There is increasing commercial interest in exploiting mineral resources from the deep seafloor, driven, in part, by a perceived need for raw materials for the ‘green transition’ (Miller et al., 2021). The mineral resources of greatest interest are (i) polymetallic nodules on the abyssal plains; (ii) seafloor massive (polymetallic) sulfides at hydrothermal vents; and (iii) ferromanganese crusts on the flanks of seamounts. The Clarion Clipperton Zone (CCZ) is an abyssal region punctuated by seamounts, with an overall average depth of 5500 m spanning approximately 6 million km² of the Eastern Pacific. Within this area, 17 contracts for exploration of seabed mineral resources have so far been granted by the International Seabed Authority, the United Nations body with responsibility for deep sea mining activities in areas beyond national jurisdiction. Thus far, mining is still not a commercial reality, but its proponents are continuing efforts to develop the industry as quickly as possible, even in the face of mounting concern as to the ecological and other risks it may pose (Miller et al., 2018; Levin et al., 2020; Niner et al., 2018; Crane et al., 2024). One argument put forward by some authors is that deep seabed mining is preferable to terrestrial mining in that it could have lower ecological and ethical impacts (Katona et al., 2022). However, it is unlikely that deep-sea mining would substantially displace terrestrial mining operations, such that, given current demand projections, these two sectors would more likely emerge as competitors rather than substitutes (Miller et al., 2021; Crane et al., 2024). The deposits targeted within the CCZ region are primarily manganese nodules, which would be harvested by seafloor collection vehicles in combination with a vertical transport system to a surface vessel where dewatering of products would facilitate the separation of ore-bearing materials (Miller et al., 2018; Levin et al., 2020). Following initial onboard processing, discharge of sediments and water would likely occur either at the seabed and/or within the water column.

Environmental concerns are diverse, including the direct removal of sessile organisms, alteration of seabed integrity, light pollution, sediment plumes and noise (for examples, see Miller et al., 2018; Christiansen et al., 2020; Drazen et al., 2020; Williams et al., 2022). The ecological impacts of nodule mining on sessile and other deep sea megafauna (typically >1 cm long) have begun to be investigated (for example, Ardron et al., 2019; Simon-Lledó et al., 2019). However, there are still major knowledge gaps relating to ecosystem functioning in the Clarion Clipperton Zone, including how the nodules themselves may be involved in dark oxygen production (Sweetman et al., 2024).

Some of the impacts from future deep seabed mining could potentially result in far-field effects in addition to direct impacts in the mining zones themselves. The potential impacts from deep seabed mining on cetaceans are likely to be varied and could relate to noise emissions, sediment plumes and their effect on food webs and to alteration of seamount ecosystems (Williams et al., 2022; Thompson et al., 2023). Williams et al. (2022) constructed predictive models of mining noise which would likely propagate through the water column and potentially across large distances from the source. The authors suggest that sounds from mid-water pumps, for example, could reach the Sound Fixing and Ranging (SOFAR) channel at ~1000 km deep, and propagate across ocean basins. Gillard et al. (2022) investigated the potential impact of sediment plumes as a result of mining in the German sector of the CCZ. The authors stated that most suspended organic particles in this region are small and settle slowly and an increase in fine lithogenic sediment particles from mining could disturb the benthic-pelagic coupling. Numerous seamounts are present in the CCZ and, given that the region overall is a large area, more information is needed on the distribution of deep water and migratory species, in addition to details of migratory pathways, diet and habitat use by cetaceans.

Few dedicated surveys for cetaceans in this area have been published, but sparse records from online repositories, such as OBIS-Seamaps (https://seamap.env.duke.edu/), show at least 16 cetacean species are present in the CCZ at some time during the year and field guides suggest that there could be up to 30 species found in this region (Carwardine, 2020; Thompson et al., 2023). Niu et al. (2021) investigated the CCZ soundscape using a static hydrophone at 300 m deep (during 2017-2018) and recorded fin whales (Balaenoptera physalus) throughout the year and blue whales (Balaenoptera musculus) during September–December and April–May. These data suggest that there may be several threatened species using this area, for at least at some time during the year.

In this study, we used passive acoustic monitoring from the vessel MY Arctic Sunrise to survey a ~41,000 km² region of the CCZ for cetaceans in August 2023. A non-systematic visual survey was also conducted to complement the acoustic survey. The survey area comprised two deep-sea mining exploration blocks, NORI-d (Nauru Ocean Resources Inc. – The Metals Company) and TOML-e (Tonga Offshore Mining Ltd. – The Metals Company), both located within the broader CCZ area. The primary aim of the survey was to provide a snapshot of cetacean presence in a region of the CCZ earmarked for future deep-sea mining.

Between 1 August and 13 August 2023, 273 hours of acoustic recordings were collected. An average speed of 9.5 kts was achieved during acoustic data collection. A total of 74 acoustic detections were made of cetacean groups during the 13 days of the survey ( Figure 1 ). A single sperm whale was detected. Two of the delphinid click events within the CCZ survey region showed the presence of spectral banding. Both events were classified as Risso’s dolphin based on their consistent spectral peaks at 22, 26, 31 and 38 kHz (Soldevilla et al., 2008; Soldevilla et al., 2017) ( Figure 2 ). A single group of common dolphins were acoustically detected, and species identity obtained from a concurrent visual sighting. The ROCCA whistle classifier was unable to provide a robust species classification (likelihood score ≥0.8) for 70 of the acoustic encounters of dolphins; therefore these have been recorded as unidentified delphinids. The deep learning humpback whale classifier did not detect any humpback whale calls with a confidence greater than 0.8, and upon manual checking, all detections were determined to have been triggered by the vessels own propulsion system.

There were six sightings of cetaceans during the survey, of which one was classified to species level ( Table 2 ). One of the other five sightings overlapped with acoustic recordings, a group of common dolphins. Throughout the 13 days of survey, weather conditions were typified by sea state of Beaufort 3, low swell height (<2 m), and good visibility (>5 km).

We provide a snapshot of cetacean presence within an ~41,000 km² of the CCZ earmarked for future commercial scale deep-sea mining, based on acoustic effort over 13 days and 4,328 km of track line during August 2023. A total of 74 acoustic detections included: a single sperm whale, two groups of Risso’s dolphins (based on spectral banding characteristics) and common dolphins. There were no visual sightings to confirm the classification of the two Risso’s dolphin encounters. However, the matching spectral characteristics of these events compared to those described by Soldevilla et al (2008; 2017), along with the typically coastal distribution of Pacific white-sided dolphins, provides a high degree of confidence to these classifications. The remaining 70 acoustic detections were classed as unidentified dolphins, as none achieved a conservative likelihood threshold of ≥0.8 (Webber et al., 2022). No large baleen whale vocalisations were detected, though two sightings of blows were attributed to unidentified whales. No beaked whales or Kogia spp. were detected either visually or acoustically during the survey.

Given previous data on the presence of beaked whales and Kogia spp. in the CCZ, it is surprising that none were detected during the survey in August 2023 (Thompson et al., 2023). The study location has a mean depth of 4259 m with few bathymetric features. Many odontocetes forage in the bentho-pelagic habitats surrounding steep slopes where waters are typically nutrient-rich. These regions may serve as ecotones where pelagic and benthic domains intersect. Beaked whales are thought to occupy such shelf break habitats, with slope being an important habitat feature, including those around oceanic islands and seamounts (Waring et al., 2001; Henderson et al., 2016; MacLeod and Zuur, 2005; Virgili et al., 2022). However, modelling work by Ferguson et al. (2005) indicated that, at least in the Eastern Tropical Pacific, beaked whale distribution may be less confined to these productive regions, and they are predicted to inhabit areas ranging from the continental slope to the deeper waters of the abyssal plains. Therefore, it is not entirely clear whether deep diving species, such as beaked whales, were absent from the area or whether the method used was not effective in detecting them – i.e., whether our result is false-negative or true-negative. The distribution of these species, and any potential seasonal changes in this deep-sea region, are unknown and clearly more survey effort is needed.

The probability of detecting an acoustically active, deep-diving whale using passive acoustic monitoring is dependent on a combination of factors, including distance of the whale from the receiver, ambient and self-noise, source level of the whale’s clicks (phase of their dive, whether in ascent or descent, group size) and the characteristics of the hydrophone array and receiver (Zimmer et al., 2008). For sperm whales, where detection probabilities have been estimated for the Arctic Sunrise, the ESHW was 3,277 m, indicating that a vocally active sperm whale within 1,500 m of the track line would be detected. Other studies with similar methods have reported ESHW’s of 4.2–10 km, much farther than those of the Arctic Sunrise (Fais et al., 2015; Gordon et al., 2020; Lewis et al., 2018). For beaked whales (and Kogia spp.), the efficiency of detection by the Arctic Sunrise acoustic monitoring system is likely to be much less than that for sperm whales. Zimmer et al. (2008) used data derived from acoustic recording tags on Cuvier’s (Ziphius cavirostris) and Blainville’s beaked whales (Mesoplodon densirostris) to determine variability in detection distances from hydrophones. The study found that whales diving within 0.7 km from the receiver were likely to be detected, but those >4 km were not. Tyack et al. (2006) suggest that foraging Cuvier’s beaked whales echolocate in regular 30-minute bouts during foraging, but can remain silent for up to 110 minutes, meaning that a listening time of 140 minutes is required to improve detection probabilities. In this survey, the Arctic Sunrise averaged 9.5 kts and, therefore, travelled ~22 nm (or 41 km) during a 140-minute period, far beyond the acoustic detection range of a foraging beaked whale. Foraging whales may have been easily missed along the track line if the ship sailed over them during the silent phase of their dive. Also key is that these whales emit regular clicks on the descent phase of a deep dive, circa 400–500 m deep above the foraging layer, with clicks orientated downwards. Therefore, clicks are likely only to be detected by hydrophones below the foraging layer at depth, and/or very close to hydrophones due to their high rate of attenuation.

There are few data on beaked whale foraging behaviour in open ocean abyssal habitats such as those surveyed here. Barlow et al. (2021) analysed the depth profiles of 19 satellite-tagged Cuvier’s beaked whales in habitats of different mean depths. The authors found that the whales generally foraged within 200 m of the bottom in depths <2000 m, but for deeper habitats of 2000–4000 m, foraging depths were ~1200 m deep, similar to those recorded in shallower waters. Several authors, however, have speculated that gouges in the seafloor in deep ocean habitats could be attributed to foraging beaked whales (Woodside et al., 2006; Auster and Watling, 2009), including those found within the CCZ (Marsh et al., 2018). Whether individual beaked whales are foraging at such depths in the CCZ is unknown, but presumably the likelihood of acoustic detection will be highly variable and dependent on the phase of their dive, the depth of the whale in relation to the vessel, ambient noise in terms of other species with overlapping frequencies (e.g. dolphins) and the speed and noise of the survey vessel.

Zimmer et al. (2008) advise that hydrophones mounted on a quiet, slow-moving platform, such as drifters or gliders, rather than a fast-moving vessel, will significantly increase the effectiveness of acoustic detection of beaked whales. The Arctic Sunrise is a relatively noisy ship in comparison to other survey vessels, with a variable pitch propeller that has high levels of cavitation noise at <8 kts (KY, pers. obs.). Survey speeds for the vessel are generally between 8–10 kts to reduce self-noise. Presumably the effect of vessel noise on detection probabilities–similar as for sperm whales—will hold for other species and our ability to detect beaked whales will depend on them being relatively close to the vessel, much less than 0.7 km as determined by Zimmer et al. (2008). They must also be vocalising toward the receiver on the towed array, when other ambient sources of sound, such as dolphins, are not masking their high frequency vocalisations. We suggest that further CCZ surveys specifically tailored to beaked whales are required, potentially using multiple drifting devices or gliders. This would help to elucidate their distributions, densities and habitat use, particularly as this group is known to be highly sensitive to noise (Klinck et al., 2012; Hooker et al., 2019; Barlow et al., 2021; McCullough et al., 2021). Such devices would also be hugely valuable for monitoring Kogia spp.

Regarding delphinid detections, only three were identified to species level with two species – common dolphins and Risso’s dolphins – though there were unidentified delphinids detected across the study area. No baleen whales were acoustically detected during the survey, though there were two sightings of whale blows that could not be identified. Noting the continuous (thoughout the year) acoustic detections of fin whales reported by Niu et al. (2021) for the CCZ, it is plausible that the blows witnessed during our survey could also have been from fin whales, but more data would have been needed to confirm this. No verified detections of humpback whales were found using the PAMGuard detector, although we would not expect significant numbers of these whales, given they would most likely be on their high-latitude summer feeding grounds at the time of the survey (Calambokidis et al., 2001). Niu et al. (2021) also reported a lack of humpback whale acoustic detections in their study of the CCZ. The number of false positive detections of humpbacks using the deep learning humpback whale classifier (Allen et al., 2021) is likely due to the detector being trained on static recorders and, as such, training recordings do not contain the same level of noise in the lower frequency range as the towed survey here.

Fragmented governance of the deep oceans in relation to deep sea mining currently limits coordinated effort to protect cetaceans and other species that inhabit these vast offshore regions (Thompson et al., 2018). Here we provide some of the first survey data on cetacean presence in the CCZ. Considering the triggering of the ‘two-year rule’ within the ISA, and the subsequent push from contractors and sponsoring governments to make commercial scale mining in the CCZ a reality in the near future, we argue that the data available on cetacean distributions, habitat use and densities in this area are currently insufficient to determine the nature and scale of potential impacts (Thompson et al., 2023). Our study is a survey of a data-poor region and should be considered when planning and regulating deep seabed mining. In addition, we highlight opportunities for collaboration across intergovernmental organisations (for example, the International Whaling Commission, Convention on the Conservation of Migratory Species of Wild Animals), scientists, non-governmental organisations and regional seas agreements to provide a coordinated response to prevent future threats of mining on cetaceans. Data on cetacean ecology and conservation are heavily biased to near-shore areas, due to the logistical challenges of studying these species in their offshore habitats. In addition, the designation of Important Marine Mammal Areas is hugely important for cetacean conservation but to fulfil selection criteria regions must be relatively data rich. Importantly, a lack of data on oceanic cetaceans should not result in destructive human activities with chronic effects being permitted to become a commercial reality in distant marine realms.