This is the accessible text file for GAO report number GAO-07-1172
entitled 'Climate Change Research: Agencies Have Data-Sharing Policies
but Could Do More to Enhance the Availability of Data from Federally
Funded Research' which was released on October 23, 2007.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
Report to Congressional Requesters:
United States Government Accountability Office:
GAO:
September 2007:
Climate Change Research:
Agencies Have Data-Sharing Policies but Could Do More to Enhance the
Availability of Data from Federally Funded Research:
GAO-07-1172:
GAO Highlights:
Highlights of GAO-07-1172, a report to congressional requesters.
Why GAO Did This Study:
Much of the nearly $2 billion annual climate change research budget
supports grants from the Department of Energy (DOE), National
Aeronautics and Space Administration (NASA), National Oceanic and
Atmospheric Administration (NOAA), and National Science Foundation
(NSF). Some of the data generated by this research are stored in online
archives, but much remains in a less accessible format with individual
researchers. As a result, some researchers are concerned about the
availability of data.
GAO analyzed (1) the key issues that data-sharing policies should
address; (2) the data-sharing requirements, policies, and practices for
external climate change researchers funded by DOE, NASA, NOAA, and NSF;
and (3) the extent to which these agencies foster data sharing. GAO
examined requirements, policies, and practices and surveyed the 64
officials managing climate change grants at these agencies.
What GAO Found:
According to the scientific community—as represented by the National
Academies and professional scientific associations—four key issues that
data-sharing policies should address include what, how, and when data
are to be shared, as well as the cost of making data available to other
researchers. First, the information necessary to support major
published results should be made available to other researchers.
However, there are statutory limits on data sharing—such as
intellectual property protections—as well as practical limits such as
the lack of appropriate archives. Second, when the appropriate
infrastructure exists, data should be made accessible through
unrestricted archives. Third, data should generally be made available
immediately or after a limited proprietary period to allow for analysis
and publication of results. Fourth, data should be made available at no
more than the marginal cost of reproduction and distribution. Finally,
the extent to which specific policies address these key data-sharing
issues may vary, depending on the type of research.
Although some program managers at all four agencies have included data-
sharing requirements in grant awards, these agencies rely primarily on
policies and practices to encourage researchers to make climate change
data available. An interagency policy, as well as numerous agency,
program, and project-specific data-sharing policies, encourages
researchers to make climate change data available. The policies range
from broad statements calling for open and timely access to data to
more detailed policies that define the mechanisms and timelines for
making the data accessible. Further, these policies often vary
according to the needs of specific research programs or projects.
Beyond their written requirements and policies, all of the agencies
also rely on unwritten practices to facilitate data sharing. For
example, two program managers withhold grant payments if data have not
been made available for use by other researchers.
While the four agencies have taken steps to foster data sharing, they
neither routinely monitor whether researchers make data available nor
have fully overcome key obstacles and disincentives to data sharing.
Because agencies do not monitor data sharing, they lack evidence on the
extent to which researchers are making data available to others. Key
obstacles and disincentives could also limit the availability of data.
For example, one obstacle is the lack of archives for storing certain
kinds of climate change data, such as some ecological data, which
places a greater burden on the individual researcher to preserve it.
Preparing data for future use is also a laborious and time-consuming
task that can serve as a disincentive to data sharing. In addition,
data preparation does not further a research career as does publishing
results in journals. The scientific community generally rewards
researchers who publish in journals, but preparation of data for
others’ use is not an important part of this reward structure.
Consequently, researchers are less likely to focus on preserving data
for future use, thereby putting the data at risk of being unavailable
to other researchers.
What GAO Recommends:
GAO recommends the agencies explore opportunities in the grants process
to better ensure the availability of data to other researchers and
determine if additional archiving strategies are warranted. In
commenting on a draft of this report, the four agencies generally
agreed with our findings and recommendations. We incorporated technical
clarifications as appropriate.
To view the full product, including the scope and methodology, click on
[hyperlink, http://GAO-07-1172]. For more information, contact John B.
Stephenson at (202) 512-3841 or stephensonj@gao.gov.
[End of section]
Contents:
Letter:
Results in Brief:
Background:
The Scientific Community Has Identified Several Key Issues That
Policies Should Address to Facilitate Data Sharing:
Climate Change Research Agencies Rely on Various Policies and Practices
to Encourage Researchers to Make Data Available:
All Four Agencies Have Taken Steps to Foster Data Sharing but Have Not
Fully Overcome Key Obstacles:
Conclusions:
Recommendations:
Agency Comments and Our Evaluation:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: Data-Sharing Policies Applicable to Federally Funded
Climate Data:
Appendix III: Examples of Data-Sharing Expectations for Different
Federally Funded Research Projects:
Appendix IV: Governmentwide Climate Change Data-Sharing Policy:
U.S. Global Change Research Program Data and Information Working Group
Data Management for Global Change Research Policy Statements:
Background:
Applicability:
Guidelines and Their Application:
Suggested Data Product Requirement for Grants, Cooperative Agreements,
and Contracts:
Compliance:
Appendix V: Comments from the Department of Energy:
Appendix VI: Comments from the National Aeronautics and Space
Administration:
Appendix VII: Comments from the Department of Commerce, for the
National Oceanic and Atmospheric Administration:
Appendix VIII: GAO Contact and Staff Acknowledgments:
Figure:
Figure 1: General Research Grant Process:
Abbreviations:
CCSM: Community Climate System Model:
CCSP: Climate Change Science Program:
DOE: Department of Energy:
GLOBEC: Global Ocean Ecosystem Dynamics:
MILAGRO: Megacity Initiative: Local and Global Research Observations:
NASA: National Aeronautics and Space Administration:
NCAR: National Center for Atmospheric Research:
NOAA: National Oceanic and Atmospheric Administration:
NSF: National Science Foundation:
OMB: Office of Management and Budget:
PCMDI: Program for Climate Model Diagnosis and Intercomparison:
REASoN: Research, Education and Applications Solution Network:
RISA: Regional Integrated Sciences and Assessments:
[End of section]
United States Government Accountability Office: Washington, DC 20548:
September 28, 2007:
The Honorable Joe Barton:
Ranking Member:
Committee on Energy and Commerce:
The Honorable Ed Whitfield:
Ranking Member:
Subcommittee on Oversight and Investigations:
Committee on Energy and Commerce:
House of Representatives:
The federal government invests nearly $2 billion annually in climate
change research, the majority of which is in the form of grants,
cooperative agreements, and other awards funding researchers at
external entities such as universities and privately owned research
institutions.[Footnote 1] Currently, electronic archives exist to
systematically preserve some but not all kinds of data produced by
federally funded climate change research. It is generally the
responsibility of researchers to make data available, regardless of
whether an appropriate archive exists for their use. Much of the data
that are not made available through archives are retained by
researchers and may be in a less accessible format. As a result, some
researchers have expressed concerns about both the availability and
long-term preservation of the rapidly growing body of climate change
data.
According to the interagency Climate Change Science Program (CCSP),
federal investment in interdisciplinary earth sciences, global
observation systems, and satellite and computing technologies has
improved our understanding of climate change. The CCSP coordinates and
directs the climate change research performed by 13 departments and
agencies. Four agencies in particular account for about 90 percent of
the annual federal climate change research budget.[Footnote 2] The
principal climate change research agencies--the Department of Energy
(DOE), National Aeronautics and Space Administration (NASA), National
Oceanic and Atmospheric Administration (NOAA), and the National Science
Foundation (NSF)--fund research through numerous programs.
Climate change data come from many specialized disciplines within the
earth, physical, biological, engineering, and mathematical sciences.
These data are often available in disparate forms, such as data output
from models forecasting climate conditions, satellite images of ocean
and land masses, and ice core samples from the Arctic region. For
purposes of this report, we define data to include factual information
or physical samples that are collected and recorded as a result of
scientific observation, experiment, analysis, or similar methods of
research. The output of models can also be considered data.[Footnote 3]
The widespread availability of such data is important to developing a
comprehensive understanding of climate change and its potential
impacts. Indeed, committees of the National Academies, professional
scientific associations, and federal research agencies regard the free
and open exchange of data as an essential part of scientific research.
A 1997 National Academies committee report stated that:
"Governmental science agencies…should adopt as a fundamental operating
principle the full and open exchange of scientific data. By 'full and
open exchange' the committee means that the data and information
derived from publicly funded research are made available with as few
restrictions as possible, on a nondiscriminatory basis, for no more
than the cost of reproduction and distribution."[Footnote 4]
In this context, you asked us to determine (1) the key issues that data-
sharing policies should address as identified by the scientific
community in order to facilitate the sharing of federally funded
climate change data; (2) the requirements, policies, and practices for
making data available to other researchers under current climate change
research awards from the four major federal climate change research
agencies; and (3) the extent to which the four agencies effectively
foster data sharing.
In conducting our work, we identified and reviewed the data-sharing
requirements, policies and practices that are part of climate change
awards--primarily grants and cooperative agreements--funded by DOE,
NASA, NOAA, and NSF. We also conducted a Web-based survey of the 64
program managers who oversee the climate change research awards at
these agencies. We received a 100-percent response rate. We also
interviewed senior officials at DOE, NASA, NOAA, and NSF who direct the
climate change research programs as well as managers from data archives
that preserve climate change data. Finally, we reviewed relevant data-
sharing requirements, policies, and practices at other federal
agencies, academic journals, and professional societies and conducted
interviews with stakeholders representing those organizations (American
Geophysical Union, American Meteorological Society, Ecological Society
of America, the journal Science, the journal Nature, the National
Academies, and the National Institutes of Health). A more detailed
description of our scope and methodology is presented in appendix I. We
performed our work between September 2006 and September 2007 in
accordance with generally accepted government auditing standards.
Results in Brief:
The scientific community as represented by the National Academies and
professional scientific associations has identified several key issues
that data-sharing policies should address, including what, how, and
when data are to be shared, as well as the cost of making data
available. The scientific community generally believes that, at a
minimum, the information necessary to support researchers' major
published results should be made available to other researchers. The
scientific community, however, also acknowledges certain statutory
limits on data sharing related to the protection of intellectual
property, privacy, and national security, as well as practical limits
to sharing, such as the lack of archival infrastructure. Nevertheless,
it is generally accepted that when the appropriate infrastructure
exists, data acquired in federally funded research should be made
accessible through unrestricted archives. In terms of timing, the
scientific community believes that data should generally be made
available immediately or after a limited proprietary period that allows
researchers to complete their initial analysis and publish their
results. The duration of such a period may be determined by the type of
research. To address cost concerns, it is generally agreed that data
should be made available at no charge or at least no more than the
marginal cost of reproduction and distribution. Finally, the way in
which specific policies address these key data-sharing issues may vary,
depending on the type of research. Data-sharing policies must take into
account their applicability to specific research projects, relevant
legal and regulatory restrictions, the existence of appropriate
archives, and the characteristics of particular research fields.
While some survey respondents at the four major climate change research
agencies reported having incorporated data-sharing requirements into
particular grant awards, each agency relies primarily on established
policies and practices to encourage federally funded researchers to
make data available. The policies we identified for all four agencies
include the interagency CCSP data-sharing policy as well as agency,
program, and project-specific policies that vary in how they address
the key issues identified above. Agencies' policies range from broad
statements calling for open and timely access to data to more detailed
policies that define the mechanisms and timelines for making the data
accessible. Further, we found that these policies vary among agencies
and often vary according to the needs of a research program or project
within the same agency. For example, the overarching data-sharing
policy for NSF requires researchers to make data available to others
but does not specify how, whereas the policy for NSF's ocean sciences
program states that researchers should submit sediment samples from the
ocean floor to particular archives for long-term preservation. We also
found that large, collaborative research projects commonly have data-
sharing policies unique to the project. For example, the AmeriFlux
program--a network of climate change researchers funded by multiple
agencies, including DOE, NASA, NOAA, and NSF--requires participants to
submit data to a particular archive within 1 year of collection and
specifies the preferred format for data submission. Beyond their
written requirements and policies, all of the agencies also rely on
unwritten practices to facilitate data sharing. For example, a majority
of program managers surveyed identified archiving as one way for
researchers to make data available. In addition, two program managers
reported that they withhold installments of grant funds if researchers
do not make data available. The use of such practices varies among and
within agencies.
While all four of the agencies have taken steps to foster data sharing,
they do not routinely monitor whether researchers make data available
from all climate change research programs and have not fully overcome
key obstacles and disincentives to data sharing. For example, one key
obstacle is the limits in the data infrastructure to preserve
particular kinds of climate change data. Data archives for certain
kinds of data in some disciplines, such as ecology, do not exist, which
places a greater burden on the individual researcher to maintain and
preserve data. Preparing data for future use is also a laborious and
time-consuming task that can serve as a disincentive to sharing data.
Furthermore, multiple data management officials said that data
preparation does not result in the same benefits, such as career
advancement, as publishing results in journals can. Officials also
noted that researchers are expected to make underlying data available
and to publish results in journals, but traditionally the scientific
community has mainly rewarded publication. Consequently, researchers
who have to compete for funding are more likely to focus on publishing
research results than preserving underlying data for future use,
thereby putting the data at risk of being lost or inaccessible to other
researchers.
We are making recommendations to the federal climate change research
agencies to improve their ability to ensure that federally funded
research data are made available to other researchers. Specifically, we
recommend that the Secretary of Commerce and the NOAA Administrator
clearly inform researchers in writing of NOAA's data-sharing
expectations. We also recommend that DOE, NASA, NOAA, and NSF consider
steps for maximizing opportunities to encourage researchers to make
data available to other researchers, including evaluating data-sharing
plans when considering grant proposals. Finally, we recommend that the
agencies evaluate whether additional strategies to facilitate permanent
archiving of relevant data are warranted. In commenting on a draft of
this report, the four agencies generally agreed with our findings and
recommendations. Some of the agencies provided technical
clarifications, which we have incorporated in this report as
appropriate.
Background:
The federal government has funded climate change programs for over 20
years, and the budget for climate change research and development--
approximately $5.9 billion in fiscal year 2006--supports a wide range
of programs. As in the past, nearly half of the fiscal year 2006
federal climate change budget funded technology programs that focus on
responses to climate change, such as developing and deploying
technologies to reduce greenhouse gas emissions or increase energy
efficiency. Less than one-quarter goes toward tax provisions related to
climate change. These provisions encourage emissions reductions
through, for example, tax incentives to encourage the use of renewable
energy. A fraction of the budget also contributes to international
assistance programs that seek to help developing countries address
climate change by, for example, improving energy efficiency technology.
Finally, the estimated $1.7 billion spent on science research programs
accounts for roughly one-quarter of the total budget for climate change
programs and are the focus of this report.
Federal climate change science programs seek to monitor, understand,
and predict climate change through both agency-led and external
research activities. In particular, the science programs seek to
advance the state of knowledge on (1) natural climate conditions and
variability; (2) forces that influence climate; (3) climate responses;
(4) the potential impacts of climate change on the environment,
population, and the economy; and (5) ways to apply this knowledge to
decision making. A total of 13 federal departments and agencies support
climate change research activities, though 4 of these departments and
agencies--DOE, NASA, NOAA, and NSF--received about 90 percent of
climate change science funding in fiscal years 2005 and 2006. NASA
accounts for the greatest portion of the climate change science budget,
about 61 percent, followed by NSF (12 percent), NOAA (9 percent), and
DOE (8 percent).[Footnote 5] Agencies may also contribute funds from
nonclimate specific accounts to the infrastructure supporting climate
change research, such as sophisticated instruments and equipment. In
particular, the climate change research budgets do not reflect all of
the funds that NASA and NOAA contribute to satellite systems and
sensors used to collect data. Also, DOE, NASA, and NOAA each have
laboratories that perform climate change research. Unlike the mission-
based agencies, NSF is a funding agency supporting all fields of
fundamental science and engineering. NSF provides about 20 percent of
all federally supported basic research conducted at U.S. colleges and
universities and generally funds this work through limited-term grants
issued to institutions supporting individuals and small groups of
researchers.
All four agencies support external climate change research, primarily
through grants.[Footnote 6] The grant review process typically begins
when a researcher, or group of researchers, responds to an agency's
formal solicitation with a written research project proposal. Such
proposals generally summarize how the researcher would use grant funds
to respond to the agency's solicitation, including how the researcher
would perform the work as well as the budget and timeline for doing so.
Some proposals may also describe plans to collect and manage data.
The agencies assess many proposals on a competitive basis. Usually a
program manager who oversees many research grants assumes
responsibility for the scientific, technical, and programmatic review
of the proposals submitted in response to the solicitation. In addition
to the intellectual and scientific merit, as well as the potential
broader impact of the proposal, agencies may use criteria such as the
past performance of the researcher, as well as the budget and
priorities for the agency's program, when determining whether to fund
the proposal. Also, agencies request written reviews or independent
panels of the researcher's peers to assess the scientific merit of
proposals in some cases. The program manager then recommends to other
agency officials which proposals the agency should fund. The agency
then compiles notification letters that formally offer the grant to the
researcher's institution and outline the terms and conditions in the
grant agreement, a legal instrument describing the relationship between
the agency and the recipient (see fig. 1 for a summary of this
process).
Figure 1: General Research Grant Process:
This figure is an illustration of the five stages of the Agency
processes and the Researcher processes in General Research Grant
Process. The data depicted is as follows:
Stage 1: Annoucement stage;
Agency process: Announce opportunity through formal solicitation;
* Provide administration and technical support; Researcher process:
Find opportunity;
* Identify potential opportunity and develop research proposal.
Stage 2: Proposal stage:
Agency process: Receive and review porposal;
* Conduct reviews (adminsitrative, budget, policy, merit, business,
application, certification, and assurances); Researcher process: Submit
proposal;
* Submit proposal package, including applicable data-sharing plans.
Stage 3: Award Stage:
Agency process: Award notification;
* Notify the researcher's institution of award decision; Researcher
process: Receive notification of award;
* Complete applicable award acceptance documents.
Stage 4: Research stage:
Agency process: Disburse payment;
* Process payments to researcher throughout the research state; Agency
process: Management and oversight;
* Review researcher progress reports; Researcher process: Request and
receive payment;
* Request disbursement of grant funds. Researcher process: Perform
research and submit progress reports;
* Comply with terms and conditions-including administrative
requirements, cost principles, and submission of progres reports- as
well as data-sharing expectations.
Stage 5: Closeout stage:
Agency process: Closeout;
* Review and reconsile final report; Researcher process: Closeout;
* Publish findings, submit final report, and make data available, as
appropriate.
Source: GAO.
[End of figure]
After agreeing to the terms and conditions of the grant, the researcher
begins the work and submits periodic progress reports to the agency.
The researcher's primary point of contact with the agency is the agency
program manager who oversees the award. When the results of their
investigation are ready, researchers usually attempt to publish their
findings and conclusions in peer-reviewed journals. The publication of
research results in journals can advance the state of science and
benefit the researcher through, for example, career advancement.
According to a senior official at the journal Science, nearly all
researchers seek to share their results and conclusions through journal
articles, but as the National Academies and agency officials have
acknowledged, the mechanisms of making the data underlying their
results available to others can vary greatly and involve many different
stakeholders. Accordingly, the expectations for data sharing can vary
by research type. In the past, researchers generally kept the data they
collected or generated under a grant award in their possession and made
them available to other researchers upon request. The development of
sophisticated tools and use of the Internet as a means to disseminate
information has greatly expanded data-sharing opportunities.
Researchers can submit some types of climate change data to federal
archives that preserve electronic data online. Some of these archives
are managed by federal programs separate from those funding climate
change research. However, archival infrastructure does not exist for
all kinds of climate change data. Indeed, the National Science and
Technology Council has established an Interagency Working Group on
Digital Data to develop and promote a "strategic plan…to ensure
reliable preservation and effective access to digital data" derived
from federally funded research. To ensure access to relevant data, some
journals have developed online databases to store data that support the
articles they publish. Researchers may also make data available by
posting them on personal or institutional Web sites and, with physical
samples, by housing the materials in facilities such as the National
Ice Core Laboratory. When no archive or other mechanisms for making
data available exists, researchers may store data in their own files
and make them available to others upon request.
Determining what data to make available from past research activities
can pose a challenge because data are not always static or discrete. A
National Academies panel on Science, Technology, and Law described data
as information that moves through many levels, ranging from raw data to
final data, during the research process. Before they can be made
available, researchers validate and perform quality assurance measures
on the data by, for example, deleting outliers or coding the data for
use in software applications. Distinguishing between raw, processed,
and final data is often a subjective determination and requires
scientific and technical expertise.
Despite these differing expectations about how and what data to make
available, the scientific community has long promoted data
sharing.[Footnote 7] In particular, the National Academies have studied
and promoted data sharing through a series of committees, symposia, and
studies. Federal research agencies that fund data collection,
professional scientific associations such as the American Geophysical
Union and the American Meteorological Society, and academic journals
such as Science and Nature have also produced a series of statements
and policies on data sharing. Other information science scholars have
published studies of data-sharing policies and practices as well.
Though some of the work produced by this community has focused on data
sharing within particular disciplines, such as the earth and life
sciences, it believes that research data should generally be shared and
available to all researchers.
The Scientific Community Has Identified Several Key Issues That
Policies Should Address to Facilitate Data Sharing:
The National Academies, professional scientific associations, and other
members of the scientific community, have identified key issues that
data-sharing policies should address, including what, how, and when
data are to be shared, as well as the cost of making data available.
The way in which specific policies address key data-sharing issues,
however, may vary depending on the type of research. Policies must take
into account their applicability to specific research projects,
relevant legal and regulatory restrictions, the existence of
appropriate archives, and the characteristics of particular research
fields.
Data-Sharing Policies Should Address What Data Are to Be Shared:
The scientific community generally believes that data-sharing policies
should address what data are to be shared and that, at a minimum, the
information necessary to support researchers' major published results
should be made available to other researchers. The National Academies
have recommended in, for example, a 1997 report from the National
Research Council that federal science agencies adopt, as a fundamental
goal, the full and open exchange of scientific data derived from
federally funded research. Various scientific associations, such as the
American Association for the Advancement of Science and the American
Geophysical Union, have also identified the open availability of data
as an issue that data-sharing policies should address and support.
These organizations have supported open access to research data because
of the many benefits of sharing data. Open access to data, according to
these organizations, maximizes the societal benefits of the scientific
endeavor. Moreover, when data are widely available, the information can
be used to provide a direct check on reported results or advance future
research in a field of study. According to officials with whom we spoke
at archives and the National Academies, data sharing can be
particularly important in the field of climate change research, because
accessing data from a variety of sources is crucial to understanding
the multivariate nature of the earth's climate. Officials also
emphasized that information made available to the wider research
community should include both the raw data or physical samples
resulting from the research as well as the metadata--i.e., information
needed to understand the content, quality, and condition of the data--
because both the raw data and metadata are essential for other
researchers to make practical use of shared information. In addition,
NOAA stated that in all cases sufficient metadata, such as data set
descriptions, should be provided so the data can be found and their
suitability for use determined.
Though the full and open exchange of data is supported as an overall
goal, the scientific community acknowledges that there are certain
legally binding limitations to the goal of openness. In particular,
there are statutory and other legal limits on data sharing designed to
protect intellectual property, privacy, and national security.
Protecting the privacy of human subjects and national security have
been acknowledged as legitimate limitations to the full and open
exchange of scientific data by the National Academies. More recently,
according to the National Academies, protecting intellectual property
has created new restrictions on data sharing. National laws and
international agreements in the area of intellectual property rights,
privacy, and national security may directly affect data access and
sharing policies. Scientific associations have also recognized these
constraints to data sharing, while also noting that the majority of
data collected with public funds are not affected by these restrictions.
Practical limitations, such as a lack of appropriate archives for
storing data, can also affect how policies address the goal of
openness. If no archive exists, then researchers may not be able to
make their data available on a long-term, low-cost basis. Moreover,
policies applicable to research that produces modeling or experimental
research data may not require all results to be shared. While modeling
activities generate large volumes of data, only a portion has an
appropriate archive or is useful to the wider research community;
therefore, some model data are generally retained by the researcher and
made available to others upon request. Furthermore, according to a
senior official at the National Academies, experimental research data
are created as a result of a specific process or analysis and can often
be recreated, so sharing the actual data and materials is not as
important as sharing information about research methodologies. Raw
observational data, on the other hand, are unique and if not made
accessible to other researchers may be lost forever. Thus, the scope
and methods for sharing data generally depend on the type of research
that was conducted. Table 2 in appendix III provides examples of how
data-sharing expectations can differ by research project type.
Data-Sharing Policies Should Address How Data Are to Be Shared:
The scientific community generally believes that data-sharing policies
should address how data are to be made available and that, when the
appropriate infrastructure exists, data acquired in federally funded
research should be made accessible through unrestricted archives. Many
existing data-sharing policies and guidelines encourage researchers to
place their data in public archives. The National Academies have
recommended in multiple reports that these data be made readily
accessible--ideally via the Internet--through repositories that are
supported by a community of researchers and in general use. Scientific
associations have also acknowledged that data-sharing policies should
be guided by the goal of making data available for the long-term via
archives.
The lack of appropriate infrastructure for the sharing and preservation
of certain kinds of data may affect specific data-sharing policies,
particularly for federal agency research programs. Some scientific
disciplines, such as ecology and hydrology, do not have the
infrastructure to facilitate data sharing. Furthermore, research
performed by individual investigators or small research groups
operating outside large research programs may not have appropriate data
archives. Without such archival infrastructure, researchers working in
these fields or research programs may not be able to easily share their
data with others.
Data-Sharing Policies Should Address When Data Are to Be Shared:
The scientific community believes that data-sharing policies should
address when data should be made available and that data should
generally be made available immediately or after a limited proprietary
period that allows researchers to complete their initial analysis and
publish their results. In an early report on data sharing by the
National Research Council, the National Academies recommended that
research data be made available by the time the initial major results
are published, except in compelling circumstances. Further, the report
maintained that data relevant to public policy should be shared as
quickly and widely as possible. Various scientific associations also
support the goal of making data available to the public as early as
possible.
While immediate open access to data is desirable, the premature
disclosure of research data may disrupt the processes of analysis,
interpretation, and peer review that normally precede such public
disclosure, according to the American Association for the Advancement
of Science. Accordingly, a federal agency scientist told us that the
research community recognizes the need for researchers to perform
quality checks on data and publish their results before releasing the
data to other researchers. Indeed, a limited proprietary period for
principal researchers is a common principle in the research community.
However, the duration of such a period may be determined by the type of
research. In particular, the length of the proprietary period in which
a researcher, or a group of researchers, has exclusive access to data
may vary by research project or discipline. Some research projects,
such as those gathering observational data from satellites, are often
expected to make their data available immediately after the standard
period of calibration of equipment and validation of observations.
Alternately, other projects, such as those involving the collection of
physical samples, cannot make their data available immediately due to
logistical constraints. Ice core samples, for instance, cannot be made
widely available due to the fact that they are difficult to transport
and must be stored in a particular, central location. Moreover, certain
projects involve a substantial investment of time and resources by the
researchers, and it is generally agreed that such researchers are
entitled to a period of exclusive use of the data they have collected.
Such projects are, in some cases, granted a proprietary period of up to
2 years to allow researchers to develop publishable results and prepare
the data for sharing. Still other projects, including those involving
multiple researchers, make their data available among original
researchers immediately and to other researchers after a limited
proprietary period. According to a 1997 National Academies report on
data access issues, the maximum length of any proprietary period should
be established by particular scientific communities. Moreover,
scientific associations have acknowledged the need for differing
proprietary periods and called for federal agencies to tailor their
data-sharing policies and expectations to specific research projects
when necessary.
Data-Sharing Policies Should Address the Cost of Making Data Available:
The scientific community generally agrees that data should be made
available at no cost or for no more than the marginal cost of
reproduction and distribution. Moreover, the process of sharing data
should seek to minimize the burden to researchers of making data
available. The National Academies and various scientific associations
recommend the full and open exchange of data, including making
federally funded research available for no more than the cost of
reproduction and distribution. According to the American Geophysical
Union, this goal is designed to balance the costs associated with
sharing data with the desire to make data easily accessible, so as to
not impose significant burdens on original or subsequent researchers.
Government agencies have often charged fees for access to data in order
to recover the costs of generating or reproducing the data. However,
with the reduced costs of capturing and storing digital data, agencies
are now often able to provide data for no cost on the Web. Indeed, in a
2003 report, the National Academies recommended making federally funded
data available for research purposes at no cost when possible.
Nevertheless, since the cost of sharing data will likely depend on the
type and format of the data, archived data not available digitally--
such as physical samples--may involve higher costs for original or
subsequent researchers.
Climate Change Research Agencies Rely on Various Policies and Practices
to Encourage Researchers to Make Data Available:
All four agencies said that they adhere to governmentwide data-sharing
guidelines and, to varying degrees, have their own agency, program, and
project-specific data-sharing policies. The manner in which these
policies address key data-sharing issues like openness, timing, and
cost vary among and within agencies based on the needs of specific
research programs. Agencies also facilitate data sharing through
unwritten practices, such as providing incentives for data sharing
through the grants process, maintaining personal contact with
researchers, and encouraging researchers to archive data.
All Four Agencies Said They Adhere to the Governmentwide Data-Sharing
Policy:
While federal statutes do not clearly specify data-sharing requirements
for external climate change researchers using federal funds, [Footnote
8] some program managers at each agency reported in our survey that
they had incorporated requirements into particular grants. The
agencies, however, have relied primarily on a number of policies and
practices to encourage data sharing among external researchers. At the
broadest level, the agencies recognize an interagency policy on climate
change data, which represents a governmentwide commitment to make
climate change data available to other researchers.[Footnote 9]
Specifically, the Data Management for Global Change Research Policy
Statements, an interagency policy under the Climate Change Science
Program (CCSP), provides guidance to the agencies on how to ensure that
researchers make federally funded climate change data available to
researchers.[Footnote 10] A related interagency research group that
predates the CCSP--the Global Change Research Program--developed the
policy in response to concerns that inadequate attention was given to
maintaining climate change data. The Global Change Research Program
observed that "proper preparation, validation, description, and care of
data sets is critical to their use by the widest possible scientific
community." The CCSP has encouraged those agencies funding climate
change research to incorporate the guidelines listed in this voluntary
policy into their data-sharing policies and practices. Senior officials
at DOE, NASA, NOAA, and NSF told us that their data-sharing policies
and practices adhere to the principles of the guidelines.
The interagency policy addresses the key issues of data sharing, such
as openness and accessibility. For example, the policy's overarching
objective calls for the "full and open sharing of the full suite of
global data sets" by all climate change researchers. The policy further
specifies what counts as data and broadly defines them as the
information "resulting from observations, the application of algorithms
to produce new data, and from the data output of models." The policy
states that metadata should be made available to allow researchers to
assess the quality of data. In addition to encouraging open data
sharing among all climate change researchers, the policy addresses
accessibility by recommending the long-term preservation of data in
archives. The policy states that agencies funding research should
develop procedures and criteria for obtaining, maintaining, and purging
data in the archives. See appendix IV for more detailed information on
how the policy addresses key data-sharing issues.
Data-Sharing Policies Vary among and within Agencies:
Our review, which included a survey of 64 program managers at the four
major climate change research agencies, identified 23 different
policies[Footnote 11]--accounting for about 80 percent of the agencies'
climate change research programs--that encourage researchers to make
data available.[Footnote 12] Although data sharing is generally
regarded as a standard practice among colleagues, the mechanics of data
sharing--such as what data to preserve and when--involve some
professional judgment. To guide researchers, the four major climate
change research agencies have policies that document these mechanics.
Agencies' written policies emphasize their commitment to data sharing
and standardize expectations for data sharing. Overall, the policies
range from broad statements calling for open and timely access to data
to more detailed policies that define the mechanisms and timelines for
making data accessible. NASA and NSF have data-sharing policies
documented at the agency level that address openness and timing and
apply to all topics of research; all four agencies have various program
and project-specific data-sharing policies.
While the policies generally underscore the importance of making data
openly available at minimal cost, we found that they vary among and
within agencies because they are often tailored to the needs of
different research programs or projects within the same agency.
Accordingly, these policies address in different ways the key issues
discussed in the previous section of this report. For example,
variations in archiving resources and the extent of quality assurance
required--such as validation and calibration--influence a policy's
recommended time frames for data sharing. See appendix II for a
complete list of the data-sharing policies.
DOE's climate change research programs have established written
policies that encourage researchers to make data available within
certain time frames and according to specific standards. One of DOE's
programs that funds the collection and analysis of measurements from
instruments through a network of researchers issued a policy that
encourages researchers to make data available quickly. This particular
DOE policy distinguishes between data that have been quality assured
and preliminary data, which have not been validated, and affords
researchers some time to work on the data before finalizing data
submissions to an archive. However, DOE expects that even the
preliminary data will be made available almost immediately. The policy
calls for "near real-time" sharing of the data among the researchers
participating on the team and for making the data available to the
research community within days to allow for routine processing and
electronic archiving.
NASA's agencywide policy briefly states that researchers should make
data available at the earliest possible time, whereas its earth science
program provides greater detail about what data to share and when to do
so. For example, NASA's earth science program policy states that
researchers do not have a period of exclusive access to the satellite
data, which are made available in the agency's data archive system as
soon as they are calibrated and validated. According to senior NASA
officials, the program formerly granted researchers a 2-year period of
exclusive use of the data but determined that the wider benefits of
making data available to all outweighed the benefit of temporary
restrictions. One NASA official, however, noted that there is a trade-
off, as the lack of an exclusive use period limits opportunities for
researchers to analyze the data and make them more user friendly.
The NSF agencywide policy states that researchers are "expected to
share with other researchers, at no more than incremental cost and
within a reasonable time, the primary data, samples, physical
collections and other supporting materials created or
gathered."[Footnote 13] In order to address the needs of specific
research programs, program-level policies often provide researchers
more detailed guidance about how to carry out the agencywide data-
sharing policy. This agencywide policy establishes a general
expectation that data are to be shared with other researchers. The data-
sharing policy for the oceans program--one of NSF's programs funding
particular climate change research--identifies particular archives for
researcher use, such as one that preserves sediment samples from the
ocean floor. Further, the agencywide policy states that data are to be
shared "within a reasonable time" and the oceans program policy states
that data should be shared as soon as possible but no later than 2
years after collection.
The Climate Observation Program is NOAA's only climate change research
program that has issued a written data-sharing policy. Similar to the
DOE example given above, NOAA's Climate Observation Program policy
states that researchers should make data available near-real-time with
associated metadata and free of charge to others. The policy further
notes that the data should be made available quickly enough to "be of
value to operational forecast centers, international research programs,
and major scientific assessments."[Footnote 14]
We found that data-sharing policies vary in part because the type of
data generated differs by program. An official with the American
Association for the Advancement of Science observed that variations in
data-sharing policies by data type reflect the differences in the ways
data are collected and accessed. According to one survey respondent,
data generated instantaneously--such as meteorological data from
instrument measurements--may not require as much preparation or quality
assurance as other forms of data, like physical samples, that may
require extensive analysis and interpretation. For example, according
to a DOE official, large atmospheric science data sets generated by DOE-
funded researchers require supercomputers for analysis and therefore
require more time and processing before the researcher can transfer the
data to someone else.
Furthermore, we found that agencies also have project-specific data-
sharing policies. The AmeriFlux program--a network of climate change
researchers funded by multiple agencies, including DOE, NASA, NOAA, and
NSF--provides another example of how the agencies tailor data-sharing
policies to the needs of particular projects.[Footnote 15] The
AmeriFlux program requires participants to submit data to a designated
archive within 1 year of collection and specifies the preferred format
for data submission. According to program officials, the 1-year limit
allows researchers to spend time preparing and documenting the data in
adherence to the standards specified by the archive that are intended
to facilitate access and use by other researchers.
We also found that large, collaborative projects like AmeriFlux usually
establish data-sharing policies for the participants. These projects
typically involve multiple funding agencies and researchers based in
different locations, some even in different countries. Similar to the
program-level policies, the project-specific policies tailor the
agencies' expectations for data sharing to the data management needs of
the project. For example, the NSF-and NOAA-funded U.S. Global Ocean and
Ecosystems Dynamics project, which involves collaboration among
physicists, biologists, chemists, meteorologists, and resource
managers, established a policy to guide participants' data sharing. The
policy describes data sharing as an iterative process and instructs
researchers to work with the data managers to assess what data would be
most important to share. The policy also encourages researchers to
submit data to an archive and to include metadata to facilitate their
use by others.
Agencies Also Rely on Unwritten Data-Sharing Practices:
While many program managers described written policies for data sharing
as essential to advancing the state of climate change science, they
also identified unwritten practices that the agencies use to encourage
and facilitate data sharing. The flexibility of practices allows
program managers to tailor data sharing to the needs of a specific
project. These data-sharing practices include using the grants process
to provide incentives for data sharing, maintaining personal contact
with the researchers, and archiving data. Our review shows that use of
practices varies among and within the agencies.
Grants Process and Personal Contact Are Used to Encourage Data Sharing:
The agencies use the grants review process to provide incentives for
data sharing, thereby encouraging researchers to make data available.
Indeed, some program managers use the evaluation of grant proposals as
an opportunity to encourage researchers to identify and plan in advance
for data management needs--such as how they will preserve and make data
available. We found that NSF expects researchers applying for grants to
present, as appropriate, a clear description of "plans for
preservation, documentation, and sharing of data, samples, physical
collections, curriculum materials, and other related research and
education products."[Footnote 16] However, the general grant guidance
materials for researchers applying for DOE, NASA, and NOAA climate
change grants do not explicitly instruct them to include data-sharing
plans in their proposals. Nevertheless, some program managers encourage
researchers to do so in practice.
DOE and NASA officials told us that program managers might encourage
researchers to include data-sharing plans on an ad-hoc basis. An
example of this practice, according to DOE, is to request the data-
sharing plan in the solicitation notice for a particular award. DOE and
NASA officials could not confirm the frequency of this practice,
however. The extent to which program managers can use data-sharing
plans as a criterion for grant award decisions appears limited because
most of the climate change research programs do not explicitly require
them.
Funding decisions made throughout the grant process are also used by
agencies to hold researchers accountable to data-sharing expectations.
Most of the program managers we surveyed reported that they consider
researchers' past data-sharing practices when deciding whether to fund
research proposals. Once the agency has awarded the grant, program
managers may use the staggered installments of grant funds as another
incentive to encourage researchers to make data available to others.
Two program managers reported in our survey that they withhold funding
installments if researchers have not made data available.
The extent to which federal climate change research agencies use
various aspects of the grant review process to encourage data sharing
varies, depending on the initiative of the program manager, in part
because there are no requirements for them to do so. For example, an
NSF official stated that the consideration of past data-sharing
activities is not a discrete factor that the agencies require program
managers to use in making award decisions.
Moreover, the agency officials told us that they have limited
information about whether researchers make data available. Some program
managers said that they attempt to determine whether researchers are
making data available by reviewing progress reports--required written
updates submitted by researchers. Progress reports inform program
managers of the status of research throughout the duration of the
grant, and a final report documents completion of the research. The
final progress report typically states whether the researcher has
published the results in a journal. However, progress reports do not
necessarily provide program managers enough information to assess the
availability of the data.
Many program managers reported in our survey that they maintain
personal contact with researchers to ensure that they make data
available to other researchers.[Footnote 17] Such personal contact can
serve to remind researchers of the importance of making data available
and help them address any difficulties in doing so. One agency official
noted that program managers often rely on personal contact to encourage
researchers to make data available to others. Some program managers
also reported that they would authorize additional funds to help ensure
that data sharing occurs.[Footnote 18]
Agencies Also Rely on Data Archiving to Foster Data Sharing:
Data archiving is one of the primary data-sharing practices used by the
federal climate change research agencies, according to our survey and
interviews with agency officials and data management experts. Archiving
refers to the long-term storage of data, most often in digital form,
but there are also some repositories that can hold physical samples. A
majority of program managers surveyed identified archiving as one way
that researchers make data available. Climate change research directors
at all four agencies also said that the agencies encourage researchers
to use archives to make data available.
In addition to encouraging researchers to archive data, the mission-
based agencies--DOE, NASA, and NOAA--support archiving practices by
operating permanent data archives that store data, such as satellite
images and measurements of indicators in the atmosphere, land, and
oceans used to understand climate change. According to senior
officials, these agencies have made notable financial investments in
these data management and preservation services.[Footnote 19] For
example, according to a DOE official, the agency contributes about $2.7
million annually to the Carbon Dioxide Information Analysis Center,
which preserves data from researchers collaborating in the AmeriFlux
Network and other climate change programs.
The archives, typically managed by programs separate from those
sponsoring climate change research, provide a number of services to
facilitate data sharing. For example, staffs operating the agencies'
data centers do not only permanently archive data in electronic
databases, but they also perform quality control measures to
standardize and make data usable to a wide audience, develop data
products to facilitate additional analysis, and help other researchers
navigate the database to find relevant information.
While part of the agencies' investment in archives support data
managers, generally neither the agency nor the data managers actively
solicit data from the researchers. One exception identified by a survey
respondent is NOAA's Climate Prediction Program for the Americas, which
employs a manager who collects data from researchers, performs quality
control measures on the data, and make data available on a Web site. A
second exception is the outreach conducted by staff at an NSF-funded
archive, the National Center for Atmospheric Research (NCAR) Coupling,
Energetics, and Dynamic Atmospheric Regions. Another survey respondent
said that NCAR data managers collect instrument measurements from
researchers funded by one of NSF's climate change programs. As part of
its efforts in maintaining the archive, NCAR sends reminders to
researchers to submit data, and as necessary, notifies NSF program
managers of researchers who have not submitted data. NSF follows up
with the researchers to ensure data are submitted to the archive.
However, not all data can be digitally archived because, for example,
data such as physical samples may not be in a form amenable to this
type of storage or because archives do not exist for the data from some
types of climate change research. Agencies recognize this challenge and
have relied on other practices to encourage data sharing. For example,
most program managers (59 of 64) reported that researchers publish
research results in journals that indicate where to find the underlying
data.
All Four Agencies Have Taken Steps to Foster Data Sharing but Have Not
Fully Overcome Key Obstacles:
While the agencies have taken steps to foster data sharing, the
effectiveness of their requirements, policies, and practices is unclear
because the agencies do not routinely monitor researchers' data-sharing
activities. As a result, the agencies lack information to assess the
extent to which researchers are making federally funded climate change
data available. In addition, we found that the agencies have not fully
overcome key obstacles and disincentives to data sharing that could
limit the availability of data.
Agencies Do Not Routinely Monitor Whether Researchers Make Data
Available:
While senior officials at all four agencies believe that researchers
share the data derived from federally funded research projects, the
effectiveness of their data-sharing requirements, policies, and
practices is unclear because the agencies do not routinely monitor
whether researchers make data available from all climate change
research programs. Instead of proactively overseeing data sharing, the
agencies rely on self-policing within the research community. That is,
they assume that researchers will adhere to the norms of data sharing
and expect members of the research community to notify them when
researchers do not make data available. According to our survey,
roughly one-third of the 64 program managers have, within the past 10
years, been notified that an award recipient did not make data
available. Nearly all of the program managers said they responded to
the reported problem, and many believed it was resolved.
Although researchers can contact the agency if other researchers
withhold data, this is not an effective way to resolve situations
involving incomplete or missing data. Several data managers told us
that documentation about the data--such as conditions under which it
was gathered--is crucial because important details about the data are
likely to be forgotten as the researcher moves on to new projects.
Furthermore, at some point, it may become too late for federal agencies
to encourage data sharing because by the time one requests access to
certain data--possibly years after the initial data collection--the
original researcher may have lost the data or failed to record
important metadata. Therefore, we believe that agencies' reliance on
self-policing by the research community does not provide adequate
assurances that researchers will fulfill the data-sharing expectations
set forth in the agencies' policies.
Senior agency officials at all four agencies told us that it is
impractical for program managers to verify data sharing because they
oversee many researchers and must focus on higher priority tasks.
Moreover, several of these officials believe that current self-policing
is effective because of the collaborative nature of climate change
research. The agencies fund many large climate change research projects
that involve multiple researchers who depend on one another to share
data in a timely manner. The researchers participating in such projects
typically submit data to an archive and also hold one another
accountable. For example, senior DOE and NASA officials reported that
they convene science team meetings wherein they coordinate activities
and receive updates from the funding agency. According to a senior DOE
official, meeting participants address data-sharing issues at these
meetings. He noted that the meetings provide a particularly effective
forum for researchers to call attention to those who have not made data
readily available. However, there appears to be greater accountability
among researchers collaborating with one another on similar projects
than among researchers who work on individual projects.
Researchers seeking data that have not been made widely available, such
as through an archive, generally need to contact the original
researcher(s) to request data. While most of the program managers we
surveyed indicated that there are several incentives for researchers to
make data available--such as maintaining informal relationships with
other researchers, obtaining recognition in the scientific community
for the work, or the potential for future collaboration--there is no
guarantee that the original researcher will have the complete data
readily available to comply with another researcher's request for data.
Furthermore, researchers face a number of practical obstacles that may
limit their ability to document and preserve data.
The Agencies Have Not Fully Overcome Key Obstacles and Disincentives to
Data Sharing:
Despite the various incentives for researchers to make data available
to others, there are several obstacles and disincentives to data
sharing that the four agencies have not fully overcome. For example,
one key obstacle is the limits in the data infrastructure, such as the
lack of archives capable of preserving certain kinds of climate change
data being generated by federally funded research. Data centers funded
by DOE, NASA, and NOAA currently archive digital data; some data
centers preserve physical samples such as ice cores or ocean sediments.
Archives currently in operation store data from some areas of climate
change, including oceans and atmospheric sciences; but according to
officials at NSF, the National Research Council, and several scientific
societies, permanent repositories are not available for other fields
within climate change, such as certain kinds of ecological and earth
sciences data.
According to several data management stakeholders, the options
available to preserve data, such as electronic archives, are limited
for climate change data developed through the use of computer models.
While there are some archives that store data from climate change
models, such as the DOE-funded Program for Climate Model Diagnosis and
Intercomparison, these stakeholders told us that permanent model data
archives are generally lacking. Furthermore, the limits in data
infrastructure for climate change data create a greater burden for
federally funded researchers to maintain and preserve data themselves.
The National Academies have raised concerns about the long-term
availability of federally funded data and observed in one report that
"data sets that commonly are gathered at great expense and effort are
not broadly available and ultimately may be lost, squandering valuable
scientific resources."[Footnote 20] The report concluded that funding
agencies should be responsible for making the data available to others.
The four agencies also have yet to effectively address key
disincentives to data sharing on the part of researchers. For example,
the time and labor required to prepare data are significant
disincentives to making data available for other researchers. One
program manager commenting on the practical obstacles to data sharing
noted that while most researchers are willing to share data, they
"resist the large additional costs of time or money to meet
requirements." Making data available often involves laborious and time-
intensive tasks to adequately document the data and to perform quality
assurance checks, such as correcting errors, to make them usable for
other researchers. For example, the National Academies have recognized
an administrative and cost burden that largely falls on the researcher
to prepare data for others' use. Researchers may also need to summarize
the data processing history, develop a codebook, and write instructions
on how to use the data files.
Moreover, researchers must weigh the trade-offs in costs and benefits,
according to one program manager, such as the limits of the program
budget and whether responding to detailed requests would impede
progress on additional research. Some of the directors of the climate
change research programs raised similar concerns about research
priorities, in light of resource constraints. A senior DOE official
noted that while the agencies can encourage researchers to make data
available, funding priorities do not typically favor the time-consuming
tasks involved in making data available. The official clarified that
when faced with budget constraints, agencies tend to target limited
funds to new visible research at the expense of data archiving. The
National Academies also recognized the bias toward new research
projects and found that among all scientific disciplines, most agencies
make data management and preservation a low priority, even when the
benefits of making data available from old projects exceed those
realized from new projects.
Furthermore, multiple data management officials pointed out that
researchers do not receive the same benefits, such as career
advancement or peer recognition, for preparing data as they do from
publishing research results in journals. These officials stated that
funding agencies and the scientific community expect researchers to
both publish their results and make underlying data available, but
researchers have traditionally been rewarded mainly for publication.
According to a National Academies report on data access, "society
fellowship and award committees generally do not place much value on
the contributions their applicants may make to the infrastructure of
science in the form of data compilation, organization, and evaluation
work."[Footnote 21] As a result, researchers who have to compete for
funding are more likely to focus on publishing research results than
preserving underlying data for future use, thereby putting the data at
risk of being lost or inaccessible to other researchers.
Our survey identified several additional disincentives that may deter
data sharing, at least temporarily, including requests for more time to
analyze data and concerns about intellectual property.
Conclusions:
Government agencies articulate expectations for recipients of federal
grants about important functions such as data sharing through written
policies. Written policies both show that the agency views data sharing
as a priority and facilitate researchers' understanding of specific
expectations about the mechanics of data sharing, which typically
involve some professional judgment to determine, for example, what data
to preserve, how to make it widely available, and the time frame for
doing so. One particular collaborative program funded by DOE, NASA,
NOAA and NSF, known as AmeriFlux, has written requirements designating
the archive where researchers must submit their data as well as the
time frame and preferred format for these submissions, all of which
facilitate efficient data sharing by its participating researchers.
This written policy helps ensure that all participating researchers
understand the expectations and make data available in a way that
advances the goals of the project. Similarly, written data-sharing
policies exist under most federal climate change research programs at
DOE, NASA, and NSF. Most of the research programs at NOAA, however,
have not documented the agency's data-sharing expectations. Agencies
such as NOAA that do not have a written policy at either the agency or
program-level have fewer assurances about a mutual understanding of
data-sharing expectations.
Federal agencies also use the grant review process to encourage data
sharing by researchers. In some cases, the agency requires researchers
to submit data-sharing plans in their grant proposals but the extent to
which they use this as a criterion for grant award decisions appears
limited. Once the agency has awarded a grant, program managers may use
the staggered installments of grant funds as leverage to encourage
researchers to make data available to others. Some program managers
have effectively withheld funding installments when researchers do not
make data available, while others review progress reports to determine
whether researchers are making data available, taking action where they
find instances of delay. In addition, during the grant review process,
some officials informally consider researchers' past data-sharing
practices in their evaluation, which conveys the importance of sharing
research results among those involved in the research process. However,
agencies have not institutionalized the use of the grants process to
further data sharing and such efforts currently depend largely on the
initiative of individual program managers who often oversee large grant
portfolios.
The four research agencies we examined have policies and employ
practices that encourage data sharing, which is ultimately the
responsibility of the researcher. The agencies generally do not monitor
and keep track of whether researchers make federally funded research
data available. While the agencies believe that their data-sharing
requirements, policies, and practices are effective, this is largely
because they rarely receive reports suggesting otherwise. However,
without data on actual data sharing by researchers, agencies cannot be
sure their policies are working or determine whether changes in these
policies are warranted. Measuring progress toward a goal of data
sharing can allow agencies to adjust their efforts over time to ensure
that data are widely available to other researchers.
There are a variety of practical obstacles and disincentives to
researchers sharing their data. Infrastructure is limited for storing
data, such as that developed through computer models; and some fields
of science, such as ecology, do not currently have archives in place
that could maintain and preserve certain data. While developing and
maintaining archives is an expensive undertaking, it is extremely
important in areas of research related to climate change. Scholars from
the National Academies and elsewhere have acknowledged the need to
consider devoting additional research funds for the preservation of
research data so that these valuable scientific resources, commonly
gathered at great expense and effort, are broadly available to foster
further research and analysis of long-term issues such as climate
change.
Recommendations:
To assist federal agencies sponsoring climate change research to better
ensure the availability of data from federally funded research, we are
making the following four recommendations.
To ensure that researchers receiving federal funds to conduct climate
change research understand NOAA's expectations for data sharing, we are
recommending that the Secretary of Commerce and the NOAA Administrator:
* Develop a set of written guidelines or use existing governmentwide
guidelines, such as those endorsed by the Climate Change Science
Program, to clearly inform researchers of NOAA's general expectations
for data sharing.
To ensure that the agencies maximize opportunities to make data
available in a manner useful to other researchers, we recommend that
the Secretaries of Commerce and Energy, the NASA Administrator, the
NOAA Administrator, and the NSF Director consider the following actions:
* Develop mechanisms for agencies to be systematically notified when
data have been submitted to archives, so that agency officials have
current information about the extent of data availability in order to
adjust data-sharing policies over time to best meet the needs of
researchers and the communities that use their data.
* Use the grant review process, where their program offices are not
currently doing so, to facilitate further data sharing by (1)
evaluating researchers' data-sharing plans as part of the grant review
process and (2) using evidence of researchers' past data-sharing
practices to make future award decisions. The use of such criteria in
the grant review process should be clearly conveyed to researchers
before they submit research proposals and after award decisions have
been made.
To ensure that researchers make climate change data available to other
researchers, we recommend that the Secretaries of Commerce and Energy,
the NASA Administrator, the NOAA Administrator, and the NSF Director:
* Evaluate whether additional strategies are warranted to facilitate
the permanent archiving of relevant data, which may include: leveraging
existing resources; devoting a greater portion of data collection funds
to archiving activities; or working with existing entities such as the
National Science and Technology Council's Interagency Working Group on
Digital Data, to develop additional data archives.
Agency Comments and Our Evaluation:
We provided draft copies of this report to DOE, NASA, NOAA, and NSF.
The four agencies generally agreed with our findings and
recommendations. In addition, several agencies offered specific
comments and technical clarifications, which we have incorporated in
this report as appropriate. The written comments submitted by DOE,
NASA, and NOAA are presented in appendixes V, VI, and VII; NSF provided
technical clarifications orally. DOE commented on the importance of
defining "data" and questioned whether we considered samples as data
for purposes of this report. Our draft report included a definition of
data that we have repeated, in an appropriate context, in an additional
section of the final report. This broad definition, which includes
research samples, allowed us to obtain a wide perspective on the
variety of data-sharing requirements, policies, and practices. As we
note in the report, each data-sharing policy may have different
definitions of what data need to be shared. We agree that policies for
physical samples will differ from those for electronic data, but we
believe that each agency should make those determinations at the
appropriate level.
As agreed with your offices, unless you publicly announce the contents
of this report earlier, we plan no further distribution until 30 days
from the report date. At that time, we will send copies to interested
congressional committees and Members of Congress, the Secretaries of
Commerce and Energy, the NASA Administrator, the NOAA Administrator,
and the NSF Director. We also will make copies available to others upon
request. In addition, the report will be available at no charge on the
GAO Web site at [hyperlink, http://www.gao.gov].
If you or your staff have questions about this report, please contact
me at (202) 512-3841 or stephensonj@gao.gov. Contact points for our
Offices of Congressional Relations and Public Affairs may be found on
the last page of this report. GAO staff who made key contributions to
this report are listed in appendix VIII.
Signed by:
John B. Stephenson:
Director, Natural Resources and Environment:
[End of section]
Appendix I: Objectives, Scope, and Methodology:
Our objectives in this study were to determine (1) the key issues
identified by the scientific community that data-sharing policies
should address in order to facilitate the sharing of data from
federally funded climate change research; (2) the requirements,
policies, and practices to make data available to other researchers
that exist under current federal climate change research awards from
the four major federal climate change research agencies; and (3) the
extent to which the major agencies effectively foster data-sharing. We
defined the major federal climate change research agencies as those
representing the bulk of federal climate change research spending. The
Department of Energy (DOE), National Aeronautics and Space
Administration (NASA), National Oceanic and Atmospheric Administration
(NOAA), and the National Science Foundation (NSF) represented nearly 90
percent of the U.S. climate change research budget in fiscal year 2006.
To address the first objective, we reviewed data sharing requirements
and policies at federal agencies that were not identified by GAO as one
of the four major federal climate change research agencies, such as the
National Institutes of Health. We also reviewed the data-sharing
policies at academic journals, including the American Economic Review,
Bulletin of the American Meteorological Society, Econometrica,
Geophysical Research Letters, Global Biogeochemical Cycles, Journal of
Applied Econometrics, Journal of Applied Meteorology and Climatology,
Journal of Atmospheric Science, Journal of Climate, Journal of
Geophysical Research, Journal of Physical Oceanography, as well as the
journals Nature and Science. We reviewed these particular journals
because they either have an explicit data-sharing requirement or were
identified in our survey of agency program managers as a leading
publisher of climate change research. We also reviewed the data-sharing
policies, statements, and reports of professional scientific societies,
including the American Association for the Advancement of Science,
American Geophysical Union, American Meteorological Society, Ecological
Society of America, Geological Society of America, International
Council for Science, and the U.S. National Academies. These
associations were chosen because they either represented a broad cross-
section of the scientific community or represent researchers in
disciplines related to climate change research. Beyond an examination
of the written policies and statements, we conducted interviews with
officials at these organizations to gather additional information on
data-sharing goals, practices, and issues. We also conducted a
literature search to identify relevant studies of data-sharing
policies, practices, and challenges. For the purposes of this report,
the scientific community refers to the general body of scientists and
its institutions as represented by the National Academies and
professional scientific associations. While no single body can be said
to speak for all of science, the National Academies and other
scientific associations such as those listed above often act as
surrogates when the opinions of the scientific community, or particular
disciplines within science, need to be ascertained. We also
supplemented our analysis of the reports and statements of these
organizations with interviews of officials, at a variety of entities,
with knowledge of data-sharing issues. Furthermore, whenever we
attribute statements to the scientific community at large, we mean that
a National Academies study and at least two of the scientific
associations listed above support those statements.
To address the second and third objectives, we identified and reviewed
the data-sharing requirements, policies, and practices that exist under
the climate change awards--including grants, cooperative agreements,
and funded field work proposals--funded by DOE, NASA, NOAA, and NSF. We
also interviewed senior officials at DOE, NASA, NOAA, and NSF who
direct the climate change research programs. For additional context on
how data sharing is carried out, we interviewed managers from data
archives that preserve climate change data, including Lawrence
Livermore National Laboratory's Program for Climate Model Diagnosis and
Intercomparison, the Long-Term Ecological Research Center Network,
NOAA's National Climatic Data Center, and Oak Ridge National
Laboratory's Carbon Dioxide Information Analysis Center.
To assist in identifying relevant data-sharing requirements, policies,
practices, and issues at the four major federal climate change research
agencies, we conducted a Web-based survey of all 64 program managers
who oversee the climate change research awards at these agencies. We
conducted the survey from April 3 to May 3, 2007. To prepare the
questionnaire, we pretested potential questions with at least one
program manager at each of the four agencies as well as a Senior Earth
Scientist with the U.S. Climate Change Science Program to ensure that
(1) the questions were clear, (2) terminology was used correctly, (3)
questions did not place an undue burden on the respondents, (4) the
information was feasible to obtain, and (5) the questionnaire was
comprehensive and unbiased. On the basis of feedback from the six
pretests we conducted, we made changes to the content and format of
some survey questions. The final questionnaire included a combination
of open-and closed-ended questions about the data-sharing requirements,
policies, and practices at the program manager's agency and specific
program.
To ensure an adequate and appropriate response to our questionnaire,
agencies provided the names and contact information for climate change
program managers. We contacted all of the program managers in advance
to ensure that we had identified the correct respondents. We also sent
letters to the program managers informing them of the approximate date
the survey would be available, and we then sent an e-mail when the
survey was available via the Internet. After the survey had been
available for 1 week, we used e-mail and telephone calls to contact the
program managers who had not completed their questionnaires. Using
these procedures, we obtained a 100-percent response rate. Because this
was not a sample survey, there are no sampling errors. However, the
practical difficulties of conducting any survey may introduce errors,
commonly referred to as nonsampling errors. For example, difficulties
in how a particular question is interpreted, in the sources of
information that are available to respondents, or in how the data are
entered into a database or were analyzed can introduce unwanted
variability into the survey results. We took steps in the development
of the questionnaire, the data collection, and the data analysis to
minimize these nonsampling errors. For instance, a survey specialist
designed the questionnaire in collaboration with GAO staff with subject-
matter expertise. Further, the draft questionnaire was pre-tested with
a number of agency program managers to ensure that the questions were
relevant, clearly stated, and easy to comprehend. When the data were
analyzed, a second, independent analyst checked all computer programs.
In several cases, we contacted respondents to clarify their responses
to the questions, but we did not otherwise independently verify the
information they provided.
[End of section]
Appendix II: Data-Sharing Policies Applicable to Federally Funded
Climate Data:
Agency: DOE;
Climate Change Program: [Empty];
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Check].
Agency: DOE;
Climate Change Program: Atmospheric Radiation Measurement Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Check].
Agency: DOE;
Climate Change Program: Climate Change Prediction Program;
Climate Change Project: Community Climate System Model (CCSM) and
Program for Climate Model Diagnosis and Intercomparison (PCMDI);
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Check].
Agency: DOE;
Climate Change Program: Terrestrial Carbon Processes;
Climate Change Project: AmeriFlux Network;
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Check].
Agency: DOE;
Climate Change Program: Program for Ecosystem Research;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Check].
Agency: DOE;
Climate Change Program: Integrated Assessment Research Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]:[Empty];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: [Empty];
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NASA;
Climate Change Program: Earth Science Research Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NASA;
Climate Change Program: Earth Science Research Program;
Climate Change Project: Research, Education and Applications Solution
Network (REASoN);
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Cryosphere Research Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NASA;
Climate Change Program: Modeling Analysis and Prediction;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Ocean Biology and Biogeochemistry Research;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Atmospheric Composition Modeling and Data
Analysis; Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Tropospheric Chemistry;
Climate Change Project: Megacity Initiative: Local and Global Research
Observations (MILAGRO);
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Terrestrial Ecology Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Terrestrial Hydrology Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Earth Sciences Program;
Climate Change Project: Land-Cover and Land-Use Change Program;
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NASA;
Climate Change Program: Upper Atmosphere Research Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NASA;
Climate Change Program: Radiation Sciences Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]:[Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NASA;
Climate Change Program: Biological Diversity;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: [Empty];
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NSF;
Climate Change Program: Ecosystem Studies;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NSF;
Climate Change Program: Sedimentary Geology and Paleobiology;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Atmospheric Chemistry Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NSF;
Climate Change Program: National Center for Atmospheric Research;
Climate Change Project: Earth Observing Laboratory;
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Paleoclimate Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NSF;
Climate Change Program: Climate and Large Scale Dynamics;
Climate Change Project: Climate Variability and Predictability Program;
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Climate Change Program: Upper Atmospheric Facilities;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NSF;
Climate Change Program: Instrumentation and Facilities Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Hydrologic Science;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NSF;
Climate Change Program: Education and Human Resources;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Oceanographic Instrumentation and Technical
Services;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Chemical Oceanography;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Physical Oceanography;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Marine Geology and Geophysics;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Biological Oceanography;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Global Ocean Ecosystem Dynamics (GLOBEC);
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Antarctic Glaciology Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Antarctic Organisms and Ecosystems;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Antarctic Earth Sciences;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NSF;
Climate Change Program: Arctic System Science Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Check];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NOAA;
Climate Change Program: Arctic Research Program;
Climate Change Project: Agency: [Empty];
Agencywide data-sharing policy[A]:[Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Atmospheric Composition and Climate;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Transition of Research Applications to Climate
Services;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Climate Change Data and Detection;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Climate Dynamics and Experimental Prediction;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Climate Observation Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Check].
Agency: NOAA;
Climate Change Program: Climate Variability and Predictability;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Climate Prediction Program for the Americas;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Climate Change Program: Global Carbon Cycle Program;
Climate Change Project: [Empty];
gencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Regional Integrated Sciences and Assessments
(RISA);
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Sectoral Applications Research Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Agency: NOAA;
Climate Change Program: Research Cooperative Institute Program;
Climate Change Project: [Empty];
Agencywide data-sharing policy[A]: [Empty];
Program or project-specific data-sharing policy[B]: [Empty].
Source: GAO analysis of survey responses.
Note: The CCSP data-sharing policy, Data Management for Global Change
Research Policy Statements, applies to each federal agency.
[A] Circle denotes agency-level policies encouraging federally funded
researchers to make data available. These policies apply to all of the
programs within the agency.
[B] Circle denotes program-or project-level policies encouraging
federally funded researchers to make data available.
[End of table]
[End of section]
Appendix III: Examples of Data-Sharing Expectations for Different
Federally Funded Research Projects:
Applicable to projects involving:[A]: Individual researcher;
Policy: Grant Policy Manual (NSF)[B];
Expectations for Openness: Researchers are expected to share data,
samples, physical collections and other supporting materials created or
gathered in the course of work under NSF grants. Data should be
released in a form that protects privacy and intellectual property;
exceptions to the openness expectation may be specified by NSF or
requested by grantees;
Expectations for Accessibility: Grantees are expected to make their
data and products widely available and usable;
Expectations for Timing: Researchers are expected to share their data
within a reasonable time;
Expectations for Cost: Researchers are expected to make their data
available at no more than an incremental cost.
Applicable to projects involving:[A]: Multiple researchers;
Policy: AmeriFlux Data Submission Guidelines (DOE)[C];
Expectations for Openness: Researchers should make available the core
suite of data;
Expectations for Accessibility: Researchers should make data available
through the central AmeriFlux data repository located at the Carbon
Dioxide Information Analysis Center;
Expectations for Timing: Researchers should make data available within
1 year of data collection. Ancillary biological data should be
submitted within a reasonable time;
Expectations for Cost: Data are provided free through the central
AmeriFlux data repository.
Applicable to projects involving:[A]: Collection of physical samples;
Policy: Division of Ocean Sciences Data and Sample Policy (NSF)[D];
Expectations for Openness: Researchers should make all environmental
data collected available. Researchers should address alternative
strategies for complying with the openness expectation when limitations
exist;
Expectations for Accessibility: Researchers should submit data to
designated National Data Centers and samples should be archived at NSF-
supported repositories and stored in a manner that preserves the
quality of the samples and respects community standards. Where no
archive exists for the data, researchers should address alternative
strategies to make data available;
Expectations for Timing: Researchers should make all data available as
soon as possible, but no later than 2 years after collection; metadata
of all marine data collected should be made available within 60 days of
the observational period/cruise; for continuing observations, data
inventories should be submitted periodically if there is a significant
change in such observations;
Expectations for Cost: Data are provided at no more than incremental
costs through data archives.
Applicable to projects involving:[A]: Collection of satellite data;
Policy: Data and Information Policy (NASA)[E];
Expectations for Openness: All Earth science data obtained from NASA
Earth observing satellites, suborbital platforms and field campaigns,
as well as source code for algorithm software, coefficients, and
ancillary data, should be made available;
Expectations for Accessibility: Data will be placed in archives that
include accessible information about the data holdings, including
quality assessments, supporting relevant information, and guidance for
locating and obtaining data;
Expectations for Timing: Data will be made available following the post-
launch checkout period. There is no period of exclusive access to NASA
Earth science data;
Expectations for Cost: NASA will charge for distribution of data no
more than the cost of dissemination. In cases where such dissemination
cost would unduly inhibit use, the distribution charge will generally
be below that cost.
Applicable to projects involving:[A]: Modeling data;
Policy: Community Climate System Model (DOE/NSF)[F];
Expectations for Openness: The CCSM project is committed to making
available the results from the model runs and the scientific data
generated in research activities. Short models runs made to examine
specific model behavior, validate a port of the code to a new platform,
or verify model functionality do not need to be made available;
Expectations for Accessibility: CCSM model data should be retained in
public archives for a period of 10 years from the date of the end of
the simulation;
Expectations for Timing: Researchers are entitled to a proprietary
period during which they can publish results but are encouraged to
share their data prior to the release deadlines; data shall become
public when a paper has been submitted or 1 year after the end of the
simulation, whichever comes sooner;
Expectations for Cost: Data will be made available for users who are
not CCSM collaborators at the marginal cost of making and shipping the
copies; however, for large data orders, special arrangements may be
made.
Source: GAO analysis.
[A] These policies are generally not exclusively relevant to the
particular types of research projects noted here. For example, a policy
labeled as applying to "individual researcher" in this column could
also be applicable to projects with multiple researchers.
[B] See section 734 of NSF's Grant Policy Manual, available at
[hyperlink, http://www.nsf.gov/pubs/manuals/gpm05_131/gpm05_131.pdf].
[C] See the Oak Ridge National Laboratory's AmeriFlux Network Data
Guidelines, available at [hyperlink,
http://www.public.ornl.gov/ameriflux/data-guidelines.shtml]. AmeriFlux
is a project that helps coordinates regional and global analysis of
observations from micrometeorological tower sites.
[D] See the General Data Policy and Sample Policy sections of the NSF's
Division of Ocean Sciences Data and Sample Policy, available at
[hyperlink, http://www.nsf.gov/pubs/2004/nsf04004/start.htm].
[E] See NASA's Data and Information Policy (part of their Earth Science
Reference Handbook), available at [hyperlink,
http://science.hq.nasa.gov/research/daac/datainfopolicy.pdf].
[F] See CCSM's Data Management Plan, available at [hyperlink,
http://www.ccsm.ucar.edu/experiments/data.mgmt.plan.050803.html].
[End of table]
[End of section]
Appendix IV: Governmentwide Climate Change Data-Sharing Policy:
This appendix presents an annotated copy of an interagency policy, the
Data Management for Global Change Research Policy Statements. This
policy was written by the U.S. Global Change Research Program Data and
Information Working Group and was later endorsed by the U.S. Climate
Change Science Program. The italicized text provides comments on key
issues in the policy, such as how and what data to share. The policy
also includes a suggested data product requirement and compliance
guidelines, about which we do not comment.
U.S. Global Change Research Program Data and Information Working Group
Data Management for Global Change Research Policy Statements:
Background:
In 1991 the Executive Office of the President, Office of Science and
Technology Policy issued the following data management for global
change policy statements. The overall purpose of these policy
statements was to facilitate full and open access to quality data for
global change research.
GAO Comment:
The policy statements reflect the scientific community's belief that
data-sharing policies should address what, how, and when data are to be
shared, as well as the cost of making data available.
Though the policy statements are applicable to all federally funded
climate change research, they are not legal requirements, as the
statements discuss under the Compliance section below. Senior officials
at the four major climate change research agencies we reviewed told us
that their data-sharing policies and practices follow the principles of
these statements.
They were prepared in consonance with the goal of the U.S. Global
Change Research Program and represent the U.S. Government's position on
the access to global change research data.
Applicability:
1. Federally funded data significantly related to the USGRP that
includes:
A. Data resulting from observations, the application of algorithms to
data to produce new data, and from the data output of models.
B. Data resulting from agency funding in whole or in part of inhouse
activities or of cooperative, grant, and contracted activities.
Included is the data an agency purchases of data from outside the
government to meet its needs*.
GAO Comment:
This section addresses what data to share by defining the information
and materials expected to be shared.
(* Such an inclusion of purchased data is included in the 2001 NAS
report "Resolving Conflicts Arising from the Privatization of
Environmental Data.")
2. While it is hoped that these guidelines would be as broadly applied
as possible, their intent is primarily focused on providing guidance
for when new data is being obtained and made available or when existing
data because of technology or other changes needs to be reformatted or
have other such changes.
Guidelines and Their Application:
Policy Statement 1. The U.S. Global Change Research Program requires an
early and continuing commitment to the establishment, maintenance,
validation, description, accessibility, and distribution of high-
quality, long-term data sets.
GAO Comment:
This policy statement addresses how and what data to share. The
statements follow the scientific community's belief that data should be
made available via unrestricted archives.
Since 1994 the USGCRP has managed a Web page, the Global Change Data
and Information System, GCDIS, that helps users find the largest amount
of USGCRP related data of any site in the world. In 1999, it also
became the largest site for data policy information. This site is at
[hyperlink, http://www.globalchange.gov/].
A. Applicable agency data should be made readily accessible to
potential users:
* Minimum application - All such data used in openly available
publications, reports, and analyses.
GAO Comment:
The statements follow the scientific community's belief that, at a
minimum, all data necessary to support researchers' major published
results should be made available. It may not always be possible or
appropriate to share all data, such as in modeling research where only
certain data outputs are relevant to the larger scientific community.
Desired application - All such significant data produced.
B. Applicable agency data should be made available via the Web:
* Minimum application - All such data used in openly available
publications, reports, and analyses that are in digital form.
* Desired application - All such significant data that's openly
available.
C. Applicable data made available on the Web should be described with
each data set having:
* Minimum application - A citation similar to those used for citing
publications in research journals and in use for data sets by the
USGCRP since 1997.
* Desired application - A citation plus a data set description that (1)
can be readily found and is adequate for users to be able to both
understand the applicability of the data to their needs and its proper
use and (2) meets at least the minimum requirements for inclusion in
the Global Change Master Directory, GCMD, and is so identified to the
GCMD.
Policy Statement 2. Full and open sharing of the full suite of global
data sets for all global change researchers is a fundamental objective.
GAO Comment:
This policy statement addresses what data to share. The statements
follow the scientific community's belief that data sharing should be
full and open.
This objective has since 1991 been repeatedly urged and defended from
compromise by the National Academy of Science, NAS. The concept has
also been widely adopted and applied both nationally and
internationally. After reviewing all these implementation actions, the
NAS recommended the following single definition "Full and open
availability is defined as being available without restriction, on a
non-discriminatory basis, for no more than the cost of reproduction and
distribution." It combines elements of this Statement with those of
Statement 6 was adopted by the USGCRP in 1997.
A. Full and open access to agency data sets should be provided to:
* Minimum application - All agency data related to the USGCRP that's
made generally available.
GAO Comment:
* As noted above, some data are not appropriate for sharing. Full and
open, in these policy statements, does not necessarily mean that every
data item or iteration of data analysis must be made available. What
data to share are generally more specifically defined by relevant
agency, program, or project policies.
* Desired application - All agency data that's made generally
available.
Policy Statement 3. Preservation of all data needed for long-term
global change research is required. For each and every global change
data parameter, there should be at least one explicitly designated
archive. Procedures and criteria for setting priorities for data
acquisition, retention, and purging should be developed by
participating agencies, both nationally and internationally. A
clearinghouse process should be established to prevent the purging and
loss of important data sets.
GAO Comment:
This policy statement addresses how to share data. The statements
follow the scientific community's general belief that, when an
appropriate archive exists, data should be shared via unrestricted
archives. The lack of appropriate archival infrastructure can be an
obstacle to data sharing.
The Federal requirement for providing adequate notice when agencies
purge significant data and information products is called for in OMB
Circular A-130 of 1997.
A. The USGCRP should be notified of any agency plans to purge data
significantly related to the USGCRP program so an interagency process
can determine the necessary remedial actions, if any.
* Minimum application - Notification at least six months prior to the
data being purged, or as soon as the agency's intent seems likely,
whichever is shorter.
* Desired application - Notification as soon as the data purging is
being seriously considered by an agency.
(It should be noted that these guidelines apply equally well in normal
times and in abnormal times, such as after the 9/11/01 event.)
Policy Statement 4. Data archives must include easily accessible
information about the data holdings, including quality assessments,
supporting ancillary information, and guidance and aids for locating
and obtaining the data.
GAO Comment:
This policy statement addresses how and what data to share. The
statements reflect the belief that data should be shared with their
metadata, i.e., information needed to understand the content, quality,
and condition of the raw data.
A. For the applicable data that agencies make available, an assessment
of its quality is needed to help assure its proper use.
* Minimum application - Identification of the source of the data so the
user has a place to check on its quality.
* Desired application - Identification of the data's quality sufficient
to assure its proper use and make unlikely its improper use.
(The requirement for the identifying the quality of data made available
is contained in OMB's "Guidelines for Ensuring and Maximizing the
Quality, Objectivity, Utility, and Integrity of Information Provided by
Federal Agencies" issued in 2001.)
B. For the applicable data that agencies make available there should be
the ability to be responsive to users questions relative to its use.
* Minimum application - A means for the user to identify the source of
the data, i.e. the specific person or organization responsible.
* Desired application - Identification of a person or organization that
will be responsive to a user's requests for help.
C. To maximize the ability of users to use the applicable data made
available, the vision is to have data from different sources be able to
be seamlessly used with data taken by other means, from different
sources, and measuring other parameters. That is, have full
interoperability.
* Minimum application - Enough data is provided with a data set so its
user can make it interoperable with other data sets.
* Desired application - Meets the preceding "Minimum application" and
the data set has at least spatial and temporal interoperability with
the other such interoperable data within the USGCRP.
Policy Statement 5. National and international standards should be used
to the greatest extent possible for media and for processing and
communication of global data sets.
GAO Comment:
This policy statement addresses how to share data.
A. In 1994 Executive order 12906 created the National Spatial
Infrastructure, NSDI, and OMB Circular A-16 its Federal Geographic Data
Committee, FGDC, management structure. For all geospatial data,
agencies are must have compatibility with their data documentation
standards. The FGDC actively tries to assure their standards are
compatible with international standards.
* Minimum Application - All applicable data when new data is being
obtained and made available or when existing data because of technology
or other changes needs to be reformatted or have other such changes.
* Desired Application - All applicable data.
B. In 1995 the parent group of the USGCRP, OSTP's Committee on
Environment and Natural Resources, instructed its participating
agencies to have their individual data and information access and
search systems be in compliance with the American National Standards
Institute, ANSI, Z39.50 10162/10163 open standards for information
search and retrieval.
* Minimum Application - All applicable data when new data is being
obtained and made available or when existing data because of technology
or other changes needs to be reformatted or have other such changes.
* Desired Application - All applicable data.
Policy Statement 6. Data should be provided at the lowest possible cost
to global change researchers in the interest of full and open access to
data. This cost should, as a first principle, be no more than the
marginal cost of filling a specific user request. Agencies should act
to streamline administrative arrangements for exchanging data among
researchers.
GAO Comment:
This policy statement addresses the cost of sharing data. The
statements follow the scientific community's belief that data should be
made available at no more than the marginal cost of reproduction and
distribution.
The Federal requirement for charging users no more than the marginal
cost of servicing their request is called for in OMB Circular A-130 of
1997.
* Minimum Application - All applicable data.
Policy Statement 7. For those programs in which selected principal
investigators have initial periods of exclusive data use, data should
be made openly available as soon as they become widely useful. In each
case the funding agency should explicitly define the duration of any
exclusive use period.
GAO comment:
This policy statement addresses when data should be shared. The
statements follow the scientific community's belief that data should
generally be made available immediately or after a limited proprietary
period that allows researchers to complete their initial analysis and
publish their results. The scientific community generally recognizes
the need for researchers to have a limited period of exclusive access
to their data to allow for analysis, interpretation, and peer review
that normally precedes public disclosure. However, the length of such a
period may be determined by the type of research.
To meet this need, in 1997 the USGCRP endorsed the following grant
language for use by its participating agencies.
Suggested Data Product Requirement for Grants, Cooperative Agreements,
and Contracts:
Describe the plan to make available the data products produced, whether
from observations or analyses, which contribute significantly to the
results. The data products will be made available to the
without restriction and be
accompanied by comprehensive metadata documentation adequate for
specialists and non-specialists alike to be able to not only understand
both how and where the data products were obtained but adequate for
them to be used with confidence for generations. The data products and
their metadata will be provided in a exchange format no
later than the final report or the publication of the data
product's associated results, whichever comes first.
* Minimum Application - All such applicable data identified as
important to the USGCRP.
* Desired Application - All such applicable data.
Compliance:
While these guidelines themselves are not requirements on the agencies,
many result from Federal requirements that do require agency
compliance. Rather the guidelines' goal is to help provide guidance to
the agencies on how best to meet the needs of users for USGCRP related
data within their resources. As such, to help users of a particular
data set made available by an agency readily understand the degree to
which it meets the guidelines, as well as to recognize the efforts an
agency to meet these guidelines:
1. Provided a data set meets all of the Federal requirements and at
least all the minimum levels of guideline application the agency should
add a single asterisk at the end of the data set's citation.
2. Provided a data set meets all of the Federal requirements and all
the desired levels of guideline application - the agency should add two
asterisks at the end of the data set's citation.
For broader compliance than for selected individual data sets:
* Minimum compliance - Endorsement of these guidelines at the highest
appropriate level in the agency.
* Desired compliance - Incorporation of these guidelines into the data
management policies of the highest appropriate level in the agency.
[End of section]
Appendix V: Comments from the Department of Energy:
Department of Energy:
Washington, DC 20585:
September 7, 2007:
John Stephenson:
Director, Natural Resources and Environment:
Government Accountability Office:
441 G St., NW, Room 2T23A:
Washington, DC 20548:
Dear Mr. Stephenson:
Thank you for the opportunity to comment on the draft Government
Accountability Report, entitled "Climate Change Research: Agencies Have
Data-Sharing Policies but Could Do More to Enhance the Availability of
Data from Federally Funded Research." We appreciate your efforts to
assist in assuring that the data and information on Climate Change is
available to the public. The Department and the partner agencies in the
Climate Change Science Program consider this a priority objective.
Overall, we agree with your principle finding and recommendation that
the agencies consider additional processes to enhance data
availability.
Below, please find additional general and specific comments for your
consideration:
General Comment:
The report should clearly define what is meant by "data." There are
several brief references to both electronic-type data and actual field
samples. For example, on page 28 the discussion implies that both are
included when referring to data throughout the report. In general,
researchers do not think of samples collected in (or from) experiments
as data but as project resources whereas the data are what results from
the analysis of the samples. A clear definition of "data" is needed up
front.
Sharing of electronic data, while complicated and potentially expensive
in its own right, is far less complicated than the sharing of actual
research samples which are also finite in the amount of materials
actually available. If the report intends to include samples as well,
then a discussion is needed of how much of a finite sample should be
available and how much should a researcher be able to keep for their
own future use. The amount will depend on what future analysis might be
done on the samples which will be virtually impossible to predict. Such
research samples should not be considered the same as data.
Specific Comments:
Page 3 reference to ice cores is unclear – Does GAO mean to say
Antarctic instead of Arctic?
Page 6, first full paragraph, "...the scientific community has
traditionally only rewarded publication." Comment/clarification: It is
not just the scientific community that rewards publication, but also
institutions, agencies and performance reviews that place a higher
value on publication of original results, and give lesser credit to
data management functions. The reward matter was discussed in broader
terms in the body of the report, but this condensed version in the "In
Brief' section seems to be an over simplification.
Highlights page, page 6, page 25, and page 26) the report says
something such as "agencies do not monitor whether researchers make
data available." This is an over simplification. Several programs, such
as the Department of Energy ARM and AmeriFlux, do monitor if data is
placed in a public archive, and also monitor if the data is accessed by
individuals outside the program. DOE recommends that such statements be
revised along the lines of "agencies do not routinely monitor whether
researchers make data available from all their research programs."
On page 8, the report says that "...agencies request written reviews
...to assess the scientific merit of proposals in some cases". For the
DOE climate research, all proposals have written reviews for scientific
merit. The last phrase might therefore be changed from "in some cases"
to "in most or all (depending on agency) cases."
Page 24, paragraph in middle of the page: The last sentence is
incomplete; it should read, "...Carbon Dioxide Information Analysis
Center, which preserves data from researchers collaborating in the
AmeriFlux Network and other Climate Change Programs." COMMENT: Only a
fraction of the $2.7 million supports the AmeriFlux data base, and the
total budget supports other data management activities. Note there is
no "and" in "Oak Ridge National Laboratory's Carbon Dioxide Information
Analysis Center" (on page 35, as well).
In several places the report states that data archiving and sharing is
not part of the reward structure for scientists. While it is perhaps
not "generally" a "significant" part of the reward system, there are
cases where important peer recognition goes along with the
dissemination of long-term data records. The statement "...agencies and
the scientific community expect researchers to both publish their
results and make underlying data available but researchers have
traditionally ONLY been rewarded for publication" (page 29) could more
accurately be stated along the lines of "...agencies and the scientific
community expect researchers to both publish their results and make
underlying data available but researchers have traditionally been
rewarded mainly for publications." Similarly, the Highlights page (near
the bottom) could say "...but preparation of data for others' use is
not A significant part of this reward structure."
We look forward to working with our partner agencies as we consider
additional opportunities and processes to make climate change research
data available to the research community.
If you have any questions, please contact me at 301-903-3281.
Sincerely:
Signed by:
Jerry W. Elwood, PhD.
Acting Associate Director of Science
for Biological and Environmental Research:
[End of section]
Appendix VI: Comments from the National Aeronautics and Space
Administration:
National Aeronautics and Space Administration:
Office of the Administrator:
Washington, DC 20546-0001:
September 17, 2007:
Mr. John Stephenson:
Director, Natural Resources and Environment:
United States Government Accountability Office:
Washington, DC 20548:
Dear Mr. Stephenson:
NASA appreciates the opportunity to comment on your draft report
entitled, "Climate Change Research: Agencies Have Data-Sharing Policies
but Could Do More to Enhance the Availability of Data from Federally
Funded Research" (GAO-07-1172).
In the draft report, GAO makes four recommendations regarding the
availability of data from Federally funded climate change research.
However, of the four recommendations made, only three (recommendations
2–4) are addressed to the NASA Administrator:
Recommendation 2: To ensure that the agencies maximize opportunities to
make data available in a manner useful to other researchers, we
recommend that the Secretary of Energy, the Administrator of NASA, the
Secretary of Commerce and the NOAA Administrator, and the Director of
the National Science Foundation consider the following actions:
* Develop mechanisms for agencies to be systematically notified when
data have been submitted to archives, so that agency officials have
current information about the extent of data availability in order to
adjust data-sharing policies over time to best meet the needs of
researchers and the communities that use their data.
Response: NASA concurs with this recommendation and already has such
mechanisms in place for its archives. NASA officials (and the public at
large) have current information about the extent of data availability
and the ability to adjust data-sharing policies over time to best meet
the needs of researchers and the user communities that use their data.
Recommendation 3: Use the grants process, where their program offices
are not currently doing so, to facilitate further data sharing by: (1)
evaluating researchers' data-sharing plans as part of the grant review
process, and (2) using evidence of researchers' past data sharing
practices to make future award decisions. The use of such criteria in
the grant review process should be clearly conveyed to researchers
before they submit research proposals and after award decisions have
been made.
Response: NASA concurs with the intent of this recommendation and
believes NASA policies facilitate data sharing. NASA employs "full and
open exchange" data policy for its satellite data and standard products
holdings.
For grantees, NASA's guidelines with regard to releasing data and
results derived through its research awards can be found in the ROSES
Guidebook for Proposers. These guidelines state that NASA may, where
appropriate, require that any data obtained through an award be
deposited in an appropriate public data archive as soon as possible
after calibration and validation.
Recommendation 4: To ensure that researchers make climate change data
available to other researchers, we recommend that the Secretary of
Energy, the Administrator of NASA, the Secretary of Commerce and the
NOAA Administrator, and the Director of the National Science
Foundation:
* Evaluate whether additional strategies are warranted to facilitate
the permanent archiving of relevant data, which may include leveraging
existing resources, devoting a greater portion of data collection funds
to archiving activities, or working with existing entities such as the
National Science and Technology Council's Interagency Working Group on
Digital Data, to develop additional data archives.
Response: NASA concurs with this recommendation. NASA's Earth Science
Program systematically evaluates the demand for data products by the
science community and users. NASA analyzes individual data collections
and develops the best methodologies for archival and distribution for
these collections. NASA then focuses on developing the most cost-
effective system for data archival and distribution of service to the
science community and the Nation. NASA's current capacity for data
archive and distribution is sufficient for all relevant data for the
foreseeable future.
Again, thank you for the opportunity to review and comment on this
draft report and for the critical insight it provides. If you have any
questions, please contact Michael Luther on (202) 358-7226.
Sincerely,
Signed by:
Shana Dale:
Deputy Administrator:
[End of section]
Appendix VII: Comments from the Department of Commerce, for the
National Oceanic and Atmospheric Administration:
The Secretary Of Commerce:
Washington, D.C. 20230:
September 10, 2007:
Mr. John Stephenson:
Director, Natural Resources and Environment:
U.S. Government Accountability Office:
441 G Street, NW:
Washington, D.C. 20548:
Dear Mr. Stephenson:
Thank you for the opportunity to review and comment on the Government
Accountability Office's draft report entitled Climate Change Research:
Agencies Have Data-Sharing Policies but Could Do More to Enhance the
Availability of Data from Federally Funded Research (GAO-07-1172). On
behalf of the Department of Commerce, I enclose the National Oceanic
and Atmospheric Administration's programmatic comments to the draft
report.
Sincerely,
Signed by:
Carlos M. Gutierrez:
Enclosure
[End of letter]
Enclosure:
Department of Commerce:
National Oceanic and Atmospheric Administration:
Comments on the Draft GAO Report Entitled "Climate Change Research:
Agencies Have Data-Sharing Policies but Could Do More to
Enhance the Availability of Data from Federally Funded Research"
(GAO-07-1172/September 2007):
General Comments:
The Department of Commerce appreciates the opportunity to review this
report. The report does a fair job in assessing the National Oceanic
and Atmospheric Administration's (NOAA's) data-sharing policies for
climate change research. The discussion of practices to encourage data-
sharing, and the associated obstacles and disincentives, is accurate
and balanced.
NOAA Response to GAO Recommendations:
The draft GAO report states, "To assist federal agencies sponsoring
climate change research in better ensuring the availability of data
from federally funded research, we are making the following four
recommendations."
Recommendation 1: "To ensure that researchers receiving federal funds
to conduct climate change research understand NOAA 's expectations for
data sharing, we are recommending that the Secretary of Commerce and
the NOAA Administrator: Develop a set of written guidelines or use
existing government-wide guidelines, such as those endorsed by the
Climate Change Science Program, to clearly inform researchers of NOAA
's general expectations for data sharing."
NOAA Response: NOAA agrees with this recommendation. There are
extensive existing guidelines available for use by NOAA to mirror or
enhance. Therefore, through the Climate Change Science Program's
Science Advisory Board, NOAA plans to address this recommendation and
take the necessary steps (i.e., develop new guidelines or enhance
existing guidelines) to more clearly inform researchers of NOAA's data
sharing expectations.
Recommendation 2: "To ensure that the agencies maximize opportunities
to make data available in a manner useful to other researchers, we
recommend that the Secretary of Energy, the Administrator of NASA, the
Secretary of Commerce and the NOAA Administrator, and the Director of
the National Science Foundation consider the following actions: Develop
mechanisms for agencies to be systematically notified when data have
been submitted to archives, so that agency officials have current
information about the extent of data availability in order to adjust
data-sharing policies over time to best meet the needs of researchers
and the communities that use their data."
NOAA Response: NOAA agrees with this recommendation and believes agency
officials should have current information about the extent of data
availability. Given other agencies are also involved, NOAA will do its
part and consider developing a mechanism to systematically notify
others when data have been archived. As part of this process, however,
several things will need to be considered, including: (1) funding
implications; (2) the need for standard reporting categories to
facilitate data collection (rather than the use of the numerous agency-
unique reporting forms and systems); and (3) the benefits of a
mechanism which would enhance the ability of both the public and
agencies to organize the data.
Recommendation 3: "To ensure that the agencies maximize opportunities
to make data available in a manner useful to other researchers, we
recommend that the Secretary of Energy, the Administrator of NASA, the
Secretary of Commerce and the NOAA Administrator, and the Director of
the National Science Foundation consider the following actions: Use the
grants process, where their program offices are not currently doing so,
to facilitate further data sharing by. (1) evaluating researchers' data-
sharing plans as part of the grant review process, and (2) using
evidence of researchers' past data-sharing practices to make future
award decisions. The use of such criteria in the grant review process
should be clearly conveyed to researchers before they submit research
proposals and after award decisions have been made."
NOAA Response: NOAA agrees with this recommendation and believes taking
the recommended actions will emphasize to researchers applying for
grants NOAA's commitment to data-sharing. NOAA plans to include past,
present, and future data sharing as an evaluation factor in its grant
opportunity announcements. NOAA will also consider including additional
rating factors in assessing grant proposals, as well as the inclusion
of additional requirements in grant awards for data sharing by the
awardee to encourage good data sharing practices. Moving forward, NOAA
will pursue avenues to more clearly convey the importance of these
factors to researchers prior to the submission of research proposals
and after award decisions have been made.
Recommendation 4: "To ensure that researchers make climate change data
available to other researchers, we recommend that the Secretary of
Energy, the Administrator of NASA, the Secretary of Commerce and the
NOAA Administrator, and the Director of the National Science
Foundation: Evaluate whether additional strategies are warranted to
facilitate the permanent archiving of relevant data, which may include
leveraging existing resources, devoting a greater portion of data
collection funds to archiving activities, or working with existing
entities such as the National Science and Technology Council's
Interagency Working Group on Digital Data, to develop additional data
archives."
NOAA Response: NOAA agrees with this recommendation. Permanent
archiving is very important and should be incorporated into the
strategic plans of archiving facilities. NOAA will consider whether
additional strategies are warranted to facilitate the permanent
archiving of data, as well as the additional resource requirements to
support such strategies.
Recommended Changes for Factual/Technical Information:
Page 12, First Paragraph:
The need for complete metadata should be identified.
We suggest adding at the end of this paragraph, "For example, in all
cases sufficient metadata, such as data set descriptions, should be
provided so the data can be found and its suitability for use
determined. This need recognizes most data will be archived, and other
researchers will need to search the archives to identify useful data."
[End of enclosure]
[End of section]
Appendix VIII: GAO Contact and Staff Acknowledgments:
GAO Contact:
John Stephenson, 202-512-3841 or stephensonj@gao.gov:
Staff Acknowledgments:
In addition to the contact person named above, Diane Raynes (Assistant
Director); Kyle Browning; Kate Cardamone; John Delicath; Carolyn
Garvey; Richard Johnson; Lynn Musser; and Katherine M. Raheb made key
contributions to this report.
(360751):
[End of section]
Footnotes:
[1] The $2 billion estimate is based on recent budget analyses from the
U.S. Climate Change Science Program (CCSP), which compiles budget data
from each agency that receives climate change funds. This estimate does
not reflect all agency activities that support climate change science,
such as operation of satellites and preservation of some kinds of
climate change data. For more information on federal climate change
spending, see GAO, Climate Change: Federal Reports on Climate Change
Funding Should Be Clearer and More Complete, GAO-05-461 (Washington,
D.C.: Aug. 25, 2005).
[2] Based on the Fiscal Year 2006 estimated budget, the most recent
year for which information about budget allocations were available.
[3] GAO is using the term data in this report to encompass both data
and metadata. Metadata refers to information needed to understand the
content, quality, and condition of the data, such as instrument
calibration.
[4] National Research Council. Bits of Power: Issues in Global Access
to Scientific Data. (Washington, D.C.: National Academy Press, 1997),
p. 10. For additional National Academies' reports on data sharing, see
Sharing Research Data (1985); A Question of Balance (1999); Resolving
Conflicts Arising from the Privatization of Environmental Data (2001);
Access to Research Data in the 21st Century (2002); The Role of
Scientific and Technical Data and Information in the Public Domain
(2003); Sharing Publication-Related Data and Materials (2003); and
Expanding Access to Research Data (2005).
[5] These figures are based on the estimated total U.S. Climate Change
Science Program budget--which includes both scientific research and
NASA space-based observations--for fiscal year 2006. Estimates for
spending in fiscal year 2007 were not available as of August 2007. See
Climate Change Science Program and the Subcommittee on Global Change
Research, Our Changing Planet: The U.S. Climate Change Science Program
for fiscal year 2007 (Washington, D.C., 2007).
[6] For purposes of this report, grant refers to cooperative agreements
as well as awards used by DOE to fund external researchers at National
Laboratories.
[7] For the purposes of this report, the scientific community refers to
the general body of scientists and its institutions as represented by
the National Academies and professional scientific associations. While
no single body can be said to speak for all of science, the National
Academies and other scientific associations, such as those listed in
appendix I, often act as surrogates when the opinions of the scientific
community, or particular disciplines within science, need to be
ascertained. We also supplemented our analysis of the reports and
statements of these organizations with interviews of officials, at a
variety of entities, with knowledge of data-sharing issues.
Furthermore, whenever we attribute statements to the scientific
community at large, we mean that the National Academies and at least
two of the scientific associations listed above support those
statements.
[8] OMB A-130 is a governmentwide policy calling for "the open and
efficient exchange of scientific and technical information." See OMB A-
130, Management of Federal Information Resources 7(k) (Feb. 8, 1996).
OMB A-130 focuses on the information activities of all agencies of the
executive branch of the federal government, but does not clearly
specify responsibility of federally funded researchers or that of the
government to foster data sharing under grants.
[9] As stated earlier, we define data to include factual information or
physical samples that are collected and recorded as a result of
scientific observation, experiment, analysis, or similar methods of
research. The output of models can also be considered data.
[10] U.S. Climate Change Science Program and the Subcommittee on Global
Research, Strategic Plan for the U.S. Climate Change Science Program
(Washington, D.C., 2003).
[11] For purposes of this report, policy refers to written, nonbinding
agency directives and guidance intended to inform agency managers,
staff, and researchers.
[12] Data-sharing policies were identified for 40 of the 51 different
programs. The number of discrete policies differs from the number of
programs because some policies applied to more than one program, and
some programs had more than one policy.
[13] National Science Foundation, NSF Grant Policy Manual, (Arlington,
VA, 2005).
[14] National Oceanic and Atmospheric Administration, Climate
Observation Program, Data Policy, [hyperlink,
http://www.oco.noaa.gov/].
[15] While multiple agencies fund the research and analysis conducted
within the AmeriFlux Network, an official at the AmeriFlux archive
reported that DOE provides nearly all of the funding to manage the data
at the archive, which is based at the Carbon Dioxide Information
Analysis Center at Oak Ridge National Laboratory.
[16] National Science Foundation, NSF Grant Proposal Guide, (Arlington,
VA, 2004).
[17] Of the 55 program managers who indicated that their program takes
steps to ensure researchers make data available, 49 said they do so by
maintaining personal contact.
[18] Thirty-seven of the 55 program managers indicating that their
program takes steps to ensure researchers make data available said they
would do so by authorizing additional funds.
[19] NSF officials reported that they have provided some financial
support for data archives but that NSF does not fund archives for
climate change data on a permanent basis.
[20] National Research Council, Preserving Scientific Data on Our
Physical Universe: A New Strategy for Archiving the Nation's Scientific
Information Resources, (Washington, D.C.: National Academies Press,
1995), p. 3.
[21] National Research Council, Bits of Power (1997), p. 61.
[End of section]
GAO's Mission:
The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting
its constitutional responsibilities and to help improve the performance
and accountability of the federal government for the American people.
GAO examines the use of public funds; evaluates federal programs and
policies; and provides analyses, recommendations, and other assistance
to help Congress make informed oversight, policy, and funding
decisions. GAO's commitment to good government is reflected in its core
values of accountability, integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each
weekday, GAO posts newly released reports, testimony, and
correspondence on its Web site. To have GAO e-mail you a list of newly
posted products every afternoon, go to [hyperlink, http://www.gao.gov]
and select "Subscribe to Updates."
Order by Mail or Phone:
The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or
more copies mailed to a single address are discounted 25 percent.
Orders should be sent to:
U.S. Government Accountability Office:
441 G Street NW, Room LM:
Washington, D.C. 20548:
To order by Phone:
Voice: (202) 512-6000:
TDD: (202) 512-2537:
Fax: (202) 512-6061:
To Report Fraud, Waste, and Abuse in Federal Programs:
Contact:
Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]:
E-mail: fraudnet@gao.gov:
Automated answering system: (800) 424-5454 or (202) 512-7470:
Congressional Relations:
Gloria Jarmon, Managing Director, JarmonG@gao.gov:
(202) 512-4400:
U.S. Government Accountability Office:
441 G Street NW, Room 7125:
Washington, D.C. 20548:
Public Affairs:
Susan Becker, Acting Manager, BeckerS@gao.gov:
(202) 512-4800:
U.S. Government Accountability Office:
441 G Street NW, Room 7149:
Washington, D.C. 20548: