Natural history citizen science crowdsourcing
Are you interested in natural history? Help us to capture label information from images of specimens from the Norwegian natural history collections in Oslo.
New crowdsourcing portal for natural history collections! Help us to record information on museum specimens from the collections of the Natural History Museum at the University of Oslo! The new transcription portal, developed by GBIF-Norway, is launched today at the 200-year jubilee party for NHM-UiO botanical garden. The presentation of the portal will take place at 17:30 in the auditorium of Lids hus (Botanical museum) at the Tøyen campus and Botanical gardens. Christian Svindseth (GBIF-Norway, NHM-UiO) has developed the computer code for the new portal. Visit the portal at: http://gbif.no/transcribe (Figure 1).
Digitization of natural history collections
The collections at the Natural History Museum in Oslo include an estimated total of more than 6 million specimens (Mehlum et al., 2011). The collections in Oslo are estimated to hold more than 65% of the specimens held by natural history museums in Norway. The digitization of the Norwegian natural history collections has high priority and has reached a level of more than 50% of the specimens recorded and added into an electronic database system. This is a high proportion digitized when compared to other large natural history collections worldwide, but the estimated efforts to complete the appropriate registration of all remaining specimens is daunting. The Natural History Museum in Oslo has started a large-scale digitization activity in 2013 where specimens are photographed and only the very minimum information of the scientific name and the country where the specimen was collected is registered.
Primary biodiversity information
Large-scale imaging of the specimens in the Norwegian natural history collections in Norway is prioritized and has started. However, only a very minimum of the label information such as scientific name (sometimes only genus) and collecting country will be captured in this project. Capturing additional information such as the collecting location (where), collecting date (when) and the verified current scientific name (what) will substantially increase the scientific value of these data records. The data on where, when and what define the so-called primary biodiversity information and is recognized as the minimum information requirement for respective scientific research. Species distribution modelling is one of the important research tools for understanding the ecology of species, and is dependent on available primary biodiversity information (where, when and what).
Why participate and contribute to citizen science transcription
* Discovery of biodiversity information: Transcription of label information and electronic registration into online databases greatly improve the discoverability of museum specimens for the purpose of scientific research and other public use.
* Education: Students from high school level to graduate and post-graduate level can engage with the photographs of the museum specimens and take part in a first class learning experience in interaction with this resource of primary biodiversity information.
* Scientific research: Scientists that study natural history need readily access to primary biodiversity information made available from museums and their online databases. Using the transcription portal they can take a direct part in making the primary biodiversity information they need for their own research available by transcribing the labels for the respective species groups and or countries that they study.
* Public good, open and free online biodiversity information: The information that we gather from the transcription portal will flow into the museum specimen database and be published to open and free data portals such as the Global Biodiversity Information Facility (GBIF), Norwegian Species Map Service (Artskart) and the Encyclopedia of Life (EOL). This valuable information for documenting historic biodiversity patterns are thus preserved not only for future generations, but also made available for ongoing current research using up-to-date and modern web technologies.
Lichen herbarium, Hildur Krog collection from eastern Africa
The first specimen collection that was loaded to the new citizen science transcription portal is the lichens collected by the Norwegian biologist Hildur Krog and others in East Africa. This collection is part of the lichen herbarium and includes more than 2 500 specimens (figure 2). Professor Hildur Krogh was originally introduced to limnology as a student of professor Eilif Dahl (1916-1993). Eilif and Hildur pioneered the work on chemical methods for identification of lichen species. Hildur was appointed curator of the lichen herbarium at the Botanical Museum of the University of Oslo in 1971. Between 1972 and 1996 Hildur Krog and T.D.V. Swinscow explored systematically the lichen genera of East Africa for the development of the flora “Macrolichens of East Africa”. With this citizen science portal, we are asking for volunteers to assist us with transcribing the label information from the herbarium specimens collected during these expeditions to East Africa. The imaging of this collection was made late 2013 and early 2014 by Silje Larsen Rekdal and Even Stensrud under the coordination of lichen curator Einar Timdal and Siri Rui (NHM-UiO) and with funding from GBIF-Norway (Figure 3).
Mycological herbarium at NHM Oslo
The Mycological herbarium includes approximately 300 000 specimens were approximately 2/3 are electronically registered into the database with label information captured. NHM-Oslo has started a large-scale activity to photograph the specimens of the collections under the coordination of Dr. Eirik Rindal. The Mycological herbarium is here one of the first collections to be photographed. During one week (in September 2013) the staff at the museum digitized around 6000 specimens from the Mycological herbarium (Figure 4). We plan to explore the new citizen science transcription portal as a tool to capture the label information for the remaining specimens not yet appropriately registered into the database. After making the first experiences with transcription of the lichen collection, we plan to also load the first approximately 40 000 specimen images from the Mycological herbarium – and later add even more collections and sets of specimen images incrementally following the progress of the digitization activity.
When is a specimen transcription complete?
Each specimen image is transcribed by at least three volunteers and the recorded information from each volunteer compared. If all three transcriptions provide the same information, the specimen transcription is flagged as completed. If all three transcriptions provide different information the specimen image will be flagged as incomplete and presented for review by new volunteers until there is a 50% agreement (on each information input box). Collection curators and museum staff will review the results, as they come in, before the information is included into the collection database and published to the Norwegian Artskart portal and the global GBIF portal.
Label information should be transcribed verbatim
We ask our volunteer citizen scientists to transcribe the specimen label information in verbatim form as close to the information printed or written on the specimen label as possible. The citizen scientists are not recommended to make their own interpretations or corrections. We do recognize that this recommendation could be a lost opportunity to collect citizen science curation and correction of the specimen database. We are working on a solution for citizen scientists to provide such interpretations and inferred additional information from the same interface. Such a specimen annotation service could provide citizen scientists to a wide variety of inferred information about the museum specimens, including eg. georeferencing with geographic coordinates or links to other systems such as sequence data deposited in GenBank or BoL or traits in EOL.
Challenge: How do we approach citizen scientist interpretations of label text? How to add annotations when the volunteers find or infer more information from other sources than only the specimen label?
Collaborator: Notes from Nature
The Notes from Nature portal provided the primary source of inspiration for the new crowdsourcing portal at NHM-Oslo. We are grateful for very valuable feedback and assistance from the Notes from Nature team including in particular director Michael Denslow at the National Ecological Observatory Network (NEON) and professor Robert Guralnick at the Museum of Natural History, University of Colorado at Boulder. Notes from Nature provides a citizen science platform to capture label information from photographs of specimens from natural history collections (Hill et al., 2012; Franzoni and Sauermann, 2013). The Notes from Nature software platform was developed and is maintained by Zooniverse and Vizzuality in collaboration with university museums and network in Florida (SERNEC), California (CalBug), Colorado (UCMNH) and the bird collection of the Natural History Museum in London (NHMUK). Notes from Nature is open source software, with the source code freely available at GitHub.
Links to some similar citizen science transcription portals
Many natural history collections are these days starting to establish similar transcription portals. One of the first of these crowdsourcing portals was the Herbaria@home from the Botanical Society of Britain & Ireland launched around 2006. The Atlas of Living Australia (ALA) Volunteer Portal provides a crowdsourcing platform for transcription of Australian collections of natural history specimens. The National Museum of Natural History (MNHN) in Paris provides a transcription portal for the collections in France. The Smithsonian National Museum of Natural History (NMNH) provides a Transcription Center with another excellent crowdsourcing portal.
Franzoni C, and Sauermann H (2013). Crowd science: The organization of scientific research in open collaborative projects, Research Policy, Available online 14 August 2013, ISSN 0048-7333, doi:10.1016/j.respol.2013.07.005.
Hill A, Guralnick R, Smith A, Sallans A, Gillespie R, Denslow M, Gross J, Murrell Z, Conyers T, Oboyski P, Ball J, Thomer A, Prys-Jones R, de la Torre J, Kociolek P, and Fortson L (2012). The notes from nature tool for unlocking biodiversity records from museum records through citizen science. ZooKeys 209: 219-233. doi:10.3897/zookeys.209.3472
Mehlum F, Lønnve J, and Rindal E (2011). Samlingsforvaltning ved NHM – strategier og planer. Versjon 30. juni 2011. Naturhistorisk museum, Universitetet i Oslo. Rapport nr. 18, pp. 1-89. ISBN: 978-82-7970-030-2. Available at http://www.nhm.uio.no/forskning/publikasjoner/rapporter/NHM-rapport-18-samlingsplan.pdf, accessed 28 May 2014.
Pensoft Publishers (2012). No specimen left behind: Mass digitization of natural history collections [special issue]. Editors: Blagoderov, V. and Smith, V. ZooKeys 209: 1-267. ISBN: 9789546426451. Available at http://www.pensoft.net/journals/zookeys/issue/209/