The Geospatial Data Curation, Management, and Discovery in Academic Libraries

The Geospatial Data Curation, Management, and Discovery in Academic Libraries

As many funding agencies began to require their funded projects to share the data publicly, researchers start to sort for resources in the libraries to curate and publish their datasets. Different with library’s traditional collections, datasets are unique products derived from research projects, which require more extensive metadata in order to make it meaningful and reusable. Also, as datasets were generated along the full research process, which datasets need to be preserved and at which stage the libraries should get involved still remains unknown. In addition, inside the libraries how librarians could manage the datasets efficiently is another question that we should pursue in building up our data services. In this research, we used geospatial data as an example to explore the questions raised above. The reason that we chose geospatial data is because it is a special but widely used data type – it allows us to reach different disciplines yet maintains the scope of this research manageable and practical. Geospatial data has been widely used in many science and technology disciplines, from its traditional application areas in Civil Engineering, Earth Science, and Agricultural studies, to the more recent application areas in computer science, industry engineering, and social science. We collected our information from multi-disciplinary collaborations, focused interviews, and training session feedbacks. From these information, we generalized the workflow of typical research projects in science and technology, and lined up the research workflow with data lifecycles. We have identified the library’s role in each research stage, and experimented different software tools, cyberinfrastructure set-ups, and practices that the library could provide in order to facilitate the geospatial data curation, management, and discovery. The initial evaluation of our practice from researchers’ feedbacks and data usage statistics have suggested that we have successfully addressed data problems in the collaborated projects.