iSGTW Feature - Montage a rising star in grid-enabled sky mosaics

Feature - Montage a rising star in grid-enabled sky mosaics

This mosaic was created using Montage and centers on the dark cloud Barnard 92. The dark clouds cataloged by Barnard contain so much dust that they obscure the light of stars embedded in them and in the line of sight behind them. The clouds thus appear as holes in the sky. Stars become visible as the dust thins out in the periphery of the clouds. Their light assumes a reddish hue because dust grains preferentially extinguish blue light. This is particularly will seen to the west (right) of the cloud. Another dark cloud, Barnard 93, is also seen along the eastern (left) side of the image. The image preserves the positional accuracy and intensities of a total of 92 input images and is an example of the types of images that astronomers use in their research.
Image courtesy of Montage

Astronomers, like most people, are impressed by pretty pictures, says Bruce Berriman.

"We say 'wow,' just as the lay public does. But we approach images in a different 'light' than the public, so to speak. We want to understand the scientific content of the images, and the influence of the instrument and background on that content," Berriman explains.

Not all pictures are created equal

Berriman is an astronomer interested in high-end computing and is part of the team that developed astronomy application Montage.

Already adopted by eleven major astronomy projects, Montage is used to develop scientific data products and support astronomical research. A bonus is that many images produced are impressive enough to excite the general public.

A grid-based application, Montage allows astronomers to compare images taken from different projects using different telescopes and satellites and with different cameras, all while preserving the scientific content-the positional accuracy of the data and the intensity of light in the original images.

"Astronomical images are delivered to the user in a variety of projections," Berriman explains. "Before you can compare these images you need to make them look as if they were taken from the same telescope using the same instrument. That way you can concentrate on the science of the images and not on the images themselves."

Montage can not only re-project images, it can find images you might want to work with, rectify their background radiation and then compose them into a science-grade mosaic of the sky.

A three-color view of the Rho Oph dark cloud constructed with Montage from deep exposures made with the Two Micron All Sky Survey (2MASS) Extended Mission.
Image courtesy of 2MASS

Astronomical benefits

Montage was written for professional astronomers as a publicly available, portable and scalable toolkit for generating science-grade mosaics. "Each task in the toolkit can be run independently using a single workstation, a cluster, a supercomputer or a computing grid," says Berriman. "We are also collaborating with our colleagues at the Information Sciences Institute, USC to use the Pegasus job execution system to run Montage on grids."

The code is portable across all common *nix-based platforms and works on operating systems including Linux, Solaris and Apple.

"A number of groups have integrated the code into their processing pipelines," says Berriman. "Others are using it for quality assurance."

Current users include the the U. S. National Virtual Observatory, which is using Montage to convert ten terabytes of data into a single mosaic; a group using Spitzer Space Telescope images to observe star formation; and even the Adler Planetarium in Chicago, who use an All-Sky mosaic measured by NASA's Infrared Astronomy Satellite in a quick-time virtual reality exhibit.

Watch this space

Just released is an on-request image mosaic service where users can create mosaics from data released by three wide-area surveys: 2MASS, the Sloan Digital Sky Survey and the Digitized Sky Surveys at the Space Telescope Science Institute. To manage requests on dedicated compute cluster, the service uses the Request Object Management Environment (ROME), a job management system developed for the National Virtual Observatory.

"The service currently restricts mosaic sizes to no more than one degree on a side, but we will extend this service to much larger images," says Berriman. "ROME will be at the heart of this extended system-it will allow seamless management of smaller jobs on our cluster, and larger jobs on a distributed system such as the Teragrid."

Montage was funded by the NASA Earth Sciences Technology Office Computing Technologies program and is maintained by the NASA IPAC Infrared Science Archive. ROME is supported by the National Virtual Observatory, which is funded by the U.S. National Science Foundation.

- Cristy Burne, iSGTW