GLAS Long-Term Archive: Preservation and Stewardship for a Vital Earth Observing Mission

Data Stewardship, preservation, and reproducibility are fast becoming principal parts of a data manager's work. In an era of distributed data and information systems, it is of vital importance that organizations make a commitment to both current and long-term goals of data management and the preservation of scientific data. Satellite missions and instruments go through a lifecycle that involves pre-launch calibration, on-orbit data acquisition and product generation, and final reprocessing. Data products and descriptions flow to the archives for distribution on a regular basis during the active part of the mission. However there is additional information from the product generation and science teams needed to ensure the observations will be useful for long term climate studies. Examples include ancillary input datasets, product generation software, and production history as developed by the team during the course of product generation. These data and information will need to be archived after product data processing is completed. NASA has developed a set of Earth science data and information content requirements for long term preservation that is being used for all the EOS missions as they come to completion. Since the ICESat/GLAS mission was one of the first to end, NASA and NSIDC, in collaboration with the science team, are collecting data, software, and documentation, preparing for long-term support of the ICESat mission. For a long-term archive, it is imperative to preserve sufficient information about how products were prepared in order to ensure future researchers that the scientific results are accurate, understandable, and useable. Our experience suggests data centers know what to preserve in most cases. That is, the processing algorithms along with the Level 0 or Level 1a input and ancillary products used to create the higher-level products will be archived and made available to users. In other cases, such as pre-launch, calibration/validation, and test data, the data centers must seek guidance from the science team. All these data are essential for product provenance, contributing to and helping establish the integrity of the scientific observations for long term climate studies. In this presentation we will describe application of information gathering with guidance from the ICESat/GLAS Science Team, and the flow of additional information from the ICESat Science team and Science Investigator-Led Processing System to the NSIDC Distributed Active Archive Center. This presentation will also cover how we envision user support through the years of the Long-Term Archive.