SECTION 3 NEW WAYS OF OBSERVING THE OCEAN
184
of the most important problems it faced. WOCE
set up a Data Management Committee to oversee
aspects of the data system. This committee provided the needed functionality for WOCE data
management and initially focused only on data
management issues. Seeds of new ideas such as
developing higher-level products and complete
information archives were planted early, and as the
field phase progressed it became apparent that
the committee must nurture these forward-looking
ideas to support successfully a synthesis of WOCE
results. Thus the Data Management Committee
was transformed into the WOCE Data Products
Committee (DPC). This involved increasing direct
participation by WOCE scientists while maintaining, ex officio, the full complement of WOCE data
management leaders in its ranks. Its primary functions were as follows:
¥ To provide oversight and advice for the incorporation of WOCE data into a coherent representation of the global circulation and its
variability, the resulting transports of heat and
fresh water, and the distribution of physical and
chemical properties.
¥ To interact with other WOCE working groups
in order to advise the WOCE Scientific Steering
Group (SSG) on the data products that will be
required for community-wide WOCE analysis
and synthesis activities.
¥ To be responsible to the SSG for the timely and
appropriate production and delivery of WOCE
data and data products to the community of
WOCE researchers.
¥ To advise the SSG on the effectiveness of all elements of the WOCE Data System (DACs, SACs,
Archive and DIU) by continuing oversight of the
System’s operation, and to recommend changes
to the system, where appropriate, and to take
advantage of technological developments.
The DPC has served as the catalyst for change
and coordination among the DACs and SACs.
Openly sharing data sources, as well as qualitycontrol and data management technologies and
experiences is a common tenet of DPC members.
The advent of network and CD-ROM based
information technologies encouraged creation of
archive-ready data products. Invited science presentations at annual DPC meetings led to improved
products, enhanced collaboration with WOCE
researchers, and served as a compass for evolutional development of the integrated WOCE Data
resource.
WOCE data-sharing policy
WOCE has developed a data-sharing policy that
strikes a balance between the requirements of the
experiment as a whole and the intellectual rights of
the individuals who contribute to the programme.
The trade-off between these different interests has
resulted in a policy that expects data to be made
publicly available within 2 years of the data set
being complete. Investigators are thus allowed a
reasonable period to complete their own analysis
and publications, while ensuring the global data set
is accumulated and made accessible for wider
examination. Data may be assembled at centres
soon after collection but for some data types will
not be released by DACs without the consent of the
originator. WOCE encourages investigators to
share data with others and to authorize general
release as soon as possible. If investigators demand
exclusive rights to data more than 2 years old, then
WOCE encourages peer pressure and funding
agencies to persuade the originator to give consent
for its general availability.
Evolution of the WOCE Data System
A number of changes and adaptations had to be
made to the data system as it developed during the
field phase of WOCE. The difficulty in assembling
data in a timely manner was not fully appreciated
during the planning of WOCE. The more time that
passes between the time of data collection and the
assembly of the data and metadata at a DAC, the
more likely it is that problems or questions about
the data will remain undocumented or unresolved.
Because investigators delayed submission of data
to assembly centres, the WOCE data-sharing policy received much attention. The data-sharing policy was refined to recognize that the need for data
assembly in a timely manner is an issue independent of the investigators’ consent to make data
publicly available.
As originally envisioned, there would have been
many SACs in WOCE – perhaps as many as one for
each data type. However, the concept of centralized
analysis centres, under data management oversight,
charged with developing and circulating data
products and analyses was not popular. Instead it
became apparent that production of any special
184
of the most important problems it faced. WOCE
set up a Data Management Committee to oversee
aspects of the data system. This committee provided the needed functionality for WOCE data
management and initially focused only on data
management issues. Seeds of new ideas such as
developing higher-level products and complete
information archives were planted early, and as the
field phase progressed it became apparent that
the committee must nurture these forward-looking
ideas to support successfully a synthesis of WOCE
results. Thus the Data Management Committee
was transformed into the WOCE Data Products
Committee (DPC). This involved increasing direct
participation by WOCE scientists while maintaining, ex officio, the full complement of WOCE data
management leaders in its ranks. Its primary functions were as follows:
¥ To provide oversight and advice for the incorporation of WOCE data into a coherent representation of the global circulation and its
variability, the resulting transports of heat and
fresh water, and the distribution of physical and
chemical properties.
¥ To interact with other WOCE working groups
in order to advise the WOCE Scientific Steering
Group (SSG) on the data products that will be
required for community-wide WOCE analysis
and synthesis activities.
¥ To be responsible to the SSG for the timely and
appropriate production and delivery of WOCE
data and data products to the community of
WOCE researchers.
¥ To advise the SSG on the effectiveness of all elements of the WOCE Data System (DACs, SACs,
Archive and DIU) by continuing oversight of the
System’s operation, and to recommend changes
to the system, where appropriate, and to take
advantage of technological developments.
The DPC has served as the catalyst for change
and coordination among the DACs and SACs.
Openly sharing data sources, as well as qualitycontrol and data management technologies and
experiences is a common tenet of DPC members.
The advent of network and CD-ROM based
information technologies encouraged creation of
archive-ready data products. Invited science presentations at annual DPC meetings led to improved
products, enhanced collaboration with WOCE
researchers, and served as a compass for evolutional development of the integrated WOCE Data
resource.
WOCE data-sharing policy
WOCE has developed a data-sharing policy that
strikes a balance between the requirements of the
experiment as a whole and the intellectual rights of
the individuals who contribute to the programme.
The trade-off between these different interests has
resulted in a policy that expects data to be made
publicly available within 2 years of the data set
being complete. Investigators are thus allowed a
reasonable period to complete their own analysis
and publications, while ensuring the global data set
is accumulated and made accessible for wider
examination. Data may be assembled at centres
soon after collection but for some data types will
not be released by DACs without the consent of the
originator. WOCE encourages investigators to
share data with others and to authorize general
release as soon as possible. If investigators demand
exclusive rights to data more than 2 years old, then
WOCE encourages peer pressure and funding
agencies to persuade the originator to give consent
for its general availability.
Evolution of the WOCE Data System
A number of changes and adaptations had to be
made to the data system as it developed during the
field phase of WOCE. The difficulty in assembling
data in a timely manner was not fully appreciated
during the planning of WOCE. The more time that
passes between the time of data collection and the
assembly of the data and metadata at a DAC, the
more likely it is that problems or questions about
the data will remain undocumented or unresolved.
Because investigators delayed submission of data
to assembly centres, the WOCE data-sharing policy received much attention. The data-sharing policy was refined to recognize that the need for data
assembly in a timely manner is an issue independent of the investigators’ consent to make data
publicly available.
As originally envisioned, there would have been
many SACs in WOCE – perhaps as many as one for
each data type. However, the concept of centralized
analysis centres, under data management oversight,
charged with developing and circulating data
products and analyses was not popular. Instead it
became apparent that production of any special