We used species records from existing studies (Wolff, 2005; Chapman et al., 2019) and expert knowledge to compile a complete list of all mollusc species described to date (up to 2021) endemic to active hydrothermal vent environments. As Red List assessments are taxon-specific, only named species were included in this study. Species were considered endemic where they had only been recorded at hydrothermal vents at time of assessment. Species that are also known from other deep-sea habitats, as well as vent-peripheral species (including all cephalopods), were excluded from this list, leaving a total of 184 vent-endemic molluscs (Supplementary Material 1). These species span five different mollusc classes: Gastropoda, Bivalvia, Monoplacophora, Polyplacophora, and Solenogastres.

We assessed the extinction risk of each species using the Red List criteria (IUCN, 2012), following IUCN guidelines and the method presented by Thomas et al. (2021). All assessments were made using Red List criteria B and/or D2 (Table 1), which use geographic distribution data and information about continuing declines (criterion B) or plausible future threats causing extreme declines (criterion D2) to determine extinction risk. Nearly all vent species are lacking data to calculate population sizes and trends or run extinction risk models required for the other criteria (Thomas et al., 2021). Literature reviews were conducted for each species and data collected included the name, location, depth, biogeographic region, and local jurisdiction of the vent fields from which each species is known. Each assessment is therefore based on the best available published distribution data available at the time and may be subject to revisions as new data become available. Vent field names and coordinates listed in the literature were compared and aligned with the InterRidge Vents Database to ensure consistent nomenclature for sites (Beaulieu and Szafranski, 2020).

As the primary anthropogenic threat to vent-endemic species (Van Dover, 2014), deep-sea mining informed the criteria B and D2 requirements of continuing declines and plausible future threats for the Vent Red List assessments (Table 1). The local mining threat for each locality was determined based on regional seabed management objectives and the regulatory frameworks at sites within the range of each species (Thomas et al., 2021). Hydrothermal vents in EEZs are regulated by national governments and those in ABNJ are regulated by the International Seabed Authority (ISA) (Thompson et al., 2018). Key considerations for threat assessment included the implementation of deep-sea mining licenses, marine protected areas (MPAs), and mining moratoria. For example, continuing decline was inferred in areas of exploratory mining contracts signed by the ISA or in the EEZs of nations that have granted mining licenses (criterion B); plausible future threat was considered in areas where there are no regulations in place to protect from future deep-sea mining (criterion D2) (Thomas et al., 2021).

To illustrate the global distribution of extinction risk to vent species, all assessment data were amalgamated to produce a list of hydrothermal vent fields with the number of species assessed under each Red List category at each site (Supplementary Material 2). As with previous studies (e.g., Tingley et al., 2019), the five Red List categories were then assigned a sequential ranked risk score (Least Concern = 1, Near Threatened = 2, Vulnerable = 3, Endangered = 4, Critically Endangered = 5), and the number of species in each category were multiplied by its risk score and summed, to produce a total assessment of threat to all species at that vent field (Eq. 1).

where i is each Red List category.

To reduce bias between sites with varying species richness, a standardized Threat Score was calculated by dividing the Ranked Sum by the total number of species across all categories at that site (Eq. 2).

For example, at the Menez Gwen vent field on the Mid-Atlantic Ridge, there are eight species, of which six species were assessed as Least Concern (risk score = 1) and two Near Threatened (risk score = 2). Thus, the Ranked Sum is (6^∗1)+(2^∗2) = 10, resulting in a Threat Score of 10/8 = 1.25 for that vent field.

The distribution of threats to hydrothermal vents is closely tied to the regulation within each geographic area. In ABNJ, hydrothermal vents within ISA mining license areas along the Mid-Atlantic and Indian Ocean Ridges (Miller et al., 2018) have a higher Threat Score than those at the Northern and Southern East Pacific Rise, with Indian Ocean vent species exhibiting the greatest proportion of threatened Red List assessments. This is influenced by the overlap in species ranges with a variety of regulatory areas: species found at sites along the Mid-Atlantic Ridge have ranges across a mosaic of ABNJ and EEZs with differing protections. Threats to vent sites within individual EEZs are highly variable and dependent on current national regulation of the seabed. Vent fields that lie within the territorial waters of countries that have granted mining licenses, such as Japan (Okamoto et al., 2019) and Papua New Guinea (Hoagland et al., 2010), generally exhibit high Threat Scores. Conversely, vent species in countries and regions that have endeavored to preserve portions of the seabed with MPAs, such as the Azores (Calado et al., 2011; Abecasis et al., 2015) and Mexico (Menini and Van Dover, 2019), are at a much lower risk of extinction.

While this study is focused on vent-endemic molluscs, our results are representative of the global distribution of mining threat for all vent-endemic taxa. Deep-sea species tend to be very data-limited, therefore future assessments of extinction risk for other vent taxa, including crustaceans and polychaetes, are expected to be reliant on the same Red List criteria as the Vent Red List (Thomas et al., 2021). Vent species exhibit high endemicity, and the primary anthropogenic threat is regionally controlled, thus the overall proportion of Red List categories for hydrothermal vent environments is unlikely to vary significantly with the addition of new taxa. The Red List is a species-based conservation tool, yet in this case it provides a comprehensive and reliable illustration of the geographic distribution of threats to a specific deep-sea habitat.

Taxonomy-driven tools such as the Red List can cultivate deep-sea conservation; the application of Red List assessments to vent-endemic molluscs provides a precedent for other deep-sea taxa and habitats (Glover et al., 2018; Sigwart et al., 2019). Red List assessments are easily understood by a wide range of stakeholders (Rodrigues et al., 2006; Betts et al., 2020), and have the potential to provide an alternative perspective to ecosystem-based management approaches and enhance deep-sea conservation initiatives beyond hydrothermal vents.

The cobalt-rich ferromanganese crusts of seamounts, for example, are a target of five mining exploration licenses granted by the ISA to Brazil, China, Japan, Russia, and the South Korea (Miller et al., 2018; ISA, 2021). Seamounts are characterized by high productivity (Clark et al., 2010) and are rich in long-lived corals that likely take decades to millennia to recover from disturbance (Gollner et al., 2017; Watling and Auster, 2017). Likewise, gas hydrate deposits that occur in conjunction with cold seeps, another important chemosynthetic habitat in the deep sea that hosts hundreds of similarly endemic species to vents (Wolff, 2005; Suess, 2018), are also under increasing consideration for commercial extraction (Chong et al., 2016; Miller et al., 2018). The insular nature, relatively well-documented global distribution, and imminent mining threat to both these habitats mean it is viable to apply a Red List approach to assess these hotspots of deep-sea biodiversity (Thomas et al., 2021).

One potential limitation to our Red List approach, as with the majority of deep-sea research, is the underlying lack of observational data for different vent populations. None of the species included in this study were assessed as Data Deficient, yet it is recognized that the distribution and range of some mollusc species at hydrothermal vents is not fully known and requires further research. The global summary of endemic molluscs at hydrothermal vent sites illustrates overall low species richness across the majority of vent fields, compared to a few species-rich sites. This may be indicative of the paucity of baseline biodiversity knowledge for deep-sea habitats (Glover et al., 2018); however, this could also be a feature of the local and regional heterogeneity seen in vent communities arising from geographic variation in tectonic activity and vent geochemistry (Van Dover, 2000; Thaler and Amon, 2019). In the face of accelerating threats, assessments of extinction risk can only be based on the best available data at the time (IUCN, 2016), and although data are lacking for many groups (Glover et al., 2018), the Vent Red List demonstrates that there is sufficient information to assess even relatively data-poor species using Red List criteria (Thomas et al., 2021).

Specimen collection and taxonomic research are fundamental to deep-sea conservation, and present a bottleneck to the application of taxon-based conservation tools (Glover et al., 2018). This can be especially problematic where there is ongoing research that could result in taxonomic revisions, such as the study of species complexes. For example, among Lepetodrilus species on the East Pacific Rise, genetic evidence from multiple studies delineate separate lineages that are currently included within nominal species (Johnson et al., 2008; Matabos and Jollivet, 2019). Assessing a species complex as a single taxon could artificially lower the threat category in a Red List assessment because the taxon represents an over-estimation of combined abundance and range for several species. Taxonomic research and timely updates following reviews are, therefore, of the utmost importance to ensure accurate measures of extinction risk. Ongoing research can be integrated into Red List assessment text and taxa can be reassessed as frequently as required (IUCN, 2012). Furthermore, based on current findings, seabed management and mining regulation appears to have a greater impact on Red List assessment outcomes than the distribution data for individual species (Thomas et al., 2021).

A more concerning trend that has recently emerged is the potential North/South divide in the extent of biological research and deep-sea mining prospects (Thaler and Amon, 2019). We found a similar pattern in this study, with the Indian Ocean having the greatest proportion of threatened Red List assessments despite exhibiting one of the lowest collective species richness counts. The high proportion of Critically Endangered assessments at the Indian Ocean vent fields is indicative of their biodiversity uniqueness as species can only be assessed as Critically Endangered under Red List criterion B if they are known from a single location (Thomas et al., 2021). In fact, along with the Southwest Pacific, the Indian Ocean hosts the greatest proportion of locally endemic species known only from a single hydrothermal vent field. Further research is however required to determine whether this is simply a factor of low sampling effort in the Southern Hemisphere.

While entire vent biotas are threatened in biogeographic regions that have no protection from deep-sea mining, such as the Indian Ocean, the threat is significantly reduced in regions that straddle different regulatory areas, like the Mid-Atlantic Ridge. Despite several ISA mining licenses issued for the Mid-Atlantic Ridge, no vent fields in this region have a Threat Score greater than three as a result of the protection provided by the Azores MPA network (Abecasis et al., 2015). This highlights the importance of implementing conservation measures like MPAs to reduce extinction risk, yet at present, other than the South Georgia and South Sandwich Islands MPA, all other vent-related MPAs are located in the northern hemisphere. In an ecosystem that has well-documented, distinct faunal communities across different biogeographic regions, the preservation of areas representative of each region is essential.

The conservation of hydrothermal vent habitats and their unique fauna requires action to lower the extinction risk of vent species. While the incorporation of new data to the Vent Red List has potential to influence assessment outcome and lead to Red List category change, this would only constitute a non-genuine change under IUCN regulations (IUCN Standards and Petitions Committee, 2019); i.e., improved knowledge on the species rather than a true change in extinction risk. Furthermore, given the relatively broad category thresholds of the Red List criteria, the addition of new species distribution data does not always affect assessment result (Thomas et al., 2021). Consequently, to improve the extinction risk of vent species, real conservation measures would need to be implemented that mitigate the threat of deep-sea mining.

One conservation method that has potential to safeguard hydrothermal vents from deep-sea mining is the implementation of MPAs (Menini and Van Dover, 2019). Globally, 70 vent-endemic mollusc species were assessed as Least Concern or Near Threatened, based on the protection afforded by MPAs, either to the entire, or a proportion of the population. These assessment outcomes demonstrate the effectiveness of MPAs to protect individual species against the threat of extinction. However, the presence of an MPA alone does not always constitute protection and both assessment and real conservation outcomes are dependent on whether the threat is sufficiently mitigated by implementation of the MPA (Edgar et al., 2014). For example, while bottom fishing is prohibited within New Zealand’s Kermadec Benthic Protected Area, there is no active legislation against other commercial activities like seafloor mining, so vent sites are left unprotected (Van Dover et al., 2011). Furthermore, even MPAs that include specific regulation for the protection of hydrothermal vents are at times too small to afford protection from nearby threats. Research predicts that mining sediment plumes can spread up to 70 km (Luick, 2012; Miller et al., 2018), therefore hydrothermal vents within this proximity, protected or not, could possibly be impacted. For example, the Endeavor Hydrothermal Vents MPA in Canada encompasses five hydrothermal vents of interest, but is within 70 km of nearby unprotected vents. Likewise, proximity to inactive vent deposits that are also potential mining targets (Van Dover, 2019) should be considered. Therefore, it is imperative that MPAs are planned with reference to all vents in an area, including inactive deposits, and are established including a generous buffer zone, as a precautionary measure.

With increasing demand for metals to support renewable energy technologies, blanket MPAs may not be the most realistic approach for policy makers. A more measured approach may be to implement a moratorium on deep-sea mining to allow for further research into the biodiversity, ecology, connectivity, and resilience of vent communities (Glover et al., 2018; Van Dover et al., 2018). Several countries (Kakee, 2020) and, more recently, large corporations (No Deep Seabed Mining, 2021), have declared their support for a mining moratorium until sufficient advances have been made to inform environmentally sound mining legislation. The Red List assessments presented in this study provide a global overview of mining threat at hydrothermal vents and support a precautionary approach for deep-sea conservation, including the implementation of a deep-sea mining moratorium.