Empowering Biodiversity Research III

• Damiano Oldoni

biodiversity Naturalis Netherlands Leiden EBR
Image by Dimitri Brosens
On this page

General

These notes are a very first DRAFT. A review will occurr soon.

Session 1 - Keynotes

Policy & Biodiversity: Research needs for EU nature policy and legislation

Presenter: Frank Vassen (European Commission DG-ENV), DG ENV.D3, Nature Protection Unit

State of Nature report every 6 years. The last one was release on 2019 and covers the period 2013-2018. Key message:

  • only 1% of habitats show a good conservation status
  • Strongest deteriorating trends: grasslands, dunes, and bog, mire and fen habitats

Key pressures on habitats and species (SoN):

  • urbanisation
  • pollution
  • hunting, illegal killing, poisoning
  • Invasive alien species
  • modification of hydrological flow regimes
  • collissions with electriicity
  • climate change: expected to become worse in next report

Natura 2000 netowrk:

  • well connected network of sites
  • 27027 sites
  • 18.6% of EU27 land area
  • large variation of coverage: from 8.3% in Denmark up to 37.9% in Slovenia
  • More than 8% of EU marine area
  • Target: 10% of such areas should be fully dedicated to biodiversity conservation

A typical problem of Central Europe NAtura2000 areas is fragmentation: one area is splitted in isolated small areas, far too small.

Key questions:

  • whatare the key pressures acting on species and habitats to reduce them?
  • how to achieve recovery?
  • How to ensure that Natura 2000 remains fit for purpose? What are the priorities?
  • How to improve the data quality on status and trends of species and habitats?

As typically in EU, there are political controversies?

  • What is driving the insect decline and the ground nesting birds?
  • What is the both direct and indirect impact of nitrogen deposition on habitats and species?
  • What is the impact of renewable energy infrastructures, marine bycatch and bottom trawling?
  • Invasive alien species: could there be novel approaches to fight them?
  • Does the Natura2000 concept work?
  • Is improvement of connectivity a really priority?
  • How much habitat is needed and where? How to define the term favourable conservation status?
  • How to deal with conflicting conservation objectives? How to set the right priorities?

We need evidence-based decision making.

Important statement: It requires more field experiments, less modelling approaches. As expected this sentence arose a question about. The speaker would like to stress that we need to find back a balance as we are now shifted to modelling way much more than data retrieving.

The definiction of broad research topics is unlikely to generate projects tthat reply to specific policy needs.

Exampel of a project creating evidence: Eurokite Life project. A project following red kites with telemetry from 2013 to 2023.

It shows that mortality due to wind farms or electricty infrastructures is way lower than expected. This research was important because in Germany the building of wind farms was blocked by bird mortality issues.

GBIF’s role in empowering biodiversity research and policy - Tim Hirsch (GBIF) keynote

Presenter: Tim Hirsch (Deputy Director, GBIF) Abstract: html

What is GBIF? Short definition: GBIF is a Global Core Biodata Resource.

About empowering biodiversity research: more than 10k scientific papers have been enabled by GBIF mediated data. 1782 papers in 2023. ~300 more than 2022. Citations are mentioned in data publishers pages via DOI system. This helps acknoledging the importance of data publishers in the research process.

There are also the Science Review GBIF publications, a proactive way of communicating research outputs by GBIF itself.

About policy, one of the 4 GBIF Strategic Framework pillars is called “Supporting policy responses and nkowlledge transfer that address urgent societal challengesaround planetary change”.

GBIF launched a Call for proposals for a systematic review of GBIF-enabled research, a fixed-ter assignment to outline and analyse patterns and trends in use of GBIF-mediated data in peer-reviewed literature.

This will help GBIF to understand how researchers use GBIF mediated data.

But why to invest on a global netowrk as GBIF? The economic valuation and assessment of the impact of the GBIF network was studied by Deloitte Access Economics. It finds that every €1 invested in global network, infrastructure and services accrues €3 in benefits to users and up to €12 to society.

What are the indirect dependencies on GIBF as foundation of policy uses? It has to be found on initiatives like:

  • IUCN: Species occurrene data required to support Red List of Threatened Species, Red List of Ecosystems.
  • GEOBON: primary data via GBIF needed to generate esserntial biodiversity variabls, data cubes.
  • TNFD: underlying data to support disclosure of business riskes, impacts and dependencies on biodiversity

GBIF is committed to the Global Biodiversity Framework. THere are some indicators where GBIF is specifically mentioned:

  • A.4
  • 6.1
  • 15.1
  • 21.1

ABout IAS indicator, these data sources are relevant:

  • national GRIIS checklists published throuch GBIF
  • Alien Species First records database (Hanno Seeens) on zenodo
  • GBIF occurrences of IAS species

GBIF contributes also to research for several Sustainable Development goals.

GBIF supports open science, see UNESCO recommendation on Open Science. GBIF has a partnership with UNESCO.

GBIF started also a closer collaboration with the Ocean Biodiversity Information System (OBIS) to support ocean science and policy.

BirdWatch – a satellite-supported service to monitor the habitat suitability of agricultural land and to evaluate the impact of agri-environmental policies on farmland birds

Presenter: Nastasja Scholz (LuP – Luftbild Umwelt Planung) keynote Abstract: html

Coordinator Horizon Project BirdWatch, a satellite-supported service to monitor the habitat suitability of agricultural land and to evaluate the impact of agri-envionrmental policies on farmlands birds.

About Belgian partners: vito and INBO. INBO is scientific advisor, vito is optimization and proposals for actions.

University Potsdam is doing species distribution modelling. Bioland is the connection point with farmers.

eurac research (Bolzano) and Sinergise (Slovenia) are involved in developing the geospatial featuers and database.

Motivations:

  • EU policies such as the Common Agricultural policy (CAP), Biodiversity, nature restoration law
  • Need for monitoring service
  • evaluation of actions
  • efficiency and impact
  • better planning process

Concept:

  • use of geospatial data (landscape parcel information, remote sensing data, parcel boundary data, …)
  • species distribution modeling
  • optimization
  • online platform

First users group was farmers, but after all commotions in the last months, now the platform’s main users group is policy makers and ngos.

Four test regions:

  • Flanders
  • Germany
  • Lithuania
  • South Tirol

The tool starts by allowing the user to choose the bird species. 10 bird species are allowed. These 10 species have been chosen as they are

  • indicator species
  • farmland birds index
  • Part of an established monitoring shceme
  • obsrevation data aivalable
  • different habitats requirements, foraging habitats
  • from commong to endangered
  • frommo to region-specific

Based on the above criteria, they could select 10 species from the 39 species classified as farmland species on EU level (Pan-European Common Bird Monitoring Scheme, PECBMS) up to 10 chosen birds

Among the existing Monitoring Schemes, they took into account the following from Belgium Common Breeding Birds Survey (BE) SOCBRU

About Environmental data: open data only, 10m resolution. Easily maintenance, low budget requirements.

Type of environmental data used:

  • crop types
  • Spectral information
  • radar data
  • green volume
  • Land cover & land use
  • soil moisture
  • parcel descriptors
  • Mowing cycles

These type of data are included in Specise Distribution Models to assess the importance of environmental components in predicting the suitability area of each bird.

vito is developing an algorithm to see whether the models are functioning.

The final goal was directed to farmers: what is the current state of habitat suitability on py parcels? While, for policy makers: how has habitat suitability of a region developerd over the years? What can be improved?

Mapping the Research Infrastructure landscape: the role of Biodiversa+

Prsenter: Rob Hendriks (Ministry of Agriculture, Nature and Food Quality, LNV) Abstract: html

The Biodiversa+ partnership in short:

  • 40 countries
  • 81 partners

  • 800M € over 7 years

What does Biodiversa+?

  • Promote and support Research & Innovation (R&I) programs and projects
  • Better connect R&I programmes and projects to policy
  • Promote and support Nature-based Solutions and valuation of biodiversity in private sectors

The Biodiversa+ Strategic Rsearch & Innovation Agenda is focussed on three topic themes:

  • Biodiversity protection and restoration
  • Transformative chnange
  • EU’s global action

Six calls are planned:

  • 2021-2022 call: protection of bioserivsity and ecosystems
  • 2022-2023: improved transnational biodiversity monitoring. 46 fouding organisations from 33 coutnires, 33 projects founded
  • 2023-24: Nature based oslutions
  • upcoming calls: Societal transofmation and on Future ecosystems/Restoration

Strenghtening internation collaborations, such as IPBES. Fostering the engagment with Research Infrastrucutres: survey the use of EU and global research infrastructure and Earth Observaiont Programmes (GBIF, ILTER, Copernicus)

Out of the survey it seems that GBIF, Copernicus, EuropaBON and GEBON are the most helpful infrastructures for biodiversa+ projects.

On the other hand, Biodiversa+ can play a role in improving the use of R&I, e.g. by increasing the findability and the availability of GBIF data. Also via the capacity building on open data and Dariwn core.

Session 2 - EBR 1

How GBIF will facilitate more varied types of data in the future

Presenter: John Wieczorek (University of Berkeley) & Tim Robertson (GBIF) - remote presentation

I will not write much about here.

See blogpost “Diversifying the GBIF Data Model”.

See webinar “Introduction to the GBIF new data model”.

See on the blogpost above links to other webinars specific for some use cases.

Last news: the ratification of the Humboldt Extension for Ecological Inventories.

Biodiversity Building Blocks for Policy

Presenter: Quentin Groom (Meise Botanic Garden)

Project website.

As part of the project, no notes are taken as nothing new has been said.

Interesting question about cloud computing platform: how to deal with big tech monopolium of cloud computing infrastructure? Still work to do. Note there is a project called EOSC infrastructure to make EU research independent from Google? Amazon, etc.

MIRRI-ERIC: taping the hidden treasure of data in culture collections

Prsenter: Ana Portugal Melo (MIRRI-ERIC)

MIRRI - ERIC stays for Microbial Resource Research Infrastructure - European Research Infrastructure Consortium.

6 member states: Belgium, Portugal, Greece, France, Latvia, Spain. Romania will be likely the 7th country.

Paths:

  • Strengthen the MIRRI-IS infrastrcuture Makes avaiable datadata-enriched microbial resources to academia and industry.
  • Capacity in data management.

Goal: valorisation of microbial biodiversity by preserving, charactersing taxonomic complexity, annotating data, promoting systematic investigation and service development.

Modalities: by facilitating access to the broadest range of high-quality microoranisms

Large-scale biodiversity assessment with eDNA and the trouble of entropy: ways forward

Presenter: Florian Leese (UNI-DUE) - remote presentation

“THe recovery of EU freshwater biodiversity has come to a halt”. To now why, we need first data, but not only data. Data to create knowledge.

While monitoring, we deal with a scaling issue: from virus to ecosystems.

DNA starts to play a role, but its importance doesn’t progress as expected.

DNA research:

  • comprehensive (from individuals, popluations, to communities)
  • scalable (100s samples in aweek), cost-effective!
  • many more organisms to species level identified
  • some false negatives and false positives
  • compatible with standard monitoring approaches

Example: Rainwash eDNA matabarcoding - canopy monitoring

You can also study biodiversity trends via archival samples.

The problem, now: lack of guidance, standardization

Example: the performance of labs differed, one key aspect: sequencing depth was not the same among labs.

We need minimum tehcnical requirements and quality control for DNA and eDNA data. Central push & funding for standards (CEN, ISO, TDWG) and guidance is needed.

Agouti: Camera Traps, from Wildlife Research to Wildlife Management & Policy

Presenter: Martijn Bollen (INBO)

Camera traps are a versatile monitoring tool: they are getting cheaper and cheaper.

Some pros:

  • non invasive
  • labour extensive
  • easy to standardize
  • well documented obs
  • operaiontl day/night

Agouti: storage, management

Functionalities:

  • import
  • annotate (with AI models)
  • analyze
  • export

After calibration using sticks with some detection points, we can calibrate cameras to calculate radius, angle and so the speed of the animal.

Via camtrap DP we have an effective data exchange format, we have also software to analyze data. We have models, e.g. Random Encounter Model which uses the speed of the detected animals to assess population densities.

So, now, the big question is “How to inform management and policy?”

Case studies: from wildlife research to wildlife management and policy

  1. Wild boar damage crops during the usmmer. ALternatives: enclose wild board in the centre of of the forest. Location: Meerdaal
  2. Wild board expand geographically beyond large forest complex. Alternatives: gain understanding of their colonization patterns. Location: Meerdal and broad surrounding

Hunting frequency high at the border, low in the core of the forest. Hypothesis: wild board tends to avoid human presence due to the applied hunting strategy.

Wild board presence expand geographically.

Session 3

Extended COL as GBIF backbone taxonomy

Presenter: Olaf Banki (Species2000)

2.1M +

Acknowledged as Global Core Biodata Resource.

ChecklistBank is a taxonomic data infrastructure for everyone

The main checklist is the Catalogue of Life. The checklists are

GBIF Backbone is an automated process. It is based on COL checklist.

We want to produce a more comprehensive checklist that can replace the GBIF Backbone.

What’s wrong with CoL? We lack some names, we lack taxonomic groups.

The idea is to inject new regional/management checklists to fill the gaps. The final output is a release and it is used by GBIF.

How to select sources?

  • License CC BY or CC0
  • Data quality: <40% issues level, such as unparasable authorship, inconsistent authorship, etc;
  • High content of genus, species or synonyms

Organization of additional sources for insertion:

  • global
  • taxonomic scope
  • global management scope
  • regional or national
  • non-linnean names
  • massive sources

PReliminary results:

  • 8M+ names
  • 40k checklsits

In 2024:

  • checklist in production
  • improving inclusion of data sources: impact on GBIF occurrrences and quality checks
  • GBIF making use of the new COL checklist version (end of 2024, probably 2025)

Presenter: Niels Raes (NLBIF)

One of the GBIF user domains is policy.

The private sector has started to deal with biodiversity and biodiversity data. The reason: Nature Restoration Law wants to push private funding.

TNFD: Taskforce on Natue related Financial Disclosures: it assesses, reports and acts on their nature releated depenrdencies impacts risks and opportunities.

Naturalis biodiversity Center, NLBIF and KPMG are collaborating to implement a new tool: THRIVE, Toolset for hierarrchical reporting and insightful Validation of Ecosystems.

The BiCIKL project traverses obstacles to FAIR and linked biodiversity data usage

Presenter: Iva Boyadzhieva (Pensoft) Abstract: html

BiCIKL

Biodiversity Community Integrated Knowledge Library

The project partners are data and data infrastructure providers. Type of data:

  • primary data (GBIF, Catalogue of Life)
  • molecular data
  • literature (Plazi, Pensoft, zenodo)

Transferring Taxonomic Knowledge through TETTRIs

Presenter: Ana Protugal Melo (RBINS) Abstract: html

TETTRIs

CETAF is a project coordinated by CETAF. It’s the first time CETAF coordinates a project.

Important Working Packages (WPs): the Taxonomic Knowledge Transfer Forum (WP8) and Stake Holder Labs (WP5), ending up in Policy Brief (WP8) and deliverables, blue print, …

TETTRIs tool: Marketplace of taxonomic resources and expertise, a centralised platform for a dynamic catalogue.

Session 3.2: MAMBO

Chair: Vincent Kalkman

Image and sound recognition for citizen scientists: current taxonomic and geographic coverage

Presenter: Vincent Kalkman (Naturalis Biodiversity Center)

Naturalis is involved in MAMBO. See Naturalis’ project page.

Tehcnology allowed to hugely increase amount of biodiversity data.

This talk speaks about the image soudn recognition behind obsidentify

Sound and image recognition allowed citizens to identify species.

What do you need:

  • smart people

  • connection with citizens

  • training data (at least 10 images, at least 5 sound samples)
  • 43M million images of species
  • 45k taxa

xeno-canto used for harvesting training sound data. Bird: model available, almost all breeding birds of Europe. Frogs and insects: work in progress

Absences due to:

  • some species are practically impossible to be distinguished by images/sound
  • lack of people/time to validate data

Actions for sound:

  • mobilize additional existing sound sources
  • can we train people to become validators of insects?

How remotely sensed point clouds can help deliver habitat condition metrics – some MAMBO trials - France

Presenter: France Gerard (UK Centre for Ecology & Hydrology)

Insect camera’s, counting insect the digital way

Presenter: Toke Høye (Aarhus University)

There is a great demand for monitoring data.

Image recognition has the potential to be cheap and fast. Time lapse cameras could replace humans for plot level observations.

Focal scenes could be:

  • biological resources: flowers, dung, fungi
  • artifical resources: light, artificial flowers, sticky traps
  • location: ant paths, perches of dragonflies

Focus of this research project is the pollinator projects.

New open acecss dataset of almost 30k common insects against complex vegetation background.

We need to not count multiple times the same individual. The image recognition needs to track the insect.

Standardizin pollinator counts in images:

  • duration of each insect visit varies greatly
  • flower preference varies among pollinator species
  • flower cover/avundance is an imprtantn metric

Opportunities:

  • phenology: connecting plant phenological variation to reproductive success
  • functional ecology: asessing thermal performance of non model organisms
  • species interactions
  • ecological monitoring

take home messages

camera traps can capture information about the abundance and seasonal dynamic of insects

  • species indetification from is doable but hcallenging
  • standardised data can be collected across may istes
  • many challenges ahead

Session 3.3: LifeWatch

Chair: Julien Radoux

Species information backbone

Presenter: Stefanie Dekeyzer (VLIZ) Abstract: html

Goals:

  • facilitate the standardization of species data
  • virtually bring together different component data bases & data systems

Coordination:

  1. System architecture & del
  2. data-management
  3. Community support

lifewatch e-lab: facilitate working with all kind of data.

The rest of the presentation was focussing on World Register of Marine Species (WoRMS), the European node of the Ocean Biodiversity Information System (EurOBIS) and the Marine Regions gazetteer, and the important outputs VLIZ is getting out of their tools. Unfortunately, no mentions of the LifeWatch Species Backbone anymore. Is this LifeWatch backbone still in development? Does it work and has been used?

VII World Conference on Marine Biodiversity 2026 - WCMB 2026 will be organised in Bruges, 16-20 November.

LTER-LIFE: Building Digital twins from data pipeline to ecosystem insights

Presenter: Geerten Hengeveld (NIOO-KNAW) Abstract: html

Integration of biodiversity monitoring data into the Digital Twin Ocean (DTO-BioFlow)

Presenter: Carlota Muñiz VLIZ) Abstract: html

Terrestrial ecosystem accounting with ecopatches

Presenter: Julien Radoux (UCL)

The tool maintained by Julien is online: https://maps.elie.ucl.ac.be/lifewatch/ecopatches.html.

Session 3.4: ARISE

Recognizing all species in all loactions through DNA and AI

Presenter: Elaine van Ommen Kloeke (Naturalis Biodiversity Center)

Arise project on Naturalis Biodiversity Center page.

Science project fiche on Naturalis page.

ARISE is a 18M research infrastructure project.

Its goal is ambitious: identifying and monitoring all multicellular species in the Netherlands.

How to collaborate more?

Impressions: no real belief on open data and open science. About camera trap, is there a link with Agouti and Camera Trap Data Package data standard?

Session 4.1: DISSCo

Chair: Frederik Leliaert

DiSSCo - European & technical perspective

Presenter: Wouter Addink (Naturalis Biodiversity Center)

Project fiche on Naturalis webpage.

keywords: unified, sustainable, links.

2018-2022: preparatory Phase to improve the overall implementation and readiness level of DiSSCo. 2023-204: Transition Phase 2025- : Construction Phase

ERIC roadmap Funders Forum: spring 2024 tehcand scientific ddescription and contribution model, members call. Now: Call for Hosting and Founding members

DiSSCo is a service-oriendented infrastructure:

  • e-Science services
  • Physical and virtual acess to collections

End user services: 17 experessions of interest to become a service provider. DISCCOver was the very first service

Digital Specimen Architecture (DSArch)

Check the services via https://sandbox.dissco.tech.

Data model based on GBIF UM with mappings to DwC and ABCD.

CANATHIST (Decolonisation) & DiSSCo FED: contributions to DiSSCo.eu

Presenter: Patrick Semal (RBINS)

CANATHIST

MapOCR: python script for georeferencing.

Share media: to do via ORTHANC, a open tool originally developed for medical science.

GBIF hosted portals for DiSSCo nodes, and university collections use cases

Presenter: Maarten Trekels (Meise Botanic Garden)

Alternative title: Increase collaboration between infrastructures

DiSSCo Flanders is a FWO project to support DiSSCo in Flanders.

Some key topics:

  • Inventory and assessment: Creating an overview of content and state of the collections
  • Best preactices and guidelins: DNA collection working group. This was created due to the presence of DNA collections at INBO.

Knowledge re-use and building:

  • based on an exercise in SYNTHESYS+ (a previous project)
  • mapped all data to the Latimer Core standard
  • creating good practices guides, e.g. about DNA collections

Some institues have already a collection management system, so we do not force to change their resources. SiSSCo Flanders tries to harmonize existing infrastructures where present.

Example: interaction among DOEDAT, GBIF, BiCIKL and DiSSCo.

Invest time and money in data standards (collaboration with TDWG) and new technologies, e.g. AI for authomatic image recognition for annotations and specimen.

Stress on using existing (local) data platforms; e.g. Botanical Collectionsand Blendeff (KUL).

Working with ELIXIR for running workflows via Galaxy.

But how to integrate the datasets from all the different sources mentioned? Push all data on GBIF and create a GBIF hosted portal for these data. Easy to maintain with basic IT skills: it’s a GitHub repo, hp-dissco-flanders. The output is the DissCo Flanders website.

We are not alone: especially for Flanders, collaborations with CLARIAH-VL and LifeWatch Belgium are needed. Easy to edit and maintain, integration with sources such as GRSciColl.

Session 4.2: ELIXIR

Chair: Gabriela Dankova

Exploration of (non-)model plant genomes using the PLAZA platform

Prsenter: Klaas Vandepoele (VIB-UGent Center for Plant Systems Biology)

We are at the beginning of a DNA sequence booming. Reason? Sequencing is getting cheaper and cheaper.

The basic genome analysis toolkit should:

  • genoem assembly
  • structural annotation: where cgenere are
  • functional annotation
  • data availability of genoeme sequences and gene annotation

Challenges:

  • How manage so many data and metadata?
  • How to integrate and compare all these genome sequences?

To deal with these challenges, PLAZA has been created many years ago. PLAZA is an highly integrated platform to translate knowledge from model species to crop.

Transform PLAZA from a platform to a technology. In this way you can choose to set up a PLAZA instance for the genome you want to work with. This has been done with ELIXIR founding. The result is FRINGE which stays for FRamework for INtegrative GEnomics.

  • setup an instance
  • nextflow pipeline

Check the PLAZA-cloud poster from the nextflow summit 2023, held in Barcelona.

Pan-genome toolkit: Pan-genomics, the pan-gene analysis is becoming the standard analysis.

How to asess genome dynamics? With Synteny Scores. It measure location variability, it is computed within context of a clade. Ideograms are used to show Synteny Scores.

Streamlining (Meta)Data Management and Submission to Repositories

Presenter: Flora D’Anna & Bert Droesbeke (VIB Data Core / ELIXIR Belgium)

Presentation focuses on the European Nucleotide Archive (ENA) data submission toolbox, one of the many services developed by ELIXIR.

This tool is a public and open-source platform. It aims to simplfy and enhance the submission process of genomic data to the ENA.

This tool has been already tested and it’s going to be extended and launched

ENA metadata templates are maintained on GitHub repo ENA-metadata-templates.

useGalaxy.be is the “Belgian Galaxy instance”. It is he public Galaxy instance of VIB Data Core and ELIXIR Belgium hosted at VSC (Flemish Supercomputer Centre) to enable accessible, reproducible and transparent computational research. It is used for running workflows.

DataHub: metadata management platform based on FAIRDOM-SEEK open source software Smaple metadata templates builder for promoting compliance to ENA standards andchecklist and ISA-JSON standard metadata framework

Export metadata in ISA-JSON format: machine actionable metadara carrier

MARS: Multi-omics adapter forrepository submissions. MARS will handle data metadata to submit to several data and metadata repositories depending.

Session 4.3: BGE

Workflows and standards delivering reference genomes for Biodiversity Genomics Europe

Presenter: Camila Mazzoni (Leibniz Institute for Zoo and Wildlife)

Achieving a step change in output and capacity building in DNA barcoding: the BIOSCAN stream of Biodiversity Genomics Europe”

Presenter: Rutger Vos (Naturalis Biodiversity Center)

Project fiche of the project.