A key part of the Linked Environments for Atmospheric Computing (LEAD) vision is that of providing seamless access to meteorological data for education and research purposes. That goal includes empowering the user community to assimilate, subset, download, visualize and mine a variety of meteorological observations, model output, radar data and satellite imagery through a simple, easy to use, and web-based LEAD portal (http://leadproject.org). Since LEAD is intended to support education and research in meteorology, it is necessary to facilitate case studies of significant meteorological phenomena by providing the ability to perform retrospective model runs and analysis of and comparison with the corresponding verification data.

In order to support these needs, the UCAR Unidata Program Center has built a relatively inexpensive medium term (~6 month, ~80 TB) archive system. This work has the additional benefit of providing a medium term archive of several data streams in the Unidata Internet Data Distribution system (http://www.unidata.ucar.edu/software/idd/) for the Unidata community. Another benefit to Unidata has been the opportunity to investigate scalability of certain technologies, including THREDDS (the Thematic Realtime Environmental Distributed Data Services – see: http://www.unidata.ucar.edu/projects/THREDDS/), and conduct stress tests on large volumes of data and longer term archives.

In this presentation, we will provide an overview of the development, capabilities, and uses of the LEAD testbed system at the Unidata Program Center.