Introduction

The urgent need for respectful, ethical, and inclusive data management is widely acknowledged1,2,3,4. Here we aim to highlight the challenges of improving visibility of biological species’ use datasets, as well as strengthening collaboration with IPLCAD– i.e., Indigenous Peoples, Local Communities and Afro-Descendants, hereafter IPLCAD, see Box 1 for definitions. Indigenous and local knowledge (ILK) systems are expressions of traditional and ancestral sciences (Box 1) recognized as crucial for tackling complex, interconnected global challenges and harnessing biodiversity’s potential to drive sustainable development5,6, https://www.wipo.int/treaties/en/ip/gratk/]. ILK about the uses of biodiversity is rarely restricted to a single cultural group, as sharing knowledge is a fundamental aspect of cultural dynamics7. Species can be used and valued by multiple groups in similar or different ways. Ecological, technological, and utilitarian ILK (along with other aspects of knowledge about biodiversity) have been passed down within and across groups for centuries and travel alongside people during migration events, which are pervasive throughout human history8. Global datasets of ILK regarding biodiversity hold undeniable value for analyzing trends and impacts of these knowledge systems worldwide, by upscaling local evidence relevant to conservation, ecology, and bioeconomy, as well as understanding the historical ecology of social-ecological systems9,10.

The importance of global datasets has grown significantly with the rise of big data and powerful processing tools and technologies, such as artificial intelligence, and the increasing push to make publicly held data more accessible1,11. However, although these datasets may incorporate information from various historical and contemporary sources12,13, challenges related to data quality, interoperability, and completeness remain14. These data often rely on original sources that did not provide information from Indigenous Peoples and other custodians of ancestral sciences–i.e., IPLCAD—or even neglect the original data’s intentionality and context. For example, specific contexts of use and knowledge custodians are frequently not associated with information about the uses of biodiversity12,15. Thus, the paradox of scarcity and abundance in IPLCAD’s data16 arises from the limited availability of information that respects and aligns with community rights and interests, contrasted with a vast amount of data on biodiversity uses rooted in ILK systems that are mislabeled, unacknowledged, or held by non-Indigenous entities–such as information on species uses that fails to credit ancestral sciences or ILK, even when derived from them16.

Over a decade ago, the Nagoya Protocol established international requirements for governing access to genetic resources and associated traditional knowledge, ensuring a fair benefit-sharing system from their use. The Protocol sparked a debate regarding the adequate management of databases related to ILK and sciences17. Central to these arguments is the requirement for the mandatory and necessary involvement of IPLCAD in decision-making processes concerning access and benefit-sharing of genetic resources and associated knowledge (e.g., articles 7, 12, 16 of the Nagoya Protocol). This includes the need to ensure effective mechanisms in the context of synthetic biology and digital sequencing of information on genetic resources (DSI18,19; and the Kunming-Montreal Global Biodiversity Framework [https://www.cbd.int/gbf]]). A significant step forward was recently made by the new World Intellectual Property Organization Treaty [https://www.wipo.int/treaties/en/ip/gratk/], which requires implementing measures to disclose the origin of genetic resources and traditional knowledge in patent applications from signatory countries20. Being able to trace IPLCAD knowledge about uses of species is necessary to implement such a requirement. Traceability of IPLCAD knowledge associated with species is essential for ensuring fair and equitable benefit-sharing17,21. This would address gaps in the access and benefit-sharing system established by the Nagoya Protocol. The protocol does not specify the need to consider holders and custodians of traditional knowledge in research reporting, however, this information is critical for traceability. Traceability is also a requirement for targets 13 (fairly share the benefits from genetic resources, digital sequence information, and traditional knowledge), 21 (make knowledge readily available to guide biodiversity action), and 22 (guarantee participation, justice, and information access for all regarding biodiversity) of the Kunming-Montreal Global Biodiversity Framework [https://www.cbd.int/gbf]. Co-creating databases and other information systems on genetic resources and associated knowledge with knowledge holders and guardians, in line with national laws, would represent major progress, according to the WIPO.

Within the context of open data and open science, IPLCAD’s control and sovereignty over databases is a crucial step toward recognizing their rights and interests in data related to their knowledge, culture, territories, and livelihoods2. Still, institutions–which are still influenced by a colonial legacy–have yet to uphold IPLCAD rights as they apply to digital environments17.

Many local, regional, and national collectives have already decided to develop databases that aggregate species information and IPLCAD´s knowledge. For example, India has broad expertise in documenting traditional knowledge, drawing on efforts to combat biopiracy and initiatives such as the People´s Biodiversity Register22, a participatory tool for reporting and managing India’s biodiversity and traditional knowledge, and maintaining the Traditional Knowledge Digital Library [22, http://nbaindia.org/content/105/30/1/pbr.html]. The “Portal de Conocimientos Tradicionales de los Pueblos Indígenas” is a database on Peruvian ILK on biodiversity, with limited information available to the general public. This initiative addresses the concern of Indigenous Peoples to maintain the confidentiality of their knowledge, aligning with the sui generis Peruvian legislation on access and benefit sharing23. At the regional level, in Brazil, the Useflora [https://useflora.ufsc.br/] database will serve as a regional synthesis of data concerning the use, management, and domestication of Neotropical plants. A core requirement of this database is the explicit acknowledgement of the knowledge holders and guardians. This database is being built through participatory governance, involving both IPLCAD researchers and non-IPLCAD researchers and aims to become an instrument to support the recognition and protection of ILK associated with socio-biodiversity and the IPLCAD custodians. The EthnoFlora DB-French Guiana project, developed by the LEEISA Lab, seeks to create a computer storage tool—whose structure and code will soon be offered in open access—for ethnobotanical data sourced from the literature, concerning IPLCAD uses and vernacular names of plants in this territory. Other potential examples include existing databases that compile ILK from published literature and other sources (e.g., ref. 14, https:mackiki.uqat.ca/) or the compilation of archival (folklore) texts (e.g., https://herba.folklore.ee). However, in such cases, the objective of making ILK openly accessible raises concerns in light of the Nagoya Protocol on Access and Benefit Sharing, as well as ongoing debates about the right of IPLCAD to exercise agency over the ownership, use, and control of their knowledge24. In addition, these and similar databases should be aligned with existing standards and initiatives – such as Indigenous Metadata Bundle25, Dublin Core for web resources26, Plinian Core for aspects of biological species information27, Darwin Core for biodiversity information28, or other TDWG (Biodiversity Information Standards) standards – and interoperable with GBIF as a global data infrastructure.

How might a global data infrastructure that respects IPLCAD data sovereignty be structured and implemented?

We propose a decentralized global data infrastructure with interoperable guidelines, protocols and metadata standards agreed upon through inclusive IPLCAD governance at global, regional, national, and local scales (see Table 1 for definitions of data infrastructure dimensions, or characteristics).

Table 1 Definitions of data infrastructure dimensions
Full size table

To promote the interoperability of existing datasets (and future ones), data aggregators can guide data integration and exchange, ensuring traceability while adhering to local determinations and decisions on data shareability. Legally, it is essential to create a framework that clearly recognizes and protects IPLCAD’s knowledge internationally, ensuring that it is not misappropriated or exploited academically or otherwise29, which is covered by the Nagoya Protocol but still lacks operability. Technically, we need a system that allows for storing and sharing such information in a FAIR - and CARE- compliant manner (see Box 2), ensuring data consistency (e.g., standardization or homogeneity in reporting). Operationalizing both CARE (Collective Benefit, Authority to Control, Responsibility, and Ethics) and FAIR (Findable, Accessible, Interoperable, and Reusable) (see Box 2) principles within existing data infrastructures would ensure that the knowledge of IPLCAD is both respected and utilized effectively, thereby addressing the historical injustices associated with data governance3. The CARE principles intend to balance the technical FAIR data management principles with social and ethical considerations. While these principles share common goals and are intended to be complementary, there are notable incongruences in their application, such as the focus on people versus data: CARE prioritizes the ethical and cultural dimensions16, whereas FAIR emphasizes technical aspects of data management, enhancing machine usability30.

Operationalizing these principles must be aligned with the complementary principles of the Codes of Ethics of institutions such as the ISE (International Society for Ethnobiology, https://www.ethnobiology.net/code-of-ethics/), the Tkarihwaié:ri (Code of Ethical Conduct to Ensure Respect for the Cultural and Intellectual Heritage of Indigenous and Local Communities Relevant to the Conservation and Sustainable Use of Biological Diversity31), and Article 31.1 of the United Nations Declaration on the Rights of Indigenous Peoples32, which states the right of Indigenous Peoples to maintain, control, protect, and develop their cultural heritage, ILK, and other cultural expressions of ancestral sciences. To accomplish this fair operationalization, IPLCAD must be able to contribute to decision-making on all data access and management aspects, including database architecture definition and its long-term governance (throughout the data management lifecycle).

Academically and politically, such a data infrastructure can be a valuable step towards decolonization, by supporting IPLCAD´s rights, contributing to data sovereignty if designed in such a way that user groups are both IPLCAD and non-IPLCAD, and including both academics and civil society. Such a data infrastructure could also serve as a foundation for IPLCAD to develop their own research projects, based on their self-identified needs and ideas, with or without the involvement of non-IPLCAD researchers. For example, in the U.S., the National Ecological Observatory Network, a large-scale ecological monitoring initiative, recognizes Indigenous sovereignty in relation to data, Indigenous governance, and promotes more equitable inclusion of Indigenous Peoples in environmental research17.

Therefore, a global data infrastructure for the knowledge and uses of biodiversity is more likely to be seen as a network of regional, national, and local databases, with each adhering to shared guidelines and common standards. A good model already exists for cross-linguistic and cross-cultural datasets, e.g., the Cross Linguistic Data Format (ref. 33, https://cldf.clld.org/), which allows researchers to link datasets and analyze them together with the same pipelines, while allowing the datasets to stand alone for different purposes (and on different servers, with different access levels). Similarly, a global data infrastructure can have differentiated mechanisms tailored to the local context, such as compliance with national access and benefit sharing legislation and IPLCAD´s rights.

Technically, we suggest a set of data standards including a) minimal taxonomic information (e.g., biological species name); b) vernacular names in the language of the species’ user(s) who originally provided the data; and metadata standards comprising the inclusion of: i) the language name in the local language and identified using the alphanumeric Glottocode (ref. 34, https://glottolog.org/); ii) community or IPLCAD identity, at the individual or group level, considering their preferences and supporting self-determination. For this last point, we recommend using the self-declared name of the group in their own language (e.g., autonym or endonym) and non-pejorative exonyms35. The use of controlled vocabularies in some of these fields can ensure interoperability between different initiatives. Examples of fields with controlled vocabulary include: biological species names, languages, and the identity of community or IPLCAD (holders) if they wish to disclose this information. When developing the controlled vocabulary for identifying holders, it is essential that it operates at the national and/or regional level. This strategy requires careful consideration of specific national and regional rights legislation related to Access and Benefit Sharing and intellectual property, alongside the cultural and biological similarities shared (if any) across countries in the region.

Whenever possible, it is desirable to include metadata information about Free Prior and Informed Consent regarding the shared data and, when applicable, about biocultural community protocols or other mechanisms for community biodiversity registers. Information about uses (purposes of use), may be included only when appropriate and transparently negotiated with IPLCAD´s knowledge holders and guardians, and their leadership, where applicable.

The definition of these minimum standards is essential for future ethnobiological and biodiversity-related field studies; however, when dealing with secondary or tertiary sources of data (e.g., data already published in scientific papers), we recognize that much data on species uses or occurrence lacks information about the language, community, and original knowledge holders or guardians. These data may have been used as sources for existing databases as mentioned above13. When these data are lacking, we suggest including a clear notice that this information was not recorded in the original studies, highlighting gaps that should be corrected in future studies, as suggested by the Local Contexts Labels and Notices system (https://localcontexts.org/labels/traditional-knowledge-labels/). Local, national, and regional databases must establish a protocol for use, reuse, and sharing, along with the necessary permissions. Other information related to this minimal set, such as detailed information about uses (e.g., traits of a given medicinal plant), may be included only when appropriate and transparently negotiated with IPLCAD to define what is relevant. Standards for cultural metadata and provenance of information can be directly embedded into the digital infrastructure of cataloguing, classification, labeling, and content management systems, with metadata fields suitable for such information. One promising tool is the use of Local Contexts Traditional Knowledge (TK) Labelling system, an extra-legal digital mechanism that repositions Indigenous cultural authority and governance over data and collections. TK Labels are digital tags that convey cultural protocols and IPLCAD rights tied to specific data and materials36. They support proper attribution, recognition, and control. TK Labels also correct misrepresentations and clarify the cultural significance and origins of collected materials across diverse types of cultural content.

Indigenous governance experts widely recognize the importance of establishing and maintaining mindful and transparent relationships with IPLCAD knowledge holders and guardians throughout the entire process, starting from their involvement in decisions about data structure and database governance2,3,16,17,21,36,37,38. These efforts should bridge the gap between knowledge production and database governance, ensuring that these groups are represented and actively involved whenever possible. By integrating diverse perspectives and promoting co-governance, such partnerships can strengthen the equitable sharing and sustainable use of biodiversity information. It is important to highlight that, since the inclusion of IPLCAD is an essential condition for building such databases, their participation also expands the beneficiary users of these databases beyond non-IPLCAD academics.

A global data infrastructure that follows these suggested minimal technical and data standards, and which assumes the participation and collaboration of IPLCAD in decision making about the architecture of the databases and in all steps of its governance could be scientifically valuable to: a) identify regions where little research has so far been recorded to support the development of future geographical priority areas39,40; b) identify regions where research efforts have been done in the past which should be revisited to provide proper attribution of knowledge rights (e.g., data previously collected about species’ use with no clear linkage with who were the users, see also8,15,40; c) conduct global analyses enhancing our fundamental understanding of nature-culture interactions (e.g., refs. 10,41); and d) identify regions and species important for both biodiversity and cultural diversity conservation and restoration efforts (e.g., ref. 42).

On a global scale, it is essential to develop a strategic action plan that is clear and operational within the framework of the CBD (Convention on Biological Diversity) and accessible across cultures, with provisions for accurate and meaningful translation into multiple languages. This plan must be co-designed through a deeply collaborative and transdisciplinary process, ensuring meaningful participation from diverse stakeholders43. While its foundation should be grounded in local realities, aligned with national legislations, and, where applicable, guided by local biocultural protocols44, it must also maintain a clear focus on the interconnectedness of global challenges, bridging local actions to broader worldwide goals. However, advancing an inclusive approach implies addressing several obstacles, including language barriers, biases that marginalize underrepresented voices, and insufficient institutional support for meaningful inclusivity44. These barriers also include access to technology, understanding of legislation, organizational level and support for leadership, and political representation of IPLCAD. Limited interdisciplinary training and awareness among researchers from different fields also emerge as an impending limitation.

Potential benefits and inherent challenges

A global data infrastructure can help achieve social-environmental justice, reduce knowledge inequalities, protect ILK and biodiversity from industrial misappropriation and ensure equitable benefit-sharing and Free, Prior and Informed Consent. Such a tool also supports local reappropriation and rematriation/repatriation efforts to prevent the loss of ILK, notably by helping lawyers and NGOs trace the knowledge back to its cultural origin. Furthermore, it can stimulate the integration of academic and ancestral sciences and other forms of knowledge in managing natural resources for the future45.

The debate about a global data infrastructure weaving IPLCAD knowledge of biodiversity also has special considerations for historical data and collections. This multi-layered approach incorporates the protection, reappropriation, and potential rematriation24,46 of IPLCAD knowledge and data to ensure that it does not disappear or become lost due to top-down systems of knowledge extraction. Historically, ILK has often been extracted, commodified, and used by external entities (governments, corporations, researchers) without the proper consent of IPLCAD communities16,24. Reappropriation involves not only regaining access to this knowledge but also the right to manage and govern its use. Databases that allow the recognition of original knowledge holders are a powerful tool to support reappropriation and rematriation, in alignment with the Global Indigenous Data Alliance, which advocates for global awareness about the importance of safeguarding Indigenous data from misuse and ensuring that Indigenous Peoples remain the primary beneficiaries of their own data (ref. 37, GIDA Global Indigenous Data Alliance, https://www.gida-global.org/).

The diverse nature of ILK presents several challenges that need to be further addressed at various levels, including the accurate identification of knowledge custodians and support for community protagonism. A significant challenge in attributing ILK to biodiversity use data lies in accurately identifying its ILK custodians, holders, and guardians. This complexity is inherent due to ILK’s communal and intergenerational nature. Even so, another challenge hindering the development of a global ILK data infrastructure on biodiversity uses is the publishing policy in many journals and ethics review committees that contributes to the invisibility of local collaborators and knowledge holders35, which conflicts with principles of attribution and transparency. Conversely, external recognition of expertise can reshape IPLCAD’s community values, within a new system of meaning altered by Western/Euro-Christian moralities, which can have a marked impact on power, knowledge, and wealth (see ref. 47).

An additional challenge is the absence of robust community-level governance over culturally significant species, which can create vulnerability to appropriation and biopiracy. Effective protection against misuse primarily relies on locally grounded biocultural protocols. Where these protocols are lacking, assessing the ethical and equitable adequacy becomes exceptionally challenging. Consequently, the risk of misuse of ILK about biodiversity uses remains a significant concern. A comprehensive approach to improving data quality and ethical practices must therefore encompass both the improvement of ethical data collection efforts and the retrospective incorporation of missing attribution information into existing published works, ensuring a consistent standard across all biodiversity-related data.

To move beyond these challenges, we highlight the need for a dedicated, multi-level working group to ensure the debate’s continuity, moving from foundational discussions to concrete implementation and interoperability. Securing the necessary financial, technical, and human resources for long-term engagement is an immediate priority for this global data infrastructure. Crucially, this work must emphasize community data governance mechanisms and include support for existing Indigenous data governance initiatives (e.g., the Indigenous Data Governance Task Group of GBIF).

To genuinely reflect a vision of such a data infrastructure of biodiversity knowledge and use, it is crucial to tackle not only the technical database challenges but also to confront and reshape core aspects of the academic system (e.g., publishing norms, ethical norms). For this, we must recognize the importance of moving beyond the perspectives of researchers, who are mostly academic-science oriented or, using the words of Bispo-dos-Santos48, “with a Western/Euro-Christian monotheistic background”. Rather, this data infrastructure should reflect, from its inception, the participation of transdisciplinary partners with diverse perspectives, including members and researchers from IPLCAD communities, who are guardians and custodians of ancestral and Indigenous sciences. Finally, more important than safeguarding ILK of biodiversity uses in a global data infrastructure—which can help mitigate language and knowledge loss for future generations—is guaranteeing the necessary conditions for these ancestral sciences to be maintained in their original territories.