GBIF4Ecology - the Global Biodiversity Information Facility

• INT58 • GBIF4Ecology – the Global Biodiversity Information Facility > D. Dmitry SCHIGEL
Content : GBIF (Global Biodiversity Information Facility) is in international research data infrastructure that mediates data on the occurrence of biodiversity. These data come from all over the world from publishers of data from museum collections, citizen science observations and machine generated data such as camera trap and environmental DNA. GBIF has been successful in mediating this occurrence – What is it? Where is it? When was it observed? – with over 2.2 billion occurrences which are documented to be used in the scientific literature.
After evaluating our place as a global research infrastructure and the expectations to accelerate data mobilization, capacity enhancement to improve science for research and policy relevance, GBIF developed a new strategic framework that specifically responds to the biodiversity crisis. In addition to the successful node-based data mobilization framework GBIF is developing thematic data use cases that will identify critical data gaps in areas of research and policy importance and work with new communities to mobilize this data into the GBIF network. These thematic topics will include area of interest to global policy initiatives, such as ecology, human health, invasive species, agricultural biodiversity, eDNA and others. Aligned with this work is the enhancement of the GBIF data model, based on DwC, to better shape data from diverse sources. The improved data model is based on the event rather than the occurrence which allows richer data integration such as allied species, interactions, measurements, and much more data that is commonly generated but not yet shared with GBIF.
We expect these advancements will encourage the integration of more and diverse data types that can be used to improve science for research and policy relevance of GBIF mediated data to better help the biodiversity community combat biodiversity loss. • INT59 • Introducing the GBIF global network and GBIF data use, tools and applications > A-S. Anne-Sophie ARCHAMBEAU
Content : This talk will introduce the GBIF global network and GBIF data use, tools and applications. Using the example of the French node, it will present how nodes interact with national research communities to help mobilize data and encourage the use of GBIF-mediated data. Some examples of how GBIF data availability advance ecological research will be presented. • INT60 • Natural history collections in their role of providing vouchered and georeferenced data sources to the GBIF network – a perspective from the GBIF Germany Node > D. Dagmar TRIEBEL
Content : Speaker: Dagmar Triebel1
Co-authors: Jörg Holetschek2, Tanja Weibulat1
Institution: 1)SNSB IT Center and Botanische Staatssammlung München/Germany, 2)Center for Biodiversity Informatics and Collection Data Integration (ZBS), Botanic Garden Berlin/Germany

Natural history collections (including herbaria) with their collection data were among the core drivers for establishing GBIF in 2001 (see https://www.gbif.org/what-is-gbif). Their recognized taxonomic expertise together with their sovereign task for long-term curation of physical specimens made them most suitable for mobilizing high-quality (multimedia) data as diverse digital collections. In addition, some of the large organizations have set up data repositories and recognized data centers for the long-term management and publication of structured primary biodiversity occurrence records and datasets. Natural history collections are important partners of the GBIF network and form a major stakeholder group among the 2,000 GBIF data publishers. They supply the majority of GBIF participant node staff and contribute to committees for expanding and adapting GBIF relevant community standards for data interoperability.
The profile of the GBIF Germany node (https://www.gbif.org/country/DE/summary) with charts on the GBIF Germany network, GBIF mediated data and numbers on structured occurrence data sources is discussed as example. Most datasets are offered under standard Creative Commons licenses. Georeferenced GBIF-mediated occurrences from collections represent a time-line of more than 200 years. They are backed by long-term stored and well-documented (historical) vouchers with stable object identifiers. Under certain regulations those physical specimens are available for future research studies. The German system of several GBIF subnodes relies on distributed data access with data pipelines agreed among collection data providers (see https://gfbio.biowikifarm.net/wiki/BioCASe_Pipeline). Data access via BioCASe Provider Software and ABCD zip archives was streamlined within the last decade. The German datasets with FAIR-compliant occurrences (GBIF data class 3) from collections are growing in number and amount of records. They are high quality data sources continuously enriched with standardized information from agreed terminologies and taxonomies and thereby appropriate for GBIF, but also for the German National Research Data Infrastructure (NFDI) and future ecological research. • INT61 • Collection data repository ready to identify and link DNA barcodes and metabarcoding data from biodiversity monitoring for GBIF > P. Peter GROBE
Content : Speaker: Peter Grobe1
Co-authors: Birgit Klasen1, Birgit Gemeinholzer2
Institutions: 1) Leibniz Institute for the Analysis of Biodiversity Change, Museum Koenig Bonn/Germany, 2) University Kassel/Germany

The ability of GBIF to represent complex data is increasing (https://doi.org/10.35035/doc-vf1a-nr22). A continuous improvement of the GBIF infrastructure makes it possible to store DNA-derived data from mass surveys in a retrievable way. This is an important feature for biodiversity monitoring for the analysis of biodiversity change. Here we need reliable data on the occurrence of organisms in space and time and their interactions over longer periods of time. The creation of such data is possible with monitoring stations equipped with different devices that provide a constant stream of data on movements (camera traps), sounds (acoustic sensors), volatile organic components (smellscapes) and through bulk collections (malaise traps, pollen collectors). While the analysis in terms of identification is now quite reliable via AI applications and thus can be performed in an automated manner to some extent, the latter method requires a great amount of manual laboratory work to extract species occurrence information from bulk insect and pollen samples (lit). The resulting data yield long tables with the sequence information and the space-time information for each occurrence of a MOTU (so-called ASV tables, https://doi.org/10.1038/ismej.2017.119). The German project "Automated Multisensor Stations for Monitoring of BioDiversity" (https://doi.org/10.1016/j.baae.2022.01.003) brings together and adapts novel techniques to automatically record species diversity. Using a combination of DarwinCore data and the DwC Extension for DNA derived Data (https://rs.gbif.org/extension/gbif/1.0/dna_derived_data_2021-07-05.xml), metabarcoding data, such as from bulk insect and pollen samples of the AMMOD project, can now be made available on GBIF.
We show here with an example data set of a pollen sample from the AMMOD project how this works, how services of established collection data repositories are involved and how the data can be retrieved on GBIF. The concepts presented are set up as use cases of the NFDI4Biodiversity (https://www.nfdi4biodiversity.org/en/) consortium. • INT62 • Roads and Politics: Quantifying geographic and political collection bias in GBIF-mediated data > A. Alexander ZIZKA
Content : GBIF-mediated data have become an irreplaceable cornerstone in ecological research. An important caveat is that the majority of data provided via GBIF were not collected systematically, but come from various sources. For instance, data from museum and herbarium collections and citizen science data contributed via iNaturalist, are based on different data types and were collected worldwide across different centuries. Hence, the availability of data differs strongly among localities and countries. Awareness of these systematic biases is essential when using GBIF-mediated data for research to obtain reliable results. Yet, these biases are easy to overlook and may be obscured by geographic scale.
Here, I briefly illustrate existing systematic biases in GBIF-mediated data caused by geographic accessibility and political conditions in countries worldwide. Furthermore, I present two open-source tools to explore these biases. SampBias (https://github.com/azizka/sampbias) is an R package to quantify the strength of bias caused by infrastructure facilitating human access (e.g., streets rivers). Bio-Dem (www.bio-dem.org) is a graphical user-interface web app to explore the link between distribution record availability and the political conditions worldwide. Data exploration using SampBias and Bio-Dem suggest that the strength of accessibility bias differs across infrastructure types, and that data availability is closely linked to the political situation in the country of collection, sometimes over decades, as in the case of former colonial ties. Accounting for sampling biases is not always possible when using GBIF-mediated data in ecological research, therefore thorough data exploration and understanding of the underlaying biasing processes are crucial, and possible even for large datasets. • INT63 • Chemical properties of key metabolites determine the global distribution of lichens > A. Andreas SCHWEIGER
Content : Speaker: Andreas H. Schweiger 1)
Co-authors: G. Matthias Ullmann2), Nicolai M. Nürk3), Dagmar Triebel4), Rainer Schobert5), Gerhard Rambold6)
Institutions: 1)Institute of Landscape and Plant Ecology, Department of Plant Ecology, University of Hohenheim/Germany
2)Computational Biochemistry, University of Bayreuth, Universitätsstraße 30, 95447 Bayreuth/Germany
3)Plant Systematics, University of Bayreuth/Germany
4)SNSB IT Center and Botanische Staatssammlung München/Germany
5)Organic Chemistry I, University of Bayreuth/Germany
6)Department of Mycology, University of Bayreuth/Germany

In lichen symbioses, fungal secondary metabolites provide UV protection on which certain lichen algae such as trebouxioid green algae sensitively depend. These metabolites differ in their UV absorbance capability and solvability, and thus vary in their propensity of being leached from the lichen body by high precipitation and temperatures, with still unknown implications for the global distribution of lichens. Close links between the chemical properties of key metabolites, the physiological requirements of lichen symbioses and their global distribution have been hypothesized but never tested. In this global study covering more than 10,00 lichen species, we show that the occurrence and chemical properties of fungal-derived metabolites are of eco-evolutionary significance for the global, latitudinal distribution of lichenized Trebouxiophyceae – the most prominent group of photobionts in lichen symbioses. This fungal-derived UV protection might represent an indirect environmental adaptation in which the lichen fungus invests to protect the trebouxioid photobiont from harsh environmental conditions and, by doing this, secures its efficient source of photosynthetic carbon.