Nodes Coordination Group

NCG#10

April 2026

The two sessions of the meeting have been blended into a single report wherever possible for clarity and completeness. The meeting slides are available here. The text of this summary was edited with the help of AI.

Introduction

NCG Co-Chairs John Nicholls (first session) and Maria Cornthwaite (second session) welcomed participants to the tenth meeting of the OBIS Nodes Coordination Group. Both Co-Chairs flagged a particularly packed agenda and encouraged participants to use the chat actively to ask questions and to exchange with each other. Nodes were also reminded to fill in the shared “So what have you been up to” sheet, which helps the NCG admins better capture Node activities.

Lightning talk (sessions 1 & 2 )

BENTH-MARC: A new database for benthos in the marine Arctic, with a user perspective on OBIS data Presented by Hanna Modin (UiT / The Arctic University of Norway / OBIS Norway)

Hanna Modin presented the early stages of BENTH-MARC, a new database compiling publicly available zoobenthic biodiversity records across the Arctic, developed in the framework of POMP (Polar Ocean Mitigation Potential), an EU project focused on biodiversity and carbon in the polar regions. Hanna Modin is not formally part of OBIS Norway but has worked extensively with OBIS data.

The workflow harvests all benthic phyla from within the Arctic Large Marine Ecosystems (PAME 2014) from OBIS, CritterBase, GBIF, OGSL, and a set of unpublished datasets, then filters out pelagic, terrestrial, and extinct taxa, cleans, harmonizes, and merges the data. The current version holds around 10,354 taxa, 3.6 million records, and 134,192 stations. Hanna highlighted four OBIS-specific issues encountered while preparing the database, framed as constructive suggestions:

> Only about 58% of occurrence records returned a populated datasetName field.The names can be retrieved manually from the OBIS website using the dataset_id, but the automatic fetch appears broken or inconsistent, which becomes a significant constraint when working with large numbers of datasets. As a result, she has tended to prefer GBIF data wherever duplicates exist between the two infrastructures.

> Conditional taxonomic downloads currently lose NaN entries at the relevant taxonomic level.This forces the user to first download all records belonging to a given group and then filter locally. For taxa such as Porifera and Bryozoa, where the highest available taxonomic resolution is often only at the phylum level, retaining NaN entries is essential.

> The eventDate column seems heterogeneous, with a wide range of formats, and even some dates set very far back in time or in the future.

> Building on the existing environment information already added from WoRMS (which she described as very useful), Hanna suggested exposing functional group information (benthic/pelagic) directly in OBIS data at harvest, since this is programmatically accessible from WoRMS and currently covers approximately 80% of accepted, extant marine species.

→ Pieter Provoost (OBIS Secretariat) addressed two of the issues during the second session.> On datasetName, he clarified that the field is only populated if the data provider populates it for every occurrence record; however, the dataset title is always available via the dataset metadata, and can be retrieved separately via a dedicated function in the robis R package and joined to the occurrence records.

> On eventDate, he explained that OBIS keeps the original (sometimes messy) value as provided, but adds a number of interpreted fields (date_year, start, mid, and end timestamps as numeric values) precisely to make downstream parsing easier in R, Python, or Excel. He invited further discussion on whether users would prefer an additional ISO-formatted output field.

→ Hanieh Saeedi (OBIS Deep Sea) asked what proportion of the BENTH-MARC records are deep-sea data, in order to compare with her own deep-sea benthos compilation. Hanna Modin offered to share the breakdown after the meeting.

Node members are encouraged to contact Hanna Modin for collaborations or further discussion at hanna.m.modin@uit.no

Insights from BBNJ PrepCom3Presented by Ward Appeltans, Head of the OBIS SecretariatWard Appeltans reported back from the third Preparatory Commission meeting (PrepCom3) for the BBNJ Agreement, held in New York from 23 March to 2 April 2026. The BBNJ Agreement entered into force on 17 January 2026, which means the first Conference of the Parties (COP-1) will be organized within one year and is now scheduled in New York from 11 to 22 January 2027. To date, 88 countries have ratified the Agreement.

What was decidedDiscussions at PrepCom3 focused heavily on the Clearing-House Mechanism (CHM), the digital backbone of the Agreement: without it, key obligations (six-month advance notification of cruises in areas beyond national jurisdiction, use of standardized batch identifiers for samples, etc.) cannot be implemented. Ward highlighted that the CHM itself is not expected to host the underlying data: the CHM would store metadata pointing back to existing resources.PrepCom3 ended with the adoption of an Outcome Document, setting a clear path toward an operational prototype of the CHM before COP-1. Development will be guided by the BNNJ Agreement Co-Chairs, the Bureau, and DOALOS. An open call for expressions of interest will be issued, inviting the four organizations referenced in Article 51 (IOC, IMO, FAO, and the ISA), as well as anyone interested, to make a proposal that must include a detailed work plan and costings. The selected technical partner will operate under the oversight of the Co-Chairs, Bureau, and DOALOS, will be required to connect and consult with relevant groups such as national focal points, non-Party users, and Indigenous Peoples and local communities, with an obligation to deliver by COP-1. Continuation will depend on the COP assessment of the prototype.

Priority functions for the prototypeThe Outcome Document identifies four priority functions to be delivered before COP-1:(a) MGR notifications and the BBNJ standardized batch identifier;(b) match-making for capacity-building and transfer of marine technology (CB/TMT), for example, connecting people from developing countries to berths and vessels;(c) notifications related to environmental impact assessments (EIAs);(d) basic infrastructure for area-based management tools (ABMTs), with some countries already pushing for a geospatial interface around marine protected areas.Ward noted that the list is ambitious, with a short timeline: the prototype, including iterative consultations, needs to be ready before the end of the year.

Relevance for OBISSeveral technical parameters in the Outcome Document map directly onto the OBIS network:

metadata standards interoperable with national, sub-regional, regional, and global databases (parameters a, f);

a central system designed for future integration of nodes, in line with the distributed OBIS architecture (parameter g);

support for low-bandwidth environments and offline submission (parameter j);

capacity-building components integrated into the prototype, connecting to OceanTeacher and OBIS training activities (paragraph 10);

and a prototype in English with the final system in all six UN languages (parameter h).

Ward also highlighted a strong push to design the CHM as a federated system with regional or national nodes rather than a purely central system. Countries see the future BBNJ nodes as places that would also offer helpdesk and support functions to assist governments and researchers with EIA or MPA proposals, or with compliance on marine genetic resources. The OBIS Nodes network is being looked at as a potential answer to that need.Key messages and next stepsOBIS is following the BBNJ process closely and is ready to support the development of the prototype if IOC is selected. The call is expected within weeks, with a probable submission deadline at the end of May. The IOC Executive Council, in late June, would still need to welcome the IOC’s proposed expression of interest. If IOC is selected, the OBIS Secretariat will likely lead the work, with significant involvement from Pieter Provoost, Steve Formel, Silas Principe, and others, from approximately July through to January 2027. Additional staff will need to be hired for both development and consultation, which will inevitably draw some resources away from “core” OBIS business in the second half of the year.→ Kit Elloran (Sea-OBIS) asked about the minimum data standards and interoperability protocols needed to align OBIS datasets with BBNJ reporting, especially for deep-sea and high-seas biodiversity. Ward Appeltans clarified that the CHM is not expected to host the underlying datasets but to keep the necessary metadata pointing to the data sources, for example, via DOIs, with the data themselves remaining in OBIS, GBIF, or other databases. The minimum metadata set has not yet been identified and will need to be defined as part of the prototype development.→ Katherine Tattersall (OBIS Australia, OBIS SG Co-Chair) shared a concrete experience from the Australian Node: the Australian delegation involved in the BBNJ negotiations had a strong legal and policy background but limited expertise in biodiversity data architecture, and approached the Node for advice. She encouraged other Nodes to proactively send a short message to their national delegations to flag their expertise, even if they do not feel “ready” to engage, as this can open the door for further discussions.

Data Coordination Group updatesPresented by Ruben Perez Perez, OBIS DCG Co-Chair (first session) and Katherine Tattersall, OBIS SG Co-Chair (second session)The DCG Co-Chairs presented progress on the group’s 2026 work plan deliverables and reminded the NCG that the DCG is an open membership and welcomes contributions from across the OBIS community.Two workshops are being planned, with provisional dates: eDNA data publishing in June 2026, and the Darwin Core Data Package format (DwC-DP) in October. Dates remain to be confirmed pending materials and trainers.The taxonomic annotation effort, in which a small group of volunteer experts reviews automated annotations against WoRMS, is paused for about a month: the person coordinating the work has left their organization and is being replaced (the work is being coordinated on behalf of OBIS by Dan Lear with a UK-based group). Volunteer training resumes in May, with the capacity to take on additional volunteers. A short plan for the work is open for review.EOV tagging work is progressing, starting with a pilot dataset to see how the tags behave through the OBIS pipelines. A plan for scaling and gradual automation is open for review. A notable milestone: the EOV controlled vocabularies are now published in the NERC Vocabulary Server (currently as PDFs); definitions still need refinement to improve interoperability, but this is already a significant step forward for EOV tagging of OBIS datasets.The modules for a more comprehensive eDNA guide have been drafted, ranging from basic to advanced content, and will support the planned eDNA workshop. A dedicated eDNA quality control session has also been organized; Nodes working with eDNA datasets are invited to bring questions to this working group.The obistools R package urgently needs reviewers and a maintainer. The package provides quality-control functions for marine Darwin Core datasets: parts of the package are showing signs of being out of step with current versions of R. The DCG and the PCG have issued a joint call for volunteers (see also the PCG update below).Several OBIS Node managers attend GBIF regional meetings, often without being aware of each other. The DCG is encouraging better coordination so that OBIS is presented in a consistent way and that materials can be shared rather than re-created. Katherine Tattersall, for example, was due to present at the Oceania regional meeting later that same week, and invited interested Nodes to get in touch.The DCG has requested a short training session on Discourse, possibly during a future NCG meeting, on how to create public and private channels in the OBIS Discourse space. The capability exists but is not widely known, and a brief walk-through would help Nodes use Discourse to its full potential, including for more sensitive discussions where a private channel is needed alongside the public ones.

Product Coordination Group updatesPresented by Silas Principe, OBIS Secretariat Researcher (first session) and Steve Formel, OBIS Secretariat (second session)

The product catalog has reached a stage where the corresponding work plan item could arguably be marked as complete. A production instance is live at products.obis.org, and a separate development instance [add link] is used to test new features before promotion. Documentation is being maintained in parallel as the catalog evolves.The main need now is volunteers to design the governance of the catalog: minimum requirements for products, which product types to expose, whether to take a phased approach, and how to ensure quality control beyond the technical layer. Interested Nodes can reach out to the PCG, contact the Secretariat through the helpdesk, or join an upcoming PCG meeting.The PCG echoed the DCG’s call to revive obistools. Steve Formel and Silas Principe are coordinating the effort and gathering interested contributors; partial involvement is welcome, and contributors do not need to be full R package developers. The first goal is to bring the package back up to speed with current R versions; from there, new functionalities could be added, in particular, helpers to work with the cloud-optimized (Parquet) versions of OBIS data.At the last PCG meeting, Sachit Rajbhandari (OBIS Australia) and Eduardo Klein Salas (IMOS) presented their work on aggregating Australian seabird data into a cloud-optimized data product directly used by Australian government partners, alongside notebooks to help users explore the data. A recording is available for OBIS Nodes and trusted partners; it has not yet been cleared for broader public dissemination, as it has not been through the comms teams of either organization.

Steering Group updatesPresented by Katherine Tattersall (first session) and Dan Lear (second session), OBIS SG Co-ChairsThe two SG Co-Chairs were invited, together with the IODE and ODIS Co-Chairs, to attend a session of the GOOS Steering Group meeting (held in India, joined virtually). Lotta Fyrberg, one of the IODE Co-Chairs, attended a number of GOOS sessions; Adam Leadbetter, ODIS Steering Group Co-Chair, presented on ODIS; and Katherine Tattersall and Dan Lear presented OBIS, its systems, and its community via the Nodes network.The reason for the invitation was to discuss the planned GOOS reform, in particular the relationship between GOOS and IODE, and the requirements of a data infrastructure to support GOOS activities. OBIS already has a strong working relationship with the GOOS BioEco panel around Essential Ocean Variables, and with the broader GOOS Biodiversity Plan discussions that started at the OBIS Steering Group in Bogotá.The OBIS SG Co-Chairs used the opportunity to encourage the use and adoption of existing IOC data architecture, with OBIS as the discovery layer for biodiversity. The core message was that a highly functional network already exists in OBIS and its Nodes, and that there is no need to re-invent that infrastructure. Instead, GOOS should engage with OBIS for the biological and biodiversity components, while continuing to cover the wider range of observations that go beyond the OBIS scope. Some Nodes may be approached by their NODCs or by GOOS representatives in the context of an upcoming consultation with Member States about data infrastructure needs. The Secretariat and the SG Co-Chairs are happy to provide background and support if needed.

Communication kit for the OBIS Nodes

Presented by Laurent Chmiel, OBIS Secretariat

Laurent Chmiel gave an update on Deliverable 2.1 of the OBIS work plan: a building-block-based communication toolkit aimed at funders, backers, and host organizations, that can be adapted to each OBIS Node. The starting point is the recognition that OBIS Nodes operate in very different institutional contexts, with different resources and challenges. The approach to building the kit is structured around five steps:

Each Node identifies up to ten of its key stakeholders, where a “stakeholder” can be a group, a person (e.g., the dean of a hosting institution), or an authority.

Each Node fills in a simple, table-based template that qualifies each stakeholder: who they are, how much influence and interest they have, what the Node expects from them, what they expect from the Node, and what the Node wants to achieve through communication with them.

The Secretariat builds a typology of stakeholder profiles.

The Secretariat builds communication blocks to support Nodes’ communication actions and objectives.

Nodes pick and choose from these blocks to fit their needs and produce their own assets. A typical use case: a Node manager hosted in a university with a ten-minute slot once a year to argue for continued or increased investment from the dean, with the kit doing the legwork on key messaging.

The template will be sent to Nodes around 6 May 2026, with a deadline for completion of 6 June 2026. Materials will initially be in English, with the intention to localize them over time; for Spanish-speaking Nodes, the template can already be sent in Spanish from the start. The Secretariat and the NCG Co-Chairs are available to support Nodes throughout the process. Further details will be shared in the next OBIS Pulse newsletter and by email.→ Ben Donnelly (OBIS-SEAMAP) asked for a concrete example of what a “communication block” would look like in practice. Laurent Chmiel illustrated this with the dean / university scenario above, noting that the difficulty of the exercise lies in the diversity of Nodes (financial, institutional, administrative). The first iteration will already be ambitious; subsequent iterations can be expanded year on year, for example, with more targeted strategies for specific actions, such as approaching a ministry for additional budget.

Wrap-up

Both Co-Chairs reminded participants to fill in the shared “what have you been up to” board between meetings, which also helps with the year-end roll-up.

The next NCG meeting is scheduled for 10 June 2026.