Introduction

It is essential to protect a considerable portion of the Earth’s biomes to slow down biodiversity loss1,2. Many bold post-2020 targets have been proposed to guide the establishment and maintenance of global networks of protected areas3,4. The Global Safety Net suggests the need to protect at least 50% of Earth’s surface to prevent further biodiversity loss and buffer the effects of climate change5. The European Union (EU)’s Biodiversity Strategy for 2030 seeks to create protected areas for 30% of EU’s land and sea territories—the “30 by 30” agenda6. Likewise, signatory countries of the United Nations Biodiversity Conference (COP 15, 7–19 December 2022, Montreal, Canada) agreed to protect 30% of the world’s land, coastal areas, and oceans by 2030. These are just a few examples among many.

As we endeavor to expand the coverage of protected areas globally, we face the challenge of incorporating ecosystems with significant knowledge gaps, especially regarding their biodiversity, into protected area plans. Subterranean ecosystems are a quintessential—though not unique—example of how a lack of direct habitat accessibility, coupled with several research impediments, hampers our understanding of biodiversity7,8,9 and the implementation of evidence-based conservation measures10. We use the term ‘subterranean’ to refer to the extensive network of interconnected underground habitats of varying sizes (ranging from small voids to large cave chambers), substates (from unconsolidated sediments to consolidated rocks), and depths (spanning from shallow to very deep). The diversity of these environments is illustrated in Fig. 1. Except for large cavities and subterranean chambers where humans can enter, most subterranean habitats are only indirectly accessible, and their boundaries are only partly known11,12,13,14. Furthermore, subterranean organisms are often difficult-to-detect, numerically rare, and exhibit narrow distribution ranges, complicating biodiversity inventorying7,8. Consequently, data on subterranean biota tend to be biased towards caves, riddled with extensive geographic and taxonomic gaps, and scattered across a myriad of disconnected publications, datasets, and personal collections—most of which are not openly available or even lost.

Fig. 1: The breadth of subterranean habitats considered in this perspective.
figure 1

The overarching classification (Subterranean [S], Subterranean–Freshwater [SF], Subterranean–Marine [SM]) is based on ref. 75. A Terrestrial caves in different substrates (e.g., karst, lava, ice, salt); B Artificial subterranean habitats (e.g., mines, bunkers, blockhouses, transport tunnels, tombs); C Shallow and deep fissured systems; D Aquifers and groundwaters (e.g., subterranean lakes, rivers, ponds); E, F Springs, wells, and other surface-subterranean ecotones (e.g., voids within vadose zone in karstic and fissured aquifers, interstitial habitats such as hyporheic zone); G Artificial aquatic subterranean habitats (e.g., tanks, aqueducts, water pipes); H Marine caves, hosting coastal pools and subterranean voids connected to marine waters; I, J Anchialine caves and pools contain tidally influenced water bodies where fresh, brackish, and salt waters mix through subterranean connections between the sea and the groundwater. Anchialine pools are also exposed to open air and sunlight. Photo credits: A, B, D, G–I uPIX Fotografia Ipogea; C, E, F Ilaria Vaccarelli; J David Brankovits.

Full size image

Despite this incomplete and sparse knowledge, subterranean ecosystems require ample protection as they, directly and indirectly, support other ecosystems (e.g., groundwater-fed springs)15, are highly biodiverse16, and deliver numerous ecosystem services17. Given the aforementioned reasons and others, we cannot afford to delay the inclusion of subterranean ecosystems further in general conservation policies and agendas18,19,20,21. Subterranean ecosystems are increasingly threatened (Box 1), and yet, even in Europe, they remain largely uncovered by the existing network of protected areas22. Furthermore, with very few exceptions, decision-makers continue postponing political actions to manage these ecosystems. We believe that, even in the face of uncertainties, it is both timely and critical to assess general principles for subterranean biodiversity protection and implement them based on precautionary principles.

To this end, we established a consortium of researchers working with various subterranean ecosystems and taxonomic groups, which the aim to foster the conservation of subterranean biodiversity throughout Europe. This consortium devised into project DarCo (see “Acknowledgments” for details), whose main objective is to collate the best available data on European subterranean biodiversity and use this knowledge to promote studies on subterranean biota, identify threats, and improve the representativeness of subterranean ecosystems into the Natura 2000 network of protected areas. In this perspective, we aim to examine the major challenges associated with designing and managing subterranean protected areas. Focusing on Europe as a case study, we discuss impediments and knowledge gaps, and advance solutions to overcome them.

Status of subterranean ecosystems protection

Area-based protection of subterranean ecosystems is deficient globally. Only 6.9% of karstic and volcanic landscapes bearing subterranean habitats overlap with around 17% of the terrestrial and inland waters surface that is covered by protected areas18. Furthermore, an estimated 85% of protected areas overlapping with aquifers do not include their catchment boundaries23. Importantly, most subterranean habitats receive protection primarily due to their location beneath areas designated for the conservation of surface species or habitats, without explicit consideration for their vertical and 3-dimensional nature (Box 2).

The EU fares better than other regions with regard to subterranean protected areas, thanks to the Natura 2000 network. This is the largest transnational coordinated network of areas of conservation in the world, aiming to preserve Europe’s most valuable and threatened species and habitats as defined in the annexes of the Habitats and Birds Directives (Council Directives 92/43/EEC and 2009/147/EC). Currently, about 18% of the EU’s land area and 8% of its marine territory are covered by the Natura 2000 network24. According to our estimations (Supplementary Text 1), 21.84% of EU subterranean habitats are indirectly covered by the Natura 2000 network, in that they overlap with protected areas set at the surface (Fig. 2). Furthermore, the Habitats Directive lists a few specialized subterranean species [e.g., the mussel Congeria kusceri (Bole) and the snail Paladilhia hungarica Soos, the beetles Leptodirus hochenwarti Schmidt, Duvalius gebhardti (Bokor), and Duvalius hungaricus (Csiki), the olm Proteus anguinus Laurenti], several facultative subterranean species (e.g., all species of subterranean-roosting bats and all cave salamanders of the genus Speleomantes), and three subterranean habitat types (Table 1) in its annexes. Since the onset of the Habitats Directive, 2407 Natura 2000 sites (8.9% of total sites) were specifically established targeting these subterranean habitat types (Table 1).

Fig. 2: The extent of (indirect) subterranean ecosystems protection across the European Union.
figure 2

The map shows the European Union surface covered by the Natura 2000 network of protected areas (orange) overlaid on areas that have subterranean habitats (blue). Methods and data sources to generate the map and associated analyses are available in Supplementary Text 1.

Full size image
Table 1 Number of Natura 2000 sites established for the direct protection of subterranean habitats across the European Union.
Full size table

Notwithstanding these positive trends, the protection of subterranean biodiversity remains deficient in most areas22,25. As for aquatic subterranean habitats, indirect protection primarily concentrates on chemical threshold values set by EU regulations (e.g., the Groundwater [2006/118/EC], Water Framework [2000/60/EC], Environmental Quality Standards [2008/105/EC], and the recast Drinking Water [2020/2184/EC] directives) to provide safe drinking water for human consumption. For groundwaters, attaining a good status entails meeting specific standards for water quantity and quality, often through the establishment of Groundwater Drinking Water Protected Areas and regulation of water abstraction. While these directives represent progress in integrating subterranean ecosystems within the same river basins under a common protection umbrella, they disregard biodiversity, ecosystem processes, and the ecological interconnections between terrestrial and aquatic subterranean compartments. Hence, it would be important to complement the current resource-oriented protection of subterranean resources with an ecosystem-oriented perspective.

Subterranean organisms and their habitat—rocks, sediments, and waters—form diverse subterranean ecosystems. While microorganisms and animals depend on certain abiotic conditions (e.g., temperature ranges, chemical composition of rocks and water) for their survival and reproduction, they, in turn, modify their abiotic environment through a range of bioengineering activities26,27. In doing so, they control important services provided to humans by subterranean ecosystems, such as drinking water supply. Policy and decision-makers need to recognize the ecological dimension of groundwaters and terrestrial subterranean habitats, thereby ensuring the preservation of subterranean biodiversity and ecosystem services for future generations.

Biodiversity data and surrogate variables

Identifying suitable sites for establishing protected areas involves delicate trade-offs between scientific knowledge (spatio-temporal distribution of biodiversity and human pressures) and short- and long-term socio-economic interests (political and societal needs). In essence, conservation practitioners aim to minimize costs and the total area dedicated to conservation while maximizing biodiversity protection across various facets—taxonomic, functional, and phylogenetic diversity, or even emergent properties like the delivery of ecosystem services. Hence, practitioners require high-quality biodiversity data as the foremost ingredient to develop cost-effective plans for designating protected areas.

As far as subterranean ecosystems are concerned, accumulation of high-resolution biodiversity data has progressed at a slow pace owing to technical obstacles in mapping and exploring subterranean ecosystems and intrinsic biological characteristics of subterranean organisms, including their rarity and high levels of endemism. Subterranean ecosystem boundaries are often unknown or inexact due to the existence of difficult-to-define transitional areas28,29,30. Subterranean organisms typically reside in hardly accessible networks of millimetric voids and fractures that extend below the surface down to aquicludes, aquifers, or the bedrock, rather than in large and accessible cavities. The scarcity of taxonomists working on subterranean taxa and the high frequency of cryptic species31,32 further complicate biodiversity inventories. As a corollary, understanding of human impacts on this biodiversity is also limited (Box 1).

While emphasizing the importance of expanding basic knowledge about subterranean biodiversity, we foresee two main approaches to avoid postponing conservation decisions. One solution is to focus prioritization assessments on the few taxonomic groups for which high-quality data are available, either because there has been a long tradition of studies (e.g., for cave-roosting bats33) or due to greater local knowledge often driven by the scientific interest of individual taxonomists (e.g., for harpacticoid crustaceans in Southern Europe34). While this approach should lead to some level of ecosystem protection, there is a main limitation associated with it. The ‘umbrella protection’ effect provided by a specific taxonomic group can be limited35,36. For example, a recent analysis showed that the conservation needs of cave-roosting bats only partially overlap with those of other subterranean ecosystem components36. Hence, for this approach to work, one would need to study correlations between the distribution of different taxonomic groups to identify the most suitable ‘biological indicators’ for prioritization exercises (see ref. 35. for an example with groundwater fauna). Otherwise, there is the risk that a biodiversity hotspot identified for a target taxon may not be such for other taxa.

A complementary solution, likely to work best at broader spatial scales, is to use environmental diversity as a surrogate of biodiversity for obtaining spatially continuous biodiversity data37,38. Over the last twenty years, subterranean ecologists have gained much experience in predicting the distribution of individual species39, the richness of communities35,40, or the range size of species41. They identified key environmental surrogates of subterranean biodiversity over a range of spatial scales42,43. In Europe, for example, the regional species richness of terrestrial and aquatic subterranean communities peaks in areas of high surface productivity and habitat heterogeneity that have not been affected by cold or arid historical events40,44. Furthermore, Quaternary climate oscillations have caused a clear pattern of decreasing species’ range size and increasing spatial turnover of communities with decreasing latitude in Europe41. A finer scales, the size of voids available to organisms (e.g., from small pores between sand grains to large cavities45), their interconnectedness, and their connectivity to the surface environment are key drivers of the richness and composition of local subterranean communities46, implying the need to protect distinct habitats within regions. Likewise, some geological formations, with certain ages or subject to particular physical-chemical phenomena, are known to be more prone to exhibit specific environmental conditions leading to species-rich subterranean assemblages47,48. Once the relation between environmental variables and the presence of some species (or functional groups49) is established, it might help in the designation of priority areas under a probabilistic framework.

From theory to practice

Conservation stakeholders typically need conservation targets—in our case, subterranean protected areas to be established in the context of the “30 by 30” agenda. These targets need to be expressed in a measurable and clearly understandable way (e.g., maps showing conservation targets using color gradients). Unfortunately, even when conservation targets are clearly identified33,34,50,51,52,53, it may be impossible to address all of them, thereby requiring the use of complementarity approaches for identifying the most pressing conservation issues (e.g., using the concept of irreplaceability54). First and foremost, the overlap between landscapes with known subterranean habitats and areas inhabited by people should be quantified to inspire discussions about realistic goals for protected area coverage. This is critical given that globally, more than a billion people live on karst55, with an estimated 25% of these people located solely in Europe. Furthermore, the ‘optimal’ protection of biodiversity should extend well beyond political borders, as these are only recognized by humans. Indeed, aquifers and karst patches are often transnational23,56, leading to logistic and legal challenges hindering effective protection57. Ultimately, pinpointing conservation targets is a complicated process that requires intersecting biodiversity data (biodiversity targets), landscape features (e.g., aquifers), estimates of ecosystem vulnerability, and actual anthropic pressures to designate areas showing risks of major biodiversity loss, and embedding this consensus information into the actual territory and existing international legal framework58.

Once priority areas for protection have been targeted, the effective establishment of subterranean protected areas should happen after consultation among multiple actors, including researchers, policymakers, and, above all, the agencies that have control of the land through ownership or legislation. This inclusive approach increases the likelihood of adoption and benefits for everyone59. Although simple to write, successfully engaging a diverse range of actors with often disparate perceptions and interests is far from trivial, especially when navigating conflicts of different stakeholders60. Failure to include some parties and their needs may result in sabotage, political manipulation, public discontent (e.g., human-wildlife conflicts, restricted resource usage, displacement of people), and ultimately the failure of conservation efforts61,62. This is also applicable to subterranean ecosystems, where successful conservation outcomes are often achieved through close cooperation among conservation scientists, the media, the public, and decision makers63 (see Box 3 for an example). Similar integrated actions are especially needed in the context of new and emerging threats and uses of groundwater ecosystems, such as changes in hydrology due to climate change or novel uses of groundwater ecosystems for heat storage or as a novel renewable energy source64,65.

A well-planned narrative can positively impact public support for wildlife conservation policies63,66,67. The development of action plans and agreements to preserve subterranean ecosystems requires the use of common (underground!) vocabulary by the transnational actors involved, i.e., steering away from scientific jargon68, to ensure a broad understanding and collective agreement. It requires identifying principles and “boundary objects” that promote participation and accountability across disciplines (e.g., ecology and hydrogeology), between scientists and decision makers, and between micro- and macro-entities and authorities (EU, member states, and regions), as the subterranean realm transcends administrative boundaries57. Furthermore, protected areas need to evolve to withstand the effects of time, including management changes (downsizing, degazetting, variability in budget and staff69,70), climate change, and population growth71,72.

As part of future research initiatives towards subterranean biodiversity conservation, sharing of credible, legitimate, and salient information by researchers and stakeholders will be key to successfully transferring scientific and technical knowledge to decision-making and implementation of protected areas73. To bolster scientific credibility, future research initiatives should assemble international and diverse panels of experts in the field of subterranean biology and conservation sciences. These initiatives should promote their legitimacy by engaging with multiple stakeholders—from conservation-makers to managers, speleological groups, and users of subterranean resources—to account for their concerns and perspectives in designing subterranean protected areas. It is crucial to deliver salient information to decision makers by combining data on multiple facets of subterranean biodiversity (e.g., taxonomic, phylogenetic, and functional diversity), aquifer intrinsic vulnerability, groundwater resource use, and anthropogenic pressure. Furthermore, future initiatives must integrate these data into a systematic conservation planning model and use it as a “boundary object” through which they are able to explore networks of protected areas that best address the concerns of different actors. Effectively protecting biodiversity and functioning of subterranean ecosystems while meeting human development is a difficult but not insurmountable task, as illustrated by the recent success of subterranean protected areas in the oceanic islands of the Azores (Box 3). Yet, we acknowledge that the task is even more challenging when it is to be achieved at the scale of a densely populated continent. The DarCo initiative will aim to fulfill the above-mentioned requirements and move subterranean biodiversity conservation forward.

Conclusions

As with all conservation actions, there are potential pitfalls in establishing subterranean protected areas. It involves testing direct management solutions with limited knowledge while adapting to the changing climate (both managerial and environmental). It requires coordination among multiple actors with often highly divergent views but a common goal, all of whom may engage at different levels of the process. It means accepting the predominantly undefined nature of subterranean ecosystems and confronting their high 3-dimensionality. It implies, quite literally, operating in the dark, defining protected areas without having all the information74.

Regardless of these challenges, it would be irresponsible to postpone political action under the agenda that ‘more knowledge is needed’; there is no single ecosystem on Earth that we will ever fully understand. As in the successful case of Azores (Box 3), it is high time to put prioritization exercises to the scrutiny of conservation practice. What is a better place to start than within the well-established infrastructure of the Natura 2000 network and the ambitious agenda of the EU Biodiversity Strategy for 2030?