Technical ReportPDF Available
Accelerating the discovery of biocollections data
Final task group report: November 2016
Leonard Krishtalka
Eduardo Dalcin
Shari Ellis
Jean Cossi Ganglo
Tsuyoshi Hosoya
Masanori Nakae
Ian Owens
Deborah Paul
Marc Pignal
Barbara Thiers
Page 2 | 41
Suggested citation
Krishtalka L, Dalcin E, Ellis S, Ganglo JC, Hosoya T, Nakae M, Owens I, Paul D, Pignal M &
Thiers B (2016) Accelerating the discovery of biocollections data. Copenhagen: GBIF
Secretariat. Available online at: http://www.gbif.org/resource/83022.
Persistent URI
http://www.gbif.org/resource/83022
This report is licensed under Creative Commons Attribution 4.0 Supported License
https://creativecommons.org/licenses/by/4.0.
Task group coordination, GBIF Secretariat
Siro Masinde (smasinde@gbif.org), Programme Officer for Content Mobilization
Cover Image
’Diversity of orders’ insect reference collection, Zoological Museum, Natural History Museum
of Denmark, Copenhagen University by Kyle Copas. Photo 2016 licensed under CC BY 4.0.
Page 3 | 41
Table of contents
Executive summary ................................................................................................................. 4!
Background ............................................................................................................................. 7!
The GBIF Task Force ............................................................................................................ 11!
Objectives .......................................................................................................................... 11!
Task Force membership .................................................................................................... 11!
Operating vision ................................................................................................................. 12!
Meetings and outreach activities ....................................................................................... 12!
A Global Survey of Natural History Collections ..................................................................... 13!
Purpose of survey .............................................................................................................. 13!
Survey methodology .......................................................................................................... 13!
Summary of survey results ................................................................................................ 13!
Selected Use Cases: Collection-based data informing solutions .......................................... 18!
Public Health: Zoonotic diseases and environmental contaminants ................................. 18!
Food Security .................................................................................................................... 19!
Invasive Species ................................................................................................................ 19!
Climate Change Impacts ................................................................................................... 20!
Extinction lessons from deep time ..................................................................................... 21!
Habitat and species loss .................................................................................................... 21!
Endangered and threatened species ................................................................................. 21!
Plants as indicators of minerals (metallophytes) ............................................................... 22!
Other major documented use-cases ................................................................................. 22!
Data Gap Analysis: Setting priorities for digitization ............................................................ 24
Next Steps ............................................................................................................................. 25
Recommendations ................................................................................................................ 26
Setting priorities for digitization and data gap analysis ...................................................... 26!
Digitization and best practices ........................................................................................... 26!
Metadata ............................................................................................................................ 27!
Partnership and collaboration ............................................................................................ 27
References ............................................................................................................................ 29!
Annex I: Acronyms and abbreviations ................................................................................... 33!
Annex II: Summary of results from a first analysis of NHC survey ........................................ 34!
Page 4 | 41
Executive summary
Biocollections and their associated data document the life on our planet, past and present.
They are fundamental for understanding, advancing and applying biodiversity science to the
discovery of knowledge and advancing human well-being. Nevertheless, the estimated 2.5
3 billion specimens of plants and animals in worldwide museums, herbaria and like
institutions remain largely underutilized, as only ~10% of their associated data has been
digitized for deployment by the educational, scientific and policy communities.
Biocollections institutions therefore face a challenging dilemma: how to prioritize and fund
the digitization of massive amounts of data associated with billions of voucher specimens of
animals, plants, fungi and other organisms that document the planet’s biodiversity and
undergird natural and human ecosystems.
To examine this issue, and as part of a broader global strategy for mobilizing primary
biodiversity data, GBIF convened a task force (20152016) to help accelerate the discovery,
digitization and access to biocollections data.
The task force’s main operating principle is the value-chain of Data Knowledge
Applications, i.e., when biocollections data is mobilized, analysed and converted by
research into knowledge, the data will be highly valued and funded by diverse constituencies
because the knowledge will inform solutions to current and future critical challenges of
human, economic and environmental well-being.
The task force’s main objectives were to:
1. Document best practices from ongoing content mobilization initiatives for small, large,
and different kinds of collections
2. Document successful business models for mobilizing resources for digitization
3. Consult with ongoing capacity-building and content-mobilization initiatives
4. Provide guidance in the development of training and outreach materials to help
institutions and interested stakeholders to implement a metadata approach for
content mobilization
5. Provide guidance on establishing priorities for digitizing biocollections in order to
serve institutional, national and global needs and achieve the greatest economies of
scale
6. Make recommendations for achieving long-term sustainability of mobilizing and
providing access to biocollections data
The task force conducted a global survey of biocollections to determine and demonstrate the
digital readiness of the world’s biocollections and their institutions, and the realized benefits
and impediments of digitization to the collection/institution. More than 800 responses from
2000 collections in 72 countries were received, with 76% at publicly funded institutions
40% at universities and 36% at non-university institutions. Key findings are encouraging.
Digitization of biocollections data is an ongoing, valued enterprise in most of the world’s
museums and herbaria.
86% (615 respondents) are currently digitizing or have completed digitizing at least
some or all of their collections. Among collection types, 13 of 15 are more than 50%
digitized. Only 1% are not digitizing their collections and have no plans to do so.
The major realized benefits of digitization are: increased use, exposure and
knowledge of the institution’s collections; more effective and efficient management
and preservation of data and associated physical specimens; enhanced data quality;
staff acquisition of new informatics skills.
Major obstacles to digitization are: lack of funding, time, credit, and/or expertise for
digitization; not an institutional priority; data has errors; effort exceeds perceived
benefit.
Page 5 | 41
The top three criteria for determining digitization priorities are research (53%),
funding/grant opportunities (51%), and select taxa (42%).
Several institutions, organizations and projects, e.g., GBIF, iDigBio, VertNet, SPNHC, ALA,
TDWG, Canadensys and so on, document community best practices for different aspects of
data mobilization that address and remove perceived barriers to digitization. Among them
are: setting smart digitization priorities, schedules, and workflows; curating and cleaning
data; adopting appropriate institutional policies and data-licensing practices to facilitate data
dissemination and reuse; and data management and archiving.
Most digitization projectsmore than 80% according to the surveyare government-funded,
and most often from one or two sources. Accordingly, institutions and digitization projects
should diversify their funding sources to minimize the impact of potential funding cuts by
governments and maximize investments in digitization. Some institutions and projects have
achieved such diversification through partnerships with information technology companies.
e.g., NHM, London. Others have developed crowdsourcing programmes to engage citizens
in transcribing specimen data labels, e.g., WeDigBio (international), Les Herbonautes at the
Muséum national d'histoire naturelle (France) and Naturalis’ crowdsourcing projects,
Glasshelder (Dutch) and LiveScience (international). Governments, the private sector,
foundations and citizen science programs are best engaged if the digitization priorities are
demand-driven by the data imperatives of human, economic and environmental well-being
that biocollections can inform (see Recommendations). At the country level, institutions
should present and highlight the value of their collection-data digitization to governments
and other entities as fulfilling the value-chain of Data Knowledge Applications.
Capturing and publishing collection metadata is a critical first step in exposing non-digitized
collections and their value to global discovery and access. Metadata also provides a
framework for institutions and biocollections to develop a comprehensive understanding of
their holdings, a consequent, prioritized digitization plan, and potential business-use cases to
recruit research and resource partners.
Biocollections are therefore encouraged to adopt a tiered strategy for worldwide collections-
data capture (and imaging where appropriate), i.e., a staged approach to digitization. Such
an approach can start with less expensive but rapid steps to capture and share metadata
about a collection-holding institution and an overview of its collections, then progress to
more expensive and time consuming steps to capture and share specimen-level data and
images at a finer granularity.
In the value-chain framework of Data Knowledge Applications, the task force’s major
recommendation is that institutions should set “demand-driven” digitization priorities to fulfil
the first link in that value chain, i.e., data-to-knowledge. Specifically, this entails mobilizing
and deploying the best biodiversity data to enable the best science for understanding and
sustaining human systems and Earth’s biological systemsand to do so in time to make a
difference. Once demonstrated that collection data is essential for smart, science-driven
solutions, data digitizationthe first link in the value chainwill be valued and supported by
entities who “demand” the second link, knowledge-to-applications.
It is clear that setting digitization priorities involves serving competing institutional, local,
regional, national and global imperatives: individual research interests; institutional
mandates; science agendas; and various environmental concerns (e.g., endangered
species, invasives, disease vectors/hosts, pollinators, pests). Moreover, each imperative
has its particular calculus of taxonomic groups, geographic areas, time periods, and
ecosystems/habitats. Overlying these permutations are the missions of different
stakeholders and funders: intergovernmental bodies (e.g., IPBES, CBD), government
agencies, NGOs, private foundations, and corporations.
Page 6 | 41
As such, the task force recommends that in a resource-limited world, a digitization strategy
of maximum efficacy will require all parties to collaborate on setting demand-driven,
overarching priorities that, simultaneously:
target the most urgent social, environmental, economic and biodiversity science
imperatives of our time;
are underpinned by sophisticated gap analyses;
include the greatest commonality among competing imperatives and interests;
tackle what is most pragmatic, first; and
promise the most immediate impacts.
Such a strategic and collaborative approach can evolve the current cottage industry of
biocollections digitization into an enterprise that is industrial strength, globally effective and
efficient, and funded by consortia of entities that value the result: governmental science,
health, natural resources, and agricultural agencies; intergovernmental agencies; and
NGOs, corporations and private foundations with missions in these and other sectors.
To enact this strategy, the task force recommends that GBIF and its partners convene a
series of high level discussions among these constituents to fund and implement these five,
long-term strategic priorities for mobilizing the remaining 90% of the world’s biocollections
data and bringing them into currency for science and society.
Page 7 | 41
Background
The discovery and access to primary biodiversity data is critical for informed decision-making
to achieve sustainable use of biotic resources and address many of the world’s key
challenges, such as the impacts of climate change, invasive species, zoonotic disease
outbreaks and food security. It is estimated that natural history collection institutions
collectively house 2.53 billion specimens that document more than 300 years of the
biological exploration of the Earth. Biocollections are the single largest source of information
on biological diversity outside nature itself (Scoble 2010, Buerki & Baker 2015, Holmes et al
2016). The physical specimens and their associated data constitute a vast biodiversity library
of the spatial and temporal occurrence of species, populations, individuals, and their
morphology and genetic traits. From the study of fossil specimens tens of millions of years
old to specimens of modern organisms, both extant and historical collections enable the
reconstruction of past and present biodiversity, and forecasting of future environmental
states.
As such, these biocollections underpin much of our knowledge and ongoing research in
biodiversity science, including irreplaceable documentation of evolutionary and ecological
patterns and processes among fossil and recent organisms, as well as current threats to and
conservation of biodiversity (Holmes et al 2016). Despite this foundational importance, the
world’s biocollections remain largely untapped, as only about 10% have been digitized.
Furthermore, only a small percentage of the digitized and imaged collections have been
optimally mobilized to make them discoverable, accessible, interoperable and reusable.
In 2009, a survey carried out by GBIF, on the challenges and concerns related to digitizing
natural history specimens, found that lack of funding was the overwhelming barrier, followed
in descending order by lack of time and staff, lack of institutional support,
infrastructural/technological constraints and challenges due to curation practices (Vollmar et
al. 2010). The survey also found an uneven digitization landscape that led to a patchy
accumulation of data at varying qualities, and based on different priorities, ultimately
influencing the fitness-for-use of the data.
In 2010, a GBIF Task Group on the ‘Global Strategy and Action Plan for the mobilization of
Natural History Collections data (GSAP-NHC) recommended the capture of essential
metadata as a first step toward facilitating access to non-digitized collections and making
them discoverable and accessible (Berendsohn et al 2010). This metadata approach
captures data on the different kinds of collections at various scales, from single units to large
groupings.
Based on recommendations of the GSAP-NHC and our own deliberations, we summarize
the value of capturing and sharing metadata for non-digitized collections as follows:
1. It is a rapid method of evaluating and assessing collections;
2. Sharing metadata makes collections discoverable and accessible;
3. It is a stepping stone towards making collections deliver immediate value and
transforming them into a global resource;
4. It enables quick reporting on gaps such as taxonomic and geographic coverage,
curation and physical state, high value series, and so on;
5. It is a framework for helping prioritize a collection’s digitization projects taxonomically,
geographically or temporally;
6. It provides institutions the knowledge of their holdings required to build the business
case for digitization;
7. It is a rapid and concise way of communicating and advertising an institution's
collection holdings and potential, which can be key in attracting partnerships and the
necessary funding for digitization.
Page 8 | 41
Since the GSAP-NHC task group report was published in 2010, much progress has
occurred, some of it in response to the recommendations of the task group. Following are
some of the noteworthy developments.
Mobilization of NHCs has remained an important component of the GBIF strategy
and the GBIFS Work Programme. The same has been promoted through GBIF
nodes and the entire GBIF community.
A GBIF metadata profile incorporating elements to describe non-digitized collections
was developed but it has not adequately served the intended purpose
To encourage scholarly credit for metadata publishing, the concept of “Data Papers”
was implemented and is now well established.
To give credit to data publishers, data citation tools such as the digital object
identifier (DOI) have been implemented in GBIF.org so that all data downloads are
traceable.
To facilitate online curation of specimens, GBIF plans to implement annotation tools
and feedback mechanisms for data publishers.
To recognize and give due credit to those involved in the enormous work of curating
specimens and managing the associated digital data, a new joint RDA / TDWG
working group on metadata standards for attribution of physical and digital collections
stewardship1 is currently in place and expects to complete its work by the end of
2017.
The number of specimen-based data records in GBIF.org have increased
tremendously and currently stands at about 125 million. This has been accelerated
by increased digitization in both small and large institutions worldwide as well as
improved infrastructure for increased collaboration, sharing and management of NHC
data. Examples among large projects include the mass digitization effort at the
Natural History Museum, London (Blagoderov et al 2012), the US iDigBio2
consortium funded through the Advancing the Digitization of Biological Collections
(ADBC) programme of the US National Science Foundation, the Canadensys3
consortium in Canada, the Atlas of Living Australia4, the Consortium of European
Taxonomic Facilities (CETAF)5 and SYNTHESYS6.
To accelerate data capture and citizen involvement, a number of GBIF nodes and
other GBIF collaborators are implementing crowdsourcing programmes especially in
transcribing specimen data labels, e.g., WeDigBio7 (International), Naturalis’ Dutch
crowdsourcing project, Glasshelder, that uses VeleHanden application8 as well as
the more international LiveScience9 project and Les Herbonautes10 at the Muséum
national d'histoire naturelle (France).
1 https://rd-alliance.org/group/metadata-standards-attribution-physical-and-digital-collections-stewardship/case-
statement
2 https://www.idigbio.org
3 http://www.canadensys.net
4 http://www.ala.org.au
5 http://cetaf.org
6 http://www.synthesys.info
7 https://www.wedigbio.org
8 https://velehanden.nl/projecten/bekijk/details/project/nat_nbc
9 http://www.naturalis.nl/en/museum/livescience/crowd-sourcing
10 http://lesherbonautes.mnhn.fr
Page 9 | 41
Much progress has been made in developing hardware, methods and tools to
industrialize the digitization of NHCs. Examples include the Digistreet conveyer belt
system employed by Naturalis Biodiversity Centre (Netherlands) for imaging
herbarium specimens (Heerlien et al 2015), and the Inselect tool - a modular, easy-
to-use, cross-platform suite of open-source software tools that supports the semi-
automated processing of specimen images generated by natural history digitization
programmes (Hudson at al 2015).
Commercial companies that can provide large scale, high quality digitization and
imaging of biological collections as well as collection management software at
reasonable cost have been outsourced for some of the mass digitization projects
carried out at some of the very large museums. For example, Naturalis
(Netherlands), Smithsonian Institution, and Muséum national d'histoire naturelle
(France) have used Picturae11 to carry out digitization. Several natural history
collections including Smithsonian Institution, New York Botanical Garden and Natural
History Museum, London, use the Emu collection management software by Axiell12
An online metadata resource for biodiversity collections, the institutions that contain
them, and associated staff members, namely, the Global Registry of Biodiversity
Repositories (GRBio)13, was set up (Schindel et al 2016).
Importantly, what has also emerged is a tiered strategy for worldwide collections digitization
(plus imaging where appropriate) and model concepts, such as Linked Open Data (LOD)
(Berner-Lee 2009) and the Digitization Maturity Model in the ALA’s Guide to Digitization
(Kalm 2012). In this strategy, the tiers are:
a) Level OneMetadata I: sharing and publishing Institution/Organization/Collection-
level information, i.e., the who and where of a collection and, broadly, its content and
history. Also, registering collections with Gerbil supports the adoption of collections
metadata standards and globally unique identifiers. Analogous to the Linked Open
Data level one “on the web”, this level is not costly or time-consuming, but is
invaluable as it is key to a collection’s discoverability.
b) Level Two - Metadata II (spindex, sensu Mason 2016): producing and publishing
species-level (or perhaps cabinet-level) collection inventories. Such collection
inventories provide excellent data for tracking collection health, space requirements,
and gaps in taxonomic, geographic, and/or temporal coverage, and follow-on
strategic planning for collection growth, conservation and digitization. For example,
recently the Academy of Natural Sciences of Drexel (ANSP) provided a Species
Index, or #spindex for short (Mason et al 2016) as a basis for a strategic specimen
digitization programme. Producing and sharing these in a standard, machine
readable and non-proprietary format, analogous to LOD levels two and three, ensure
global access to data for planning and funding initiatives.
c) Level ThreeSpecimen Data I (skeletal data): capturing and publishing at least
skeletal-level data (with or without locality georeferencing) and perhaps images for
each specimen or lot, preceded by a strategically chosen level of pre-digitization
curation (Nelson et al 2012). Where possible, all shared data is mapped to currently
accepted data standards, while advancing standards development when needed.
Imperfect or erroneous data are then exposed to machine algorithms and the
expertise in the worldwide community for effective, efficient correction and
improvement.
11 https://picturae.com
12 http://alm.axiell.com
13 http://grbio.org
Page 10 | 41
d) Level FourSpecimen Data II (richer data): locality data is georeferenced where
possible and the specimen record is enriched with other data, such as: field notes,
grey literature, note cards, etc. Field notebooks may be captured at Level Three or
Level Four, whichever is most appropriate for the particular collection digitization
project.
e) Level FiveSpecimen Data III (born digital data): The data associated with all new
collections is “born digital” and incorporated into the existing, georeferenced
database, with collection management documents, e.g., specimen labels, generated
from the database. Digital records may also have links to Embank accessions, BCoL
IDs, etc., and must include globally unique identifiers, which makes Linked Open
Data a reality, a “web of data” (Berners-Lee 2009).
Ultimately, for global digitization, collections worldwide should collaborate (see
Recommendations) to achieve critical mass, least redundancy, and economies of scale in
their digitization, and to meet global, demand-driven imperatives that depend on collection
data for solutions.
Page 11 | 41
The GBIF Task Force
Objectives
As part of a broader global strategy for mobilizing primary biodiversity data, GBIF convened
a task force to help accelerate the discovery and access to biocollections, especially those
yet to be digitized. The task force commenced its work in March 2015 and runs up to the end
of 2016. The objectives of the task force are to:
1. Document best practices from ongoing content mobilization initiatives, taking
into account their applicability to large and small collections as well as
different kinds of collections (e.g., wet, dry, mounted, pinned).
2. Document successful business models for mobilizing resources for
digitization.
3. Consult with ongoing capacity-building and content-mobilization initiatives
(e.g., SEPDD14, GBIF-BID Project15, ALA16, adagio17), specialist groups and
the GBIF community in order to bring together the different stakeholders and
catalyse activities around metadata capture.
4. Provide guidance in the development of training and outreach materials to
help institutions and interested stakeholders to implement a metadata
approach for content mobilization.
5. Provide guidance on establishing priorities for digitizing biocollections in order
to serve institutional, national and global needs and achieve the greatest
economies of scale.
Task Force membership
The task force is composed of eight members with diverse international experience and
expertise. In addition, the task force has three ex officio members. As a nuclear team, it is
consulting widely with stakeholders including experts, institutions, initiatives and projects as
well as potential funders. The members are:
Leonard Krishtalka (Chair), Director, Biodiversity Institute, University of Kansas, USA
krishtalka@ku.edu
Barbara Thiers, Director of the Herbarium and Vice President for Science
Administration, New York Botanical Garden, USA
Deborah Paul, Digitization and Workforce Training Specialist, iDigBioHUB, Florida
State University, USA
Eduardo Dalcin, Biodiversity Informatics Expert, Rio de Janeiro Botanic Gardens,
Brazil
Ian Owens, Director of Science, Natural History Museum, London, UK
Jean Ganglo, Professor of Forestry, University of Abomey-Calavi, Benin
Marc Pignal, Natural History Museum, Paris, France
Masanori Nakae, Curator, National Museum of Nature and Science, Tsukuba, Japan
o Siro Masinde, Programme Officer for Content Mobilization, coordinator and contact
person for the task force at the GBIF Secretariat, Copenhagen.
o Shari Ellis, Consultant to the task force and iDigBio External Evaluator
14 http://www.sud-expert-plantes.ird.fr/sepDD
15 http://www.gbif.org/bid
16 http://www.ala.org.au/
17 https://www.idigbio.org/
Page 12 | 41
o Tsuyoshi Hosoya, Division Head, Fungi and Algae Research, National Museum of
Nature and Science, Tsukuba, Japan - alternate to Masanori Nakae
Operating vision
The task force adopted the Data-Knowledge-Application value chain framework as its
operating vision. It is imperative to demonstrate how digitizing and sharing biocollections
data contributes to this value chain, especially if all stakeholders are to be persuaded to
contribute significant resources to accelerate the mobilization of biocollections data. The
natural history collections community has made good progress in recent years in the first link
in this value chain, mobilizing data through digitization and publishing to GBIF and other data
portals, and converting that data to published knowledge through research, mostly in
academic institutions.
However, the knowledge-to-application link in the value chain remains weak, despite the
number of compelling, documented use cases. Two possible reasons among others are: (1)
the number of institutions devoted to this link are much fewer, less networked and more
poorly funded than institutions devoted to the data-to-knowledge link; and (2) the results of
data-to-knowledge, published in professional journals are highly inaccessible to the
governmental and NGO communities devoted to the knowledge-to-application component.
Clearly, it is incumbent on the biocollections community, working with other communities
across the value chain, to demonstrate the knowledge-application link if it is to rally the
business community, major donors, funding agencies and governments to support and
accelerate the digitization of primary specimen data.
Meetings and outreach activities
Task force members have held 16 monthly group meetings. Other meetings and
consultations have been held in between as and when needed. Task force members have
also participated at other conferences and workshops and continue to consult widely with
stakeholders on pertinent issues, such as setting strategic institutional, national and global
digitization priorities, as well as making compelling cases that can attract buy-in from
potential funders in the public and private sectors. Table 1 summarizes the main meetings
and outreach activities in which the members have participated.
Table 1. Meetings and outreach activities by GBIF task force members
Event and attendee’s initials
Venue
16 group meetings
Virtual
TF face-to-face meeting (LK, BT, DP, MN, ED, IO, SE, SM)
Washington DC
Gerbil outreach (BT)
Washington DC
SPNHC (BT, DP, SE)
Gainesville, Florida
Biodiversity Collections Network - BCoN (LK)
Field Mus., Chicago
TDWG (LK, DP, JG, SM)
Nairobi
iDigBio Summit V (LK, BT, DP, MN, ED, IO, SE, SM)
Washington DC
Entomological Coll. Network - ECN presentation (DP)
Minneapolis
Amer. Inst. Biol. Sc.-AIBS & BCoN, capacity bldg (BT)
Washington DC
SPNHC TF symposium (LK, SM, DP, BT, MN, IO)
Berlin
Entomological Coll. Network ECN presentation (DP)
Orlando, Florida
Page 13 | 41
A Global Survey of Natural History Collections
Purpose of survey
In late 2015, the task force carried out a global survey on natural history collections. The
purpose of the survey was to enable the task force to determine and demonstrate: (1) The
digital readiness of the world’s biocollections and their institutions; (2) the benefits to the
collection/institution that digitization engenders; and (3) the impediments to collection data
digitization.
Survey methodology
The survey questionnaire was prepared through meetings, consultations and research, and
mainly administered online through Qualtrics18 software. A Microsoft Word version was
however also made available for respondents that had difficulty in filling out the online
survey. We distributed the survey to individuals affiliated with collections worldwide through
a variety of channels including listservs and member lists such as Index Herbariorum, and
GBIF nodes, as well as personal contacts with institutions. We also announced the survey at
relevant meetings, workshops and conferences attended by potential target respondents.
The survey was translated into Japanese and distributed throughout Japan via the
community of “Science Museum Network (S-Net)”. Those contacted were requested to help
with the further distribution of the survey in order to help maximize the global reach. Table 2
summarizes the key distribution channels.
Table 2. Main distribution channels for the global NHC survey
Channel
GBIF Node Managers
GRBio
Herbaria-L
iDigBio
Index Herbariorum
MUSEUM-L
NHColl
SERNEC
S-Net (http://science-net.kahaku.go.jp/)
Taxacom
TDWG
Conferences, e.g., TDWG (Nairobi, 2015), iDigBio V Summit (Washington DC, 2015), Entomological Collections
Network Conference (Minneapolis, 2015), etc.
Summary of survey results
We present a summary of the survey results here, but a more detailed report (Annex 2) from
a first analysis was distributed to the survey respondents that expressed interest in being
18 https://www.qualtrics.com/
Page 14 | 41
contacted directly. More than 800 individuals completed at least a portion of the survey, of
which 617 were complete enough to be counted. Note that the number of respondents who
answered each question varied either because in the case of the online version, some
questions were automatically skipped based on prior responses, or the respondents elected
to not answer the question. Respondents represented almost 2000 collections distributed
over 72 countries. Map 1 shows the locations of the survey respondents based on the
registered IP addresses from where the online survey was completed. The distribution and
density of respondentsand by extension the locations of natural history collections
mirrors that for the global distribution of the internet infrastructure19 as well as the GBIF
species occurrence map20.
This is the most comprehensive survey of NHCs that has been carried out in the past six
years, as it reached all continents and achieved a large number of responses. A similar,
2009 survey (Vollmar et al 2010), but of more limited scope elicited 201 responses, mostly
from North America (62%) and Europe (22%), with none from Africa. Of the respondents in
the current survey, 76% were at publicly funded institutions40% at universities and 36% at
non-university institutions. Almost all (92%) of respondents were primarily curators or
collection managers with 10% as head of research and collections.
Key findings from the task force survey
Of the usable 617 responses, 86% are currently digitizing or have completed
digitizing some or all of their collections.
13 out of 15 collection types report more than 50% data capture.
Very few respondents (1% or 5 individuals) reported they are not digitizing and have
no plans to do so.
Major obstacles to digitization were: funding, time (lack of), size of task, not an
institutional priority, data has errors, limited expertise in databasing or processing
specimen data, no credit (tenure, reappointment) for digitization effort, effort exceeds
perceived payoff.
19 http://internetcensus2012.bitbucket.org/images/worldmap_16to9
20 http://www.gbif.org/occurrence
Map 1. Location of survey respondents by IP address, CartoDB by Kevin Love (iDigBio)
Page 15 | 41
Major priorities for collection digitization are research (53%), funding/grant
opportunities (51%), and select taxa (42%).
The major realized benefits of digitization are: increased use, exposure and
knowledge of the institution’s collections; more effective and efficient management
and preservation of data and associated physical specimens; enhanced data quality;
staff acquisition of new informatics skills.
Responses by taxonomic collections included vascular plants (20%), bryophytes (10%),
fungi (10%), algae (9%), arthropods (8%), mammalogy (6%), ornithology (5%), herpetology
(5%), and ichthyology (5%), with the remaining representing malacology, marine
invertebrates, terrestrial invertebrates, and fossil invertebrates, vertebrates and non-vascular
plants.
It is clear that the value of databasing and publishing collection data is now embedded in the
communitymost collections are digitizing or trying to do so. To that point, 86% (615
respondents) indicated that they are currently databasing some of or have completed
databasing their collections. The percentage varies across collection type, as does the
mean portion of the collection that has been digitized. But, on average, 13 of 15 taxonomic
collection types are more than 50% databased. Also, across organismal groups, the average
of all collection types digitized is more than 50%, except arthropods at 38% and invertebrate
fossils at 47%.
Barriers to digitization were raised by 587 respondents, most citing either funding or time,
which is similar to the findings reported by ITHAKA for digitization of special collections
(Maron and Pickle 2013). The top 10 barriers to digitization, for our survey respondents
(n=587), were:
1. Funding and other resources (80%)
2. Personnel time/effort (80%)
3. Task is overwhelming (40%)
4. Not an institutional priority (35%)
5. Collection data has errors (30%)
6. Limited digitization expertise among personnel (25%)
7. Insufficient information on digitization process (15%)
8. No benefit to job advancement, tenure (14%)
9. Not a priority of the individual in charge of the collection (12%)
10. Effort exceeds payoff (10%)
Virtually all of the cited barriers to digitization after the first two (funding; effort) fall into four
categories:
1. Size of task is overwhelming.
2. Digitization is not institutional priority because of perceived mismatch between
effort required and benefit achieved.
3. Sentiment that data cannot be digitized because of the number of errors
4. Personnel require greater experience and expertise in digitization workflows and
applications
The TF determined sharing collection metadata can help overcome the major barrier of
insufficient resources, as it makes an institution’s collections discoverable to funding by
various stakeholders and potential funders. For example, the British Library (BL) currently
has more than 30 digitization efforts; all funded by various foundations and other entities with
a vested interest in particular materials held by the BL (from a presentation by the British
Library, at the Cisco Pitstop (Jackson 2016)).
Page 16 | 41
How are collections deciding what to digitize?
With 519 respondents, the top three variables driving digitization priorities include: research
(52%), funding/grant opportunities (51%), and taxonomic focus (42%). Other criteria include
partnership in a larger community effort (24%), and a geographic focus (23%). About 18%
cited opportunistic digitization, and about 5% health and human services.
Who is doing the digitization?
Collections rely on a variety of personnel to perform digitization tasks, usually staff, students,
volunteers, and rarely, third-party organizations (210 %). Most often, digitization is being
completed by paid personnel (90% respondents), as opposed to paid staff (53%) or students
(59%; paid or unpaid) “frequently” doing the digitization.
How are collections funding digitization?
Of the 523 respondents to this question, 87% cited receipt of some funding, of which 69%
reported external sources, and 61% regular institutional sources. Slightly more than half
(53%) cited only one source of funding (internal, external, ad hoc), whereas 36% received
funds from two sources and 30% from three. Most of the external funding (80%) is from
government agencies, with 72% receiving funds from just one kind of external entity (i.e.,
industry, foundations, government, other).
Benefits of digitization
The top 8 perceived benefits of digitization (516 respondents) are: increased use of
collections, increased exposure, better knowledge of holdings, better management of data,
digital data preservation, enhanced data quality, new skills for staff, and better management
of physical specimens. Thirty percent reported new communities using the data, 35% saw
increased publicity and reduced physical handling of the collection, and 40% cited increased
use of their collection data in research and publications, as well as increased public
awareness of the importance of collections.
Institutional commitment
Of the 516 respondents, 72% confirmed their institution’s commitment to continued
digitization. Of 422 respondents, 50% reported that their institutions were providing staff
training and seeking continued funding. More than 60% indicated plans for long term
archival storage, and more than 70% for long term data curation.
Metadata
What collection metadata do we need as a community? The survey requested information
Figure 1. Metadata being provided
Page 17 | 41
on the kinds of metadata collections are currently providing and consider important (see
Figure 1 for results).
With regard to the relative importance of sharing different kinds of metadata, more than 50%
cited taxonomic and spatial data as critical, with 45% indicating type-specimen data as
important. Metadata values ranked as critical by 20%30% of respondents included: percent
collection digitized, notable publications, percent georeferenced, place name coverage,
name of collection manager, collection size, and name of curator. Respondents (506) report
most metadata being shared via GBIF (43%), Index Herbariorum (36%), and institutional
websites (21%).
Long term plans
Encouragingly, more than 80% of 513 respondents indicated their institution/organization
intends to digitize their entire collection(s), and more than 30% that they plan to prioritize
digitization in response to research needs/requests. About 12% will focus on strategically
important or unique collections, such as type specimens or endemic species. Others (about
12 %) cited new collections.
Page 18 | 41
Selected Use Cases: Collection-based data informing solutions
Discovery and access to primary biodiversity data are indispensable in ensuring informed
decision-making on human, environmental and economic well-being (Kremen et al 2008;
Gaiji et al 2013; Peterson et al 2015). The principle is that Digitally Accessible Knowledge
(DAK), e.g., primary biodiversity data that is digital, published and therefore accessible
worldwide, can be integrated into the broader global storehouse of biodiversity information
(Sousa-Baena et al 2013a) for fulfilling the knowledge-to-application link of the value chain.
The Society for the Preservation of Natural History Collections (SPNHC) outlines the
following use case themes that illustrate how DAK of collection-based biodiversity informed
critical solutions for science and society. We illustrate each theme with selected examples.
Public Health: Zoonotic diseases and environmental contaminants
Perhaps the most compelling examples of the importance of natural history collections are in
the area of public health and safety (Suarez & Tsutui 2004). In several important cases,
collections have been used to track the history of infectious diseases, identify their sources
or reservoirs, and pinpoint geographic areas for potential interdiction.
Zika, dengue and chikungunya viruses
Zika, dengue, and chikungunya are mosquito-borne diseases that have recently re-emerged
in epidemic proportions across wide geographical areas hitherto not known to harbour them
(Kraemer et al 2015; Bogoch et al 2016; Cao-Lormeau and Musso 2014). These diseases
are spread by two mosquito species, Aedes aegypti and Aedes albopictus. Essential for
health planning, is mapping the global distribution of these vectors using all known sources
of data, including specimens in natural history collections that indicate the geographical
determinants of their ranges. When Kraemer et al (2015) mapped the global distribution of
these mosquitoes; it showed that they are more widespread than previously known, thus
predicting an increased risk of new infections in areas hitherto thought to be infection-free
zones.
Hantavirus
In the 1990s, a Hantavirus (Bunyaviridae) caused a pulmonary infection that was fatal for
most people who contracted it. Public health officials could not identify the vector of this
disease until evidence of the virus was found in museum tissue collections of deer mice from
the American Southwest (Yates 2002).
Ebola
The 2014 outbreak of the Ebola virus in West Africa was the largest of this deadly disease
and the first in this region. A public health priority is the ability to identify the potential spread
of this zoonotic disease and geographic areas at greatest future risk. Natural history
collections are central to this task. Three different species of bats are suspected to play an
important role in the life cycle of Ebola and similar viruses. Using data from natural history
collections, researchers have determined the geographic range of these three species.
Niche modelling of these data enabled researchers to pinpoint the geographic areas and
communities at highest risk for future outbreaks across Central and West Africa. These data
will help prioritize surveillance for Ebola outbreaks and improve the diagnostic capacity in
these at-risk regions, which have a combined human population of 22 million people (Piggott
et al 2014).
Anthrax
In 2001, anthrax (Bacillus anthracis) was sent through the mail to media outlets and
government officials in the U.S., resulting in five deaths. Researchers from the Centers for
Disease Control and Prevention (CDC) compared isolates from the 2001 anthrax attack in
the United States with stored museum specimens to differentiate and identify the strain used
in these attacks (Hoffmaster et al 2002).
Page 19 | 41
Lassa Fever
Lassa Fever (LF) is a zoonotic disease caused by Lassa virus (LASV), a member of the
Arenaviridae family. Introduction of the virus into humans occurs through direct or indirect
contact with excreta of the natural reservoir, the rodent Mastomys natalensis, although
precise modes of transmission are not well characterized (Fichet-Calvet and Rogers 2009,
Peterson et al 2014). Peterson et al (2014) applied ecological niche modelling, based on
museum collection data, of the rodent’s geographic occurrence to 107 data records of LF
from an initial dataset of 111 records collected by Fichet-Calvet and Rogers (2009) in seven
West African countries: Nigeria, Benin, Côte-d’Ivoire, Burkina Faso, Sierra Leone, Guinea,
and Liberia, LF prevalence areas. Their results indicate that West Africa and particularly the
southern humid forest habitats of the sub region as well as the drier Sahel zone of the
continent are suitable for LASV transmission and therefore are at risk for LF. These areas
are the priority for a health surveillance infrastructure so that health officials and decision-
makers can detect and stem the spread of the disease. The study also revealed that
additional data is needed from field and museum collections to add geographic and
ecological resolution to the risk assessments in West Africa, including Benin, Togo, and
Ghana.
Environmental contaminants—Mercury, DDT, Atrazine
Environmental contamination affects human health as well as ecosystem health. Museum
specimens have been used to track ambient mercury levels over time (Berg et al 1966). In a
classic study from the 1960s, eggs from museum collections were critical in establishing the
link between the chlorinated hydrocarbons in DDT to the sharp decline in bird species.
Specifically, measurement of bird eggs collected over a century demonstrated a marked
decrease in shell thickness that coincided with the widespread use of DDT (Radcliffe 1967,
Hickey and Anderson 1968). More recently, museum collections were used to demonstrate
that sexual abnormalities in frogs increased after the widespread adoption and use of
atrazine as an herbicide (Hayes et al 2002).
Food Security
Agricultural diseases
Phytophthora infestans, the cause of the potato blight that triggered famine in Ireland in the
1840s, continues to cause damage to potato fields around the world, resulting in huge losses
annually to the world’s third largest food crop. Yoshida et al (2013) compared the genomes
of herbarium specimens of P. infestans with that of modern strains and determined that
outbreaks in the 19th century were caused by a single lineage, which is not the direct
ancestor of the strains that have come to dominate more recent global populations.
Bioterrorism
Collections can be used to determine whether or not emerging pests have spread naturally,
accidentally or deliberately by comparing specimens of pests across temporal and
geographic ranges. Bioterrorism through deliberate introduction of agricultural pests has
been identified as a threat by the US National Research Council, which has stressed the
need for “reference specimens and other taxonomic information for rapid and accurate
identification” of newly discovered pests (NRC 2003).
Invasive Species
Cheatgrass
Invasive species in the U.S. cause environmental damage and loss totalling more than $130
billion per year. Cheatgrass, Bromes tectorum, one of the most damaging invasive species,
crowds out wheat plantations and fodder crops. The grass has limited nutritional content
and the long, sharply pointed fruit can penetrate the skin of livestock, causing injury or
infection. Cheatgrass is native to Europe and Central Asia, where it does not exhibit the
Page 20 | 41
invasive tendencies that it does in North America. Comparison of genetic information
between plants from the species’ native range and those from historical and modern
herbarium collections in North America (Novak & Mack, 2001), revealed that cheatgrass was
introduced multiple times to North America from different areas of its native habitat. When
these different strains came in contact with one another in North America, they interbred,
creating novel strains with invasive qualities.
Argentine ant, and green alga
Museum specimens have also been used to elucidate the invasion history of the Argentine
ant (Linepithema humile) (Suarez et al 2001), and the green alga Codium fragile (Provan et
al 2007).
Climate Change Impacts
Food stocks
Climate change is expected to cause dramatic shifts in the distribution of species, with
serious implications for natural ecosystems, crop plants, their pollinators, and food supply.
Studies using museum collections demonstrate distributional shifts and extinctions in
butterflies (Parmesan, 1996), and changes in nesting times in tree swallows (Dunn &
Winkler, 1999). More recently, Jones et al (2014) used GBIF’s collection-based species
occurrence data to model predicted changes in distribution of aquatic food species around
Great Britain. The results project decreases in species diversity and catch weight, which will
reduce the profitability of the fishing industry and threaten its decline. Accordingly, the
authors recommend changes in the British fishing industry that may help to offset a future
drop in revenue.
Bees
Over the past 40 years, drier weather has limited the growth of some populations of alpine
plants in the Rocky Mountains of North America, making it more difficult for bees to obtain
nectar. The paucity of nectar-producing flowers as well as the warmer temperatures favour
bees with shorter tongues that can access a broader range of flowers. Miller-Strotman et al
(2015) documented this phenomenon by measuring the tongues of specimens of bees in
museum collectionsbees now inhabiting this region indeed had shorter tongues. If this
trend continues, it could lead to the extinction of plants whose flowers can only be pollinated
by bees with long tongues.
Baobab tree
The baobab, with more than 300 product uses and ensuing commercial value in the EU and
US, is one of the most important trees to be conserved and domesticated in Africa, given the
impact of this industry on African economies and livelihoods. Sanchez et al (2011) using
available DAK (480 records) on the African baobab tree (Adansonia digitata) and niche
modelling, found that under IPCC scenarios of climate change only a percentage of the
present distribution of the species in Africa will remain viable in the future.
Their results informed useful strategies for baobab conservationin-situ in protected areas,
ex-situ in seed banks, and through sustainable use of the species. The existence of only 480
records for African countries, of which less than 100 belong to West Africa, indicates that
field and collection-data on baobab occurrences in West and East African countries need to
be digitized and published to the research community to increase the resolution of baobab
distribution models and conservation strategies under different scenarios of climate change.
African palms
Palms (Arecaceae) are a multi-use resource for many African economies and communities
but especially in West Africa. For example, they are the source of income through the palm
oil and wine trades, local palm alcohol production and sales, and multiple uses of palm
branches and leaves. Blach-Overgaard et al (2010, 2015) applied ecological niche modelling
Page 21 | 41
on 1920 occurrence records of 29 palm species obtained mainly from herbarium specimens,
to assess the degree to which African continental-scale palm species distributions are
controlled by climate, non-climatic environmental factors such as habitat and human impact,
or non-environmental spatial constraints such as biotic interactions and/or dispersal
limitations. They found that, at the continental scale, climate, especially water-related
factors, constitutes the only strong environmental control of palm species distributions in
Africa. Furthermore, due to the strong response of palm distributions to climate in
combination with the importance of non-environmental spatial constraints, African palms will
be sensitive to future climate change in that their ability to track suitable climatic conditions
will be spatially constrained. As with the baobab, the collection-based modelling studies of
the African palms can inform in-situ, ex-situ and other species conservation measures.
Extinction lessons from deep time
Among the many studies based on paleontological and recent collections and their
associated data are analyses published in Nature (Barnosky et al, 2011) and Science
(Ceballos et al, 2015) indicating that the Earth’s sixth mass extinction may well be underway
at the present time, given the documented species losses over the past few centuries and
millennia. Current extinction rates were shown to be higher than expected compared to
those documented in the fossil record.
In another study (Barnosky 2008), analysis of megafaunal collection data and climate
records revealed that an increase in human biomass and impacts, along with climate change
were the fingerprints on the Quaternary megafaunal extinction and its subsequent ecological
threshold event. Humans have since become the dominant ecological species which, with
higher rates of climate change, will induce extinctions across taxa of all body sizes and
possibly a near-future biomass crash that will have a severe impact on humans and their
domesticates.
Habitat and species loss
The greatest threat to biodiversity and its contribution to ecosystem function is habitat loss
(Millennium Ecosystem Assessment, 2005). Studies involving museum collections have
successfully documented such shrinking habitats and the effects on their biodiversity. The
loss of prairie habitat has led to the decline of its small mammals (Pergams and Nyberg
2001).
Based on a review of thousands of herbarium specimens of lichens, Lendemer and Allen
(2014) identified a previously unknown biodiversity hotspot in the Mid-Atlantic Coastal Plain
of North America. This region, which is already under threat from encroaching development,
is also projected to be completely inundated due to climate-induced sea level rise within the
next century. Development pressures have seriously reduced the availability of corridors
through which species in the areas affected by sea level rise could migrate to higher ground.
Endangered and threatened species
Since 2010, the Brazilian National Centre for Flora Conservation (CNCFlora) is responsible,
at the national level, for assessing the conservation status of the Brazilian flora and
developing recovery plans for species threatened with extinction. CNCFlora is the Red List
Authority for plants in Brazil and adopts the standards and procedures recommended by the
International Union for the Conservation of Nature (IUCN). So far, CNCFlora has assessed
the extinction risk of 5,165 species of the Brazilian flora (11.2% of the national flora). For this
assessment, the CNCFlora has built up a database of species occurrences from two main
sources: Rio de Janeiro Botanical Garden Virtual Herbarium and speciesLink.
The case of one species, Abatia microphylla (Family Salicaceae) reveals a risk assessment
in which every herbarium occurrence record is critical. Initially, the evidence of only four
valid records for two different localities caused this species to be listed as "Critically
Page 22 | 41
Endangered". A further query of GBIF’s records for Brazil identified two new records of this
species from one new locality, effectively an "Extension of Occurrence" of 273,8Km2, and a
revised listing of "Endangered".
The CNCFlora continues working on the assessment of all species of Brazilian floraca.
40,000 speciesbased on the best knowledge available. Whereas 80% of the Brazilian
herbaria records are digitized, only ~20% are reliably georeferenced, effectively relegating
12% of the 5,165 assessed species to IUCN’s "Data Deficient" category. Sousa-Baena et al
(2013b) however demonstrated that about 40-54% of the 934 Data Deficient angiosperm
species that were listed at the time had considerable digitally accessible knowledge
available. The problem was knowledge deficiency because the available data remained
unanalysed and dormant for conservation decision-making.
Plants as indicators of minerals (metallophytes)
Miners and scientists have long known that certain plant species can be a signal for ore-
bearing rocks (Brooks et al 1985; Ernest 2006). For example, Lychnis alpina, a small pink-
flowering plant in Scandinavia, and Haumaniastrum katangense, a white-flowered shrub in
central Africa, are both associated with copper. Haggerty (2015) discovered that Pandanus
candelabrum is closely associated with kimberlite pipes that are mined for diamonds in
Liberia. Using herbarium specimens to map the distribution of Pandanus candelabrum can
help in diamond prospecting since this species is restricted to the diamond-bearing
kimberlite dykes and is not found even in the alluvium covering the adjacent dikes.
Other major documented use-cases
Following are other major studies and compendia that demonstrate the use of collections
and associated data for important research and findings across a broad suite of subjects.
The links to the publications are given after each summary.
GBIF-mediated data (2012-2015) use cases
Since 2012, the GBIF Secretariat systematically reviews and compiles peer-reviewed
research using and applying GBIF-mediated data. These are published annually in the GBIF
Science Review. The following links to the literature reviews for 2012-2015 document more
use cases mostly addressing the data to knowledge value chain.
201621 GBIF Science Review: http://www.gbif.org/resource/82873
2014 GBIF Science Review: http://www.gbif.org/resource/82191
2013 GBIF Science Review: http://www.gbif.org/resource/80915
2012 GBIF Science Review: http://www.gbif.org/resource/80847
GBIF Chapman report (2005) use cases
Chapman (2005) in a GBIF report on the uses of primary species occurrence data discusses
a wide range of use cases of natural history collections data.
http://www.gbif.org/resource/80545
UK Natural Science Collections Association, (NatSCA 2005) use cases
In 2005, the Natural Science Collections Association, UK, published a report entitled, “A
matter of life and death: Natural science collections: why keep them and why fund them?”,
in which they emphasized the importance of NHCs and the uses they serve.
21 As of 2016, the name reflects the year of publication of the Science Review rather than the papers summarised
herein. As such, the 2016 Science Review chiefly summarises articles published in 2015.
Page 23 | 41
http://www.natsca.org/sites/default/files/publications-full/A-Matter-Of-Life-And-Death.pdf
US Interagency Working Group on Scientific Collections (IWGSC 2009) use cases
The IWGSC report published in 2009 documents a number of use cases for the US federal
scientific collections, both biological and non-biological. The US federal collections are
viewed as part of the global scientific infrastructure and enterprise. The use cases illustrate
the impact of scientific collections in the following areas: economy and trade, environmental
change over time, environmental quality, invasive species, scientific treasures, food and
agriculture, public health and safety, national security and unanticipated uses.
https://usfsc.nal.usda.gov/sites/usfsc.nal.usda.gov/files/IWGSC_GreenReport_FINAL_2009.pdf
Virginia Tech - Biological collections as a resource for technical innovation
Virginia Tech recently launched a three-year project (2015-2017) funded by the US National
Science Foundation that addresses the unanticipated uses of biological collections -
biocollections inspiring engineering innovation. The project aims at giving scientists and
engineers from diverse backgrounds specific suggestions as to how natural history
collections could be leveraged for engineering innovation. Secondly it will also provide policy
makers and the general public a well-justified outline of the innovative and economic
potential of natural history collections as well as estimates for the effort that would be
required to realize this potential.
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1521072.
Outcomes for this project will be archived at http://bist.centers.vt.edu.
Page 24 | 41
Data Gap Analysis: Setting priorities for digitization
For all of biodiversity scienceand knowledge in generaldata gap analysis (DGA) enables
us to “know what we don’t know” and then to prioritize the filling of those gaps according to
strategic imperatives (Arturo et al 2015). With regard to biocollections institutions, worldwide
they are faced with the challenging dilemma of how to prioritize the digitization of massive
amounts of data associated with millions of voucher specimens of animals, plants, fungi and
other organisms that document the planet’s biodiversity. Setting digitization priorities,
informed by data gap analyses, is essential to having the best biodiversity data enable the
best science for understanding and advancing social, economic and environmental well-
beingand to do so in time to make a difference. Indeed, setting such gap-based priorities
is incumbent on biocollections if they are to speed the flow of data-to-knowledge-to-
application in the value chain.
The Task Force recognizes that setting digitization priorities involves serving competing
institutional, local, regional, national and global imperatives. These include, but are not
limited to: individual investigator research interests; institutional mandates; science agendas;
and various pressing environmental concerns (e.g., endangered species, invasive species,
zoonotic diseases, pollinators, pests). Moreover, each imperative has its particular calculus
of taxonomic groups, geographic areas, time periods, and ecosystems/habitats. Overlying
these permutations are the missions of different stakeholders and funders:
intergovernmental bodies (e.g., IPBES, CBD), government agencies, NGOs, private
foundations, and corporations. In a resource-limited world, a digitization strategy of
maximum efficacy will require all parties to collaborate on setting overarching priorities that,
simultaneously: (1) target the most urgent environmental imperatives of our time; (2) are
underpinned by sophisticated data gap analyses of those imperatives; (3) include the
greatest commonality among competing interests; (4) tackle what is most pragmatic; and (5)
promise the most immediate impacts (see Recommendations).
To that end, the Task Force convened a symposium on “Setting Global and Local
Digitization Priorities” at the SPNHC conference in Berlin, June 20-25, 2016
(http://www.spnhc2016.berlin/page40.html). The five presentations in the symposium
centred on setting digitization priorities in a variety of situations from the global, regional,
national and institutional level, and across different collection sizes and themes to satisfy a
variety of competing needs. A summary of the results from the NHC survey (section 2 in this
report) including how respondents have set digitization priorities, was also presented.
The task force also convened a side meeting to plan the review of the Natural Collections
Description (NCD) standard and metadata needs for NHCs.
Page 25 | 41
Next Steps
The Task Force plans to deliver on the following goals and action plans by the end of its
tenure.
Draft a priority-setting framework for individual biocollections institutions.
Based on the draft framework, convene a series of meetings with stakeholders to
develop strategic frameworks for helping biocollections deliberate and set their
digitization priorities.
In partnership with the RDA/TDWG joint working group on metadata standards for
the sciences, evaluate the application of NCD standard
Develop roadmap documents to assist institutions in mobilizing biocollections
metadata
Help form a closer-working cooperative network of global biocollection entities and
societies to achieve a critical mass for planning, policy impact, and generating
resources.
Hone and tailor biocollections use-cases for specific communities (researchers,
corporations, foundations, policy makers, educators, etc.) to demonstrate the benefit
of published, vouchered biodiversity data for science, society, governments, and the
private sector across a series of thematic imperatives.
Convene major summits of government, corporate and foundation institutions to
develop a funding mechanism to complete the strategic, priority-based digitization of
biocollections data worldwide.
Page 26 | 41
Recommendations
Setting priorities for digitization and data gap analysis
1. The community, stakeholders and individual biocollections should establish
collaborative, integrated priorities for digitization of biocollections data based on
the value framework of data-knowledge-application. Within this framework,
priorities should be demand-driven by global, national, regional, and local
concerns and required research that simultaneously: (a) target the most urgent
social, environmental, economic and biodiversity science imperatives of our time;
(b) are underpinned by sophisticated gap analyses; (c) include the greatest
commonality among competing imperatives and interests; (d) tackle what is most
pragmatic, first; and (e) promise the most immediate impacts.
Such a demand-driven approach addresses both links in the value chain of data-
to-knowledge-to-application. Fulfilling the first linkdata-knowledgeprovides
the raw, vouchered data that catalyses research results that inform solutions.
Fulfilling the second linkknowledge-to-applicationrecruits demand-driven
investments for such solutions from governments, foundations, and the corporate
sector. Mathematical algorithms may be a useful tool for calculating and
modelling such priorities (see Butts et al 2010).
Extensive gap analyses of existing digitized data will identify the critical
taxonomic and geographic data gaps and data enhancements (e.g.,
georeferencing) that need to be filled to address the strategic priorities. The
GBIF DGA report by Arturo et al (2016) provides such DGA methodologies.
Critical geographic gaps can be inferred from GBIF's overall occurrences plot
densities - http://www.gbif.org/occurrence. Community data aggregators such as
GBIF and iDigBio should work together on providing robust DGAs.
2. The community, led by GBIF and international and national partners, should
convene a series of high-level summits among biocollections, governments,
corporations and foundations to develop, fund and implement the five, long-term
strategies (Recommendation 1) for mobilization the remaining 90% of the world’s
biocollections data and bringing them into currency for science and society. A
component of this investment to be explored is an international fund for
accelerating digitization of biodiversity data in regions with the largest data gaps.
Digitization and best practices
3. Biocollections should employ the proven best practices and community standards
to digitize their collections, with examples and guidance from GBIF, iDigBio,
SPNHC, ALA, TDWG, etc. One of the best practices is adoption of a tiered
strategy for worldwide collections digitization (plus imaging where appropriate)
and model concepts, such as Linked Open Data (LOD) (Berner-Lee 2009) and
the Digitization Maturity Model in the ALA’s Guide to Digitization (Kalm 2012).
Specifically, the five tiers in this strategy are:
a) Level OneMetadata I: Make the collection globally discoverable by
publishing the Institution/Organization/Collection-level information, i.e., the
who and where of a collection and, broadly, its content and history, and by
registering collections with GRBio.
b) Level Two - Metadata II: Produce and publish species-level or cabinet-level
collection inventories as a basis for strategic planning of collection growth,
conservation and follow-on digitization.
c) Level ThreeSpecimen Data I: Capture and publish skeletal-level data (with
or without locality georeferencing), mapping data to currently accepted
Page 27 | 41
standards, and exposing imperfect or erroneous data to machine algorithms
and global expertise for correction and improvement.
d) Level FourSpecimen Data II: Georeference locality data and enrich
specimen records with other data, such as: field notes, grey literature, note
cards, etc.
e) Level FiveSpecimen Data III: Data associated with all new collections is
“born digital”, i.e., immediately captured and incorporated into the existing,
georeferenced database. Links to GenBank accessions, BCoL IDs, etc., must
include globally unique identifiers.
Global collaboration, where possible, can help meet Recommendation 1 (see
above) to address global imperatives and to achieve critical mass, least
redundancy, and economies of scale.
4. Biocollections should work with community partners to remove their perceived
barriers to their digitization efforts. The three major barriers cited by respondents
in the TF survey have been successfully eased or removed by numerous
institutions and biocollections. For example, setting strategic priorities and
schedules for data capture will ease the perception that the “task is
overwhelming.” Productive collection-based research and funding from
government and private entities can reverse the perception that the
efforts/resources expended exceed the benefits of digitization. And digitization of
collections is the fastest, most efficient method of identifying and correcting
collection data errors, often en mass, so that both the physical specimens and
their data can be readily deployed and trusted for the very purposes those
collections were intended to serve. Many biocollections-focused organizations,
e.g., iDigBio, BCoN, NSCA, SYNTHESYS3, SPNHC, TDWG, etc. can provide the
expertise, training, and tools to advance efficient and effective collection data
capture and publication to the worldwide community.
Metadata
5. Natural history collections should publish their rich metadata for both digitized
and non-digitized specimens in order to make them discoverable and advertise
their value to science and society.
6. To accomplish this, the community should develop robust yet user-friendly APIs
for metadata, and evolve metadata standards, including resolving the semantic
problem with “metadata” as it means different things to different people. Current
metadata standards could be adapted and emended, for example, Ecological
Metadata Language (EML) with an extended profile for NHC, and mapping data
from Natural Collections Description (NCD) to EML.
7. GBIF’s dataset metadata should be presented in a structured way so that it can
be searchable by using various parameters
8. The community, led by GBIF, SPNHC, TDWG, GRBio, iDigBio and other entities,
should develop an educational campaign on the importance of metadata, how it
advertises the collection to the world of users and its value, how to capture and
report critical metadata fields, etc.
9. Following the lead of the IWGSC, government agencies should improve and
provide access to the documentation of the contents of their scientific collections.
Partnership and collaboration
10. Current biocollections-centred organizations should strongly consider integration
into a federated union with greater critical mass, impact and effectiveness
nationally and internationally. For example, organizations that focus mainly on
Page 28 | 41
physical specimens and their data (e.g., SPNHC), data standards (TDWG),
collections metadata (GRBio), policy and funding (NSCA), data portals (GBIF,
VertNet, BISON), and regional/national efforts (iDigBio, Sibs22, NBN23, ALA, etc.)
have extensively overlapping missions and interests that would be less dispersed
and more efficiently fulfilled through federated cooperation. One model in this
domain is the World Federation of Culture Collections.
11. Following the example of the IWGSC (2009), agencies should collaborate much
more closely in setting and implementing the policies, procedures and protocols
for managing their scientific collections.
22 http://www.sibbr.gov.br/
23 https://data.nbn.org.uk/
Page 29 | 41
References
Arturo, A., Chavan, V. and Otegui, J. 2016. Best Practice Guide for Data Gap Analysis for
biodiversity stakeholders. http://www.gbif.org/resource/82566
Barnosky, A. D. 2008. Megafauna biomass tradeoff as a driver of Quaternary and future
extinctions. Proc. Nat. Acad. Sci. 105, Supplement 1: 11543-11548
Barnosky, A. D., Matzke, N., Tomiya, S., Wogan, G.O.U., Swartz, B., Quental, T. B.,
Marshall, C., McGuire, J., Lindsey, E. L., Maguire, K., Mersey, B. & Ferrer, E. 2011. Has
the Earth’s sixth mass extinction already arrived? Nature 471:51-57.
Berendsohn, W.G., Chavan, V., Macklin, J. 2010. Summary of Recommendations of the
GBIF Task Group on the Global Strategy and Action Plan for the Digitisation of Natural
History Collections. Biodiversity Informatics, [S.l.], v. 7, n. 2, oct. 2010. ISSN 15469735.
Available at: https://journals.ku.edu/index.php/jbi/article/view/3989. Date accessed: 06
may 2016. doi:http://dx.doi.org/10.17161/bi.v7i2.3989.
Berg, W., Johnels, A., Sjostrand, B., and Westermark, T. 1966. Mercury contamination in
feathers of Swedish birds from the past 100 years. Oikos 17:7183.
Berner-Lee, T. 2009. Linked Data. https://www.w3.org/DesignIssues/LinkedData.html
Blach-Overgaard, A., Balslev, H., Dransfield, J., Normand, S. & Svenning, J. C. 2015.
Global-change vulnerability of a key plant resource, the African palms. Sci. Rep. 5,
12611; doi: 10.1038/srep12611
Blach-Overgaard, A., Svenning, J. C., Dransfield, J., Greve, M. & Balslev, H. 2010.
Determinants of palm species distributions across Africa: the relative roles of climate,
non-climatic environmental factors, and spatial constraints. Ecography 33, 380391;
DOI: 10.1111/j.1600-0587.2010.06273.x
Blagoderov, V., Kitching, I.J., Livermore, L., Simonsen, T.J., and Smith, V.S. 2012. No
specimen left behind: industrial scale digitization of natural history collections. ZooKeys
209:133-146. doi:10.3897/zookeys.209.3178
Bogoch, I.I., Brady, O.J., Kraemer, M.U., German, M., Creatore, M.I., Kulkarni, M.A. et al.
2016. Anticipating the international spread of Zika virus from Brazil. Lancet 387:335336.
http://dx.doi.org/10.1016/S0140-6736(16)00080-5
Brooks, R.R., Malaisse, F., and Empain, A. 1985. The heavy metal tolerant flora of
southcentral Africa: A multidisciplinary approach. A.A. Balkema, Rotterdam and Boston.
Available at: http://trove.nla.gov.au/work/18240023
Buerki, S. and Baker, W. 2016. Collections-based research in the genomic era. Biological
Journal of the Linnean Society 117:5-10. http://dx.doi.org/10.1111/bij.12721
Butts, S.H., Bazeley, J.A., and Briggs, D.E.G. 2010. A curatorial assessment for stratigraphic
collections to determine suitability for incorporation into a systematic collection.
Collection Forum, 24(1-2):46-51. Available at: www.spnhc.org/media/assets/cofo-24.pdf
Cao-Lormeau, V.M., and Musso, D. 2014. Emerging arboviruses in the Pacific. Lancet 384:
15711572. http://dx.doi.org/10.1016/S0140-6736(14)61977-2
Ceballos, G., Ehrlich, P.R., Barnosky, A.D., García, A., Pringle, R.M. and Palmer, T.M.,
2015. Accelerated modern humaninduced species losses: Entering the sixth mass
extinction. Science advances, 1(5), p.e1400253. DOI: 10.1126/sciadv.1400253
Chapman, A.D. 2005. Uses of Primary Species-Occurrence Data, version 1.0. Report for the
Global Biodiversity Information Facility, Copenhagen. 100 pp. Available online at:
http://www.gbif.org/resource/80545
Page 30 | 41
Dunn, P. O. and Winkler, D.W. 1999. Climate change has affected the breeding date of tree
swallows throughout North America. Proceedings of the Royal Society of London B 266.
24872490.
Ernest, W.H.O. 2006. Evolution of metal tolerance in higher plants. For. Snow Landsc. Res.
80, 3: 251274. http://www.wsl.ch/publikationen/pdf/7764.pdf
Fichet-Calvet, E., Rogers, D.J. 2009. Risk maps of Lassa fever in West Africa. PLoS
Neglected Tropical Diseases 3: e388. doi: 10.1371/journal.pntd.0000388
Gaiji, S., Chavan, V., Arino, A.H., Otegui, J., Hobern, D., Sood, R., and Robles, E. 2013.
Content assessment of the primary biodiversity data published through GBIF network:
Status, challenges and potentials. Biodiversity Informatics, [S.l.], v. 8, n. 2, July 2013.
doi:http://dx.doi.org/10.17161/bi.v8i2.4124
Haggerty, S. 2015. Discovery of a kimberlite pipe and recognition of a botanical diagnostic
indicator in NW Liberia. Economic Geology 110: 851-856. DOI:
10.2113/econgeo.110.4.851
Hayes, T., Haston, M., Tsui, M., Hoang, A., Haeffele, C., and Vonk, A. 2002. Herbicides:
Feminization of male frogs in the wild. Nature 419:895896.
Heerlien, M., van Leusen, J., Schnoerr, S., de Jong-Kole, S., Raes, N., and van Hulsen, K.
2015. The natural history production line: An industrial approach to the digitization of
scientific collections. ACM Journal of Computing and Cultural Heritage 8(1), Article 3, 11
pages. DOI: http://dx.doi.org/10.1145/2644822
Hickey, J. J. and Anderson, D.W. 1968. Chlorinated hydrocarbons and eggshell changes in
raptorial and fish-eating birds. Science 162:271273
Hoffmaster, A. R., Fitzgerald, C.C., Ribot, E., Mayer, L.W., and Popovic, T. 2002. Molecular
subtyping of Bacillus anthracis and the 2001 bioterrorism-associated anthrax outbreak,
United States. Emerging Infectious Diseases 8:11111116.
Holmes, M.W., Hammond,T.T., Wogan, G.O., Walsh, R.E., LaBarbera, K., Wommack, E.A.,
Martin, F.M., Crawford, J.C., Mack. K.L, Bloch, L.M., and Nachman, M.W. 2016. Natural
history collections as windows on evolutionary processes.Molecular Ecology 25(4):864-
8. doi: 10.1111/mec.13529.
Hudson, L.N., Blagoderov, V., Heaton, A., Holtzhausen, P., Livermore, L., Price, B.W., et al.
2015. Inselect: Automating the Digitization of Natural History Collections. PLoS ONE
10(11): e0143402. doi:10.1371/journal.pone.0143402
IWGSC (National Science and Technology Council, Committee on Science, Interagency
Working Group on Scientific Collections). 2009. Scientific Collections: Mission-Critical
Infrastructure of Federal Science Agencies. Office of Science and Technology Policy,
Washington, DC.
Jackson, J. 2016. Digital Collections: the Cisco Pitstop | Digital Museum
https://blog.nhm.ac.uk/2016/02/26/digital-collections-the-cisco-pitstop/ 26 February 2016.
Jones, M. C., Dye, S.R., Pinnegar, J.K., Warren, R and Cheung, W.W.L. 2014. Using
scenarios to project the changing profitability of fisheries under climate change. Fish and
Fisheries 16: 603-622. DOI: 10.1111/faf.12081
Kalms, B. 2012. Digitisation: A strategic approach for natural history collections. Atlas of
Living Australia, CSIRO Ecosystem Sciences, Canberra, ACT, Australia.
http://www.ala.org.au/wp-content/uploads/2011/10/Digitisation-guide-120326.pdf
Kraemer, M.U.G., Sinka, M.E., Duda, K.A., et al. 2015. The global distribution of the
arbovirus vectors Aedes aegypti and Ae. albopictus. eLife, 4 (2015), p. e08347.
http://dx.doi.org/10.7554/eLife.08347
Page 31 | 41
Kremen, C., Cameron, A., Moilanen, A., Phillips, S.J., Thomas, C.D., Beentje, H., Dransfield,
J., Fisher, B.L., Glaw, F., Good, T.C., Harper, G.J., Hijmans, R.J., Lees, D.C., Louis Jr,
E., Nussbaum, R.A., Raxworthy, C.J., Razafimpahanana, A., Schatz, G.E., Vences, M.,
Vieites, D.R., Wright, P.C., Zjhra, M.L. 2008. Aligning Conservation Priorities Across
Taxa in Madagascar with High-Resolution Planning Tools. Science 320: 222-226. DOI:
10.1126/science.1155193
Lendemer, J. C. & Allen, J. L. 2014. Lichen biodiversity under threat from sea-level rise in
the Atlantic Coastal Plain. Bioscience 64: 923-931.
Maron, N. L., & Pickle, S. 2013. Appraising our Digital Investment: Sustainability of Digitized
Special Collections in ARL Libraries. Retrieved from http://sr.ithaka.org?p=22363
Mason, Jr. S. C., Betancourt, I., Gelhaus, J. K. 2016. A digital species index of the
entomology collection at the Academy of Natural Sciences of Drexel University.
Presented at Entomological Collections Network Meeeting September 23, 2016.
Orlando, Florida, USA
Millennium Ecosystem Assessment. 2005. Ecosystems and Human Well-Being: Biodiversity
Synthesis. World Resources Institute.
Miller-Struttmann, N. E., Geib, J., Franklin, J.D., and Galen, C. 2015. Functional mismatch in
a bumble bee pollination mutualism under climate change. Science 349: 1541-1544.
DOI: 10.1126/science.aab0868
Nelson, G., Paul, D., Riccardi, G., Mast, A. (2012) Five task clusters that enable efficient and
effective digitization of biological collections. ZooKeys 209: 19-45. doi:
10.3897/zookeys.209.3135
Novak, S. and Mack, R. 2001. Tracing plant introduction and spread: Genetic evidence from
Bromus tectorum (Cheatgrass). BioScience 51: 114-112.
NRC (National Research Council). 2003. Countering Agricultural Bioterrorism. Washington
(DC): National Academies Press.
Parmesan, C. 1996. Climate and species' range. Nature 382:765766.
Pergams, O.R. and Nyberg, D. 2001. Museum collections of mammals corroborate the
exceptional decline of prairie habitat in the Chicago region. Journal of Mammology 82:
984-992.
Peterson A.T., Moses, L.M., Bausch, D.G. 2014. Mapping Transmission Risk of Lassa Fever
in West Africa: The Importance of Quality Control, Sampling Bias, and Error Weighting.
PLoS ONE 9(8): e100711. doi:10.1371/journal.pone.0100711
Peterson, A. T., Soberón, J., and Krishtalka, L. 2015. A global perspective on decadal
challenges and priorities in biodiversity informatics. BMC Ecology, 15, 15.
http://doi.org/10.1186/s12898-015-0046-8
Piggott, D. M., Golding, N., Adrian, M., Zhi, H. et al. 2014. Mapping the zoonotic niche of
Ebola virus disease in Africa. DOI: http://dx.doi.org/10.7554/eLife.04395
Provan, J., Booth, D., Todd, N.P. and Maggs, C.A. 2007. Tracking biological invasions in
space and time: elucidating the invasive history of the green alga Codium fragile using
old DNA. Diversity and Distributions 14: 343-354.
Ratcliffe, D. A. 1967. Decrease in eggshell weight in certain birds of prey. Nature 215:208
210.
Sanchez A. C., Osborne, P.E. and Haq, N. 2011. Climate change and the African baobab
(Adansonia digitata L.): the need for better conservation strategies. African Journal of
Ecology 49: 234245. doi:10.1111/j.1365-2028.2011.01257.x
Page 32 | 41
Schindel, D., Miller, S., Trizna, M., Graham, E., Crane, A. 2016. The Global Registry of
Biodiversity Repositories: A Call for Community Curation. Biodiversity Data Journal 4:
e10293. doi: 10.3897/BDJ.4.e10293
Scoble, M. 2010. Rationale and Value of Natural History Collections Digitisation. Biodiversity
Informatics, [S.l.], v. 7, n. 2, oct. 2010. ISSN 15469735. Doi:
http://dx.doi.org/10.17161/bi.v7i2.3994.
Sousa-Baena, M.S., Couto Garcia, L. & Peterson, A.T. 2013a. Completeness of digital
accessible knowledge of the plants of Brazil and priorities for survey and inventory.
Diversity and Distributions, 2013, pp.113.
Sousa-Baena, M.S., Couto Garcia, L. & Peterson, A.T. 2013b. Knowledge behind
conservation status decisions: Data basis for “Data Deficient” Brazilian plant
species.Biological Conservation 173: 80-89.
http://dx.doi.org/10.1016/j.biocon.2013.06.034
Suarez, A. V. and Tsutui, N.D. 2004. The value of museum collections for research and
society. Bioscience 54: 66-74.
Suarez, A. V., Holway, D.A. and Case, T.J.. 2001. Patterns of spread in biological invasions
dominated by long-distance jump dispersal: Insights from Argentine ants. Proceedings of
the National Academy of Sciences 98:10951100
Vollmar, A., Macklin, J.A., Ford, L. 2010. Natural History Specimen Digitization: challenges
and concerns. Biodiversity Informatics, 7: 93 112. DOI:
http://dx.doi.org/10.17161/bi.v7i2.3992
Yates, T. L. 2002. The ecology and evolutionary history of an emergent disease: Hantavirus
pulmonary syndrome. BioScience 52:989998.
Yoshida, K., Schuenemann, V.J., Cano, L.M., Pais, M., Mishra, B. et al. 2013. The rise and
fall of the Phytophthora infestans lineage that triggered the Irish potato famine. DOI:
http://dx.doi.org/10.7554/eLife.00731
Page 33 | 41
Annex I: Acronyms and abbreviations
ADBC
Advancing the Digitization of Biological Collections
ALA
Atlas of Living Australia
API
Application programming interface
BCOL
Barcode of Life
BCoN
Biodiversity Collections Network
BID
Biodiversity Information for Development
BISON
Biodiversity Information Serving Our Nation, USA
Canadensys
Network of Canadian biological collections
CBD
Convention on Biological Diversity
CETAF
Consortium of European Taxonomic Facilities
GRBio
Global Registry of Biodiversity Repositories
iDigBio
Integrated Digitized Biocollections, USA
IPBES
The Intergovernmental science-policy Platform on Biodiversity and Ecosystem Services
IWGSC
Interagency Working Group on Scientific Collections
MUSEUM-L
A general purpose, cross-disciplinary electronic discussion list for museum professionals,
students, and all others interested in museum related issues, under the auspices of the
International Council of Museums
NBN
National Biodiversity Network, UK
NCD
Natural Collections Description
NGO
Non-Governmental Organization
NHC
Natural History Collections
Nicol-L
Natural History Collections List of Society for the Preservation of Natural History Collections
NSCA
Natural Science Collections Alliance, USA
RDA
Research Data Alliance
SERNEC
SouthEast Regional Network of Expertise and Collections, USA
SiBBr
Sistema de Informação sobre a Biodiversidade Brasileira (Information System on Brazilian
Biodiversity)
SPNHC
Society for the Preservation of Natural History Collections
SYNTHESYS
An EU-funded project creating an integrated European infrastructure for natural history
collections
Taxacom
Biological Systematics Discussion List
TDWG
Biodiversity Information Standards also known as Taxonomic Databases Working Group
TF
The GBIF Task Force
VertNet
Database of vertebrate biodiversity data from natural history collections
WeDigBio
Worldwide Engagement for Digitizing Biocollections
Page 34 | 41
Annex II: Summary of results from a first analysis of NHC survey
GBIF Task Force Report on Survey Subsection: A Global Survey of Natural History Collections.
http://www.gbif.org/newsroom/news/accelerating-discovery-of-biocollections-data
Introduction
As part of a broader global strategy for mobilizing primary biodiversity data, GBIF convened
a Task Force (TF) to help accelerate the discovery and access to both digitized and non-
digitized collections24. In the initial meetings the TF realized that it was important to get data
on the current state of NHCs in order to inform future discussions and recommendations.
Some of the pertinent questions that we sought to understand included the following:
o What methods and models are proving most successful for worldwide collections
digitization?
o What metadata are key to sharing so that other stakeholders can discover
collections?
o Who is doing the digitization?
o What are the obstacles most often encountered?
o What variables are driving what gets digitized?
With no available data, the GBIF Task Force on Accelerating the Discovery of Biocollections
Data decided to carry out a global survey as the best way to find out the state of affairs of
NHCs worldwide.
Purpose of survey
In the late 2015, the task force designed and administered a global survey on natural history
collections. The purpose of the survey was to enable the task force to determine and
demonstrate: (1) The digital readiness of the world’s biocollections and their institutions; (2)
the benefits to the collection/institution that digitization engenders; and (3) the impediments
to collection data digitization.
Survey methodology
The survey questionnaire was prepared by the TF through meetings, consultations and
research, and mainly administered online through Qualtrics software. An MS Word version
was made available for respondents that had difficulty in filling out the online survey. We
distributed the survey to individuals affiliated with collections worldwide through a variety of
channels including listservs and member lists including Index Herbariorum, and GBIF nodes,
as well as personal contacts with institutions. We also announced the survey at relevant
meetings, workshops and conferences attended by potential target respondents. The survey
was translated into Japanese and distributed throughout Japan via the community of
“Science Museum Network (S-Net)”. Those contacted were further requested to help with
the further distribution of the survey in order to help maximize the global reach. Table 1
summarizes the key distribution channels.
24 http://www.gbif.org/newsroom/news/accelerating-discovery-of-biocollections-data
Page 35 | 41
Table 1 Main distribution channels for the global NHC survey
Taxacom
Herbaria-L
iDigBio
GBIF Node Managers
SERNEC
NHColl
GRBio
Index Herbariorum
TDWG
MUSEUM-L
S-Net (http://science-net.kahaku.go.jp/)
Conferences, e.g., TDWG (Nairobi), iDigBio V Summit (Washington DC), ECN, etc.
Summary of survey results.
We present a summary of the survey results here but a more detailed report with further
analyses, interpretation, and discussions will be made available at a later date as part of the
final report of the GBIF Task Force. Over 800 responses representing nearly 2000
collections distributed over 72 countries were received (Map 1). Deleting very incomplete
responses resulted in 617 usable responses. The survey had a wide geographic reach
across all the major continents but with much higher response rates from Western Europe,
North America in particular the USA and Japan because the survey was translated into
Japanese. Note that the number of respondents that answered each question varied either
because either in the case of the online version, they were not shown the question based on
prior responses, or the respondents elected to not answer the question. We present a
summary of the key results.
Map 1 shows the locations of the survey respondents based on the registered IP addresses
from where the online survey was completed.
Map 1. Survey respondents by IP address, by Kevin Love, iDigBio, using CartoDB
Page 36 | 41
Table 2. Summary of Key data from the global NHCs survey
86% (615 respondents) indicate they are currently digitizing or have completed digitizing at least some or all
of their collections.
76% of respondents were individuals based at publicly funded institutions, with 40% universities and 36%
non-university institutions. Nearly all (92%) of respondents were primarily curators or collection managers
with 10% as head of research and collections.
13 out of 15 collection types, when averaged, report their collections over 50% electronically databased.
Very few respondents (1% or 5 individuals) reported they are not digitizing and have no plans to do so.
The top 10 obstacles to digitization are: funding, time (lack of), size of task, not institutional priority, data has
errors, limited expertise, lack of digitization process knowledge, no credit (tenure, reappointment) for this
work, not priority for those in administration, not a good payoff for needed effort
The top three priorities when deciding what to digitize are research (53%) and funding / grant opportunities
(51%), followed by taxonomic priorities (42%).
Benefits of digitization - top eight responses: increased use of collections, increased exposure, better
knowledge of holdings, better management of data, digital preservation, enhanced data quality, new skills for
staff, better management of physical specimens
Details
Collections responding
Collection type response rate varied (see Figure 1). Vascular plant collections represent
nearly 20% of total respondents followed by Bryophytes (10%), Fungi (10%), Algae (9%),
Arthropods (8%), Mammalogy (6%), Ornithology (5%), Herpetology (5%), and Ichthyology
(5%) with the remaining representing Malacology, Marine Invertebrates, Terrestrial
Invertebrates, Invertebrate Fossils, Paleobotany, Vertebrate Fossils, and other.
Figure 1. Collection types represented by our survey respondents (number of collections n=1992)
Page 37 | 41
Staff responding to survey
For n=615 respondents, curators made up 57%, collection managers 35%, faculty 34%,
head of research and collections 13%, information manager 10%, director/CEO 8%, and
other 4%.
Digitization and databasing trends
Overall, our data seems to indicate that most collections are digitizing at least some part of
their collections or trying to do so. Of the 615 respondents who answered the question, 86%
indicated that they are currently or have completed databasing their collections. The
percentage varies across collection type, as does the mean portion of the collection that has
been completed.
If we average answers for “percent of collection databased” and group by collection type, we
are encouraged to note that 13 out of 15 collection types are over 50% databased,
collectively (Figure 2). In other words, if one vascular plant collection said it was 25%
digitized, and another vascular plant collection said it was 75% digitized, then between the
two collections - they are 50% digitized.
When looking at these averages of the percent of collection databased, for each organismal
group, it’s notable that the averages of all collection types is over 50% for all groups except
arthropods which averages 38% and invertebrate fossils which averages 47% (Figure 3).
Published collections
The mean percentage of collections that have been published is higher than the mean
percentage of those databased, but that is because the published statistic reflects those that
have already been databased. It is based on a smaller number of collections. To be exact,
the number of those who reported publishing their collection data is one-third smaller than
those who reported databasing their collections.
Figure 2. If digitizing all or part of a collection, what percent is databased?
Page 38 | 41
We can illustrate with vascular plants. We have 375 vascular plant collections in the data set
(or, more precisely, 375 people provided information about vascular plant collections). Of
these 274 report databasing their collections. On average, 55% of the collections are
databased. This is averaged across all collections that provided an actual percentage. A
smaller number of individuals, 215, reported publishing their collection data. On average,
65% of the collection data is published (Figure 4).
Barriers to digitization
If many are digitizing, or trying to do so, what barriers get in the way of digitizing? If we know
what these barriers are, how can we use this information going forward? (See
recommendations in interim and final report). Although only 1% of respondents (n=5
individuals) indicated that there are no plans to digitize their collections, far more people
(n=587) offered to share what they experience as obstacles to digitization most notably
lack of funding/resources and lack of time. See similar insightful data from ITHAKA for
digitization of special collections (Maron & Pickle 2013).
Figure 3. Averaging the percent of collection databased, across each collection type
Figure 4. Percent of collections published (partial or complete)
Page 39 | 41
Table 3. The top 10 barriers to digitization, for our survey respondents (n=587)
1. Funding / resources not available (80%)
2. Lack of time among personnel (80%)
3. Size of task is overwhelming (40%)
4. Not an institutional priority (35%)
5. Collection data has errors (30%)
6. Limited expertise among personnel (25%)
7. Insufficient information on digitization process (15%)
8. No benefit to reappointment, tenure (14%)
9. Not a priority of the individual in charge (12%)
10. Not a good effort / payoff ratio (10%)
The reasons beyond the first two, can be grouped into 3 categories as suggested here, in
order to address them (Table 4). (See recommendations in upcoming interim and final
report).
Table 4. Obstacles to digitization - 3 groups
size of task is overwhelming
not an institutional priority, no benefit, not a priority, not good effort / payoff ratio,
not priority, lack of perceived need, deemed not valuable
data has errors, limited expertise, lacking information on the digitization process
How are collections deciding what to digitize?
With 519 respondents, the top three variables driving the decisions for what to digitize include:
research priority (52%), funding/grant opportunities (51%), and taxonomic priority (42%). Other
reasons selected were: partnership in a larger community effort (24%), and geographic priority (23%).
About 18 percent reported their digitization is opportunistic with no target priority, and about 5% said
health and human service needs drive their digitization decisions. (Figure 5)
Figure 5. How collections are prioritizing digitization choices
Page 40 | 41
Who is doing the digitization?
Collections rely on a variety of personnel to perform digitization tasks usually including staff, students,
volunteers, and rarely, third party organizations (2 to 10%). Most often, digitization is being completed
by paid personnel, with 90% of respondents reporting that paid staff or students “frequently” perform
the tasks. While 53% of respondents reported paid staff “frequently” do the digitization, 59% reported
that students (paid, unpaid, or volunteers) “frequently” digitize their collections. (Figure 6)
Funding
How are collections paying for this work? Of the 523 respondents who answered the question, 87%
indicated that they had received at least some funding to support digitization. Of these, 69% reported
that they received external funds with 61% receiving regular institutional funding. Slightly more than
half (53%) reported receiving only one of these types of funding (i.e., external, institutional), while
36% received funds from two types of sources and 30% from three. Most of the external funding
(80%) is from government sources with nearly three-quarters of respondents (72%) receiving funds
from just one type of external source (i.e., industry, other, foundations, government).
Benefits of digitization
The top 8 responses (from 516 respondents) given for benefits of digitization include: increased use
of collections, increased exposure, better knowledge of holdings, better management of data, digital
preservation, enhanced data quality, new skills for staff, and better management of physical
specimens. Thirty percent report new communities using the data, 35% see increased publicity and
reduced physical handling of the collection, and 40% share awareness of an increased use of their
collection data in research and publications and an increased public awareness of the importance of
collections.
Institutional Commitment
Of the 516 individuals responding to this question, 72% indicated their institutions were committed to
sustainability (or responded positively to a later question that asked about ways their institutions were
sustaining digitization). Of 422 respondents, 50% are providing staff training and seeking continued
funding. Over 60% have long term archival storage plans in place, and over 70% have plans in place
for long term curation of the data.
Figure 6. Who is doing the digitization work?
Page 41 | 41
Metadata
Metadata is the information about your collection/s (e.g., taxonomic, geospatial and occurrence
coverage, collection contacts, etc.). As a community, what metadata do we need to effectively move
forward with strategic digitization and data mobilization? To address this, we first need to know what
metadata collections currently provide and consider important. 419 respondents answered the
question, see Figure 7.
We also asked respondents to rank the importance of sharing each type of metadata. Over 50%
marked taxonomic and geographic range of the data as critical to share and 45% marked Type
Specimen data as important. Metadata values ranked as critical by 20% to 30% of respondents
included: percent collection digitized, notable publications, percent georeferenced, place name
coverage, name of collection manager, collection size, and name of curator. 506 Respondents report
most metadata is being shared via GBIF (43%), Index Herbariorum (36%), and Institutional Websites
(21%). (See recommendations in interim and final report for providing and accessing metadata). Note
that reason number one in Table 4, a lack of funding and / or resources, provides a perfect reason to
seek digitization opportunities by sharing metadata to increase discoverability / visibility.
Long term plans
Encouragingly, over 80% of 513 respondents indicate their institution / organization intends to digitize
their entire collection/s. Over 30% of these same respondents share they plan to focus on digitizing
specific areas of their collections in response to research needs / requests. About 12% indicate they
are planning to digitize the parts of their collections that they find strategically important or unique
such as type specimens or endemic species. Others (about 12 %) share they plan only to digitize new
additions to the collections.
Next Steps
Please be on the lookout for our Interim Report for our recommendations and more details about our
findings. Some of this work was presented in detail at The Society for the Preservation of Natural
History Collections (SPNHC) 2016 Conference in Berlin in June and Botany 2016 in August and
looking for community input and feedback. We plan to release our final report near the end 2016. If
you have questions or you’d like to discuss any of these points further, please do contact us through
our Task Force Chair: Leonard Krishtalka (krishtalka@ku.edu) or our GBIF Programme Officer for
Content Mobilisation, Siro Masinde (smasinde@gbif.org).
Survey subsection authors
Shari Ellis, iDigBio Project Evaluator, Deborah L Paul, iDigBio Digitization and Workforce Training
Specialist
Reference
Maron, N. L., & Pickle, S. (2013). Appraising our Digital Investment: Sustainability of Digitized Special
Collections in ARL Libraries. Retrieved from http://sr.ithaka.org?p=22363. Complete ITHAKA Survey
Results here: http://www.sr.ithaka.org/wp-content/uploads/2015/08/digitized-special-collections-
report-slides-15feb13.pdf
Annex 2 was emailed to survey respondents on 16 August 2016
Figure 7 Metadata provided by collections
... In the recent past, criteria for the prioritization of digitization have been inventoried several times, for example by the Global Biodiversity Information Facility (GBIF; Frazier et al. 2008 andKrishtalka et al. 2016), Naturalis Biodiversity Center ( Heerlien et al. 2015) and the Atlas of Living Australia (Kalms 2012). Based, among others, on these past inventories, criteria used to prioritize digitization were summarized and categorized. ...
... That is because respondents that answered the question about imaging with "Yes, 100%", did not always answer the previous question about databasing with "Yes, 100%". While databasing is often considered to be the first step in digitization, followed by imaging ( Krishtalka et al. 2016), when it comes to herbarium specimens, it sometimes is the other way around. Here, the first step in digitizing can be imaging of the sheets (specimens) and then databasing them based on that image which could explain this discrepancy. ...
... Scientific relevance was most often indicated as the most important criteria category, which is in accordance to Krishtalka et al. (2016). Despite the acknowledged priority of scientific needs, in reality this process has been often driven by other factors (e.g. ...
Technical Report
Full-text available
Anno 2017 the task of mobilizing data from biocollections ahead of us is still enormous (data of 90% of the biocollections still needs to be mobilized). It is imperative for stakeholders, individual keepers of natural science collections, the community at large, and even for funding agencies, not only to tackle this backlog as quickly as possible, but do it in the best possible order. To establish the best possible order for digitizing biocollections a demand driven framework is required based among others on criteria used to digitize biocollections.
... Across the globe, there are 3,400 herbaria store botanical collections which are an invaluable record of the world's biodiversity (Krishtalka et al., 2016;Thiers, 2019). Herbaria have unique advantages over other botanical repositories such as germplasm banks or botanical gardens in producing good quality data for native flora conservation, among other uses (Funk, 2018). ...
Article
Full-text available
Botanists, a section of the broad universe of researchers in Biology, are intensive users of herbaria. Presumably, all botanists use herbaria, with greater or lesser frequency and intensity, in the development of their research. In this article, we will try to prove this statement. For this purpose, an institutional history of Botany and herbaria in Argentina is presented. This study will also show that there are other fields of knowledge in which the herbarium has a role as an input, or data source, for research (e.g. agronomy, ethnobotany, medicine). On the other hand, it will be demonstrated that, in addition to the uses of the herbarium in basic science, this institution has a crucial role in the knowledge and preservation of biodiversity, and in the improvement of species for commercial use.
Book
Full-text available
This best practice guide elaborates on processes of conducting Data Gap Analyses (DGA) of biodiversity data, which can help prioritize data mobilization activities. The guide illustrates general practices by summarizing 16 actual DGA exercises, with six focused on GBIF-mediated data discussed in detail. This guide is intended to support biodiversity data stakeholders including: Biodiversity information systems/networks Biodiversity data publishers Biodiversity organizations Research groups Information managers National/regional/thematic Biodiversity Information Facilities (BIFs) National/regional funding agencies
Article
Full-text available
The Global Registry of Biodiversity Repositories is an online metadata resource for biodiversity collections, the institutions that contain them, and associated staff members. The registry provides contact and address information, characteristics of the institutions and collections using controlled vocabularies and free-text descripitons, links to related websites, unique identifiers for each institution and collection record, text fields for loan and use policies, and a variety of other descriptors. Each institution record includes an institutionCode that must be unique, and each collection record must have a collectionCode that is unique within that institution. The registry is populated with records imported from the largest similar registries and more can be harmonized and added. Doing so will require community input and curation and would produce a truly comprehensive and unifying information resource.
Article
Full-text available
Dengue and chikungunya are increasing global public health concerns due to their rapid geographical spread and increasing disease burden. Knowledge of the contemporary distribution of their shared vectors, Aedes aegypti and Aedes albopictus remains incomplete and is complicated by an ongoing range expansion fuelled by increased global trade and travel. Mapping the global distribution of these vectors and the geographical determinants of their ranges is essential for public health planning. Here we compile the largest contemporary database for both species and pair it with relevant environmental variables predicting their global distribution. We show Aedes distributions to be the widest ever recorded; now extensive in all continents, including North America and Europe. These maps will help define the spatial limits of current autochthonous transmission of dengue and chikungunya viruses. It is only with this kind of rigorous entomological baseline that we can hope to project future health impacts of these viruses.
Article
Full-text available
Ebola virus disease (EVD) is a complex zoonosis that is highly virulent in humans. The largest recorded outbreak of EVD is ongoing in West Africa, outside of its previously reported and predicted niche. We assembled location data on all recorded zoonotic transmission to humans and Ebola virus infection in bats and primates (1976-2014). Using species distribution models, these occurrence data were paired with environmental covariates to predict a zoonotic transmission niche covering 22 countries across Central and West Africa. Vegetation, elevation, temperature, evapotranspiration, and suspected reservoir bat distributions define this relationship. At-risk areas are inhabited by 22 million people; however, the rarity of human outbreaks emphasises the very low probability of transmission to humans. Increasing population sizes and international connectivity by air since the first detection of EVD in 1976 suggest that the dynamics of human-to-human secondary transmission in contemporary outbreaks will be very different to those of the past.
Article
Full-text available
The world’s natural history collections constitute an enormous evidence base for scientific research on the natural world. To facilitate these studies and improve access to collections, many organisations are embarking on major programmes of digitization. This requires automated approaches to mass-digitization that support rapid imaging of specimens and associated data capture, in order to process the tens of millions of specimens common to most natural history collections. In this paper we present Inselect—a modular, easy-to-use, cross-platform suite of open-source software tools that supports the semi-automated processing of specimen images generated by natural history digitization programmes. The software is made up of a Windows, Mac OS X, and Linux desktop application, together with command-line tools that are designed for unattended operation on batches of images. Blending image visualisation algorithms that automatically recognise specimens together with workflows to support post-processing tasks such as barcode reading, label transcription and metadata capture, Inselect fills a critical gap to increase the rate of specimen digitization.
Article
Full-text available
Biological collections are at the front line of biodiversity research, informing taxonomy, evolution, conservation and sustainable livelihoods. In April 2014, we organised a meeting at the Linnean Society (UK) discussing the impact of next-generation sequencing (NGS) methods on collections-based research. Here, we explore the main themes of this meeting and outline the incredible potential of NGS to reinvent collections-based research. Among the many opportunities at the interface of genomics and collections, we focus specifically on (1) the genomic characterisation of biological collections, (2) the enhancement and development of DNA-based identification, (3) the tree of life and (4) interdisciplinary research addressing the most pressing environmental challenges of our times. Across the world, biological collections are at risk, primarily due to declining funding and shifts in scientific fashions. We encourage all users of collections to embrace the genomic era, not only because of the unparalleled scientific potential that it presents, but also because new cross-disciplinary synergies will reinvigorate and secure the collections for future generations.
Article
Full-text available
Climate change decoupling mutualism Many coevolved species have precisely matched traits. For example, long-tongued bumblebees are well adapted for obtaining nectar from flowers with long petal tubes. Working at high altitude in Colorado, Miller-Struttmann et al. found that long-tongued bumblebees have decreased in number significantly over the past 40 years. Short-tongued species, which are able to feed on many types of flowers, are replacing them. This shift seems to be a direct result of warming summers reducing flower availability, making generalist bumblebees more successful than specialists and resulting in the disruption of long-held mutualisms. Science , this issue p. 1541
Article
Phytophthora infestans, the cause of potato late blight, is infamous for having triggered the Irish Great Famine in the 1840s. Until the late 1970s, P. infestans diversity outside of its Mexican center of origin was low, and one scenario held that a single strain, US-1, had dominated the global population for 150 years; this was later challenged based on DNA analysis of historical herbarium specimens. We have compared the genomes of 11 herbarium and 15 modern strains. We conclude that the 19th century epidemic was caused by a unique genotype, HERB-1, that persisted for over 50 years. HERB-1 is distinct from all examined modern strains, but it is a close relative of US-1, which replaced it outside of Mexico in the 20th century. We propose that HERB-1 and US-1 emerged from a metapopulation that was established in the early 1800s outside of the species' center of diversity.
Article
Natural history collections provide an immense record of biodiversity on Earth. These repositories have traditionally been used to address fundamental questions in biogeography, systematics, and conservation. However, they also hold the potential for studying evolution directly. While some of the best direct observations of evolution have come from long-term field studies or from experimental studies in the lab, natural history collections are providing new insights into evolutionary change in natural populations. By comparing phenotypic and genotypic changes in populations through time, natural history collections provide a window into evolutionary processes. Recent studies utilizing this approach have revealed some dramatic instances of phenotypic change over short time scales in response to presumably strong selective pressures. In some instances evolutionary change can be paired with environmental change, providing a context for potential selective forces. Moreover, in a few cases, the genetic basis of phenotypic change is well understood, allowing for insight into adaptive change at multiple levels. These kinds of studies open the door to a wide range of previously intractable questions by enabling the study of evolution through time, analogous to experimental studies in the laboratory, but amenable to a diversity of species over longer timescales in natural populations. This article is protected by copyright. All rights reserved.