Assembling an infrastructure for historic climate data recovery: data friction in practice
| Pages | 791-806 |
| DOI | https://doi.org/10.1108/JD-08-2018-0130 |
| Date | 08 July 2019 |
| Published date | 08 July 2019 |
| Author | Jo Bates,Paula Goodale,Yuwei Lin,Penny Andrews |
| 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 |
Assembling an infrastructure for
historic climate data recovery:
data friction in practice
Jo Bates and Paula Goodale
Information School, University of Sheffield, Sheffield, UK
Yuwei Lin
University of Roehampton, London, UK, and
Penny Andrews
Information School, University of Sheffield, Sheffield, UK
Abstract
Purpose –The purpose of this paper is to adopt an assemblage theory lens to examine the socio-material
forces shaping the development of an infrastructure for the recovery of archived historical marine weather
records for use in contemporary climate data sets.
Design/methodology/approach –The authors adopted a data journeys approach to research design,
conducting in-depth semi-structured interviews with climate scientists, citizen scientists and a climate
historian who were engaged at key sites across the journey of data from historical record to the International
Comprehensive Ocean-Atmosphere Data Set database. Interview data were complemented by further
qualitative data collected via observations of working practices, a digital ethnography of citizen scientists’
online forums, and documentation relevant to the circulation and governance of climate data across emergent
data infrastructures. Data were thematically analysed (Ryan and Bernard, 2003), with themes being informed
primarily by the theoretical framework.
Findings –The authors identify and critically examine key points of friction in the constitution of the data
recovery infrastructure and the circulation of data through it, and identify the reflexive and adaptive nature
of the beliefs and practices fostered by influential actors within the assemblage in order to progress efforts to
build an infrastructure despite significant challenges. The authors conclude by addressing possible
limitations of some of these adaptive practices within the context of the early twenty-first century neoliberal
state, and in light of current debates about data justice.
Originality/value –The paper draws upon original empirical data and a novel theoretical framework that
draws together Deleuze and Guattari’s assemblage theory with key concepts from the field of critical data
studies (data journeys, data friction and data assemblage) to illuminate the socio-material constitution of the
data recovery infrastructure within the context of the early twenty-first century neoliberal state.
Keywords Data friction, Climate data, Critical data studies, Data assemblage, Data infrastructure,
Historical records
Paper type Research paper
Introduction
Climatology requires long-term data from many locations, consistent across both space and time.
This requirement implies a lengthy chain of operations, including observation, recording,
collection, transmission, quality control, reconciliation, storage, cataloguing, and access. Every link
in this chain represents an information interface subject to data friction. (Edwards, 2010, p. 84)
Over the last decade, climate scientists have engaged in a variety of international efforts to
enhance the quality and quantity of meteorological data that constitute important global
climate data sets such as the International Comprehensive Ocean-Atmosphere Data Set
(ICOADS). Funding has been provided to aid the construction of some data sets, and various
meteorological agencies have endorsed principles aimed at the full integration of weather
and climate data networks in order to create a robust and sustainable climate data
infrastructure. However, there remain many barriers to realising such a vision, and as
Journal of Documentation
Vol. 75 No. 4, 2019
pp. 791-806
© Emerald PublishingLimited
0022-0418
DOI 10.1108/JD-08-2018-0130
Received 13 August 2018
Revised 14 December 2018
Accepted 27 December 2018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0022-0418.htm
791
Historic
climate data
recovery
Edwards (2010, p. 302) observes, a particularly “daunting task”is that of “refining and
reconstructing the historical record”.
The current climatological record is relatively strong in some geographical regions such
as the land regions of the northern hemisphere where detailed observational records can
exist back to the late nineteenth century. However, the uneven development of
meteorological observation and data archiving infrastructures around the world means
that the historical data of many regions is sparse or non-existent. In order to deal with these
gaps in the historical record, climate scientists can either generate synthetic data or recover
archived data that are not yet integrated into international climate data sets. Scientists
argue that by increasing the quantity and quality of observational data held in databases,
the baseline for state of the art climate “reanalysis”will be enhanced, improving the
integrity and reliability testing of climate models. One source of such data are historical ship
logbooks in which crew members recorded detailed accounts of weather conditions at sea,
and which are now stored as archival documents in different sites around the world.
While early efforts to integrate meteorological observation data from different archival
sources concentrated on variables such as pressure, more recently workhas been directed at
synthesising marine and terrestrial surface temperature observation records from the last
200 years. The Atmospheric Circulation Reconstructions over the Earth (ACRE) initiative
aims to bringtogether people and organisationsfrom around the worldin order to achieve this
ambition. In this paper, we focus on the efforts of those who are engaged, both directly and
indirectly, through the ACRE initiative in work to develop a robust infrastructure for the
recovery of historical marine surface temperature data from regions such as the Southern
Ocean, SouthPacific and Arctic. Our particular interestis in efforts to organise the recoveryof
meteorological observation records from the archived logbooks of historical ships, primarily
those of the UK’s RoyalNavy and the US Navy (c. 1800–1950). The intention of recoveringthis
data is that they can be integrated into the ICOADS, and ultimately be used by climate
modellers and application developers to better predict and respond to changes in the Earth’s
climate. While there are significant quantities of meteorological records available to be
recovered from archives, there are many challenges faced by climate scientists working to
discover, digitise and transcribe such data and integrate them into the ICOADS.
The socio-material factors that influence efforts to develop data infrastructures for
gathering, preserving and using historical meteorological records have been observed by
Edwards (2010) and Bates et al. (2016). In our previous work, we identify the movement of
historic meteorological observation data from ship logbooks into the ICOADS database as
an example of a “data journey”. In our work on “data journeys”(Bates et al., 2016), we
developed a methodological approach for illuminating the socio-material “life of data”as
they move across different data infrastructures, from sites of production through to multiple
sites of re-use in different sectors. By following data on their journey between different sites
of practice, we were able to examine some of the complex socio-material forces shaping the
constitution of data movements across space and time. An aim of developing the data
journeys approach was to move beyond Science and Technology Studies (STS)-influenced
rich descriptive accounts of data infrastructures and the intra-network politics of their
development as seen, for example, in the work of Edwards (2010), Leonelli (2013) and
Ruppert et al. (2015). Instead, we aimed to address how data come to be constituted and used
across multiple interconnected sites of data practice, and how the movement of data
between such sites brings social actors into qualitatively new forms of relation with one
another. The findings from our work on data journeys identified that shifting sites and
moments of “friction”(Edwards, 2010) in the circulation of data were important to examine
in order to develop critical insight into the power dynamics shaping the movement of data
between different actors, and thus emergent social relations. In later work, we further
developed Edwards’concept of “data friction”to identify how the interrelated socio-material
792
JD
75,4
To continue reading
Request your trial