IEEE Geoscience and Remote Sensing Magazine - September 2017 - 74

means of advancing their studies; nor are there agreed-upon
methodologies and practices to be employed in using these
technologies. Instead, the choices made, the technologies
embraced, and the practices adopted are influenced by the
discipline of study, organizational traditions, lack of access
to a common cyberinfrastructure, variations in terms of
resource availability (human and financial), and other cultural factors, such as accepted practices and user requirements. This has led to a divergent, rather than convergent,
set of systems and practices that do not consistently enable
efficient and effective interoperability. The driving forces for
monsoon events, for examAN INCREASING AMOUNT
ple, act on time scales ranging
OF ENVIRONMENTAL
from months to hundreds of
RESEARCH IS NOT
thousands of years, and the proREPRODUCIBLE, IF
cesses connected to these pheJUDGED BY THE STANDARD
nomena make it crucial to take
OF BEING ABLE TO
a multidisciplinary approach
EXACTLY DUPLICATE AN
when studying them, but often
EXPERIMENT AND ARRIVE
these data are frustrating and
AT THE SAME RESULTS.
incompatible (Figure 3).
There are efforts underway
to overcome the interoperability dilemma, such as plans for the GEOSS and EarthCube, the latter endeavor sponsored by the National Science Foundation [8]. Progress is limited but positive with
respect to understanding and addressing these challenges.
However, the most difficult issue to consider is the disparity
between cultural and technological change. The pace of change
within the social and cultural context is glacial compared
to the current pace of technical innovation. These differences manifest themselves in various ways, such as placing
greater value and effort on gathering more data (for which
technology is a significant enabler) than on sharing data.
The unrelenting advance of widely available technologies

adds a factor of complexity. Because more technical choices
continually become available, interoperability among various approaches becomes more difficult to maintain. This
challenge must be recognized and overcome to fully open
up the epic tale of the Earth system.

REPRODUCIBLE RESEARCH
An increasing amount of environmental research is not reproducible, if judged by the standard of being able to exactly
duplicate an experiment and arrive at the same results. This
is because the environment being observed is dynamically
changing. Compounding the challenge are the increasingly sophisticated methods used in experiments, which are
often not adequately applied and/or documented. Geoscience cannot blithely neglect the spirit of the golden rule of
reproducibility, even if the letter of the rule is impossible
to enforce under many circumstances. The need to adhere
to best principles in the conduct of science is more urgent
now than ever.
But there are circumstances under which exact reproducibility of scientific experiments is not possible, such as
the bit-wise reproduction of a 100-year Earth system simulation [9] or Eulerian measurements of the upper ocean. It
is common for the provenance of the data used in research
activity not to be completely known. The fact that scientific experiments cannot be exactly reproduced does not diminish their value, but it does necessitate a qualitative and
quantitative assessment of the data. The geoscience community must rethink the appropriate context and standards
that should be applied when reproducing the outcomes
from research endeavors.
There has never been a greater opportunity to capture and rapidly advance our understanding of the Earth
system. This opportunity is created, in part, by the community's ability to harness the ongoing advances in technology and enable their application to achieve societal
benefits. There is no doubt that geoscience
has harvested the fruits of the technology
and knowledge evolution. The discipline
believes it is ready and able to tackle future
challenges. However, the barriers identified-which constitute only a partial list-
become onerous in the milieu of exponential technology and data growth.
For instance, it is relatively easy to creatively apply technology that enables a varied set of observations to be taken simultaneously. The difficulty is in deriving a
holistic understanding of the environmental
system being observed and identifying the
underlying interdependences when there is
insufficient knowledge of the processes in
play. There is a need for experiment design
that ensures its outcomes will be of mutual
FIGURE 3. A freighter under monsoon rain clouds. (Photo courtesy of M. Mohtadi,
benefit to all pertinent subdisciplines of geoCenter for Marine Environmental Sciences, Universität Bremen.)
science, not just to the group initiating the

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

SEPTEMBER 2017

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - September 2017