Lists of sessions by track

Last updated 10 April 2026

This is the list of proposed sessions for TDWG 2026 Conference organized by tracks to facilitate searching the most suitable session if you are submitting an abstract. Alternatively you can browse the list of sessions organized by session type.

Everybody is welcome to attend any session! Restrictions apply only to abstract submission.

AI and Robot-ready

SYM13 Responsible AI, Open Digital Curation, and Round-Tripping for Biodiversity Data

Session type

Symposium (Open)

Organizers

Deborah Paul, Species File Group, Prairie Research Institute, University of Illinois, United States; Siobhan Leachman, Landcare Research group of the New Zealand Bioeconomy Science Institute, New Zealand; Sabine von Mering, Museum für Naturkunde, Germany

We propose a session focused on open digital curation as a continuous, inclusive process across the biodiversity data lifecycle. Presentations would illustrate how human and AI assistance can be made transparent, attributable, versioned, and reusable during data mobilization and enrichment (e.g. annotations and linking, taxonomic determinations, georeferencing improvements, use of persistent unique identifiers).

Particular attention would be paid to round-tripping: ensuring that annotations and AI-generated and human enhancements flow back to source systems (e.g., collection management systems and institutional repositories), rather than remaining disconnected products. By emphasizing openness, provenance, and interoperability, the session would explore practical approaches for integrating human work and AI responsibly and sustainably while maintaining curatorial authority and trust.

With an emphasis on natural history data from museums and herbaria – including specimens, images, and genomic resources – the session presentations would feature case studies and methods that bridge institutional practice, community engagement, and open science infrastructure. Contributions would address challenges such as scaling AI-enabled curation, validating and crediting annotations, aligning with standards like Darwin Core, and designing workflows and user interfaces (i.e. UI/UX) that support equitable participation and vetting by collection staff, researchers, and external contributors.

SYM25 From Mobilizing Data to AI-Ready Knowledge: Infrastructure for Multimodal Biodiversity Data

Session type

Symposium (Open)

Organizers

Hilmar Lapp, Neuromatch, Inc., United States; Rob Guralnick, University of Florida, United States; Kelsey Huelsman, NASA, United States; Eric Sokol, National Ecological Observatory Network, United States; David Bloom, University of North Carolina at Greensboro, United States

More ecological and biodiversity data are online than ever before, in part thanks to the enormous decades-long efforts by the biodiversity community in mobilizing observational data, supporting shared standards, and building aggregation networks. Publicly accessible data are also more varied than ever before, including images of many modalities, videos, and streaming sensor data. However, using this sea of data for increasingly ML-heavy and AI-driven research pipelines remains fraught. Significant ecological context, locality detail, trait information, and interaction data are buried in unstructured media. Records are often incomplete, unevenly georeferenced, and documented inconsistently across platforms. Metadata regarding terms of use, provider expectations, and desired round-tripping of data amendments are unclear or missing. Without downloading massive amounts, assessing the quality and fitness-for-use of data is often difficult or impossible. As AI-enabled and agent-driven methods mature, developing shared, community-aligned infrastructure and practices that transform multimodal biodiversity data into interoperable, provenance-aware, and research-ready knowledge becomes a central challenge. Contributions to this session will explore how standards, human–AI collaboration, uncertainty representation, ontology integration, and data governance frameworks can enable AI-orchestrated pipelines to understand fitness-for-use, respect provider expectations, and track provenance.

SYM30 AI for Biodiversity Data

Session type

Symposium (Open)

Organizers

David Williamson, Norwegian University of Science and Technology (NTNU), Norway; Wouter Koch, Artsdatabanken, Norway; Arianna Salili-James, Natural History Museum London, UK; Sanson Poon, Natural History Museum London, UK; Rukaya Johaadien, University of Oslo, Norway; Michal Torma, University of Oslo, Norway; Maarten Trekels, Meise Botanic Gardens, Belgium; Kit Lewers, University of Colorado Boulder, USA

Biodiversity researchers today are offered access to vast troves of data: billions of images and data points in online repositories, citizen scientists contributing millions of new observations every year, countless specimens in museum collections around the world, and an endless supply of studies, reports, and expert knowledge in articles and books stored in libraries or published online.

But data alone is not enough. To have any meaning it must be collated, organised, standardised, evaluated and interpreted. With such quantities of information, machine learning and artificial intelligence tools are vital to our ability to parse and understand the data available to us, and essential if we are to scale our research to meet the urgent biodiversity challenges the world faces today.

This session invites talks on cutting-edge applications of AI technology in automating the collection, preparation and interpretation of biodiversity data. We welcome contributions on topics such as novel applications of AI in data collection and mobilisation, large-scale studies enabled by AI, automation to extract data and new insights from historical collections and archives, and systems that improve our ability to arrange and interpret large biodiversity datasets.

We strongly encourage speakers to include the financial and computational costs of model training and operation, in addition to performance metrics, in discussing their work to allow assessment of its feasibility in wider use.

SYM46 AI-Readiness Metrics and Metadata for Biodiversity Data

Session type

Symposium (Open)

Organizers

Yasin Bakış, Tulane University, United States; Cem Meydan, Cornell University, United States

Artificial intelligence is transforming biodiversity research, yet many datasets remain hard for automated systems to interpret, validate, and integrate. “Robot-ready” (AI-ready) data requires more than FAIR: structured, machine-actionable metadata, clear semantics, provenance, quality metrics, and reproducible workflows.

This session will examine emerging approaches for defining, measuring, and operationalizing AI-readiness in biodiversity data infrastructures. We invite contributions addressing: (1) quantitative AI-readiness metrics and validation frameworks; (2) extensions or refinements to Darwin Core and related TDWG standards to support machine reasoning; (3) automated metadata generation and enrichment; (4) integration of image-, trait-, or sensor-derived data into standards-compliant records; (5) provenance capture and workflow reproducibility; (6) semantic alignment and ontology-driven interoperability; and (7) benchmarking and evaluation of datasets for AI applications.

We especially welcome case studies from repositories, aggregators, and infrastructures implementing automated quality checks, semantic annotation, or containerized workflows, including cross-domain links to climate or geoscience data.

The goal of this session is to identify concrete standards gaps, promote interoperable solutions, and advance a shared community framework for AI-ready biodiversity data aligned with TDWG principles.

SYM47 Large Language Models for Biodiversity Data Discovery, Integration, and Curation

Session type

Symposium (Open)

Organizers

Yasin Bakış, Tulane University, United States; Bahadır Altıntaş, Abant Izzet Baysal University, Turkey; Cem Meydan, Cornell University, United States

Large language models (LLMs) and other generative AI systems are changing how researchers interact with data and digital infrastructure. In biodiversity informatics, these technologies create new opportunities for improving data discovery, automating metadata generation, assisting with data curation, and enabling natural language interfaces for biodiversity databases.

This session will explore emerging applications of large language models in biodiversity data infrastructure and research workflows. Topics may include natural language querying of biodiversity repositories, extraction of biodiversity information from literature and specimen labels, AI-assisted metadata generation and validation, and integration of LLM-based tools with biodiversity data standards such as Darwin Core.

We encourage case studies demonstrating how LLM-based tools can support biodiversity data discovery, improve metadata completeness, and assist with data publishing and curation workflows. Contributions addressing challenges such as reliability, provenance, bias, and integration of generative AI with structured biodiversity data are also welcome.

The goal of the session is to examine how large language models may reshape biodiversity data access and curation while identifying opportunities and limitations for their responsible use.

LT09 Advancing Environmental Safeguard Practices through Robot-Ready Data Standards

Session type

Session of Lightning Talks (Open)

Organizers

Abdulrahim Abubakar, Ministry of Environment, Kwara State Nigeria, Nigeria; Mohammed Abubakar, Federal University, Lafia, Nasarawa State, Nigeria

The integration of environmental safeguards within the biodiversity conservation framework is critical in today’s rapidly changing ecosystems. This organised session aims to explore how the implementation of robot-ready data standards can enhance environmental management and social impact assessments, particularly in projects funded by international organisations like the World Bank. Drawing on over 10 years of experience in environmental management, social impact assessment, and policy implementation, we will discuss the role of automated data collection and machine learning in improving the efficiency and compliance of environmental safeguards. Participants will delve into the latest advancements in data standards that facilitate the machine-readability of environmental data, making it accessible for researchers, policymakers, and conservation practitioners. The session will feature lightning talks from leading experts, including case studies that highlight best practices in using these standards to ensure environmental compliance across various projects. Through interactive discussions and workshops, we will address challenges faced in standardising environmental data and foster innovative solutions to promote effective data collection, sharing, and analysis. Our goal is to facilitate collaboration within the environmental community and push forward innovative practices that support sustainable management of biodiversity.

LT17 Bots, Bits, and Biodiversity

Session type

Session of Lightning Talks (Open)

Organizers

Jack Hollister, Natural History Museum, London, United Kingdom; Sanson Poon, Natural History Museum, London, United Kingdom; Arianna Salili-James, Natural History Museum, London, United Kingdom; Qianqian Gu, Natural History Museum, London, United Kingdom

This session will consist of a series of short, rapid presentations on robotic and automated approaches to biodiversity data acquisition. It will focus on the design and use of physical and hybrid systems that generate structured, machine-actionable datasets and their alignment with TDWG Biodiversity Information Standards.

The session explicitly welcomes contributions across the full development spectrum, including early-stage prototypes, experimental platforms, field-deployable systems and production infrastructures.

Relevant topics include, but are not limited to:

• robotic specimen handling and imaging

• automated sensing platforms in the laboratory or field

• high-throughput digitisation systems

• workflows for generating training and validation data for machine learning

• automated or semi-automated annotation and quality control

• feedback loops between data acquisition and model performance

• methods for structuring outputs in compliance with TDWG standards

By using a rapid-fire format, the session aims to maximise the number and diversity of case studies, enabling comparison of different technical approaches and maturity levels. The discussion will synthesise common challenges, identify opportunities for standardisation, and explore how emerging automated workflows can contribute to scalable and sustainable biodiversity data infrastructures.

DS03 From Field to AI: Making Community Biodiversity Data Research- and Robot-Ready

Session type

Discussion session (Open)

Organizers

Idriss Adoum Idriss, Reseau des Jeunes pour le Climat au Tchad, Chad; Pelagie MOLBIGA, Youth for Adaptation Finance for Africa, Burkina Faso; Fatimah Zannah Moustapha, Pan African Climate Justice Alliance, Nigeria

Biodiversity data standards are increasingly designed to support automation, artificial intelligence, and large-scale data integration. Yet a significant share of biodiversity data—especially from the Global South—originates from community-led monitoring, restoration projects, and local ecological knowledge systems that are rarely research- or robot-ready by design.

This symposium explores how community-generated biodiversity data can be transformed into research- and robot-ready datasets using TDWG standards, FAIR data principles, and interoperable workflows. Drawing on concrete case studies from Sahelian biodiversity and restoration initiatives, including HINNA and GreenSahel TreeTech, the session highlights practical challenges such as data heterogeneity, limited infrastructure, multilingual metadata, and gaps between local knowledge and global standards.

Speakers will share approaches for structuring field-level observations with TDWG standards, integrating low-tech and community-collected data into AI-ready pipelines, and improving data quality and interoperability without excluding local actors. Particular attention will be given to ethical and inclusive data governance, ensuring that robot-ready biodiversity data strengthens both global research and local decision-making.

By connecting biodiversity informatics with grassroots action, this session demonstrates how data standards can empower communities while advancing scalable, machine-actionable biodiversity knowledge.

DS31 Panel Discussion: Critically Assessing the Role of AI in Biodiversity Research

Session type

Discussion session (Closed)

Organizers

David Williamson, Norwegian University of Science and Technology (NTNU), Norway; Wouter Koch, Artsdatabanken, Norway; Rukaya Johaadien, University of Oslo, Norway; Michal Torma, University of Oslo, Norway; Arianna Salili-James, Natural History Museum London, UK; Sanson Poon, Natural History Museum London, UK; Maarten Trekels, Meise Botanic Gardens, Belgium; Kit Lewers, University of Colorado Boulder, USA

AI promises much to scientists working with biodiversity data. There has been a recent surge in projects based on LLMs and related technologies: systems that go beyond species identification from images and also identify species traits, agents answering complex ecological questions in natural language based on real data, machine transcription that not only reads handwritten labels but understands them. Such projects, and many more, hope to harness AI in answering some of the most pressing questions in biodiversity research, multiplying the efforts of scientists through more efficient research, larger studies, and novel insights into data.

But while much is promised, critics might say that little has been delivered. Studies focus on the exciting potential of new models, but often lack performance metrics to prove their abilities. Are the data centres in which our models run destroying the same ecosystems we study? Does public backlash to AI risk losing the trust of citizen scientists whose data is used in training models without attribution or consent? Is AI amplifying our abilities, or are we outsourcing our thinking and neglecting hard-won expert knowledge? How do we reconcile a reliance on closed models and secret training data with our obligations to FAIR and open science?

This discussion session invites a panel of experts to discuss these topics with our audience, with the aim of co-creating guidelines for effective and ethical use of AI in biodiversity research.

DS49 What Makes Biodiversity Data “Robot-Ready”? Standards, Infrastructure, and AI

Session type

Discussion session (Open)

Organizers

Henry Bart, Tulane University, United States; Yasin Bakis, Tulane University, United States

This discussion session will examine what technical and infrastructural conditions are

required for biodiversity data to become “robot-ready” usable by automated systems

and AI pipelines without manual intervention.

Despite the scale of global biodiversity repositories, many datasets remain difficult for

automated systems to interpret because key metadata fields are incomplete, data

quality indicators are rarely available, and existing standards lack terminology designed

for AI workflows. As a result, researchers and data scientists often spend significant

effort preparing data before it can be used in computational pipelines.

The session will begin with short invited perspectives from biodiversity data

infrastructure providers, standards developers, and AI researchers. These perspectives

will highlight gaps in metadata completeness, missing data quality metrics, and the lack

of machine-actionable descriptors needed for automated analysis.

The discussion will then focus on identifying a community framework for robot-ready

biodiversity data, including: 1) essential metadata fields, 2) standardized data quality

metrics, 3) terminology for AI and machine-generated annotations, and 4) priorities for

extending existing standards.

The goal of the session is to identify community priorities and outline a roadmap toward

robot-ready biodiversity data standards. Outcomes will inform a collaborative

community paper and future standards work within the TDWG community.

WM48 IQAIR Task Group Working Meeting: Image Quality and AI-Readiness Metadata for Biodiversity Media

Session type

Working meeting (Open)

Organizers

Yasin Bakış, Tulane University, United States; Henry Bart, Tulane University, United States

The TDWG Image Quality and AI-Readiness (IQAIR) Task Group focuses on developing metadata concepts and vocabulary needed to describe biodiversity media in ways that support automated processing and AI-based analysis. As digitization efforts continue to produce large volumes of specimen images and other multimedia resources, standardized descriptors for image quality, content, and machine usability are increasingly important for enabling computational workflows.

This working meeting will bring together members of the IQAIR Task Group and interested participants to advance ongoing work on defining image quality and AI-readiness metadata for biodiversity media. The meeting will review progress on candidate metadata terms, examine alignment with existing TDWG standards such as Audiovisual Core and Darwin Core, and discuss use cases involving automated trait extraction, image analysis, and machine-generated annotations.

Participants will discuss practical implementation issues, including how repositories and aggregators can incorporate image quality descriptors, automated metadata extraction, and machine-actionable annotations into existing data infrastructures.

The goal of the meeting is to refine vocabulary proposals, clarify use cases, and identify next steps for developing community recommendations that support AI-ready biodiversity media within the TDWG standards framework.

Biodiversity Literature & Knowledge Extraction

SYM08 BHL at 20: A New Chapter for Biodiversity Literature and Data

Session type

Symposium (Open)

Organizers

Nicole Kearney, Biodiversity Heritage Library, International; David Iggulden, Royal Botanic Gardens, Kew, United Kingdom

In 2026, the Biodiversity Heritage Library (BHL) celebrates both its 20th anniversary and its first year as an independent consortium. Following its transition away from long-term hosting at the Smithsonian Institution, BHL is now operating from a stable foundation, with its people, expertise, and services firmly in place and renewed momentum across its global network.

This session will explore where BHL stands today and what this new phase makes possible. Presentations will examine the outcomes of the transition, including updated governance structures, evolving funding approaches, strengthened partnerships, and technical developments that position BHL for sustained growth and innovation within the biodiversity data ecosystem.

Speakers will present emerging applications of BHL content and services, such as artificial intelligence and machine learning, large-scale text and image mining, knowledge graphs, annotation systems, and efforts to improve persistent identifier reliability and stewardship. Talks will demonstrate how BHL functions as essential infrastructure connecting literature with taxa, specimens, institutions, and datasets, from the 1100s to the current year.

The session will also reflect on two decades of BHL’s impact, from mobilizing historical biodiversity knowledge to supporting modern data integration and reuse, inviting contributors and attendees to consider how BHL can continue to evolve as a core component of global biodiversity information infrastructure.

SYM45 Planning the Libroscope: creating research ready biodiversity from scientific publications

Session type

Symposium (Closed)

Organizers

Donat Agosti, Plazi, Switzerland; Laurence Bénichou, Muséum National d’Histoire Naturelle, France

Scientific publications systematically present original or synthesised research and incorporate it into the formal scholarly record through specialist journals and books. They use standardised vocabularies, cite prior work, and—particularly in biodiversity science—reference studied specimens and DNA sequences, thereby linking the literature to natural science collections (NSC) and DNA sequence databases. However, although they contain critical factual data, publications are written for human consumption. They are rarely semantically enhanced with machine-actionable links that would make them Artificial Intelligence–ready.

The Disentis Roadmap is a decadal initiative to liberate global biodiversity knowledge from scientific literature, endorsed by 110 scientists and organizations. Its primary goal is to deliver open machine-actionable data by building the Biodiversity Libroscope instrument. Envisioned as a shared framework, the instrument aims to systematically discover, structure, publish, liberate, and connect biodiversity data from published sources in FAIR-compliant formats. It builds on existing projects, infrastructures, and use cases to define the standards needed for global collaboration, interoperability, and scalability.

This symposium will present exemplary contributions demonstrating key elements of the Libroscope conversion workflow, followed by a critical discussion of the approach, adopted standards, and strategies for meaningful community engagement.

WKS14 New approaches and tools to maximise the value of the Biodiversity Heritage Library

Session type

Workshop (Closed)

Organizers

Roderic Page, University of Glasgow, United Kingdom; Nicole Kearney, Melbourne Museums, Australia

This workshop aims to be a first step in rethinking how we can interact with the Biodiversity Heritage Library (BHL), and how we extract information to build interfaces and datasets appropriate for a user landscape dominated by mobile devices and new AI tools. It will be run as a series of live demos of ideas for interfaces and information extraction tools. Participants will be invited to explore the demos and provide feedback on them. We will also encourage participants to share their views on what BHL could do to increase its value to their research.

Topics would include:

• Taxonomic namestree-based taxonomic browsing using classifications such as the Catalogue of LifeGoogle n-gram like tools tracking changes in taxonomic names over time

• Geographymap-based interfaces to explore BHL content geographicallyfinding images of maps in BHL content

• Imagesimage extraction and classification tools using LLMsimage search using embeddings (find images that look like this)

• Entitiesextracting entities from text, such as taxonomic names, people, places, and museum specimens

• AI-friendly interfacesmodel content protocol (MCP) interfaces for AI tools to query BHL using natural language

We think that the chance for people to explore new tools, and discuss those with active members of BHL will be beneficial for BHL as it restructures and rethinks its platform, and to the wider biodiversity data community for whom BHL is a valuable, if underutilised, resource.

WKS33 Liberating and providing AI-ready data from the Natural History Literature

Session type

Workshop (Closed)

Organizers

Chris Le Coquet, Muséum national d’Histoire naturelle, France; Laurence Bénichou, Muséum national d’Histoire naturelle, France; Donat Agosti, Plazi, Switzerland; Patrick Ruch, Swiss Institute of Bioinformatics, Switzerland; Rainer Krug, Swiss Institute of Bioinformatics, Switzerland

Ever since the first attempts to observe, document and describe the natural world—followed by the establishment of the discipline as a scientific field—literature has served as the primary channel for the validation, dissemination and archiving of natural history knowledge, resulting in an ever-growing corpus of 500 million pages. However, this huge amount of information is often inaccessible—lost in the corner of a library or locked in a PDF—and forms what we commonly refer to as “unknown known knowledge”.

To unlock and transform it into FAIR, AI-ready, interlinked objects, the community has produced open source tools for the extraction and annotation of data from scanned or digital-born PDFs based on established standards and persistent identifiers. Meanwhile, the community has also produced XML tools (e.g., JATS, TEI) to address a similar challenge in prospective literature. Over the past decade, this unlocked knowledge has greatly contributed to a rich ecosystem of increasingly interconnected data infrastructures, creating new opportunities for research and bioinformatics communities.

This workshop will complement the proposed Libroscope symposium. As a starter, the participants will be introduced to the stakes and methods of literature FAIR-isation with a strong focus on data cleaning and linking. Then, the main course will have the participants explore new opportunities offered by the Libroscope using human interfaces as well as APIs to retrieve liberated AI-ready data.

WKS34 Liberating Data from Historical Biodiversity Literature with The Unified Corpus Explorer (UCE)

Session type

Workshop (Closed)

Organizers

Marius Böyng, University Library J. C. Senckenberg, Goethe-University, Germany; Gerwin Kasperek, University Library J. C. Senckenberg, Goethe-University, Germany; Katrin Peikert, University Library J. C. Senckenberg, Goethe-University, Germany; Mevlüt Bagci, Text Technology Lab, Goethe-University, Germany; Jana Hoffmann, Senckenberg Nature Research, Germany

Biodiversity research relies on data that are deeply structured, semantically explicit and interoperable across systems. This workshop introduces the Unified Corpus Explorer (UCE) as a tool for transforming biodiversity literature into machine-actionable research data.

Hosted by the Specialised Information Service Biodiversity Research (BIOfid), the workshop demonstrates how data from historical and contemporary texts, with a strong focus on publications from the 19th and 20th centuries, can be systematically mobilised through digitisation, NLP- and LLM-based annotation, and semantic enrichment using domain-specific ontologies. These sources contain essential information on species distributions, ecosystems and spatio-temporal dynamics that are largely absent from modern databases, yet are critical for establishing robust baselines for research on biodiversity change.

Participants will explore how UCE enables interactive access to richly annotated text corpora, supports standardized representations of extracted entities and relations and facilitates their integration into interoperable data workflows. The workshop highlights how text-based biodiversity information can be aligned with community standards and made accessible not only for human interpretation, but also to improve interoperability and reusability.

The workshop is aimed at researchers, interested in unlocking biodiversity knowledge from literature and embedding it into scalable, standards-oriented data ecosystems.

WKS43 How Literature Services can Support & Benefit from Biodiversity Publication and Data Standards?

Session type

Workshop (Closed)

Organizers

Emilie Pasche, Switzerland

Rapid growth of biodiversity literature makes it difficult to retrieve precise, relevant, and interoperable knowledge. Traditional bibliographic catalogues mainly rely on metadata and abstracts, limiting fine-grained discovery and large-scale reuse.

This hands-on workshop explores how literature services can support biodiversity publication and data standards. Using BiodiversityPMC, developed within the SIB Literature Services (SIBiLS), we demonstrate how structured, machine-readable publications enable advanced indexing, semantic enrichment and concept-based retrieval across heterogeneous sources.

The SIBiLS platform integrates taxonomic treatments from Plazi, biodiversity-related documents from Zenodo and full-text articles from biodiversity journals (e.g., EJT, Pensoft). By leveraging structured formats, persistent identifiers and domain ontologies, it transforms publications into AI-ready contents that can be cross-linked and queried at scale. We will also demonstrate how pre-computed literature selection through Triage supports Biodiversa+ communities.

Through guided exercises, attendees will perform advanced searches via the web interface, explore semantic annotations and test selected API calls for programmatic access, including MCP calls from AI bots.

The session also fosters discussion on feedback loops between publishers and literature platforms, showing how literature services can strengthen biodiversity data standards toward interoperable ecosystems.

Collections Digitisation & Management

SYM20 Meeting Biodiversity Data Standards Using Specify Software

Session type

Symposium (Open)

Organizers

Theresa Miller, Specify Collections Consortium, United States

As the biodiversity community moves toward pursuing FAIR principles and the concept of the Extended Specimen, the need to produce high quality, biodiversity standard compliant data has never been greater.  However, implementing these standards can present a challenge for collections managers and institutions.

This session aims to showcase the role Specify software has played in empowering collections staff to digitize, manage, and export their collections’ data according to Biodiversity Data Standards.  The session will feature presentations from Specify users at institutions across the globe who will highlight data management workflows using Specify tools and techniques.

In addition to user presentations, the Specify Collections Consortium (SCC) will also deliver an overview of Specify’s core and newer features that allow users to meet evolving Biodiversity Data Standards and ensure data quality, standardization, and interoperability.

SYM29 From Cabinets to Clouds – Bridging the gap between collection- and database management

Session type

Symposium (Open)

Organizers

Caitlin Thorn, Museum für Naturkunde Berlin, Germany; Nora Lentge-Maaß, Museum für Naturkunde Berlin, Germany; Frederik Berger, Museum für Naturkunde Berlin, Germany; Mareike Petersen, Museum für Naturkunde Berlin, Germany; Sabine von Mering, Museum für Naturkunde Berlin, Germany

Physical collections and digital databases are often managed in parallel by collection managers and data managers. As dedicated development teams become more common in institutions, opportunities for misalignment between those working in the collection cabinets and those in the data clouds arise.

Across institutions, this relationship is handled in different ways. A wide range of approaches for coordinating collection digitisation pipelines exist, reflecting differences in institutional structures and ongoing integration of new development roles into long-established collection management practices and workflows. In some cases, workflows are closely aligned across teams, while others are more loosely defined. These differences offer an opportunity to learn how institutions balance collection management expertise with emerging digital practices.

We aim to bring together a range of examples to learn from each other’s institutional workflows to implement biodiversity data standards and contribute to global biodiversity knowledge. Topics could include experiences and best practices in facilitating functional relationships between collection and data managers; decision-making processes for data standards; having one expert in two domains vs. two experts; enriching the collection management system vs. linking data; and many more. This session aims to identify common issues and explore best practices for enabling a cohesive workflow from the cabinets to the cloud.

SYM42 Solutions for Research Collection Management Systems Challenges

Session type

Symposium (Open)

Organizers

Jonas Grieb, Senckenberg - Leibniz Institution for Biodiversity and Earth System Research, Germany; James Macklin, Agriculture and Agri-Food Canada, Canada; Christian Bölling, Museum für Naturkunde Berlin, Germany; Sam Leeflang, Naturalis, Netherlands; Wouter Addink, Naturalis, Netherlands; Michaela Grein, Überseemuseum Bremen, Germany; Volker Lohrmann, Überseemuseum Bremen, Germany; Etta Grotrian, Überseemuseum Bremen, Germany; Claus Weiland, Senckenberg - Leibniz Institution for Biodiversity and Earth System Research, Germany

Collection Management Systems (CMS) face evolving challenges in managing, curating, publishing, and integrating increasingly complex research data from diverse sources. This session seeks contributions that address practical and conceptual solutions for modern, research-oriented CMS, with a focus on enabling richer, interoperable, and sustainable data flows across the global biodiversity data landscape. We welcome talks on data mobilization for research infrastructures, e.g., GBIF, OBIS, and DiSSCo. We especially seek submissions focused on emerging standards such as Darwin Core Data Package (DwC-DP), and alignment with TDWG and related domain standards to support semantic consistency and FAIR principles. Presentations exploring implications of the Darwin Core Conceptual Model - including clearer treatment of relationships among material entities, organisms, and occurrences - are encouraged. We also invite contributions on the maintenance and use of persistent identifiers and the relationship between material samples and primary literature in the context of material sample management systems and their data consumers. Further topics include machine-assisted curation workflows, community-driven development, capacity building, long-term maintenance, CMS-as-a-Service, and governance models. By bringing together CMS developers, data architects, curators, and aggregators, the session aims to advance practical solutions to current and emerging biodiversity data exchange challenges.

WKS35 From Skeletal to Scientific: A Hands-on Workshop for Measuring Digitization Depth with the MIDS Standard

Session type

Workshop (Closed)

Organizers

Mathias Dillen, Meise Botanic Garden, Belgium; Elspeth Haston, Royal Botanic Garden Edinburgh, United Kingdom; Caitlin Chapman, iDigBio / Florida Museum of Natural History, United States

The digitization of natural history specimens renders their associated data more accessible, both to human and robot users. As collections accelerate their digital transition, the need for objective metrics to measure the “completeness” of a specimen’s digitization has become paramount. The Minimum Information about a Digital Specimen (MIDS, https://github.com/tdwg/mids) standard provides a framework of four levels to categorize the depth of digitization, helping institutions prioritize workflows, track progress and report to funders and researchers.

This workshop provides a practical, hands-on introduction to the MIDS standard. Participants will learn how their own collection data (Darwin Core or ABCD formats) are mapped to MIDS information elements and use the MIDSCalculator tool to generate scores and visualizations. By the end of the session, attendees will understand how to use MIDS as a diagnostic tool for the digitization level of collections, including why certain levels were achieved.

Participants should bring a laptop and should install a local version of the MIDSCalculator tool in advance. Instructions can be found on https://github.com/AgentschapPlantentuinMeise/MIDSCalculator and any issues encountered can be flagged in that repository. Participants are also encouraged to acquire a sample Darwin Core Archive (DwC-A) or ABCD file for specimen data (biology, geology, paleontology), either from their own institution or from a third party.

Community Science, Regional Networks & GBIF Connectivity

SYM27 The ASEAN Biodiversity Dashboard as a GBIF-Hosted Portal for Monitoring Species Occurrence and Biodiversity Trends

Session type

Symposium (Open)

Organizers

Christian Elloran, ASEAN Centre for Biodiversity, Philippines

Effective management of ASEAN Heritage Parks (AHPs) and other protected areas is essential for conserving Southeast Asia’s biodiversity and sustaining vital ecosystem services. To support this goal, the ASEAN Biodiversity Dashboard was established as a regional, open-access platform to monitor biodiversity trends and species occurrences across ASEAN Region.

Developed within a GBIF-hosted portal framework and built on the infrastructure of the Global Biodiversity Information Facility (GBIF), the Dashboard enables standardized data publication, integration, and visualization using interoperable biodiversity data standards. By consolidating datasets from ASEAN Member States and global partners, it strengthens evidence-based policymaking, spatial analysis, and digital decision-support tools for conservation planning.

The platform allows users to assess species distributions, identify data gaps, and support reporting under global biodiversity commitments, including the Kunming-Montreal Global Biodiversity Framework. By linking regional data to the global knowledge network, the ASEAN Biodiversity Dashboard enhances transparency, collaboration, and coordinated action for biodiversity conservation across Southeast Asia.

SYM28 From Standards to Implementation: Connecting Observation Data in Asia to GBIF Infrastructure

Session type

Symposium (Open)

Organizers

Kumiko Totsu, National Institute for Environmental Studies, Japan

Across Asia, biodiversity observation networks generate large volumes of primary data through research and long-term monitoring. While international biodiversity data standards are well established, connecting locally managed observation data to global infrastructures such as GBIF can involve practical challenges.

This session will provide a forum to share experiences and perspectives on linking observation networks in Asia to GBIF-compatible data workflows. The discussion will consider how data transformation processes, alignment with interoperable biodiversity data standards (e.g., Darwin Core), and publication mechanisms can be incorporated into routine research and monitoring in feasible and sustainable ways. It will also examine practical gaps between established standards and their real-world implementation in diverse institutional contexts.

We welcome contributions from researchers, data managers, infrastructure developers, regional networks, and policy stakeholders to exchange experiences and identify areas for improved coordination.

We aim to gather insights on current approaches, common obstacles, and emerging tools that support more streamlined and automated data mobilization. By fostering dialogue across roles and regions, the session seeks to support better alignment between data production, data publication, and evidence-based decision-making practices. It will also explore pathways toward more scalable and interoperable biodiversity data infrastructures.

LT16 Community-Powered, Research-Ready: Citizen Science for a Digital Future

Session type

Session of Lightning Talks (Open)

Organizers

Ram Dayal Vaishnav, The Naturalist School, India; Vijay Barve, GBIF, India

Citizen science has evolved from a niche data-gathering tool into a powerhouse of global biodiversity monitoring. However, as the research community moves toward “Robot-ready” and machine-actionable data, a critical gap remains: how do we maintain the authentic, local, and human-centric nature of community-based work while meeting the rigorous requirements of modern informatics?

This session invites presenters to share their experiences, successes, and challenges in managing Citizen Science programs across all scales. We seek to explore the “bridge” between the volunteer in the field and the digital infrastructure that hosts their data. Topics for discussion include:

• Program Design: How local initiatives are adapting to produce higher-quality, interoperable data.

• The Volunteer Experience: Maintaining engagement and agency as data becomes more automated.

• Bridging the Gap: Case studies on translating “unstructured” community observations into valuable research assets.

• Inclusivity & Technology: Ensuring that the drive toward standardized, “Robot-ready” data does not exclude under-resourced communities or traditional knowledge holders.

By sharing diverse perspectives—from software developers to community organizers—this session aims to foster a dialogue on how we can empower humans and “robots” to work together for the future of biodiversity conservation.

DS40 National and policy-relevant species checklists: why official, updated and validated data matter

Session type

Discussion session (Closed)

Organizers

Diana Hernandez, Catalogue of Life, Mexico; Camila Plata, Catalogue of Life, Colombia; Olaf Banki, Catalogue of Life, The Netherlands

National and policy-relevant species checklists, like protected or alien species lists, and national inventories, are essential inputs for biodiversity policy and decision making, serving as regulatory tools and foundations for governmental strategies. National checklists are typically scientifically driven instruments supporting natural resources management, while policy-relevant such as IUCN, CITES, or GRISS build on scientific evidence and incorporate additional economic, regulatory, or societal considerations.

Variation in origin, scope, structure, licensing, and quality has limited their interoperability and use in global biodiversity infrastructures, despite ongoing progress toward improved standardization.

Through ChecklistBank, the Catalogue of Life (COL) already hosts over 1,000 national and 400 policy-relevant lists, positioning it as a key platform for improving access, visibility, and reuse.

This session will combine short presentations with an open, moderated discussion to explore the needs around generating and using these lists, and how COL can support processes from production through publication.  Discussion topics include translating heterogeneous lists into shared data standards, improving metadata and validation for policy use, and documenting taxonomic differences transparently. By bringing together taxonomists, data managers, and policymakers, we seek pathways to build trust, improve integration, and translate taxonomic work into sustained policy impact.

WKS24 Mobilizing Botanical Checklists and Red Lists to GBIF

Session type

Workshop (Closed)

Organizers

Lycka Kamoen, Leiden University

Naturalis Biodiversity Center, Netherlands; Niels Raes, Naturalis Biodiversity Center, Netherlands

The MetaPlantCode (MPC) project develops harmonized best practices for collecting, analyzing, and publishing plant metabarcoding data, with a strong focus on making these data findable, accessible, interoperable, and reusable through GBIF.

A key challenge in plant metabarcoding is that many species cannot be distinguished solely by their DNA barcode sequences. Closely related taxa often share (near‑)identical barcodes, making species‑level assignment difficult. To overcome this limitation, researchers must integrate contextual information such as the geographic distributions of potential species. These data are typically found in national checklists, floras, and red lists, but many of these resources are not yet accessible or interoperable.

To address this gap, the MPC project has developed a practical manual that guides users through the complete workflow of preparing, structuring, and publishing botanical checklists and red lists via the GBIF Integrated Publishing Toolkit (IPT). The guidelines build on GBIF best practices and provide step‑by‑step instructions for data preparation, Darwin Core mapping, extension use, metadata completion, and quality control. Although developed in the context of plant metabarcoding, the manual is widely relevant for national GBIF node managers, biodiversity information specialists and conservation organizations. This workshop will demonstrate key steps in the workflow and support participants in applying the guidelines to their own resources.

WKS50 Living Atlases around ALA: From Deployment to Contribution and Future Technical Directions

Session type

Workshop (Closed)

Organizers

Javier Molina, Atlas of Living Australia / CSIRO, Australia; Francisco Pando, Real Jardín Botánico-CSIC, Spain; Vicente José Ruiz Jurado, GBIF.ES - Consejo Superior De Investigaciones Científicas (CSIC), Spain

The Living Atlases community, coordinated by the Atlas of Living Australia, is a global, open-source network supporting over 20 biodiversity data portals across multiple continents. Since 2013, it has enabled widespread adoption through shared infrastructure, standards, and collaboration among GBIF nodes and partners.

As the network expands, it faces evolving technical and organisational challenges, including supporting new deployments, improving shared codebases, and adapting to architectural changes.

This workshop will cover the Living Atlases lifecycle: deployment strategies and common pitfalls, best practices for contributing to the shared codebase, and emerging technical challenges such as interoperability and sustainability. It aims to strengthen collaboration, support new adopters, and align future priorities across the community.

Darwin Core, Standards Development & Implementation

SYM11 Data Quality: From Standard to Practice

Session type

Symposium (Open)

Organizers

Lee Belbin, Blatant Fabrications Pty Ltd, Australia; Arthur Chapman, Australian Biodiversity Information Services, Australia

The utility of biodiversity data is defined by its fitness for use, a state achieved only when data is measured against rigorous, transparent benchmarks. The Biodiversity Data Quality (BDQ) standard is central to this work. BDQ provides the necessary foundation for consistent tests and assertions across the community. This session explores BDQ and related ‘data quality’ concepts and applications.

The session will provide an overview of BDQ by Lee Belbin, the BDQ ontology by Paul Morris, BDQ vocabularies by Arthur Chapman and BDQ-Darwin Core relationships by John Wieczorek. Related presentations are encouraged.

A standard only gains value through its implementation and the subsequent communication of its results. Following the presentation session, a hands-on workshop led by Paul Morris will guide participants through the technical realities of implementing the BDQ standard and interpreting its responses. This session will aim at ensuring data managers and researchers leave with the skills to apply these principles to their own datasets.

SYM22 State of the Collections Description Interest Group. What’s Happening and What’s Not!

Session type

Symposium (Open)

Organizers

Sharon Grant, The Field Museum of Natural History, USA; Ben Norton, USA

This session will:

1) Bring together the active task groups under, and maintenance groups affliated with the Collections Description Interest Group. Key topics will include Latimer Core, MIDS and Expeditions.

2) Invite talks from groups who are interested in the work of the Interest Group and have ideas and thoughts around new work that could be taken up.

SYM23 To see or not to see - that is the question: Managing restricted access data (RASD)

Session type

Symposium (Open)

Organizers

Tania Laity, Atlas of Living Australia - CSIRO, Australia; Cameron Slatyer, Atlas of Living Australia - CSIRO, Australia; Simon Sherrin, Atlas of Living Australia - CSIRO, Australia; Kristin Eschenfelder, University of Wisconsin-Madison, USA; Martin Kaehrle, University of Wisconsin-Madison, USA

This session seeks to explore the issues around solutions that data providers and conservation managers have encountered in handling Restricted Access Species Data (RASD) as well as providing an overview of the process of moving towards the ratification of RASD metadata extensions. Sensitive data (also known as Restricted Access Species Data - RASD) has been a long-running issue for data managers and is challenging for conservation decision-makers. The TDWG Sensitive Data Extension SubGroup has been working for the past 12 months on developing vocabularies for RASD treatments and reasons and a DWC RASD extension and DWC DP extension. The Task Group has proposed moving towards using RASD as a term (rather than sensitive species data) to escape many of the pre-conceived notions of what sensitive means.  The intention of the extension is to improve the visibility of the reasons why data has been modified and to standardise the reasons and generalisations that have been applied to data, in alignment with FAIR principles. Session attendees will gain knowledge of a range of approaches for addressing challenging RASD decisions and will gain greater understanding of the vocabulary development efforts.

Session organisers will consider abstracts on the themes of issues around solutions that data providers and conservation managers have encountered in handling RASD and those providing an overview of the process of moving towards the ratification of RASD metadata extensions.

SYM26 From WDSRPD to GeoSchemes: Recording Biodiversity Distributions for Science and Policy

Session type

Symposium (Open)

Organizers

Paco Pando, CSIC / GBIF.ES, Spain; Serge Gofas, Departamento de Biología Animal, Universidad de Málaga, 29071 Spain, Spain; Pierre Bonnet, AMAP, University of Montpellier, CIRAD, CNRS, INRAE, IRD, France; Alexandre Monro, Royal Botanic Gardens, Kew, United Kingdom; Britt Lonneville, Flanders Marine Institute (VLIZ), Belgium

The TDWG GeoSchemes task groups are finalising an updated World Geographic Scheme for Recording Species Distributions (WGSRPD). The new version expands the scope of the original standard to cover all taxonomic groups, not only plants, and to include the marine domain. GeoSchemes will provide a versioned, and citable framework for representing biogeographic areas following  political boundaries.

By bringing together standard developers and implementers / users, the session will foster dialogue on how GeoSchemes can support interoperable, policy-relevant biodiversity data across terrestrial and marine realms. It aims to present recent developments in the TDWG GeoSchemes from WGSRPD, as well as  their use in recording, exchanging, and interpreting biodiversity distribution information for scientific research, policy, and data management.

The session explicitly invites users of biodiversity distribution schemas—across data infrastructures, collections, publishing workflows, and policy-facing applications—to share practical implementation experiences. We expect that contributions will highlight current uses, integration challenges, and limitations encountered in real-world contexts. The discussion aims to identify emerging needs, inform maintenance and governance priorities, and help delineate future directions for WDSRPD and GeoSchemes as community-driven TDWG standards.

SYM32 May the Data Be Structured: Linking Descriptions, Identification and AI

Session type

Symposium (Open)

Organizers

Adeline Kerner, National Centre for Scientific Research (CNRS), France; Wouter Koch, Norwegian Biodiversity Information Centre, Norway

Taxonomic identification relies on explicit, character-based knowledge derived from published descriptions. Beyond assigning names, taxonomy depends on structured representations of morphological, ecological, and other diagnostic information that make comparisons possible and reproducible. Identification keys and structured knowledge bases play a central role in organizing this information and making it accessible to both specialists and non-specialists.Standards such as Structured Descriptive Data (SDD) provide a formal framework to encode taxonomic descriptions in a computable and interoperable way. They support the generation of identification keys, ensure traceability to the scientific literature, and allow long-term reuse of curated data. In this sense, structured knowledge bases are not only tools for identification, but infrastructures for preserving and mobilizing taxonomic knowledge.Recent developments in Artificial Intelligence create new opportunities, but also new challenges. AI can assist in accessing or navigating taxonomic information, yet it still depends on well-structured, high-quality data. This session will focus primarily on the role of SDD and structured knowledge bases in identification workflows, and will explore how they can interact with emerging AI approaches while maintaining transparency, scientific rigor, and interoperability.

WKS39 Using the new DarwinCore Data Package to promote FAIR data principles in ecological research

Session type

Workshop (Closed)

Organizers

Kelsey Huelsman, ERT, NASA Goddard Space Flight Center, United States; Eric Sokol, NEON, Battelle, United States; Dhruva Kathuria, UMBC/NASA GSFC, United States; Robert Guralnick, University of Florida, United States

Data aggregators like GBIF have increased access to large volumes of multimodal biological data, for which increased computing power can be leveraged to enable large-scale biodiversity analyses. These data, however, often lack details and context needed to confidently include them in ecological analyses. Planned, detailed, plot-level or transect surveys conducted by researchers have been much less widely mobilized to data aggregators like GBIF and are therefore under-utilized in these computing-enabled, timely analyses. This session will explore current efforts and challenges in data standard harmonization among data providers, aggregators, and users that promote accessibility and interoperability via the new DarwinCore (DwC) data package. Participants will 1) learn to apply the DwC standard and relevant extensions (e.g., Humboldt) to in situ survey data, 2) explore how DwC can (or cannot) provide necessary details for reuse, including application-specific fitness-for-use and credit attribution throughout the data lifecycle, and 3) increase the interoperability of their data with complementary data streams, including long-term studies, climate datasets, and remote sensing products. With increased computing power and advances in AI, improved ecological models and understanding are on the horizon. However, without high-quality in situ data, we cannot fully leverage this opportunity to answer questions about the past, present, and future of Earth’s ecosystems and biodiversity.

Data Infrastructure, Resilience & Governance

SYM18 Building resilient data infrastructures for biodiversity science

Session type

Symposium (Open)

Organizers

Giuditta Parolini, Museum für Naturkunde Berlin - Leibniz Institute for Evolution and Biodiversity Science, Germany; Eva Alonso Vizcaino, Naturalis Biodiversity Center, Netherlands

Data infrastructures in science are facing increasing challenges. Cyberattacks, aggressive data scraping, loss of technical expertise, sudden funding cuts, political decisions limiting data access, and environmental hazards are just a few of the potential problems. Therefore, data repositories need strategies to build resilience as part of their daily operations and have plans in place for risk mitigation.

Biodiversity science is especially affected by these challenges, as it relies on very different data, which are hosted by a multiplicity of small and large repositories located all over the world. Some of the smaller repositories, like DiscoverLife.org, are reporting AI bot crawlers as an “existential threat”  for their knowledge-sharing mission, established repositories like the Biodiversity Heritage Library are not immune from governance and funding challenges, and datasets used in environmental and climate research have been decommissioned and are not

available anymore due to a political agenda.

The symposium is an opportunity to present how data infrastructures in biodiversity science are addressing resilience and risk mitigation in the face of current challenges. It is organised as an open session to invite contributions from all biodiversity science organisations—academic institutions, charities promoting conservation initiatives, citizen science projects, etc.—interested in the topic.

SYM36 Designing Institutional Knowledge Data Science Centres for Biodiversity

Session type

Symposium (Open)

Organizers

Peter Grobe, Leibniz Institute for the Analysis of Biodiversity Change, Germany; Lars Vogt, Leibniz Institute for the Analysis of Biodiversity Change, Germany; Carina Goretzky, Leibniz Institute for the Analysis of Biodiversity Change, Germany; Björn Quast, Leibniz Institute for the Analysis of Biodiversity Change, Germany; Sheikh Mastura Farzana, Leibniz Institute for the Analysis of Biodiversity Change, Germany

Natural history museums and biodiversity research institutes are evolving from collection holders into stewards of AI-ready biodiversity knowledge. As DiSSCo builds pan-European infrastructures for FAIR Digital Objects and extended specimens, a key question arises: how can institutional biodiversity knowledge centres reinforce—rather than replicate—these global efforts?

We envision an Open Knowledge Space for Data Centers: a semantic, web-based platform linking biodiversity knowledge with collection data, research outputs, and AI-generated insights. Designed for machines and humans, it aims to serve as a biodiversity hub where data and knowledge are FAIR, CLEAR, and AI-ready—intuitive and meaningful for researchers, citizens, and policymakers.

This session brings together scientists and infrastructure architects to co-design institutional knowledge spaces that:

• Integrate semantic models and digital twins capturing data, workflows, and epistemic knowledge

• Align semantically and technically with DiSSCo

• Offer AI-ready services with operations for validation, prediction, and analysis

• Balance machine-actionability with human interpretabilitywithintuitive interfaces

We welcome contributions on:

• Prototypes of semantic data pipelines from collections to knowledge graphs

• Interactive dashboards for exploring biodiversity knowledge

• Governance models connecting collections, data science, and infrastructure

• Lessons learned from building institutional knowledge spaces

SYM38 Sustaining the Geo-Biodiversity Data Ecosystem: Standards-Driven Approaches to Long-Term Resilience

Session type

Symposium (Closed)

Organizers

Libby Ellwood, iDigBio, United States; James Macklin, Agriculture and Agri-Food Canada, Canada; Jutta Buschbom, Statistical Genetics, Germany; Kerstin Lehnert, Columbia University, United States; Andrew Bentley, University of Kansas, United States; Ely Wallis, Commonwealth Scientific and Industrial Research Organisation, Australia; Pamela Soltis, University of Florida, United States; José Fortes, University of Florida, United States

Over the past several decades, geo- and biodiversity sciences have been reshaped by digitization, data mobilization, and data interlinking efforts from which arose the community-developed Digital Extended Specimen concept, a rallying framework akin to transformative large-scale scientific innovations like the Human Genome Project. These initiatives have delivered substantial physical and digital infrastructure, new tools and workflows, and an increasingly connected global community.

As funding cycles end, challenges emerge: sustaining infrastructure, maintaining software and data quality, retaining expertise, coordinating governance across fragmented systems, and supporting workforce development. Machine-actionable, FAIR-compliant data require sustaining the entire ecosystem—data providers, aggregators, platforms, repositories, users, and underlying standards.

This session frames geo- and biodiversity data infrastructure as a distributed socio-technical system where standards enable coordination, reduce duplication, and support research while shaping infrastructure needs. The presentations and discussion in this symposium will explore how coordinated approaches combined with research- and robot-ready data standards can align social, physical, and digital infrastructures across domains, leverage economies of scale, strengthen governance and collaboration, and build relevance beyond geo- and biodiversity sciences.

DS19 Strengthening resilience in biodiversity data infrastructures: From practice to global coordination

Session type

Discussion session (Closed)

Organizers

Sharif Islam, Naturalis Biodiversity Center, Netherlands; Giuditta Parolini, Museum für Naturkunde, Germany; Eva Alonso, Naturalis Biodiversity Center, Netherlands; Wouter Addink, Naturalis Biodiversity Center, Netherlands; Jose Alonso, Naturalis Biodiversity Center, Netherlands; Sam Leeflang, Naturalis Biodiversity Center, Netherlands

Recent alignment between DiSSCo and GBIF shows how shared standards (Darwin Core Data Package, and DiSSCo’s openDS specification) can enable interoperability without information loss. Specimen-level data can now move more seamlessly across systems when conceptual foundations are aligned. Building on this, a broader question emerges: how can collaboration move beyond bilateral technical integrations toward a truly coordinated biodiversity data ecosystem? Researchers increasingly need to discover and combine diverse data types — collections, observations, environmental measurements, molecular data — yet these remain siloed across specialised infrastructures like DiSSCo, GBIF, BHL, eLTER, ELIXIR, and LifeWatch. Using the DiSSCo-GBIF collaboration as a starting point, this session invites participants from technical, scientific, and policy communities to explore what it would take to enable findability, usability, and interoperability at scale. Where should alignment happen: conceptual models, identifiers, metadata, vocabularies, access mechanisms, or governance? How can infrastructures collaborate without converging into a single data model? And what practical steps can reduce fragmentation for users today?

The session combines short scene-setting inputs with open discussion focused on shared challenges, emerging opportunities, and actionable next steps — fostering a shift from project-level thinking toward a more resillent and connected biodiversity data landscape.

WKS12 Strengthening National Biodiversity Data Infrastructures through International Coordination

Session type

Workshop (Open)

Organizers

Anton Güntsch, Freie Universität Berlin, Botanic Garden and Botanical Museum Berlin, Germany; David Fichtmueller, Freie Universität Berlin, Botanic Garden and Botanical Museum Berlin, Germany; Tanja Weibulat, Bavarian State Collections of Natural History, Germany; Victor Heijke, Naturalis Biodiversity Center, The Netherlands

Inspired by a symposium at the TDWG conference 2022 in Sofia, nine national biodiversity data infrastructures (NBDIs) from around the world have jointly developed and published a shared definition of their roles in relation to international services and initiatives. This work resulted in the identification of ten essential functions that capture the core activities of national infrastructures and highlight their relevance for science, policy, and practice (https://doi.org/10.1093/biosci/biae109).

Building on this foundation, the proposed workshop aims to further consolidate collaboration among NBDIs by exploring the establishment of an NBDI Interest Group under the umbrella of TDWG Biodiversity Information Standards. Participants will discuss potential areas of activity, priorities, and modes of collaboration, with the goal of co-developing a draft charter for the Interest Group.

The workshop is designed as an interactive forum, providing ample space for open discussion and exchange of perspectives, while also offering opportunities for short presentations that showcase national experiences and strategic developments, ongoing initiatives, and emerging challenges. By bringing together NBDI coordinators, managers, technical experts, and other key stakeholders, the workshop will strengthen international collaboration and foster alignment between NBDIs and the broader TDWG community.

Digital Twins & Emerging Infrastructures

SYM15 Operationalizing Biodiversity Digital Twins within Data Space Ecosystems

Session type

Symposium (Open)

Organizers

Claus Weiland, Senckenberg – Leibniz Institution for Biodiversity and Earth System Research, Frankfurt am Main, Germany; Dag Endresen, Natural History Museum, University of Oslo, Oslo, Norway; Tomas Martinovic, IT4Innovations, VSB - Technical University of Ostrava, Ostrava, Czech Republic; Gabriela Zuquim, CSC - IT Center for Science, Espoo, Finland; Sharif Islam, Naturalis Biodiversity Center, Leiden, Netherlands; Hanna Koivula, CSC - IT Center for Science, Espoo, Finland; Niels Raes, Naturalis Biodiversity Center, Leiden, Netherlands; Dmitry Schigel, Global Biodiversity Information Facility - Secretariat, Copenhagen, Denmark; Taimur Khan, Helmholtz Centre for Environmental Research - UFZ, Halle (Saale), Germany

This session explores the transition from prototype to operation of Digital Twins (DTs) in biodiversity, building upon the foundational approaches elaborated in projects such as Biodiversity Digital Twin (BioDT), Integration of biodiversity monitoring data into the Digital Twin Ocean (DTO-BioFlow) and Biodiversity Meets Data (BMD). As the landscape of environmental monitoring matures, we examine how these technological advancements are being institutionalized within architectures like the Green Deal Data Space, the EU flagship initiative Destination Earth (DestinE), Research Infrastructures such as LifeWatch ERIC, eLTER, DiSSCo, and intergovernmental organizations involving GBIF and OBIS to support long-term environmental governance.

We invite contributions detailing the practicalities of this shift - specifically regarding data models, FAIR workflow and provenance preservation, metadata standardization, and the operational expansion of DT approaches. By fostering a dialogue between diverse DT initiatives, this session aims to define a roadmap for a sustainable digital ecosystem that empowers key policy frameworks like IPBES with scalable, evidence-based tools, ensuring that DT innovations evolve into permanent assets for European and global biodiversity decision-making.

SYM44 Standards in Practice: Shaping the Development of Coastal Biodiversity Digital Twins

Session type

Symposium (Open)

Organizers

Laura Slaughter, SINTEF Digital, Norway

Coastal Digital Twins are dynamic, data-driven virtual representations that depend on continuous data integration from heterogeneous sources, including remote sensing, acoustic monitoring, eDNA, and autonomous underwater vehicles. Coastal twins combine biodiversity data from terrestrial and marine environments. For these systems to function autonomously and at scale, they must be built on standards that enable machine readability, interoperability, and reproducibility. In this session, we will examine how standards support automated data ingestion, semantic harmonization, provenance tracking, real-time updating, and cross-domain integration. Emphasis will be placed on machine-actionable metadata, persistent identifiers, vocabularies, and ontologies. Contributions are invited that showcase implementations, conceptual frameworks, workflows, and lessons learned in making coastal biodiversity data infrastructures digital twin-ready. Topics may include: automated quality control, integration of multimodal data, semantic alignment across marine and terrestrial domains, AI-ready data pipelines, and governance challenges in highly automated systems.

Earth Sciences & Geological Collections

SYM06 Rocks, Minerals, and Metadata: Integrating Earth Sciences into the TDWG Community

Session type

Symposium (Open)

Organizers

Ben Norton, Yale University, United States of America

Natural history collections extend well beyond biological specimens to encompass minerals, rocks, meteorites, fossils, and other geological materials fundamental to understanding Earth systems. Yet data standards development has historically prioritized biodiversity, leaving geological collections underserved and often isolated from broader aggregation and discovery infrastructures.

This symposium examines the current landscape and future directions for the standardization of earth sciences collection data. We will present progress on the Mineralogy Extension to Darwin Core, now approaching public review after two years of community development, and explore how existing standards can accommodate the distinct requirements of geological specimens.

Presentations will address practical implementation challenges, conceptual models for geological collections, and opportunities for advancement. We invite contributions from collection managers, curators, data architects, and researchers across the earth sciences collection community.

WKS10 Geological Collections with Darwin Core: Compound Specimen Model and Mineralogy Extension

Session type

Workshop (Closed)

Organizers

Ben Norton, Yale University, United States of America

Earth sciences collections present digitization challenges that traditional biodiversity standards were not designed to address. Geologic material — naturally occurring matter generated by geologic processes, encompassing minerals, rocks, meteorites, and fossils — requires vocabularies and data structures that go beyond those developed for biological specimens. This workshop introduces a standards-based workflow that leverages the Compound Specimen model to represent part-whole and associated-object relationships using Darwin Core and the Mineralogy Extension, a community-developed vocabulary now entering public review.

Participants will map real earth sciences collection datasets to Darwin Core and Mineralogy Extension terms, working through examples drawn from mineral, rock, and fossil collections. Facilitators will demonstrate how the Compound Specimen model handles objects comprised of multiple identifiable components — such as a thin section linked to its parent rock sample, or a mineral specimen associated with chemical analyses. The session closes with a structured feedback exercise on the Mineralogy Extension and an introduction to the next phase of integrating earth sciences collections into the broader TDWG community.

No prior experience with the extension is required. Familiarity with Darwin Core is helpful but not essential.

Publishing, Communications & Community Engagement

SYM07 Publishing and Communications Promoting Biodiversity, Natural History Collections, People, Data and Data Standards

Session type

Symposium (Open)

Organizers

Shelley James, Department of Biodiversity, Conservation & Attractions, Western Australian Herbarium, Australia; Julia Percy-Bower, Department of Biodiversity, Conservation & Attractions, Western Australian Herbarium, Australia; Deborah Paul, Illinois Natural History Survey, University of Illinois, Champaign, USA; Elycia Wallis, Atlas of Living Australia, National Collections & Marine Infrastructure, CSIRO, Australia

Publish or perish, the catchphrase used commonly in the academic world, is increasingly relevant for biodiversity data, collections and standards. The pressure to publish applies to peer-reviewed journals, databasing and sharing information, publishing and promoting data standards, documenting best practices, highlighting expertise in collections (e.g. Bionomia), creating harvestable and linked information, and producing success metrics and new collaborations. Social media, blogs, websites and formal publications are all ways to share and amplify the needs, values and potential of our work. Can we better promote the exchange of knowledge, ideas and best practices between our leaders, biodiversity data professionals, researchers, allies and enthusiasts above and beyond our local audiences and conferences for lasting impact? Are there tools and software available to make publishing in all formats less challenging, time consuming or stressful? We invite submissions from across the globe, from any discipline, data or physical, software and standards developers, publishers, education or policy that demonstrate or explore examples of publishing impact and success within our domain.

All TDWG 2026 speakers will submit an extended abstract to the Pensoft journal BISS; we encourage you to gain experience as a reviewer for the session, and explore publishing a full paper in this or other relevant journals, like Natural History Collections and Museomics, Biodiversity Data Journal or RIO.

Taxonomy & Freshwater Biodiversity

SYM37 A consensus taxonomic reference for improved freshwater biodiversity data and knowledge

Session type

Symposium (Open)

Organizers

Astrid Schmidt-Kloiber, BOKU University, Vienna, Austria; Koen Martens, Royal Belgian Institute of Natural Sciences (RBINS), Aquatic & Terrestrial Ecology, Belgium; Vanessa Bremerich, Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB), Germany; Helen Dallas, Freshwater Research Centre (FRC), South Africa; Olaf Bánki, Catalogue of Life Foundation, The Netherlands

Freshwaters are among the least comprehensively documented habitats, despite showing some of the fastest rates of biodiversity decline. Fragmented taxonomic knowledge, heterogeneous data standards and limited interoperability between infrastructures continue to constrain global assessments and policy uptake. This session will bring together initiatives within the freshwater community to relaunch and align key components of the freshwater data landscape, including the Freshwater Animal Diversity Assessment (FADA), the Freshwater Information Platform (FIP) and freshwater data mobilisation efforts in FWBON, GBIF and Catalogue of Life. We will explore how curated, realm-specific taxonomic checklists and biodiversity data can be mobilised as interoperable machine-actionable resources adhering to global Biodiversity Information Standards feeding into emerging freshwater specific infrastructures and reproducible workflows.

The session will focus on operationalising freshwater taxonomy through standards, sustainable infrastructures, and linking taxonomic backbones with trait, occurrence, and monitoring databases and conservation status information from the IUCN Red List of Threatened Species. We also invite discussion on community-driven approaches, including regional networks, e.g. the Freshwater Biodiversity Information System (FBIS), governance models, and long-term sustainability, aiming to advance a coherent, FAIR, and globally integrated freshwater biodiversity data ecosystem.

Unconference, Contributed Orals and Posters

Contributed Oral

Session type

Symposium (Open)

Organizers

Elie Mario Saliba, Lebanon

There are many symposia to choose from for the TDWG 2026 conference, but you may have a presentation to give that doesn’t clearly fit in any of the symposia already proposed. Contributed Oral sessions provide a way for people to offer talks on any topic of interest to the TDWG audiences. Talks related to the conference theme of Research and Robot-ready Biodiversity Data Standards are particularly welcomed.

Posters

Session type

Poster (Open)

Organizers

Javier Molina, Atlas of Living Australia / CSIRO, Australia; Elie Mario Saliba, Lebanon; Meghan Balk, Natural History Museum, University of Oslo, Norway

All those submitting poster abstracts will be required to create a virtual poster and will have the option to upload a short video to accompany it. 

Please choose this session type if you wish to present a poster rather than give an oral presentation.

TDWG 2026 Unconference

Session type

Unconference (Closed)

Organizers

Javier Molina, Atlas of Living Australia / CSIRO, Australia; Deborah Paul, Species File Group, Prairie Research Institute, University of Illinois, USA; Chandra Earl, Neon Biorepository, Arizona State University, USA; Shelley James, Department of Biodiversity, Conservation & Attractions. Western Australian Herbarium, Australia; Esteban Marentes Herrera, SiB Colombia, Colombia; Nicky Nicolson, RBG Kew, UK

The Unconference is a participant-driven session designed to explore ideas inspired by the main TDWG 2026 Conference event and this year’s theme “Research and Robot-ready Biodiversity Data Standards”. Attendees can propose topics, join breakout discussions, and collaboratively examine issues of shared interest.  The format provides an opportunity for all participants, in person or online, to connect with both existing and new colleagues, explore subjects in greater depth or propose new areas that haven’t been mentioned yet throughout the conference - sowing the seeds for new collaborations.

We invite participants to share their ideas on a topic board during the conference week, where others can also express their interest in the topic. With a little guidance from the organizers at the start of the session, we will work together to select and finalize discussion group topics and self-organize into breakout groups to learn, find synergy and gain new insights.

The session will end with breakout group leaders reporting insights or discussion summaries back to the whole group (e.g., what each group discovered, something learned). Participants are encouraged to carry discussions forward beyond the session though short writeups or position statements to invite broader engagement and continue collaboration.