Home » Technology » Scientific Computing » Analyzing Astronomical Data Can Drive You (to) AIPS!

Analyzing Astronomical Data Can Drive You (to) AIPS!

© Adam Hughes

The importance of scientific visualization has often been discussed in this space in the past, and with good reason. Experiments and simulations produce reams of data that must be analyzed and understood by the scientists producing them, which can be a daunting task. And even though researchers involved in a project may develop the ability to visually scan numbers and glean some meaning, communicating their significance to others is often nearly impossible without some sort of graphical representation. Many products and protocols have been developed for converting numbers to pictures, but when it comes to analyzing astronomical data, many scientists can often be found playing with the AIPS.

In the late 1970's, researchers at the National Radio Astronomy Observatory (NRAO) in Charlottesville, VA, recognized the need for some set of standardized methods for handling radio interferometric telescope data. From this ideal, they developed the Astronomical Image Processing System, or AIPS. The Astronomical Image Processing System is a software package that not only allows scientists to create images from data, but also permits the editing and calibration of the constituent data. It appears that AIPS has been widely accepted by the radio interferometric community, as some version or another of the software is estimated to be running at some 250 research sites. The AIPS package is primarily used for Very Long Baseline Array and Very Large Array processing, but is also employed in many other astronomical visualization efforts.

Even without knowing much about the physics of the problems for which AIPS is commonly employed, a brief survey of its features is quite impressive and gives one a general feel for the quantity and quality of programming effort which has been involved, estimated at well over 60 man years.

The AIPS package has evolved into the primary software package for displaying and analyzing radio images of two or three dimensions. Moreover, it serves as both an interactive tool and a batch processing program, allowing maximum flexibility for the user. In addition, there are several megabytes of documentation provided to guide the researcher in his visualization efforts.

In addition to the above features, AIPS has been developed with optimum performance and portability in mind. Written primarily in Fortran and C (generally fast computational languages), AIPS was written in a general way to make porting from one computer platform to another relatively seamless and painless. As a general rule, when developing with an eye toward portability, minimal changes should be needed to the source code, with differences being more or less confined to compiler directives. AIPS appears to have done a nice job here, too, limiting source

The copyright of the article Analyzing Astronomical Data Can Drive You (to) AIPS! in Scientific Computing is owned by . Permission to republish Analyzing Astronomical Data Can Drive You (to) AIPS! in print or online must be granted by the author in writing.