News Archives

Strip Mining the Sky

February 1, 2005

  • Date: February 1, 2005 
  • Time: 11:00 – 12:15 p.m. 
  • Location: Woodward 149

Professor John McGraw <mcgraw@lodestar.org>
Department of Physics and Astronomy, UNM

The universe is a mighty big place. We know a great deal about it, and our understanding of the content, structure, texture and extent of the universe is accelerating, as apparently is the universe itself!

UNM’s Department of Physics and Astronomy is engaged in creating a unique, unbiased, and precise survey of the sky that will shed light (quite literally) on many of the fundamental issues of astrophysics, including the population of nearby stars and brown dwarfs, the structure of our Milky Way Galaxy, the nature of lethal activities in the environs of massive black holes in the centers of other galaxies, and the discovery of distant supernova explosions.

The telescope that will acquire our data is unique in that it has no moving parts – it does not track the stars and galaxies. Rather, our telescope is fixed to the Earth, and uses charge-coupled devices under computer control to produce a continuous multicolor image of the sky that runs from sunset to sunrise every clear night.

Every night for seven years the telescope will shovel in data from the same long, thin strip of sky, creating a data mine unprecedented in astronomy. Each night the telescope will produce 150 to 400 Gbytes of image data. And therein lies the rub. How do we discover a supernova explosion amongst a Sagan (“billions and billions”) of stars?

We have some ideas, but quite frankly, we’re counting upon the ideas, experience and skill of newly-found collaborators who will attend this talk in UNM’s CS Department to help us solve our wonderful problems.

Other CS-related problems range through real-time (i.e. one nanosecond precision) detector control, remote observing and observatory control, algorithms, data structures, data bases and data mining. Analysis of the data mine will necessitate application of cutting edge computer science techniques, including, as appropriate, artificial intelligence, adaptive systems, and scientific visualization.

The telescope, its data and data systems will be described, as will the scientific programs that drive its global design. You are cordially invited to attend, and then to dig in.