19th International CODATA Conference
Category: Poster, Data Archiving
The Threat of Lost Libraries
Dr. R. Elizabeth Griffin
(Elizabeth.Griffin@nrc.gc.ca)
Dominion Astrophysical
Observatory, Herzberg Institute for Astrophysics,
Progress in understanding the physics of celestial objects relies heavily on
observations of Change. Short-term changes can be studied through new observations,
but what of changes that happen slowly over decades, sometimes almost imperceptibly,
occasionally violently, often serendipitously? These
are areas in which our present knowledge is seriously weak, and we must do all
in our power to improve the situation. Digital archives are the answer. Archived
data have become unique, powerful and essential tools for scientific research.
Astronomy is fortunate in possessing a rich heritage of some 3 million historic photographic observations. The collections are distributed in observatory plate stores around the world, and together they constitute an enormously important and, for the large part, unrepeatable resource, not only for astronomy research but also for Earth and Geo-Sciences (see "Studies of the Earth's Ozone from Historic Stellar Spectra"), and to the International Virtual Observatory (paper by Alex Szalay). They are best visualised as libraries of observations, requiring parallel indexing and access.
The scientific need for worldwide access to precious or time-sensitive observations is undisputed; we list examples of new science based wholly or largely on archival material. Digitization is the only practical mode of access. A wide spectrum of researchers can re-use digitized observations for many purposes, often quite different from the original ones, thus rendering them much more cost-effective.
Because that heritage is not in the digital form required by modern research, it is real danger of being lost. Once the relevant expertise also vanishes, the libraries of observations may go for good. The plates are deteriorating with time, though the worse immediate threat is human indifference.
Some initiatives are now
trying to rescue the observations digitally; we describe efforts in