HR: 14:25h
AN: IN23C-04    [Abstracts]
TI: Madrigal - Lessons Learned from 25 years of Evolution from a Single-Instrument Database to a Distributed Virtual Observatory
AU: * Holt, J M
EM: jmh@haystack.mit.edu
AF: MIT Haystack Observatory, Route 40, Westford, MA 01886 United States
AU: Rideout, W
EM: wrideout@haystack.mit.edu
AF: MIT Haystack Observatory, Route 40, Westford, MA 01886 United States
AU: van Eyken, T
EM: Tony.van.Eyken@eiscat.se
AF: EISCAT Scientific Association, PO Box 432, Longyearbyen, N-9171 Norway
AB: Madrigal is a distributed, open source virtual observatory which has been operational for 25 years. During that time it has evolved from a simple database system for the Millstone Hill Incoherent Scatter Radar to a full-featured virtual observatory distributed among five major sites. Madrigal is interoperable with the CEDAR Database and, in addition to being the primary data repository for incoherent scatter radar data, contains data from many other ground-based space science instruments. Madrigal features a well-defined metadata standard, real-time capability, an interactive Web interface, provision for linking ancillary information such as html pages and figures to data, interactive plotting and a complete Web-services interface. A number of important lessons have been learned from the Madrigal project: systems such as Madrigal depend critically on robust data and metadata standards; they need to be a community project; they must permit user interface improvements to be shared across the community; they require a standard, robust interface; scientific efforts using systems such as Madrigal can lead directly to improvements in the system. An example of the last has been the development of several climatological models from Madrigal data. Several features of Madrigal, such as a global search capability, were added in response to requests from the model developers. The models have recently been incorporated into Madrigal and provide a powerful basis for event discovery based on deviations of data from the climatological average. Madrigal will never completely solve the VO problem, but it will make life much easier for future VO projects.
UR: http://www.openmadrigal.org
DE: 0910 Data processing
DE: 6929 Ionospheric physics (1240, 2400)
SC: Earth and Space Science Informatics [IN]
MN: Fall Meeting 2005