Page 34 - gt1511

We need to talk: Facilitating communication between field-
based geoscience and cyberinfrastructure communities
GSA TODAY | NOVEMBER 2015 Matty Mookerjee*, Daniel Vieira, Dept. of Geology, Sonoma State
University, Rohnert Park, California 94928, USA; Marjorie A. Chan,
Dept. of Geology & Geophysics, The University of Utah, Salt Lake
City, Utah 84112, USA; Yolanda Gil, Information Sciences Institute,
University of Southern California, Marina del Rey, California 90292,
USA; Charles Goodwin, Dept. of Applied Linguistics, University of
California at Los Angeles, Los Angeles, California 90095, USA;
Thomas F. Shipley, Dept. of Psychology, Temple University,
Philadelphia, Pennsylvania 19122, USA; and Basil Tikoff, Dept. of
Geoscience, University of Wisconsin–Madison, Madison,
Wisconsin 53706, USA
A UNIQUE OPPORTUNITY FOR CYBERINFRASTRUCTURE
It is increasingly important to integrate datasets and models from
multiple geoscience subdisciplines in order to significantly advance
our knowledge of how the planet works. To facilitate interdisci-
plinary investigations, geoscientists need a cyberinfrastructure that
will easily access and combine datasets from all of the current and
future geo-community databases. To this end, NSF introduced the
EarthCube initiative (www.earthcube.org) to “create a community-
driven data and knowledge management system that will allow for
unprecedented data sharing across the geosciences.” The ultimate
goal of EarthCube is to transform Earth science investigations by
promoting efficient data access, incorporating cyberinfrastructure
into our scientific workflow, and allowing for increased sophistica-
tion of analyses and models (Gil et al., 2014; Kelbert, 2014; Richard et
al., 2014). A significant strength of EarthCube is its potential to
create sustained communication across the subfields within the
Earth sciences, allowing scientists to ask new types of questions, and
providing the means to address previously unanswerable ones.
Examples of specific use cases are available on the EarthCube
webpage; however, using machine learning to extract data from
published articles (e.g., DeepDive [http://deepdive.stanford.edu])
and curating useful software/scripts (e.g., GeoSoft [http://www.isi.
edu/ikcap/geosoft/]) are two widely applicable examples of
EarthCube outcomes.
While the technical issues of interconnecting all existing
community databases are significant challenges, an even more
fundamental issue needs to be addressed: Not all communities
have a database or the institutional support to manage one. In
order for EarthCube to be successful, data from all subdisciplines
need to be represented in the data management system. In
GSA Today, v. 25, no. 11, doi: 10.1130/GSATG248GW.1.
*E-mail: matty.mookerjee@sonoma.edu
34
particular, the field-based geosciences have lagged behind other
subdisciplines with respect to developing a cyberinfrastructure for
their datasets. This is likely due to the fact that most field data is
collected and recorded in an analog format (e.g., Brunton
compass, field notebook, and sketches) and through various
personalized conventions. The additional step of digitizing these
data is often onerous. In order to facilitate the development of
cyberinfrastructure for the field-based geosciences, digitization
processes must be incorporated into the typical geoscience work-
flow in a way that is as unobtrusive as possible (e.g., digital field
notebooks, digital compasses, voice recognition software, digital
pens, etc.). These potential solutions need to be developed in
tandem with the cyberinfrastructure for managing these datasets.
This is why it is critical to get the cyberinfrastructure and the
field-based geoscience communities together and communicating
effectively. Field-based geoscientists need the cyberinfrastructure
community to advise them on the efficient collection of data for
optimal digitization, while keeping them grounded in what is
technically feasible. The geoscience community must engage and
communicate their current and anticipated needs along with their
specific data formats and requirements in order to design an effec-
tive data management system.
BRINGING CYBERINFRASTRUCTURE RESEARCHERS INTO
THE FIELD
In order to facilitate the necessary communication between
field-based geoscience and cyberinfrastructure communities, we
proposed the currently funded NSF EarthCube project: “(EC3)
Earth-centered communication for cyberinfrastructure:
Challenges of field data collection, management and integration.”
With this project, we brought together various field-based geolo-
gists with computer scientists and a cognitive psychologist in a
field setting. In August of 2014, 32 field-trip participants (12
computer scientists, 10 geoscientists, four graduate students, three
undergraduates, two applied linguists, and one cognitive scientist)
traveled to Yosemite and Owens Valley, California, USA, in order
to discuss cyberinfrastructure-related issues. There is no better
place to gain an appreciation for the field geologist’s workflow
than in the field itself. For the same reasons that we bring students
into the field to explain fundamental concepts in the Earth
sciences, the field provides an excellent venue for engaging with
computer scientists about the multiple scales and interconnec-
tions of geological data, data collection strategies and techniques,