This is a 4-year project to modernize and upgrade the data infrastructure for the Coupling, Energetics and Dynamics of Atmospheric Regions (CEDAR) research program. CEDAR is a NSF supported research program that began in 1986. Initially focused on instrumentation, it soon broadened to include theory, modeling and coordinated observational campaigns. A database has been an important component of CEDAR since its inception. Since the beginning, the CEDAR Database format has been the standard for archiving and distributing data from the wide variety of instruments and models employed by the CEDAR community. Currently, the primary CEDAR data archive is housed at the National Center for Atmospheric Research, but from the start there have been two primary distribution mechanisms - that provided by NCAR and the Madrigal Database, a distributed system that is employed by all incoherent scatter radars used for CEDAR research. The result of this effort will be a merged system with Madrigal providing both data archiving and distribution. Madrigal will be extended to include the concept of one or more archiving nodes that will automatically download and archive new and modified data when it appears on any of the distributed nodes. The main new archiving node will be located at the Haystack Geospace Science Center, with off-site backup at the MIT Data Center. In addition to the archiving capability, a number of other new features will be added to Madrigal in response to user and administrator requests. The most important of these will be the replacement of the existing scaled integer data format with a modern floating point, self-describing format. This will greatly simplify adding new data types to Madrigal and will also allow Madrigal data sets to be directly accessed from a wide variety of existing standard analysis software.
Community infrastructure building and service is at the heart of the proposed activity. There are numerous societal benefits to enhanced understanding of atmospheric effects and sustained, easy access to CEDAR data is vital to this research community. Madrigal is already used by the CEDAR community as an efficient and popular tool for training of undergraduates and graduate students. Madrigal is an open-source project, which means that software and developer documentation are on-line so anyone can contribute capabilities. The web services component of the system, coupled with the use of a standard scientific data format will enable anyone with internet access to participate in the analysis of CEDAR data.
