Biodiversity Data Journal : Data Paper (Biosciences)
PDF
Data Paper (Biosciences)
Research Infrastructure Contact Zones: a framework and dataset to characterise the activities of major biodiversity informatics initiatives
expand article infoVincent Stuart Smith‡, Lisa French‡, Sarah Vincent‡, Matt Woodburn‡, Wouter Addink§,|, Christos Arvanitidis¶,#, Olaf Bánki¤,§, Ana Casino«, Francois Dusoulier», Falko Glöckler˄, Donald Hobern˅, Martin R. Kalfatovic¦, Dimitrios Koureas§,|, Patricia Mergenˀ,ˁ, Joe Miller₵, Leif Schulmanℓ,₰, Aino Juslén₱
‡ Natural History Museum, London, United Kingdom
§ Naturalis Biodiversity Center, Leiden, Netherlands
| Distributed System of Scientific Collections - DiSSCo, Leiden, Netherlands
¶ LifeWatch ERIC, Seville, Spain
# Institute of Marine Biology, Biotechnology and Aquaculture, Heraklion, Crete, Greece
¤ Catalogue of Life, Amsterdam, Netherlands
« Consortium of European Taxonomic Facilities, Brussels, Belgium
» Muséum national d'histoire naturelle, Paris, France
˄ Museum für Naturkunde Berlin, Leibniz Institute for Evolution and Biodiversity Science, Berlin, Germany
˅ International Barcode of Life, Guelph, Canada
¦ Smithsonian Institution Libraries and Archives / Biodiversity Heritage Library, Washington, United States of America
ˀ Meise Botanic Garden, Meise, Belgium
ˁ Royal Museum for Central Africa, Tervuren, Belgium
₵ GBIF, Copenhagen, Denmark
ℓ Finnish Environment Institute, Helsinki, Finland
₰ University of Helsinki, Helsinki, Finland
₱ Finnish Museum of Natural History, Helsinki, Finland
Corresponding author: Vincent Stuart Smith (vince@vsmith.info)
Academic editor: Quentin Groom
Received: 01 Mar 2022 | Accepted: 06 Aug 2022 | Published: 16 Sep 2022
© 2022 Vincent Smith, Lisa French, Sarah Vincent, Matt Woodburn, Wouter Addink, Christos Arvanitidis, Olaf Bánki, Ana Casino, Francois Dusoulier, Falko Glöckler, Donald Hobern, Martin R. Kalfatovic, Dimitrios Koureas, Patricia Mergen, Joe Miller, Leif Schulman, Aino Juslén
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation: Smith VS, French L, Vincent S, Woodburn M, Addink W, Arvanitidis C, Bánki O, Casino A, Dusoulier F, Glöckler F, Hobern D, Kalfatovic MR, Koureas D, Mergen P, Miller J, Schulman L, Juslén A (2022) Research Infrastructure Contact Zones: a framework and dataset to characterise the activities of major biodiversity informatics initiatives. Biodiversity Data Journal 10: e82953. https://doi.org/10.3897/BDJ.10.e82953
 Open Access

Abstract

Background

The landscape of biodiversity data infrastructures and organisations is complex and fragmented. Many occupy specialised niches representing narrow segments of the multidimensional biodiversity informatics space, while others operate across a broad front, but differ from others by data type(s) handled, their geographic scope and the life cycle phase(s) of the data they support. In an effort to characterise the various dimensions of the biodiversity informatics landscape, we developed a framework and dataset to survey these dimensions for ten organisations (DiSSCo, GBIF, iBOL, Catalogue of Life, iNaturalist, Biodiversity Heritage Library, GeoCASe, LifeWatch, eLTER ELIXIR), relative to both their current activities and long-term strategic ambitions.

New information

The survey assessed the contact between the infrastructure organisations by capturing the breadth of activities for each infrastructure across five categories (data, standards, software, hardware and policy), for nine types of data (specimens, collection descriptions, opportunistic observations, systematic observations, taxonomies, traits, geological data, molecular data and literature) and for seven phases of activity (creation, aggregation, access, annotation, interlinkage, analysis and synthesis). This generated a dataset of 6,300 verified observations, which have been scored and validated by leading members of each infrastructure organisation. The resulting data allow high-level questions about the overall biodiversity informatics landscape to be addressed, including the greatest gaps and contact between organisations.

Keywords

methods, data visualisation, coordination, alignment, community, biodiversity informatics

Introduction

Biodiversity informatics – the application of informatics techniques and technologies to collate, harmonise, manage, share and use data and information on the world’s biota – has progressed considerably in the last two decades (Gadelha et al. 2020). These developments have made unprecedented volumes of data readily available for the scientific community and other stakeholders. For example, GBIF now offers over 2.2 billion occurrence records (gbif.org; 5 August 2022), providing a comprehensive map of past and present species distributions. iBOL, a global repository of DNA barcodes, now includes 779 thousand barcode index numbers (a proxy for species, boldsystems.org; 5 August 2022). Likewise, the Biodiversity Heritage Library has digitised over 60 million pages of biodiversity literature (biodiversitylibrary.org; 5 August 2022), while citizen science projects like iNaturalist have over 2.3 million observers contributing contemporary data on taxon occurrences (inaturalist.org/observations; 5 August 2022).

Data from these and related infrastructures are critical to addressing many of science and societies greatest challenges, including the interconnected crises of biodiversity loss and climate change. For example, much of the science that underpins policies designed to tackle biodiversity loss comes from data mediated by these infrastructures. Consequently, there is an ever more pressing need to tackle the barriers that hinder the acquisition of more data. As new needs emerge, especially in responding to the growing data needs, ever more coordination is required in the development of new infrastructures. One such gap relates to the provision of data from natural science collections. Their collections provide unique and critical insight into historical distributions of species and are the gateway to a rich wealth of additional information associated with these specimens. At present, very little (perhaps just 5%) of the estimated 1.5 billion specimens in these collections has any digital record (Cocks et al. 2020). Within European collections, this gap is being addressed by a consortium of natural science collections, who are coordinating in their efforts to make these collections digitally accessible and provide a common digital gateway to facilitate collections access. This network, the Distributed System of Scientific Collections, or "DiSSCo" for short, is in the process of establishing its service infrastructure as it moves to become a legal entity under the European Union "European Research Infrastructure Consortium (ERIC)", which is a specific legal form that facilitates the establishment and operation of European Research Infrastructures (Addink et al. 2019).

As part of the efforts to formalise DiSSCo, a working group was commissioned by the DiSSCo Interim General Assembly to examine the strategic position of DiSSCo with related research infrastructures. Global data infrastructures tend to have specialised niches representing only a narrow segment of the multidimensional biodiversity informatics space. They differ by data type(s) handled and data life cycle phase(s) supported. They may deal with only one data type (e.g. Fishbase, a global species database on fish) or support one or a few links in the data mobilisation chain (e.g. GBIF that collates, integrates and distributes digital data, but does not digitise analogue data). Specialisation may represent a reasonable division of labour at the macro level and be the only feasible way to advance service generation. However, in many countries, the mosaic pattern has been repeated at the national, regional or continental level, necessitated by funding and/or jurisdictional constraints. When this occurs, there is a risk of duplication or actions becoming siloed, hindering effective development.

The worldwide biodiversity informatics landscape is, therefore, composed of numerous elements, which have invested much effort in connecting to provide a complete service array to end-users, but have not always succeeded in avoiding redundancy. The strategic positioning working group of DiSSCo was tasked with examining potential links with related data infrastructures to prove an evidence base that would refine DiSSCo's niche of operation. To minimise these risks for DiSSCo, the working group developed a methodology to examine the niche of DiSSCo and nine related infrastructures. In doing so, we have not only built a comprehensive picture of these infrastructures' activities, both now and into the future, but also a methodology for examining their interrelationships.

General description

Purpose: 

We sought to characterise the current and planned activities performed by major organisations involved in biodiversity informatics, through a quantitative assessment that described not only the many dimensions of their activities, but also their relative technological maturity (referred to as a ‘maturity index’). This maturity index (Table 1) addresses the fact that these organisations are at different stages in their life cycle and many are yet to realise the full maturity of their ambitions.

Table 1.
Download as
CSV
XLSX

Definitions of terms provided to the biodiversity data infrastructures and used during the assessment process.

Category Term Definition

General

 Organisation An entity – such as a company, an institution or an association – comprising one or more people and having a particular purpose. In the context of this framework, this is the entity whose activity is being scored.
Type A high-level class of information associated with a physical specimen held within a natural science collection.
Phase A stage with the data processing lifecycle.
Infrastructure The set of fundamental content, facilities, systems or services necessary for a community to function.
Maturity Index A measurement system used to assess the maturity level of a particular activity, domain or technology.
Evidence Examples relevant to the major 'Type' (not 'Phase') of activity, given as short unstructured text remarks and/or web links to further information.

Scope

 Specimens An evidential record of an individual, item or part of a natural science collection.
Collection registry/description Metadata used to describe any set of individuals, items or parts (specimens) that form a whole or part of a natural science collection.
Observations (opportunistic) An evidential record of an unplanned encounter with an individual organism at a particular time and place.
Observations (systematic) An evidential record of an encounter with an individual organism at a particular time and place as part of a programme of study.
Biological taxonomy/classification Any activities associated with the branch of science that encompasses the description, identification, nomenclature and classification of organisms.
Biological descriptions/traits The non-molecular phenotype of a biological entity, in the form of a text description, statement, multimedia or dataset.
Geology Any aspect of the characterisation (including Earth or planetary system science) of rocks and minerals of any origin, in the form of a text description, statement or dataset.
Molecular Any aspect of the structure, function, evolution, mapping and editing of an organism's DNA or RNA nucleotides.
Literature Any non-fiction scholarly writing or metadata associated with such writing, concerning any aspect of the natural world.

Phase

 Create The first stage in the data life cycle in which an initial digital representation is created.
Aggregate The bringing together of a group, body or mass composed of many distinct parts or individuals.
Access The "ability to access" and benefit from some system or Accept entity.
Annotate The addition of extra information associated with a particular point in any data, information or knowledge.
Interlink The connection of things (e.g. entities in a database).
Analyse To subject to scientific analysis.
Synthesis The combining of often diverse conceptions into a coherent whole to create new knowledge.

Category

 Data/Content Factual information used as a basis for reasoning, discussion or calculation.
Standards The rules (format and meaning) by which data are described, recorded and exchanged.
Software Any set of programmes, procedures and routines associated with the operation of a computer system.
Hardware Tools, machinery and other durable equipment (e.g. computers and storage) associated with any phase of activity.
Policy/Culture The community networks and agreed practices to make our activities an openly shared, freely available, connected resource.

Maturity Index

 P0 - No activity/inapplicable No current/planned activity or inapplicable to an organisation's operations.
P1 - Planned Named a strategy, roadmap or outline development as a proof of concept (evidenced through documentation or a prototype solution).
P2 - Presence Addresses part of the domain/problem set served, sometimes as a dependency to addressing other issues and in use (evidenced through the use of the solution beyond the developing organisation).
P3 - Performance Addresses a majority/full scope of the domain it serves and in widespread use (evidenced through the richness of feature set and widespread use).
P4 - Predominance A domain leader to which all other innovators would aspire to or work with, addressing the full scope of the domain and sustained through continuous improvement (evidenced through market share).

This framework aimed to allow the results to be actionable, providing insights on where there is likely to be the greatest future contact between the shared ambitions and deepest gaps across the overall landscape of biodiversity informatics activities. At every stage in the data collection process, rigorous efforts were made to standardise the data such that it can be directly compared across each organisation. However, despite the granularity of the framework used to gather data, there may be considerable differences in the activity of organisations operating in precisely the same niche. In such instances, this 'Contact' between the activities of different organisations should signal the need for further investigation, rather than an immediate inference of duplication.

Project description

Title: 

Distributed System of Scientific Collections (DiSSCo) Interim General Assembly: Research Infrastructure Contact Zones Task Force

Personnel: 

Wouter Addink, Christos Arvanitidis, Olaf Bánki, Ana Casino, François Dusoulier, Lisa French, Falko Glöckler, Donald Hobern, Aino Juslén, Martin Kalfatovic, Dimitrios Koureas, Patricia Mergen, Joe Miller, Leif Schulman, Vincent Smith, Sarah Vincent, Matt Woodburn

Design description: 

A task force was commissioned by the DiSSCo Interim General Assembly to examine the activities and alignment of DiSSCo in relation to the fragmented and complex landscape of related biodiversity informatics organisations and infrastructures. A new framework was developed to survey infrastructures across five categories (data, standards, software, hardware and policy), for nine types of data (specimens, collection descriptions, opportunistic observations, systematic observations, taxonomies, traits, geological data, molecular data and literature) and for seven phases of activity (creation, aggregation, access, annotation, interlinkage, analysis and synthesis). This work was inspired by an early model recently published by the Finnish Biodiversity Information Facility, which depicts biodiversity informatics organisations by the data type supported and the data life cycle phases covered (Schulman et al. 2021).

Funding: 

This work was partially supported by the Horizon 2020 Framework Programme of the European Union: H2020-INFRADEV-2019-2020 – DiSSCo Prepare – Grant Agreement No. 871043.

Sampling methods

Description: 

A subset of large infrastructures active in biodiversity informatics across Europe and willing to take part, became the focus of our research. These are arguably those infrastructures operating conceptually and geographically closest to the domain of DiSSCo. Nevertheless, the selection excludes many other potentially relevant groups and, in the absence of a global infrastructure registry, it is impossible to fully know how many infrastructures might be missing from this survey. The ten organisations that agreed to participate were: the Distributed System of Scientific Collections (DiSSCo, https://www.dissco.eu/), the Global Biodiversity Information Facility (GBIF, https://www.gbif.org/), the International Barcode of Life (iBOL, https://ibol.org/), the Catalogue of Life (CoL, https://www.catalogueoflife.org/), iNaturalist (https://www.inaturalist.org/), the Biodiversity Heritage Library (BHL, https://www.biodiversitylibrary.org/), the Geoscience Collections Access Service (GeoCASe, https://geocase.eu/), LifeWatch (https://www.lifewatch.eu/), the integrated European Long-Term Ecosystem, critical zone and socio-ecological Research Infrastructure (eLTER, https://elter-ri.eu/) and ELIXIR (https://elixir-europe.org/). By necessity, many of these organisations have global reach, but we particularly focused on related European Strategy Forum on Research Infrastructures (ESFRI) organisations given their proximity in activities, governance model and funding to DiSSCo. Additional organisations within and beyond Europe became interested in the research during the course of our data collection, but we agreed to constrain our initial research to limit the frequency of changes to the survey structure.

Sampling description: 

Each infrastructure organisation was sent a personalised data collection template, alongside an extensive Frequently Asked Questions document that outlined the rationale for the work and the methodology. Each organisation was asked to evidence their results and did not have sight of other infrastructures' scores, corresponding to the relative maturity of their activities, during the data collection phase. Considerable efforts were made to allay any concerns about the nature of the survey to minimise the risk of organisations over- or underestimating their scores or declaring a pattern of activity that exceeds their stated actions or ambition. An extensive glossary of terms (Table 1), tightly and clearly defining all the parameters being scored, was included in the data collection template to ensure that there was a common understanding of activities across each organisation and, thus, reinforce the standardisation and comparability of the datasets.

A personalised dataset for each infrastructure was initially populated with preliminary data, based on the Task Force’s initial understanding of each infrastructure's activities. This helped to frame expectations and minimise the data collection burden on the part of the individual(s) completing the survey. In some cases, this preliminary dataset proved a close match to the final verified data submitted and, in a few cases, significantly over- or underestimated the scope and maturity of activities. Regardless, all data providers significantly evidenced their submissions, providing confidence that the data received are a fair and close match to current or planned activities. In several cases, this was further clarified through follow-up discussions with the data provider.

Every effort was made to standardise the interpretation of the terms being assessed. However, some may still be subject to differences in understanding, leading to minor discrepancies in how certain activities were scored. For example, some infrastructures considered their use and development of High-Performance Computing infrastructure, in the context of the survey questions covering hardware development, while others excluded this from their hardware considerations.

Step description: 

A contact zones database was developed to store the survey responses. This was subsequently used to support the data visualisations and analysis of the results. The overall database schema can be found in Fig. 1.

Figure 1.  

Structure of the Research Infrastructure Contact Zones Database.

Geographic coverage

Description: 

The research infrastructures described by this dataset have a mix of global and European coverage.

Usage licence

Usage licence: 
Creative Commons Public Domain Waiver (CC-Zero)

Data resources

Data package title: 
Research Infrastructure Contact Zones
Resource link: 
https://doi.org/10.5281/zenodo.6138822
Number of data sets: 
7
Data set name: 
flattened_research_infrastructure_contact_zones.tsv
Data format: 
TSV
Description: 

TSV containing the denormalised, flattened view of the Contact Zones database. Definitions of the terms found in the category, scope and phase columns are found in Table 1.

Column label Column description
infrastructure Name of infrastructure.
category Name of category.
scope Name of scope.
phase Name of phase.
level_current Maturity index score: current level.
level_ambition Maturity index score: ambition level.
Data set name: 
tbl_scores.tsv
Data format: 
TSV
Description: 

TSV exports of the normalised database tables shown in Fig. 1. These tables include linking fields and other utilitarian/structural elements of the data not included in the flattened version of the data.

Column label Column description
score_id Score ID.
infrastructure_id Infrastructure ID.
category_id Category ID.
scope_id Scope ID.
phase_id Phase ID.
stage Indicates if the maturity index score relates to current level or long-term ambition.
level_definition_id Maturity index level definition ID
Data set name: 
tbl_infrastructure.tsv
Column label Column description
infrastructure_id Infrastructure ID.
infrastructure Name of infrastructure.
last_updated Date the survey was completed or updated.
scored_by Name of individual who completed the survey.
Data set name: 
tbl_category.tsv
Column label Column description
category_id Category ID.
category Name of category.
sort_order Default sort order for category.
Data set name: 
tbl_scope.tsv
Column label Column description
scope_id Scope ID.
scope Name of scope.
sort_order Default sort order for scope.
Data set name: 
tbl_phase.tsv
Column label Column description
phase_id Phase ID.
phase Name of phase.
sort_order Default sort order for phase.
Data set name: 
tbl_level_definition.tsv
Column label Column description
level_definition_id Maturity index level definition ID.
level Name of maturity index level.
level_no Maturity index level number.
level_definition Definition of maturity index level.
level_title Full maturity index level title, including name and number.

Additional information

Infrastructure Summaries: current and future scope

The activity levels when viewed by scope shows the subject matter and area of interest of the infrastructures. Table 2 shows the current levels of activity by the scope for each infrastructure, as well as their future ambition. This Table includes a ranking of scope by infrastructure. The changes between current and future (ambition) levels is visualised in Fig. 2.

Table 2.
Download as
CSV
XLSX

Infrastructure priorities: scope-focus per infrastructure, ranked by proportion of activities at level P2-P4.

Infrastructure Rank - Current Top Scopes - Current Activities P2-P4 - Current Scope % - Current Rank - Ambition Top Scopes - Ambition Activities P2-P4 - Ambition Scope % - Ambition
BHL 1 Literature 29 62% 1 Literature 29 57%
BHL 2 Biol. taxonomy/classification 7 15% 2 Biol. descriptions/traits 7 14%
BHL 3 Specimens 5 11% 2 Biol. taxonomy/classification 7 14%
BHL 4 Biol. descriptions/traits 4 9% 3 Specimens 6 12%
BHL 5 Geology 2 4% 4 Geology 2 4%
BHL Observations (systematic) 0 0% Collection registry/description 0 0%
BHL Molecular 0 0% Observations (opportunistic) 0 0%
BHL Collection registry/description 0 0% Observations (systematic) 0 0%
BHL Observations (opportunistic) 0 0% Molecular 0 0%
Catalogue of Life 1 Biol. taxonomy/classification 27 64% 1 Biol. taxonomy/classification 27 43%
Catalogue of Life 2 Literature 15 36% 2 Literature 22 35%
Catalogue of Life Observations (opportunistic) 0 0% 3 Specimens 6 10%
Catalogue of Life Observations (systematic) 0 0% 4 Molecular 4 6%
Catalogue of Life Specimens 0 0% 5 Collection registry/description 1 2%
Catalogue of Life Molecular 0 0% 5 Biol. descriptions/traits 1 2%
Catalogue of Life Collection registry/description 0 0% 5 Observations (opportunistic) 1 2%
Catalogue of Life Geology 0 0% 5 Observations (systematic) 1 2%
Catalogue of Life Biol. descriptions/traits 0 0% Geology 0 0%
DiSSCo 1 Specimens 10 91% 1 Specimens 30 22%
DiSSCo 2 Collection registry/description 1 9% 2 Biol. taxonomy/classification 20 15%
DiSSCo Observations (opportunistic) 0 0% 3 Collection registry/description 19 14%
DiSSCo Observations (systematic) 0 0% 4 Biol. descriptions/traits 17 12%
DiSSCo Biol. descriptions/traits 0 0% 5 Geology 16 12%
DiSSCo Molecular 0 0% 5 Literature 16 12%
DiSSCo Biol. taxonomy/classification 0 0% 6 Molecular 15 11%
DiSSCo Geology 0 0% 7 Observations (systematic) 2 2%
DiSSCo Literature 0 0% 8 Observations (opportunistic) 2 2%
ELIXIR 1 Molecular 35 26% 1 Molecular 35 24%
ELIXIR 2 Biol. taxonomy/classification 25 18% 2 Biol. descriptions/traits 27 18%
ELIXIR 3 Biol. descriptions/traits 23 17% 3 Biol. taxonomy/classification 25 17%
ELIXIR 4 Observations (systematic) 19 14% 4 Observations (systematic) 21 14%
ELIXIR 4 Literature 19 14% 5 Literature 19 13%
ELIXIR 5 Specimens 15 11% 6 Specimens 18 12%
ELIXIR 6 Collection registry/description 1 1% 7 Collection registry/description 3 2%
ELIXIR Geology 0 0% Observations (opportunistic) 0 0%
ELIXIR Observations (opportunistic) 0 0% Geology 0 0%
eLTER 1 Literature 23 25% 1 Observations (systematic) 34 18%
eLTER 2 Observations (systematic) 20 22% 1 Geology 34 18%
eLTER 3 Observations (opportunistic) 15 16% 2 Literature 33 17%
eLTER 4 Molecular 14 15% 3 Observations (opportunistic) 31 16%
eLTER 5 Geology 13 14% 4 Molecular 23 12%
eLTER 6 Biol. descriptions/traits 6 7% 5 Biol. descriptions/traits 17 9%
eLTER 7 Biol. taxonomy/classification 1 1% 6 Biol. taxonomy/classification 11 6%
eLTER 7 Specimens 1 1% 7 Specimens 6 3%
eLTER Collection registry/description 0 0% 8 Collection registry/description 1 1%
GBIF 1 Observations (systematic) 27 13% 1 Specimens 27 12%
GBIF 1 Observations (opportunistic) 27 13% 1 Biol. taxonomy/classification 27 12%
GBIF 1 Collection registry/description 27 13% 1 Collection registry/description 27 12%
GBIF 1 Specimens 27 13% 1 Observations (systematic) 27 12%
GBIF 1 Biol. taxonomy/classification 27 13% 1 Observations (opportunistic) 27 12%
GBIF 2 Biol. descriptions/traits 25 12% 2 Molecular 25 11%
GBIF 2 Literature 25 12% 2 Literature 25 11%
GBIF 2 Molecular 25 12% 2 Biol. descriptions/traits 25 11%
GBIF 3 Geology 4 2% 3 Geology 12 5%
GeoCASe 1 Geology 24 38% 1 Geology 26 32%
GeoCASe 2 Specimens 16 25% 2 Specimens 18 22%
GeoCASe 3 Biol. taxonomy/classification 15 23% 3 Biol. descriptions/traits 17 21%
GeoCASe 4 Biol. descriptions/traits 8 13% 4 Biol. taxonomy/classification 16 20%
GeoCASe 5 Collection registry/description 1 2% 5 Collection registry/description 3 4%
GeoCASe Observations (systematic) 0 0% 6 Literature 2 2%
GeoCASe Literature 0 0% Observations (opportunistic) 0 0%
GeoCASe Molecular 0 0% Molecular 0 0%
GeoCASe Observations (opportunistic) 0 0% Observations (systematic) 0 0%
iBOL 1 Molecular 34 30% 1 Molecular 34 28%
iBOL 2 Observations (systematic) 28 25% 2 Biol. taxonomy/classification 28 23%
iBOL 3 Biol. taxonomy/classification 27 24% 2 Observations (systematic) 28 23%
iBOL 4 Specimens 23 21% 3 Specimens 23 19%
iBOL Biol. descriptions/traits 0 0% 4 Biol. descriptions/traits 7 6%
iBOL Observations (opportunistic) 0 0% Geology 0 0%
iBOL Collection registry/description 0 0% Literature 0 0%
iBOL Geology 0 0% Observations (opportunistic) 0 0%
iBOL Literature 0 0% Collection registry/description 0 0%
iNaturalist 1 Observations (opportunistic) 21 28% 1 Biol. descriptions/traits 21 28%
iNaturalist 1 Biol. descriptions/traits 21 28% 2 Observations (opportunistic) 21 28%
iNaturalist 2 Biol. taxonomy/classification 19 25% 3 Biol. taxonomy/classification 19 25%
iNaturalist 3 Observations (systematic) 12 16% 4 Observations (systematic) 12 16%
iNaturalist 4 Molecular 2 3% 5 Molecular 2 3%
iNaturalist Specimens 0 0% Geology 0 0%
iNaturalist Collection registry/description 0 0% Literature 0 0%
iNaturalist Geology 0 0% Specimens 0 0%
iNaturalist Literature 0 0% Collection registry/description 0 0%
LifeWatch 1 Biol. taxonomy/classification 35 12% 1 Biol. descriptions/traits 35 12%
LifeWatch 1 Biol. descriptions/traits 35 12% 1 Biol. taxonomy/classification 35 12%
LifeWatch 2 Observations (opportunistic) 34 12% 2 Observations (systematic) 34 12%
LifeWatch 2 Observations (systematic) 34 12% 2 Observations (opportunistic) 34 12%
LifeWatch 3 Geology 33 12% 3 Geology 33 12%
LifeWatch 4 Molecular 31 11% 4 Molecular 31 11%
LifeWatch 5 Literature 30 11% 5 Literature 30 11%
LifeWatch 6 Specimens 27 9% 6 Collection registry/description 27 9%
LifeWatch 6 Collection registry/description 27 9% 6 Specimens 27 9%
Figure 2.  

Activity counts with a Maturity Index of 2 and above for each infrastructure within each scope.

Biodiversity Heritage Library

The Biodiversity Heritage Library is a worldwide consortium and aims to make biodiversity literature openly available through digitisation. This is reflected in their scoring in the contact zones analysis (Fig. 2). Most of their current activities at a Maturity Index of 2 and above (P2 and above) are within the Literature scope (62%, 29 activities) and this remains the focus of BHL's future ambitions.

Catalogue of Life

The mission of the Catalogue of Life is to provide a freely accessible list of species across all taxonomic groups. It currently has a tight remit, with P2 and above activities within two scopes: Biological taxonomy/classification (64%, 27 activities) and Literature (36%, 15 activities) (Fig. 2). Catalogue of Life has ambitions to slightly broaden this scope, with some presence in all scope areas apart from Geology and aims to increase its activity within Literature (from 15 activities to 22).

DiSSCo

DiSSCo is a new European research infrastructure for natural science collections, aiming to digitally unify European natural science assets (e.g. specimen collections) through their digitisation. It is currently within its preparatory phase, via the DiSSCo Prepare project and this is reflected in the low number of activities currently rated at P2 and above (11 activities), with most of these in the Specimens scope (91%, 10 activities) (Fig. 2). In future, DiSSCo aims to dramatically increase its P2 and above activities from 11 to 137, including the scope of its activities on Specimens (22%, 30 activities), Biological Taxonomy/Classification (15%, 20 activities) and Collection Registry/Description (14%, 19 activities).

ELIXIR

ELIXIR aims to coordinate and develop life science resources in Europe, with a particular focus on molecular/genomic bioinformatics resources. It has P2 and above activities in most of the scope areas, including Molecular (26%, 35 activities), Biological Taxonomy/Classification (18%, 25 activities) and Biological Descriptions/Traits (17%, 23 activities) (Fig. 2). ELIXIR aims to keep this broad remit in future, with only a slight increase in the activities in which it has at least a presence - from 135 to 148.

eLTER

eLTER is a new European research infrastructure in its preparatory phase of development. It aims to improve the scientific understanding of terrestrial, freshwater and transitional water ecosystems through a socio-ecological approach to studying these systems. eLTER's current activities are highest in Literature (25%, 23 activities) and Observations (Systematic) (22%, 20 activities) (Fig. 2). In the future, Observations (Systematic) remains important (18, 34 activities), but there is an increasing focus on Geology (from 13 to 34 activities). eLTER will continue to have many P2 and above activities within the context of Literature (17%, 33 activities) and will also increase its activities within the Observations (Opportunistic) scope (31 activities, 16%).

GBIF

GBIF is a global network and data infrastructure that provides open access to data about life on Earth, as well as common standards and open-source tools to enable the sharing of information about where species have been recorded. It currently has a large number of activities at P2 and above (214 out of 315 possible activities) (Fig. 2). GBIF has the least concentration of activities within the scope of Geology. GBIF plans to continue this spread of activities in future and aims to increase its presence in Geology (from 4 to 12 activities).

GeoCASE

GeoCASE is designed to make data on collections of minerals, rocks, meteorites and fossils easily accessible online. In this regard, GeoCASE aims to be the Earth Science counterpart to GBIF. This mission is reflected in the P2 and above activities that GeoCASe have recorded in this dataset, with most of its ambition scores within the scope of Geology (28%, 24 activities), Specimens (25%, 16 activities) and Biological Taxonomy/Classification (23%, 15 activities) (Fig. 2). GeoCASE plans to maintain its presence in these areas, as well as increase activity within Biological Description/Traits (from 8 to 16 activities).

iBOL

iBOL is a global research alliance that builds DNA barcode reference libraries, sequencing facilities and informatics platforms with the aim to discover and identify multicellular life. iBOL’s current P2 and above activities are within four scope areas: Molecular (30%, 34 activities), Observations (systematic) (25%, 28 activities), Biological Taxonomy/Classification (24%, 27 activities) and Specimens (21%, 23 activities) (Fig. 2). This continues in future, with not much change in the activities it aims to be at P2 and above, although it does aim to slightly expand into the Biological Description/Traits scope (7 activities).

iNaturalist

iNaturalist allows naturalists and citizen scientists to record their observations of biodiversity via mobile apps or through their website, with their research-grade findings shared with GBIF. The majority of iNaturalist’s P2 and above activities currently focused on Observations (Opportunistic) (28%, 21 activities), Biological Descriptions/Traits (28%, 21 activities) and Biological Taxonomy/Classification (25%, 19 activities) (Fig. 2). iNaturalist is a well-established infrastructure with a relatively narrow and distinct niche and does not aim to widen the breadth of its P2 and above activities in future.

LifeWatch

LifeWatch is a European Research Infrastructure Consortium (ERIC) that provides e-services to biodiversity and ecosystem researchers, helping to address planetary challenges. LifeWatch currently has a P2 and above in most activities relevant to the biodiversity informatics domain, with a presence in most activities in every scope (286 out of 315 activities). Although LifeWatch does not plan to significantly increase this breadth in the future (Fig. 2), this survey was completed before the development of a new five-year Strategic Working Plan, launched in June 2022.

Measuring ambition: how the development of these infrastructures will change the biodiversity informatics landscape

Fig. 3 shows the current activity levels and the future ambitions of each infrastructure, with a count of the number of activities each infrastructure has at Maturity Index Level of P2 (presence) and above. GBIF and LifeWatch have the highest number of activities and could be considered generalists, with both showing a presence in over 200 current activities. iBOL and ELIXIR both have a presence of over 100 activities, with a degree of specialisation, with DiSSCo and eLTER planning to fall within a similar range of activities as they deliver on their development roadmaps. BHL, CoL, GeoCASe and iNaturalist are much more specialised, each operating with a narrow focus of activities (all below 85 activities).

Figure 3.  

Change in Total Activities (Maturity Index 2 and above) by Infrastructure.

DiSSCo and eLTER are both new infrastructures in the early stages of development and show the greatest difference in activity between current levels and future ambition. Two of the more specialist infrastructures, GeoCASe and Catalogue of Life, also have proportionally ambitious plans to expand their activities compared to other infrastructures. The introduction of two new infrastructures, which aim to actively expand their activity levels, will likely result in a changing dynamic from the current landscape and require new collaborations and coordination. Consideration of existing mechanisms of collaboration with specialist organisations like GeoCASe and Catalogue of Life will also be beneficial as they start to broaden their activities.

An analysis of the current and future activities within each scope shows the changing nature of the research infrastructure landscape (Fig. 4). Observations (both opportunistic and systematic) and Geology are the scopes with the lowest current activity levels and this is likely to remain the case in future. Biological Taxonomy/Classification has the highest activity levels, both now and in future. The highest increase in activity can be found in Biological Description/Traits and there may be a need to strengthen collaboration between infrastructures in this space going forward.

Figure 4.  

Change in the total activities (Maturity Index P2 and above) across all infrastructures from current to ambition.

It is also possible to look at how the landscape plans to shift in future and whether there is an overall increase in maturity levels in the activities for each scope. All scopes show an increase in the number of activities that will be at the P3 'Performance' and P4 'Predominance' levels in future (Fig. 5). Within the Geology and Biological Descriptions/Traits scope, there is also an increase in activities rated at 'P2 - Presence'. This is due to these two scopes being relatively immature in comparison to other scope areas, with more potential activities that are currently absent (P0).

Figure 5.  

Change in the count of activities within each scope rated at Maturity Index Level between current and ambition.

There is a notable shift in Observations (Systematic) with activities moving from P2 'Presence' to P4 'Predominance' and the Molecular scope also showing a large movement towards activities rated P4 'Predominance'. This change is primarily through an increase in activity by a small number of infrastructures (Fig. 6) and collaboration in this space between these infrastructures would likely be beneficial. A large number of infrastructures have moved to P3 'Performance' within Biological Descriptions/Traits and, as mentioned above, there is a high increase in activity within this scope. This is likely to be an area where collaboration and cooperation between the majority of infrastructures will be required to ensure alignment and synergy in activity.

Figure 6.  

Change in the count of infrastructures within each scope by Maturity Index Level between current and ambition.

Future perspectives

This dataset and methodology for quantitatively assessing present and planned infrastructure activities hold considerable promise to support cooperation and planning amongst biodiversity informatics research infrastructures. In the first instance, expanding the dataset by adding closely related infrastructures and networks, such as TDWG (Biodiversity Information Standards; https://www.tdwg.org/), MIRRI (Microbial Resource Research Infrastructure; https://www.mirri.org/), iDigBio (Integrated Digitized Biocollections; https://www.idigbio.org/) and ALA (Atlas of Living Australia; https://www.ala.org.au/), would be beneficial to provide a more complete picture of the biodiversity informatics landscape. A more complete assessment of eligible infrastructures might draw on recent reviews of the biodiversity informatics domain, starting with the requirements set out in the OECD Megascience report from 1999 on biological informatics (OECD 1999).

Another limitation of our approach is that the methodology is dependent on self-assessments that were only validated and reviewed by the survey team. This could be improved through a larger community survey by asking more independent network stakeholders to assess a research infrastructure's coverage and maturity and by comparing these results with the self-reported scores of the research infrastructure. In the longer term, more automated methodologies, such as those used by various FAIR-metrics working groups within the Research Data Alliance (https://www.rd-alliance.org/), European Open Science Cloud (https://eosc.eu/) and Go-FAIR (https://www.go-fair.org/) communities, might provide inspiration for building more objective evaluation criteria.

More refined data visualisations, dynamically constructed off of a growing dataset of research infrastructures would also be useful to support the strategic development of service provision by these infrastructures, as well as identifying gaps in the landscape. Further generalisation of the method, including an expansion of the Scope, Phase and Category terms to encompass activities in other domains beyond biodiversity informatics, has the potential to broaden the application of this approach, potentially providing an evidence base when considering strategic investments in a much wider range of research infrastructures. For example, this approach has the potential to support investment decisions by funders (e.g. ESFRI, the European Strategy Forum on Research Infrastructures; https://www.esfri.eu/), which is a strategic instrument used in Europe to develop the scientific integration of research infrastructures. The dataset and tool also have potential for associated infrastructures like EOSC, the European Open Science Cloud (https://eosc-portal.eu/). EOSC's efforts to address the cloud-computing need of other infrastructures, may benefit from a deeper understanding of the current and future of potential user communities when planning targetting application of their services.

Acknowledgements

The authors would like to thank Ken-ichi Ueda of the California Academy of Sciences, who scored the iNaturalist dataset; Mark Frenzel of the Helmholtz Centre for Environmental Research (UFZ) who scored the eLTER dataset; and Jerry Lanfear, Chief Technical Officer at ELIXIR for the ELIXIR scores.

Author contributions

Contribution types are drawn from CRediT - Contributor Roles Taxonomy

Conceptualisation: Leif Schulman, Vincent Smith, Matt Woodburn

Data curation: Sarah Vincent, Matt Woodburn

Formal Analysis: Sarah Vincent

Investigation: Lisa French, Vincent Smith, Sarah Vincent, Matt Woodburn

Methodology: Vincent Smith, Matt Woodburn

Project administration: Lisa French, Aino Juslén, Vincent Smith

Supervision: Aino Juslén, Vincent Smith

Validation: Wouter Addink, Christos Arvanitidis, Olaf Bánki, Falko Glöckler, Donald Hobern, Martin Kalfatovic, Joe Miller, Vincent Smith

Visualisation: Sarah Vincent

Writing - original draft: Lisa French, Vincent Smith

Writing - review & editing: Wouter Addink, Christos Arvanitidis, Olaf Bánki, Ana Casino, François Dusoulier, Falko Glöckler, Donald Hobern, Aino Juslén, Martin Kalfatovic, Dimitrios Koureas, Patricia Mergen, Joe Miller, Leif Schulman, Vincent Smith, Sarah Vincent, Matt Woodburn

References

login to comment