Metadata describes what a digital resource is and is about. Through structured and systematic representation of the resource objects, metadata provides a powerful tool for discovering, browsing, navigating, and retrieving information resources and establishes inventories for scientific data archiving and preservation. However, creating metadata for access to and management of scientific data is a challenging task and requires careful planning and collaboration between metadata experts and scientists.

This presentation will review the roles of metadata in scientific data access and management. For scientists, metadata is the gateway to the vast data repositories, through which they can effectively search for, navigate, and retrieve data cross disciplines and from multiple data nodes. Metadata can also support data analysis and visualization functions. For scientific data curators, metadata will function as an annotated map for documenting the history of data sets as well as necessary information for ownership, conditions of access, distribution, and technical details.

As important as metadata is, its creation and maintenance remains challenging both economically and technically. Much of the metadata creation for scientific data is largely manual work and needs trained personnel to ensure the quality of metadata. This makes the process not only expensive but also time-consuming. Even though technologies today have the capability of performing some automatic processing, There is a great deal of work to be done in developing applications tailored to the need for representing and managing scientific data. This presentation will discuss what kinds of work need to be done in creating metadata for access to and management of scientific data.