Date of Award

1-1-1998

Degree Type

Thesis

Degree Name

Master of Science

Faculty

Faculty of Communications, Health and Science

First Advisor

Dr Gulten Wagner

Second Advisor

Dr Ron Oliver

Third Advisor

Dr Julie Johnson

Abstract

A group of seventy six scientists and data managers in the Australian research agency CSIRO were surveyed to establish their needs and preferences in relation to information systems for Earth observation data. After study of available alternatives, three prototype Earth observation information management systems were installed and the user response was evaluated through interview of fifteen of the group. The prototypes consisted of web-based client servers which permitted users to interrogate databases of Earth observation datasets; to search for information about sensor or satellite performance, and to retrieve data and information products. The chosen systems were CILS, the CEOS (Committee on Earth Observation Satellites) Information Location System; IDN, the CEOS International Directory Network; and JMS, NASA's Information Management System of EOSDIS, the Earth Observing System Data and Information System. For this study, no special effort was taken to populate the system directories and inventories with local data holdings, and the prototypes were essentially mirror sites of operational data management systems used in other parts of the world. While some of the interviewed scientists expressed enthusiasm for web based spatial information management approaches, all indicated that improvements should be sought in the prototypes to make them more user-oriented, intuitive, and responsive. Most of the interview group were experienced remote sensing researchers who had developed their own contacts with overseas peers and data providers. Several in this category expressed the vithem, unless the scientists changed discipline, application or geographic area of interest. On the other hand, several individual research projects or organisational units of CSIRO, as a result of these trials, were considering utilising one of more of the prototypes - particularly the IMS - to address their current unfulfilled requirements for data management. The study also found that while all fifteen of the interviewees felt they could benefit in some way from electronic information retrieval and spatial data management systems of the type assessed, it seemed unlikely that the target organisation would ever assign a sufficient priority to implement any of them in a systematic manner. The biggest impediment to an organisation-wide approach to spatial data management for Earth observation was the low priority assigned to information management, because this activity was considered "supporting" or "non-core" in relation to the central objective of scientific research. Results indicated that a piecemeal, decentralised or federated approach was the only means by which systems of this type could feasibly be introduced into the operating environment of CSIRO, in the absence of a major external forcing mechanism. This observation was compared to the evolution of EOSDIS, which had demonstrated a marked change from a centralised to a federated paradigm due to user preferences similar to those observed in the CSIRO case.

Share

 
COinS