These proportions are roughly consistent with those in our image, although the NORI-D animal appeared to be considerably longer than 40 mm. In T. pacifica, many of the parapodia and the anterior segments of the tail region have small pigment spots (Izuka, 1914), which were either not present or not discernable in our in situ image. In Monterey Bay, T. pacifica “probably does not occur near the surface” and does occur in “deeper hauls,” that is, a maximum of 1,300 m depth above 2,000 m seafloor depth (Dales, 1955), consistent with the NORI-D occurrences at 837–875 m.

Similarly, Tomopteris apsteini can reach 40 mm in length (Rosa, 1908) with 12–24 trunk segments (Rosa, 1908; Izuka, 1914; Day, 1967) and up to 15 tail appendages, with the last segment of the tail being “very slender and bare” (Rosa, 1908). The parapodia of the tail are sexually dimorphic: after the first three pairs, they are “barely visible or absent” in females, whereas in males the parapodia continue and diminish in size (Rosa, 1908). We observed only a bare section of the tail in the NORI-D female, although a transitional region might have been present and obscured by the cluster of juveniles. The original description does not specify the relative lengths of the tail and main body (Rosa, 1908), but Izuka (1914) reported the length of the tail (11.0 mm) exceeding that of the main body (9.0 mm). The prostomium of T. apsteini is not notched and may even protrude between the two horns, and the second cirrus is described as three-fourths the length of the entire body (Rosa, 1908), subsequently reported as three-fourths the length of the trunk (Day, 1967), which would be more consistent with the NORI-D image.

Tomopteris nationalis can reach 20 mm in length (Day, 1967), with a long tail bearing three or four rudimentary parapodia and measuring 18% the length of the main body (based on 2 mm of tail in 13 mm total length) (Apstein, 1900; Day, 1967). The overall size and relative length of the tail are considerably smaller than in the NORI-D animal. The prostomium of T. nationalis is shallowly notched (Day, 1967), and the second cirrus has been reported as half the body length (Rosa, 1908) or equal to the combined body and tail length (Day, 1967).

Tomopteris duccii reportedly is very similar to T. nationalis (Dales, 1957), with a total length of 20 mm including a tail of 5 mm (i.e., 33% of the main body length; Rosa, 1908). The body is described as long and lanceolate, with the parapodia tapering in size until the tail is completely bare. The prostomium lacks a “frontal incision” and the second cirrus measures approximately 40% the length of the body (Rosa, 1908).

Another similar species, Tomopteris aloysii sabaudiae, has a total length of 15 mm including a tail of 3 mm (i.e., 25% of the main body length; Rosa, 1908). The last half of the tail is completely bare of parapodia (Rosa, 1908), and overall the tail region is less well differentiated than in T. nationalis (Dales, 1957). The prostomium is “slightly depressed” and the second cirrus measures two-thirds the length of the body (Rosa, 1908).

Several Tomopteris species occurring in the Pacific have been described as lacking a tail region (Izuka, 1914; Dales, 1955; 1957; Day, 1967) and can be excluded from identification: T. cavallii Rosa, 1908, T. elegans Chun, 1887, T. ligulata Rosa, 1908, T. planktonis Apstein, 1900, and T. septentrionalis Steenstrup, 1849. Several Pacific species described as having a tail region can most likely be excluded based on other characteristics. Tomopteris dunckeri Rosa, 1908 is noted as having pinnules along the tail, “different from those of the trunk but not rudimentary,” with the last pair protruding freely beyond the tip (Rosa, 1908); additionally, the prostomium of T. dunckeri is notched and the antennae often have a frilly margin (Rosa, 1908; Day, 1967). Tomopteris euchaeta Chun, 1887 attains sizes consistent with the NORI-D animal (up to 150 mm long), but adults have up to 39 body segments and second cirri shorter than the body, whereas juveniles have approximately 15 segments and second cirri two to four times the body length (Day, 1967); neither of these combinations seems consistent with 15 body segments and second cirri shorter than the body in the apparently mature female at NORI-D. Tomopteris krampi Wesenberg-Lund, 1936, once categorized as “without a tail” (Dales, 1955), has rather a “very short tail” and has been recorded only to 26 mm in length (Dales, 1957; Day, 1967), whereas in our image the tail alone measured approximately 22 mm and nearly half the main body length. Tomopteris nisseni Rosa, 1908 also has a relatively short tail bearing reduced parapodia (Rosa, 1908; Dales, 1955; 1957) and frequently shows “a dark brown or violet-brown gut visible through the translucent body-wall” (Dales, 1955), in contrast to the complete transparency of the NORI-D animal.

We also acknowledge that Tomopteris sp. USNM_IZ_1181815 could represent one of the many described species not presently recorded from the Pacific, or an undescribed species. To date, most of the available Tomopteris sequences in GenBank have not been matched to the names of described species, and only five described species of Tomopteris have representative sequences available, so further integrative taxonomic work will be important.

In addition to the mother with juveniles, a similar NORI-D individual demonstrated protective circular swimming behavior with an egg case, and we provisionally identified it as the same species (Figure 10h and Table 1). Protection of tomopterid egg masses has been previously documented, for example, in an unpublished report by Simmons and Von Thun (2009) and annotated ROV images (Monterey Bay Aquarium Research Institute (MBARI), 2013). Our observations appear to be the first record extending parental care to juveniles. In an educational video featuring the latter ROV images, very small juveniles are visible within the egg mass and are accompanied by narration: “Tomopterid worms create round egg masses to protect their babies until the young worms grow large enough to break free and begin hunting in the open ocean” (Monterey Bay Aquarium Research Institute (@MBARIvideo), 2015). In the NORI-D example, the juveniles remained protected by the mother even though they were large enough that no remnants of the egg mass were apparent. We do not know whether juveniles of this size are sustained by yolk reserves, independent feeding, or nutrients provided by the mother such as captured prey.

The mother with juveniles was collected within the severely hypoxic region of the oxygen minimum zone (Table 1). This finding was surprising, given the constant rapid swimming of the mother and the high oxygen consumption rates reported for Tomopteris pacifica and T. nisseni, which reach similar sizes as the NORI-D adult and show similar constant active swimming (Thuesen and Childress, 1993). The mother with eggs occurred slightly deeper than the mother with juveniles, under conditions of intermediate hypoxia (Table 1). Additional observations linked to oxygen data would be helpful to assess whether brooding status affects the depth preferences of this species.

Distribution: This work extends the known range of Tomopteris sp. USNM_IZ_1181815 from Monterey Bay to NORI-D, representing a total latitudinal range of approximately 2,900 km and an extension of approximately 1,400 km from the previous southernmost record in the Gulf of California (Figure 1a). The known depth range for this species remains 235–1,238 m (Figure 1b).

Tomopteridae gen. indet.

Material examined: One individual was imaged in situ, but not collected, on ROV Odysseus Dive 32 at 1,077 m depth (Figure 12a and Table 1).

Morphology: This tomopterid did not have a visible tail-shaped posterior projection and therefore probably represents a morphospecies distinct from Tomopteris sp. USNM_IZ_1181815, although we cannot exclude ontogenetic variation given the tail-less appearance of the juveniles of Tomopteris sp. USNM_IZ_1181815.

Remarks: This individual occurred under conditions of intermediate hypoxia (Table 1). Additional morphospecies are likely present at NORI-D, based on the variety of body shapes observed but not collected or linked to high-quality images.

Typhloscolecidae Uljanin, 1878

Typhloscolecidae gen. indet.

Material examined: One individual, possibly belonging to Travisiopsis Levinsen, 1885, was imaged on ROV Odysseus Dive 32 at 1,077 m depth (Figure 12b and Table 1).

Remarks: This individual occurred under conditions of intermediate hypoxia (Table 1). Little is known about the natural history of Typhloscolecidae, but the few studied examples are parasitoids on chaetognaths (Jumars et al., 2015). Chaetognaths, including the known prey species Eukrohnia hamata (Möbius, 1875), occur from 1,000 m to at least 1,500 m depth at NORI-D, where they comprise approximately 10–17% of the relative abundance of soft-bodied midwater zooplankton (Nauru Ocean Resources Inc., 2022b). These chaetognaths may host additional diversity and abundance of typhloscolecids, although the latter can be small (<3 mm long) compared to their hosts (up to 21 mm long; Øresland and Pleijel, 1991) and may be difficult to observe in situ.

Polynoidae Kinberg, 1856

Polynoidae gen. indet.

Material examined: An apparently benthopelagic morphospecies was observed within 25 m of the seafloor, at depths of 4,171–4,270 m. A representative individual was imaged within 1 m of the seafloor on ROV Odysseus Dive 37 at 4,232 m depth (Figure 12c and Table 1).

Remarks: Although depths >4,000 m fall outside the scope of the bathypelagic zone as defined by the NORI-D Environmental Impact Statement (Nauru Ocean Resources Inc., 2022a), we acknowledge the possibility that this species could also range higher into the midwater.

Nine distinct midwater annelid species were documented (Figure 1b and Table 1). Five species were vouchered, and four of these represent genetically confirmed southern range extensions from previously documented northeastern Pacific localities (Figure 1a): Oregon to the Gulf of California for Swima bombiviridis (total range approximately 4,000 km; extension of approximately 1,600 km), Monterey Bay for Swima fulgida and Flota sp. KJO-2008 (extension of approximately 2,900 km), and Monterey Bay to the Gulf of California for Tomopteris sp. USNM_IZ_1181815 (total range approximately 2,900 km; extension of approximately 1,400 km). Poeobius meseres, previously confirmed to have a latitudinal range of 8,800 km encompassing the CCZ (Seid et al., 2020), was also observed at NORI-D.

Potentially impacted activities for midwater animals include feeding, growth, reproduction, respiration, behavior, physiology, and survival (van der Grient and Drazen, 2022; Stenvers et al., 2023). The five species in this work with available genetic data show connectivity to California or beyond, suggesting the potential for species-level resilience to localized impacts if the remainder of their range is not imperiled.

Several of the pelagic annelid species documented in this work occurred above the expected 1,200 m depth of the NORI-D discharge plume depth (Figure 1b) and thus may be buffered from potential impacts (Poeobius meseres, Tomopteris sp. USNM_IZ_1181815, Pelagobia sp. indet., and the unidentified lopadorrhynchid, tomopterid, and typhloscolecid). Elsewhere in their geographic ranges, however, Poeobius meseres and Tomopteris sp. USNM_IZ_1181815 occur to depths of 3,975 m and 1,238 m, respectively (Figure 1b), so some portion of the NORI-D populations likely also occur below the expected discharge depth. Poeobius meseres feeds opportunistically on sinking detritus, primarily fecal pellets and phytoplankton, using a mucus web (Uttal and Buck, 1996). For individuals occurring below the discharge depth, elevated sediment concentrations might dilute the nutritional content of ingested matter. For Tomopteris sp. USNM_IZ_1181815, the parental behavior and reproductive success of deep-dwelling individuals could potentially be impacted if sediment adheres to the egg mass and affects its buoyancy characteristics. Tomopteris of the large size documented at NORI-D are thought to be predatory, as are Pelagobia and likely other lopadorrhynchids (Jumars et al., 2015). The transcriptome of a close relative of the Tomopteris species collected at NORI-D suggests that the long acicular cirri may function in light detection and chemosensing of prey (Burns et al., 2024), although no data are currently available on the visual and/or olfactory sensitivity of Tomopteris to suspended sediment.

Three species (Swima bombiviridis, S. fulgida, and Flota sp. KJO-2008) occurred below the expected depth of the discharge plume (Figure 1b). All three species are eyeless (Osborn et al., 2009; 2011; Osborn and Rouse, 2011), although water clarity may facilitate predator evasion by S. bombiviridis and S. fulgida, as the bioluminescent “bomb” branchiae of these worms produce an intense glow when detached and are interpreted as a defensive mechanism to distract visual predators (Osborn et al., 2009; 2011). Swima are thought to be predators (Haddock and Choy, 2024) or perhaps suspension-feeders, using their palps for mechanical or chemical sensing (Jumars et al., 2015). The S. bombiviridis and S. fulgida specimens collected in the Gulf of California were encountered approximately 700–900 m above a high-temperature hydrothermal vent emitting dissolved metals and hydrocarbons (Paduan et al., 2018) with no obvious ill effects, although these vents are noted for having “little to no particulate plume” (Paduan et al., 2018) and are not directly comparable to a downward-oriented mining discharge plume. The feeding mode of Flota sp. KJO-2008 and its relatives is unknown; based on the anatomy of preserved specimens, these worms have been suggested to use a mucus net as with Poeobius meseres, or potentially predatory behavior (Salazar-Vallejo and Zhadan, 2007).

Although the NORI-D specimens of Swima bombiviridis, S. fulgida, and Flota sp. KJO-2008 were collected approximately 110–750 m above the seafloor (Table S1), these species are also expected to occur closer to the anticipated benthic collector plume. Off California, S. bombiviridis and S. fulgida have been observed within sight of the seafloor but not interacting with it, at altitudes of 1–444 m and 30–340 m, respectively (Osborn et al., 2009; 2011). Relatives of Flota sp. KJO-2008, Buskiella abyssorum and B. vitjasi, can occur to 5,700–6,235 m depth (Salazar-Vallejo and Zhadan, 2007; Miura, 2014). At NORI-D, deeper occurrences have been confirmed by collection of Flota sp. in net tows at 100–300 m above the seafloor (JC Drazen and colleagues, personal communication, 03/03/2025) and collection of Swima spp. by benthic landers at 5–50 m above the seafloor (GN Ellis and colleagues, personal communication, 03/03/2025), that is, within the expected 50 m maximum vertical range of the benthic collector plume. The unidentified benthopelagic polynoid, observed 1–25 m above the seafloor, occurred within the 2 m zone of maximum suspended particle concentrations from the benthic collector plume and would be appropriate to include in benthic impact studies.

The overall NORI-D checklist was consistent with the dominant polychaete taxa reported in midwater ROV surveys off California: Poeobius, Pelagobia, Tomopteris, and Swima (Haddock and Choy, 2024). Additional annelid species at NORI-D may be uncovered with further review of the survey imagery, and future genetic sampling of the unidentified morphospecies documented only from images would be informative.

Extensive geographic ranges spanning multiple ocean basins are commonly reported in holopelagic annelids (Berkeley and Berkeley, 1964; Dales and Peter, 1972; Fernández-Álamo, 2021b), but genetic assessment including the type locality is important to verify such ranges, as putative cosmopolitan species such as Pelagobia longicirrata Greeff, 1879 and Typhloscolex muelleri Busch, 1851 have been found to contain cryptic species complexes (Kolbasova et al., 2020; Kolbasova and Neretina, 2021; Kolbasova et al., 2023). Most regional studies of holoplanktonic polychaetes in the eastern tropical Pacific have been based on trawl sampling at <200 m depth (Dales, 1957; Fernández-Álamo, 2000; Fernández-Álamo et al., 2003; Fernández-Álamo and Sanvicente-Añorve, 2005; Jiménez-Cueto et al., 2012; Fernández-Álamo, 2020). The deep-sea diversity of these groups likely remains largely uncharacterized.

In a global biogeographic classification of the mesopelagic zone (Sutton et al., 2017), NORI-D is considered part of the Eastern Tropical Pacific ecoregion. This region is generally thought to contain many endemic taxa adapted to the extreme oxygen minimum zone (Sutton et al., 2017), although two species at NORI-D (Poeobius meseres and Tomopteris sp. USNM_IZ_1181815) are neither endemic nor restricted to mesopelagic or hypoxic depths, as they occur beneath the lower oxycline (i.e., 1,000 m) at other Pacific localities. The three other species genetically analyzed in this study (Swima bombiviridis, S. fulgida, and Flota sp. KJO-2008) are currently known only from depths >2,500 m (Osborn and Rouse, 2008; Osborn et al., 2011).

We provide the first pelagic annelid checklist for the CCZ, complementing the current CCZ megafaunal checklist, which was compiled from literature and international biodiversity databases and exclusively benthic in scope (Rabone et al., 2023). This benthic checklist includes 339 polychaete species, consistent with a minimum count of 291 benthic polychaete species identified by DNA taxonomy (Stewart et al., 2023). The true count of benthic polychaete species in the CCZ is estimated to exceed 550 (Stewart et al., 2023).

Although pelagic taxa were intended to be excluded from the benthic CCZ checklist, we noticed that six holopelagic annelid entries (Lopadorrhynchus, Pelagobia, Lopadorrhynchidae, Alciopina, Alciopini, and Typhloscolecidae) were included in the final list (table S1 in Rabone et al., 2023). Because these entries were annotated with “sediment” rather than “pelagic” as their habitat, they may reflect contamination of sediment samples with surface plankton (as has been acknowledged for box core studies at NORI-D; Nauru Ocean Resources Inc., 2022b), incomplete data curation, misidentification, or ingestion by benthic species. Images, sequences, or detailed remarks would be required to compare the lopadorrhynchids and typhloscolecid in the benthic checklist to the deep pelagic morphospecies in this study.

The western and eastern regions of the CCZ have been classified under different pelagic provinces (Sutton et al., 2017; Washburn et al., 2021) and may yield distinct species checklists. Some benthic polychaete species show minimum ranges of nearly 5,000 km within the CCZ, similar to the ranges of deep pelagic species in this work, although connectivity could not be assessed for most of the benthic species as they were recorded only from a single site (Stewart et al., 2023). The genetically confirmed ranges of certain benthic polychaete species occurring in the CCZ even extend into the Atlantic Ocean (Meißner et al., 2023). Particulate organic carbon flux also varies across the CCZ (Washburn et al., 2021) and is thought to drive the observed higher diversity of benthic polychaetes in the eastern CCZ (Stewart et al., 2023). ROV surveys of the deep pelagic zones of the western and central CCZ would offer important biodiversity comparisons.

Within the eastern CCZ, the Federal Institute for Geosciences and Natural Resources (BGR, Germany) and Global Sea Mineral Resources (GSR, Belgium) have also performed collector tests and submitted environmental impact statements to the ISA, as has the Government of India for a contract area for polymetallic nodules in the Indian Ocean (International Seabed Authority, 2024). These statements did not include midwater surveys or literature reviews because the proposed activities were limited to the seafloor and no major impacts to the midwater were expected (Bundesanstalt für Geowissenschaften und Rohstoffe, 2018; Global Sea Mineral Resources NV, 2018; Ministry of Earth Sciences, Government of India, 2020). Thus, these initial midwater biodiversity surveys for NORI-D establish a precedent for future environmental impact studies. The BGR exploration contract area adjoins NORI-D to the north and the GSR exploration contract area is located approximately 1,000 km northwest of NORI-D. Based on the established ranges of species in this work, we would expect the pelagic annelid checklists of these three regions to be similar.

In this first pelagic faunal checklist available to date for the CCZ, we report nine holopelagic annelid species at NORI-D and extend the known geographic ranges of these species as much as 2,900 km. We emphasize the importance of high-quality reference DNA sequences, live images, and vouchered specimens in delineating biogeographic ranges, ensuring interoperability of taxonomic data, and underpinning environmental DNA surveys (Howell et al., 2020; Bianchi and Gonçalves, 2021; Govindarajan et al., 2021). The combination of ROV sampling and careful DNA taxonomy has been instrumental in unmasking cryptic diversity and describing new species of soft-bodied midwater fauna (Osborn et al., 2011; Lindsay et al., 2023; Montenegro et al., 2023; Robison and Haddock, 2024). We expect that further deep pelagic surveys of the CCZ and globally will reveal extensive additional diversity.

DNA sequences are available on GenBank (accession numbers PQ655456–PQ655464). Specimens and associated data have been deposited with the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) and the Scripps Institution of Oceanography Benthic Invertebrate Collection (SIO-BIC).

The supplemental files for this article can be found as follows:

File S1. Animation of parental care for juveniles in the pelagic annelid Tomopteris sp. USNM_IZ_1181815. Fifty in situ images of the mother of specimen lot SIO-BIC A13760 from ROV Odysseus Dive 39 were combined in chronological order and cropped to the region of interest using Ezgif.com (Open Idea Ltd, 2012). Animation speed may not accurately reflect the speed of in situ motion because the images were captured manually at arbitrary intervals. Image timestamps span 20211215T15:49:22Z to 20211215T15:49:50Z (20211215T07:49:22 to 20211215T07:49:50 local). Latitude: 10.9811, longitude: –116.2231, depth: 843–844 m. (GIF)

Table S1. Locality and event details for pelagic annelid species documented at NORI-D and nearby Mexico. (DOCX)

We thank Tiffany Bachtel, Leah Bergman, Kioshi Mishiro, and Erik Thuesen for assistance at sea and Kate Burns, Avery Hiley, and Marina McCowin for DNA extraction and sequencing of SIO-BIC specimens. Mehul Sangekar is also thanked for his help in reconstructing the environmental metadata for ingestion into the video annotation system. We thank the scientific parties of Environmental Campaigns 5b and 5e, project personnel from The Metals Company and Maersk Supply Services, captain and crew of Maersk Launcher, and pilots of ROV Odysseus for supporting operations at sea. For collection of the Gulf of California Swima specimens, we thank the Schmidt Ocean Institute, captain and crew of R/V Falkor, pilots of ROV SuBastian, chief scientist Robert Zierenberg, and the scientific party of cruise FK210922.

This project was partially supported by funding from The Metals Company (TMC) in a collaboration between the University of Hawaii and JAMSTEC. Authors JM and DJL received support from TMC through its subsidiary Nauru Ocean Resources Inc. (NORI). NORI holds exploration rights to the NORI-D contract area in the CCZ regulated by the International Seabed Authority and sponsored by the government of Nauru. This is contribution TMC/NORI/D/021. The authors declare that this study received funding from The Metals Company Inc. and its subsidiary Nauru Ocean Resources Inc. (NORI). The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article, or the decision to submit it for publication.

The authors have declared that no competing interests exist.

Contributed to conception and design: CAS, DJL, GWR.

Contributed to acquisition of data: CAS, JM, DJL, GWR.

Contributed to analysis and interpretation of data: CAS, GWR.

Drafted and/or revised the article: CAS, JM, DJL, GWR.

Approved the submitted version for publication: CAS, JM, DJL, GWR.

How to cite this article: Seid, CA, Montenegro, J, Lindsay, DJ, Rouse, GW. 2025. Deep midwater Annelida from the NORI-D exploration area of the Clarion-Clipperton Zone, Pacific Ocean. Elementa: Science of the Anthropocene 13(1). DOI: https://doi.org/10.1525/elementa.2024.00087

Domain Editor-in-Chief: Jody W. Deming, University of Washington, Seattle, WA, USA

Guest Editor: Jeffrey Drazen, University of Hawaii at Manoa, Honolulu, HI, USA

Knowledge Domain: Ocean Science

Part of an Elementa Special Feature: Deep-Sea Mining of Polymetallic Nodules: Environmental Baselines and Mining Impacts from the Surface to the Seafloor