Objectives of PEN-TCN project and the role of the FLAS Herbarium

March 19th, 2013

Overall objective of the project

The objective of the project is to digitize the label information and images of  North American lichen and bryophyte specimens (Canada, the continental United States and Mexico).  This  project is a contribution towards the Lichens, Bryophytes and Climate Change NSF Thematic Collections Network.  You can check more on the initiative in the following site: http://lbcc1.acis.ufl.edu/

The role of the FLAS Herbarium

Our role as part of the consortium is to contribute with the digitization of the Florida collections that are important because they potentially fill gaps of the tropical and sub-tropical ecosystems that may be especially susceptible to climate change effects.  We are going to image and process the specimens of the following Florida Herbaria:

Acronym         Institution                              Approximate number

FLAS              University of Florida               32,000 specimens

FTG               Fairchild Tropical Garden         1,900 specimens

FTU               University of Central Florida     1,500 specimens

USF               University of South Florida        1,500 specimens

UWFP            University of West Florida         320 specimens

We project that we will have to process and digitize 33,000 specimens in a year to complete the project in three years.

Phases_project6

Projected phases of the NSF-PEN project implemented in the FLAS herbarium of the Florida Museum of Natural History

We plan to complete this project in three years.  The first phase will deal with specimen preparation and digitization, and will take place in the FLAS herbarium that is part of the Florida Museum of Natural History of the University of Florida.  We are currently on the second year of the project, and we are editing the information that was uploaded in the portals and populating different field of the database using the data generated by optical character recognition, the later,  performed in the herbarium of the University of Madison, Wisconsin.  The third year will be devoted to finishing the edition of the data and georeferencing it.

 

Specimen preparation

April 25th, 2013

To all the FLAS specimens we added an archival grade card (5 1/2′” x 5 ¾” archival grade cards) to give additional support to the specimens, and to provide a flat paper surface on which to apply the barcode label and determination labels that were otherwise stapled and covering the labels. The barcode labels were place on the upper right side of the cardboard so they can be easily scanned and visible on the boxes that hold the specimens.

The barcodes used for this project were previously printed using Digitek labels (B274 Polypropylene Base Material, P236 – Adhesive, L107 .5 mil clear matte laminate) and code 39 printed.  In the cases in which FLAS accession numbers were previously designated, barcodes that match the FLAS catalog number were applied.  The specimens lacking accession numbers were assigned barcodes in a sequential manner.

ex_2pck

Examples of a lichen (left) and moss (right) package ready to be imaged.

The retroactive barcoding has been labor intensive and is one of the bottlenecks of this project as we have only one set of barcodes that have to be shared by all our students.  Additionally, matching barcodes is prone to errors, and the incorrect assignment of barcodes in one point cascades when another sample needs to be barcoded.

The most common problems that we have found while matching barcodes to accession numbers are that the latter are not unique, especially in old collections, where the same barcode has been used multiple times by mistake.  In older collections we also have found that the same numbers have been given to mosses and hepatics. In more modern collections, the most common error is to mistakenly use the same barcodes for gifts that enter the collection at the same time that other specimens are being accessioned.

We strongly recommend sequential barcoding.

FLAS first challenge: consolidating and preparing the collections

April 22nd, 2013

The FLAS collection was initially organized under two curatorial styles.  The modern collection was kept in envelope-type packets, placed vertically and organized in archival trays.  Another part of the collection, encompassing most of the historical collections that date back to the 1800’s (Severin Rapp collection), were kept in envelope-type packets mounted on non-archival sheets, organized in folders and stored flat.

Because of the different orientation of the specimens these two collections have been kept spatially separated, making difficult the revision of material to visiting scientists.  As for this project, keeping the two collections separated made the workflow for capturing and transferring images time-consuming.

We prepared the collection by cutting these specimens, alphabetizing them, applying the barcode and adding the card and then, intercallating them with the  collection.

Old_spec copy

The old collection was preserved in folia (left), and we cut and alphabetize these specimens prior to their integration with the modern collection

We started by processing the lichen collection and using it as a pilot collection to identify the work that was needed to conduct the project.  Early in this stage we had to  revise our workflow, and in the following figure you can see how intricate the process was.

The following figure contrasts the procedure that we (A) initially envisioned, with (B) procedure we used.

Flowchart_lichens

(Left) Initial workflow of the project contrasted with the (right) optimized workflow of the project

 

After identifying our new challenges, one of our goals was to consolidate our lichen and bryophyte collection and for this purpose we proceed to cut the envelopes mounted on folia of the old collection, and intercalate them with the modern collection .  This step was time demanding because in the old collection hand written labels are prevalent and are difficult to read and by consequence, difficult to alphabetize.

Collections_lichens

A) Claudia Zafra and Brontie Herrera sort specimens; B) Andrea Amorelli and Jenny Barber working with specimens of the old collection.

Check more of our awesome students here

The Bryophyte collection had yet another challenge.  This collection was divided in Anthoceros and mosses and liverworts.  Additionally the moss collection was divided into tropical and temperate.
In this case, our strategy was slightly different, check the following figure

Flowchart_bryophytes_envelopes2

Optimized workflow used for the bryophyte collection

NEWS! We have completed barcoding, alphabetizing and imaging the hepatic, the moss and the lichen collections.  We are ready to start processing the other herbaria from Florida that are part of this grant. (May 13, 2013)

Imaging system

April 20th, 2013

Our imaging setup is composed of a Bencher Copymate III Fluorescent Tabletop Producer with two cool, full spectrum fluorescent lamps that are daylight balanced at 5200°K for superb color clarity and operate at 12,000 Hz to eliminate flickering. We opted for the fluorescent system because it is cool, and it offers a full spectrum / daylight balanced light. The copystand includes an aluminum column that supports the 35 mm Canon EOS T3i digital camera. We are currently using a Canon EF 50mm f/2.5 compact macro lens with autofocus. Additionally, the camera is connected to an AC Canon Adapter kit, in this way we do not depend on the batteries for energy, and we use the batteries as a backup.

The system is connected to the PC computer of the project, and we capture the images using the Canon EOS Utility software coupled with the Imaging Workflow Application provided by on the LBCC – Lichens, Bryophytes and Climate Change Web site.

In order to capture the image, we have found useful small earth magnets (7 mm diameter) to fix the specimens to the copystand.  We  painted them white with an industrial grade paint marker.

Imaging of bryophyte/lichen packeted specimens averages from 80-100 per hour.  Two images are necessary when there are annotations covering portions of the label.  If a group has many annotations the imaging process is considerably slower.

DSCN6004