Biological Investigations of Marine Antarctic Systems and Stocks

Project Description
The BIOMASS Programme was established in the late 1970's for the study
of the Antarctic marine ecosystem and its living resources. Data
were collected during 3 major field experiments between 1981 and
1985. The first focused on extended spatial coverage, whilst the
second and third were concerned with repeat sampling at pre-defined
locations to give a temporal sequence. Data collected from these
field experiments were transferred to a central BIOMASS Data Centre
to enable their standardisation for integrated analysis. The Data
Centre was also responsible for running a series of data analysis
workshops. With the end of the BIOMASS Programme in 1991, the data
set and its supporting documentation are being prepared for
distribution to those scientists who took part in BIOMASS and to any
other investigators who request copies. The BIOMASS Data Centre
faced many problems in standardising, integrating and documenting
the data supplied by individual researchers into a coherent data!
set. The majority of these problems were managerial rather than
technical. There was a lack of integration of the data management
with the objectives of the science programme. For example, the need
for a BIOMASS Data Centre was identified in 1979, but it was not
finally established until 1986. Once established, the Data Centre
did not always respond to the scientific requirements of the
programme. There was an over reliance on software that was developed
within the Data Centre instead of using commercially available
products. Time was spent creating and testing software, which would
have been better spent supporting data analysis.

Problems were experienced in persuading individual researchers to
contribute data to the Data Centre. Researchers often found that the
effort involved in submitting their data to the Data Centre was much
greater than the benefits they gained. Ensuring that the data were
validated and of the required quality was also difficult. The task was
hampered by the lack of supporting information about the data
themselves (the meta-data). Restricted access to certain data sets
reduced the effectiveness of the BIOMASS Data Centre and it operated
for much of its life with a very restrictive data access
protocol. This was designed to protect some data sets before their
originators had published their own analyses, but hampered the
distribution of data to the wider BIOMASS community.

The lessons that have been learned from BIOMASS about the management
of complex, large-scale, biological data sets will be of great use to
future programmes. Increasing the quality of data holdings, especially
by the inclusion of meta-data, will increase the chances of
successfully networking databases together to support biodiversity and
other research.