Astrometry and the virtual observatory

A primary goal of the Virtual Observatory (VO) is the integration of data from many instruments and wavelengths and creating source and objects lists to facilitate pan-chromatic science. Astrometry is essential to define the positional coincidences (either static position or motion in the solar system, galaxy, or other system) that mark multiple observations of flux as likely to be originating from the same physical source. Support for VO is based on the community experience with data archives like that of the Hubble Space Telescope (HST) and the extremely high science value of large, homogeneous surveys, for example the Sloan Digital Sky Survey. These experiences have clarified the fundamental role that advanced information technology plays in astronomy research. Groups around the world have drawn lessons from these experiences that have changed the way that astronomical data is collected and managed. We discuss the Canada-France-Hawaii Telescope Legacy Survey and the WFPC2 Association Stacks project. These projects demonstrate some of the ways in which data management has changed and we point out the essential role that astrometry plays in migrating the data output of these projects into VO-like systems and user interfaces. High quality astrometry has been under-valued in the past because data management issues focussed on the needs of a single Principal Investigator. It is now clear that data have little value for large-scale VO projects if they lack reliable astrometry.