Data rescue archive weather (DRAW). Preserving the complexity of historical climate data
| Published date | 09 July 2018 |
| Pages | 763-780 |
| DOI | https://doi.org/10.1108/JD-10-2017-0150 |
| Date | 09 July 2018 |
| Author | Eun G. Park,Gordon Burr,Victoria Slonosky,Renee Sieber,Lori Podolsky |
| Subject Matter | Library & information science,Records management & preservation,Document management,Classification & cataloguing,Information behaviour & retrieval,Collection building & management,Scholarly communications/publishing,Information & knowledge management,Information management & governance,Information management,Information & communications technology,Internet |
Data rescue archive weather
(DRAW)
Preserving the complexity of historical
climate data
Eun G. Park and Gordon Burr
School of Information Studies, McGill University, Montreal, Canada
Victoria Slonosky
Department of Geography, McGill University, Montreal, Canada
Renee Sieber
Department of Geography and School of Environment, McGill University,
Montreal, Canada, and
Lori Podolsky
McGill University Archives, McGill University, Montreal, Canada
Abstract
Purpose –To rescue at-risk historical scientific data stored at the McGill Observatory, the objectives of the
Data Rescue Archive Weather (DRAW) project are: to build a repository; to develop a protocol to preserve the
data in weather registers; and to make the data available to research communities and the public. The paper
aims to discuss these issues.
Design/methodology/approach –The DRAW project adopts an open archive information system
compliant model as a conceptual framework for building a digital repository. The model consists of data
collection, conversion, data capture, transcription, arrangement, description, data extraction, database design
and repository setup.
Findings –A climate data repository, as the final product, is set up for digital images of registers and a
database is designed for data storage. The repository provides dissemination of and access to the data for
researchers, information professionals and the public.
Research limitations/implications –Doing a quality check is the most important aspect of rescuing
historical scientific data to ensure the accuracy, reliability and consistency of data.
Practical implications –The DRAW project shows how the use of historical scientific data has become a
key element in research analysis on scientific fields, such as climatology and environmental protection.
Originality/value –The historical climate data set of the McGill Observatory is by nature unique and
complex for preservation and research purposes. The management of historical scientific data is a challenge
to rescue and describe as a result of its heterogeneous and non-standardized form.
Keywords Metadata, Data repository, Climate data, Data rescue, Historical scientific data, Weather data
Paper type Research paper
Introduction
Historical scientific records are unique and highly valued by historians, scientists,
librarians, archivists and users. Historical scientific data have been accumulated over time
and some of these data are at risk of being lost, often locked in heterogeneous formats. As a
recent Nature article remarked, “old photos, logbooks and papers are a goldmine for fields
such as ecology and climatology”hidden in the past (Kwok, 2017, p. 419); these data need to
be rescued for preservation and research purposes. In order to be compiled into large,
research-friendly data sets, the data must be reformatted, digitized, and then entered into a
machine-readable form. Within the procedures of preserving the data, the creation of
metadata is crucial to managing the conversion of hand written or typed data that are
recorded in registers, logbooks and diaries.
Journal of Documentation
Vol. 74 No. 4, 2018
pp. 763-780
© Emerald PublishingLimited
0022-0418
DOI 10.1108/JD-10-2017-0150
Received 25 October 2017
Revised 18 March 2018
19 March 2018
Accepted 20 March 2018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0022-0418.htm
763
Data rescue
archive
weather
The example we present here is the Data Rescue: Archives and Weather (DRAW) project
for old climate data that were stored at the McGill Observatory beginning in the nineteenth
century. This study explores the procedures for building a repository and digitizing and
describing non-standardized handwritten records of historical climate data in the context of
its inherently changing format and complexity. It also explores issues concerning
organizing and rescuing digitally captured and transcribed scientific data in an approach to
preserving this data and promoting access to scientific research.
This study is organized into the following sections. The second section provides a
literature review of historical scientific data rescue and relevant issues. The third section
explains the context for the McGill Observatory’s historical climate data. The fourth section
explains the procedures for building a repository and describing the data through the
development of multiple sets of metadata. The final section discusses major issues and
challenges in managing the rescue of historical scientific data and concludes with a plan for
further research.
Literature review: the digitization and preservation of historical climate data
Climate data are usually taken from observations of the weather at specific times and
recorded manually or automatically on preprinted forms on paper or in software programs
during specific observation periods (e.g. Tan et al., 2004; Jeffrey et al., 2001). Historical
climate data refers to information on past weather or climate, and the definition of
“historical”can depend on the context, ranging from records that are ten years old to those
that are centuries old. According to the International Environmental Data Rescue
Organization (2014), the world loses thousands of pieces of critical historical climate data
and records every day. Some historical climate data from earlier times is at risk because of
the careless storage practices by holding institutions or as a result of the old, deteriorating
or obsolescent forms of media they are stored on. Natural and environmental hazards, theft
or vicious destruction can also sometimes happen to the data (World Meteorological
Organization (WMO), 2017). Data rescue indicates “the process of securing data at risk of
being lost”and “ensuring that data can be easily accessed and used”(WMO, 2017,
DATA Rescue). To effectively preserve historical scientific data, WMO addressed the four
key activities of data rescue: “cataloguing and storing of paper media to minimize the
deterioration; imaging paper records for improving preservation and accessibility; digitizing
data into computer compatible form; and continuous updating of digital records to
up-to-date media and formats”(2017, DATA Rescue). According to WMO (2017), in the data
rescue procedures, first, the storage of (original) paper media and producing descriptions of
the data in metadata schema are important to minimize the deterioration of data until more
effective measures can be taken. The next activities are scanning or imaging from original
media to digital formats (i.e. computer readable formats) to improve preservation and
guarantee accessibility to the data. Lastly, the data should be continuously updated to
current media and formats so that it is accessible in the long term or permanently. Similarly,
Paul Conway (2010) refers to the term digitization for preservation as “activities that result
in the creation of digital products worthy of long-term preservation”(p. 65). The activities
involved include conversion processes, selection, the creation of descriptions, and digital
collection building. In managing scientific data, Karasti et al. (2006) explained that while
data curation activities “ingest, archive and [deliver]”large-volume and more homogeneous
collections, data stewardship activities include “data planning to sampling, from data
archive to use and reuse of data and information infrastructure”for more specialized and
heterogeneous collections (p. 352).
The importance of recovering and preserving historical data that are at risk cannot be
stressed enough for institutions and for society as a whole. The benefits of preserving and
analyzing historical scientific data are considerable. Preserving historical scientific data
764
JD
74,4
To continue reading
Request your trial