The climate is changing rapidly. The answer to the question of how rapidly, however, depends directly on how long relevant data can be analyzed. Archival data, then, which predates the satellite and digital era, is essential to the study of climate change over time. The National Snow and Ice Data Center (NSIDC) at the University of Colorado Boulder is responsible for managing, archiving, and disseminating cryospheric and polar data. Most of these data are digital. However, hidden within the NSIDC is a collection of historical archival materials that record the earth's glaciated regions prior to modern data gathering methods. With funding from the Council on Library & Information Resources, NSIDC and the CU Boulder Libraries are digitizing the entirety of the archive's print glacier photograph collection in order to curate and thereby enable future scientific discoveries related to climate change, and to help tell the story of a warming planet to the public and policy makers. But we must contend with legacy information architectures and metadata formats. This poster presents how we intend to provide digital images and associated metatdata in two digital platforms, one of which is co-discoverable with born-digital cryospheric data, the other of which is co-discoverable with collections more likely to reach humanists and social scientists. This project is a rapid prototype for future collaborations, which we hope will lead to a more comprehensive access and preservation model for both analog and digital data in glaciological and related sciences.
Irreplaceable Data in Historical Photographs
ICSU CODATA includes at Data at Risk Task Force which “is concerned with the plight of many sets of scientific data which are not in modern electronic formats and whose information is therefore not accessible to the research that needs it” (ICSU, 2015). Such is the case with NSIDC’s Analog Archives Collection, which houses legacy sea ice charts, maps, explorer diaries, and more than 20,000 glacier photograph prints dating back to the 1850s.
William O. Field compiled one subset of this collection, roughly 5,000 prints, including many taken by Harry F. Reid, "America’s first geophysicist." Collected by Field after Reid’s death in 1945, the materials also include approximately 600 glass plate negatives and prints, 22 expedition notebooks, and about 1 cubic foot of manuscript materials (including drafts of Reid’s map of Glacier Bay). Other subsets include 13 Rocky Mountain National Park Glacier Survey Reports and the accompanying 264 glacier photographs; 79 terrestrial photographs taken by Fred D. Ayres in Peru during the 1950s; 360 images of Colorado's Arapaho Glacier taken in the early 1900s by Junius Henderson (first curator of the CU Museum); and over 1,200 photographs of Greenland glaciers, donated by the U.S. Coast Guard. Images from southern Colorado, Glacier National Park, the Cascades, and much of Alaska across several decades are also included in the collection.
These images are valuable scientific data, most strikingly apparent in repeat photography projects, particularly those of Alaskan glaciers taken by USGS geologist Bruce Molnia (USGS, 2012), but including others, such as those shown in Figure 1. The images were also used in Ken Burns’s documentary “The National Parks: America’s Best Idea” and NOVA’s “Extreme Ice.” They have been used in artist books, text books, even children’s books, in addition to journal articles, dissertations, and exhibits (1).
Figure 1: Images of Okpilak Glaicer, Alaska. On the left: 1907 (Leffingwell, 1907). On the right, 2004 (Nolan, 20014). How much mass did the glacier lose? What was the impact on the surrounding terrain?
Perhaps more importantly, these images and other analogue data must be rescued and made accessible for techniques not yet invented by researchers not yet born. There are current users, for example, interested in the technical aspects of how these images, both digitized and born-digital, can be analysed to obtain geophysical information. How might a researcher go about determining focal length, for instance, to be able to deduce the height of a glacier front in a picture? How might that information be used to analyse other properties of the glacier and surrounding terrain? Could additional geospatial metadata be added to the images over time in order to enable GIS analysis? Would other metadata, such as descriptions of glacial moraine or vegetation cover, contributed by scientists analysing the images or their locations, further future discovery?
Unless these data are rescued and migrated into modern formats, we may never know.
Data Still at Risk, Still Being Rescued
These data were rescued many times over the decades, thanks to the foresight of scientists who understood their value. They were physically rescued, but for years these collections were largely unorganized and inaccessible until funding for an archivist was secured. Archivists and librarians at NSIDC brought order to and housed the items in more secure locations, began writing finding aids for them, and even digitized over 14,000 prints. But funding for the archivist was subsequently lost, putting the collection again at-risk. As the Data at Risk Task Force writes, [u]nfortunately, analogue media are perishable and many samples are already fragile; furthermore, data that cannot be accessed are assumed to be ‘unwanted’ and risk getting thrown away” (Data at Risk, 2015). We are convinced, however, that these materials constitute an irreplaceable contribution to the human record and our relationship to the planet.
Due to this conviction, we have secured approximately $150,000 in funding to complete the digitization of the remaining 9,000 images, and to provide all images in both the NSIDC’s Glacier Photograph Collection Database, and the CU Libraries Digital Library. The former will allow co-discoverability with born-digital data and geospatial metadata searches, and the latter supports the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH), in order to facilitate sharing of the images with other institutions and repositories. Discovery in the CU Digital Library will be alongside other special collections, including aerial photographs and maps of Colorado. The project will also address long-term digital preservation issues by storing master images on CU Boulder’s PetaLibrary, a National Science Foundation-supported service for storage and archiving of digital data.
Ultimately, these are next-step solutions, but are not end-goal or optimal. They will enable us to explore access, metadata, and preservation techniques that will facilitate future, broader collaborations that will bridge the crevasse between analog and digital data description and access in glaciology and related disciplines.
We believe this is a critical issue that must not be forgotten in data science, a discipline appropriately focused on scientific data that is born digital. These images are but one of vast treasure troves of analogue data stored in labs, libraries, offices, archives, and researchers’ homes scattered all over the world. Related, perhaps, to Session 62, and another submission from CU Boulder Libraries colleagues related to Florissant fossil beds, it will be a unique contribution particularly suitable as a poster.
The authors would like to thank the Council on Library & Information Resources for their support in this project.
The authors declare that they have no competing interests. Guidelines for competing interests are here: http://datascience.codata.org/about/competinginterests/
Examples of research utilizing the collection include: Molnia, Bruce, 2007. “Late nineteenth to early twenty-first century behavior of Alaskan glaciers as indicators of changing regional climate,” Global and Planetary Change (56); Barclay, David J.; Wiles, Gregory C., Calkin, Parker E., 2009. “Holocene glacier fluctuations in Alaska,” Quaternary Science Reviews (28); Connor, Cathy, et al., 2009. “The Neoglacial landscape and human history of Glacier Bay, Glacier Bay National Park and Preserve, southeast Alaska, USA,” Quaternary Geochronolgy (4).
ICSU CODATA 2016. Data at Risk Task Force. Accessed May 26, 2016. Available at: http://www.codata.org/task-groups/data-at-risk.
United States Geological Survey 2012. Glacier and landscape change in response to changing climate: Repeat photography of Alaksan Glaiciers. Accessed May 26, 2016. Available at: https://www2.usgs.gov/climate_landuse/glaciers/repeat_photography.asp.
Leffingwell, Ernest. 1907 Okpilak Glacier: From the Glacier Photograph Collection. Boulder, Colorado USA: National Snow and Ice Data Center.
Nolan, Matt. 2004 Okpilak Glacier: From the Glacier Photograph Collection. Boulder, Colorado USA: National Snow and Ice Data Center. Digital media.