Grant Supports Interdisciplinary Program as Researchers Focus on Astrostatistics

13 June 2001-- A three-year, $1 million grant from the National Science Foundation (NSF) has provided funding for researchers at Penn State and collaborators at two other universities to further study the emerging, interdisciplinary field of astrostatistics.

"As the nature of observational astronomy has changed, going from single astronomers watching a handful of objects to teams of astronomers locating millions of objects as part of large digital sky surveys at a variety of wavelengths, the necessity for statistical analysis has increased," said Jogesh Babu, professor of statistics at Penn State and principal investigator for the NSF grant. "An enormous amount of data exists--so much that even basic statistical measures such as the median and quartiles of a set of data can be difficult to determine."

Research under the grant has been motivated by a recent National Virtual Observatory (NVO) initiative to link archival datasets and catalogues from many existing astronomical surveys. The effective use of such integrated massive datasets involves more than just access to and extraction of information. Scientific understanding requires sophisticated statistical modeling of the selected data.

Despite the challenges, Babu and his collaborators believe the pairing of astronomy and statistics provides a logical extension for both fields of study. Those collaborators include: Eric Feigelson, professor of astronomy and astrophysics at Penn State, and colleagues at California Institute of Technology, home of the NVO, and Carnegie Mellon University, with collaborators in its departments of astronomy, computer science, and statistics.

During the past 14 years, Babu and Feigelson have led the effort to develop advanced statistical methods to serve the research needs of observational astronomers. They co-authored a book titled Astrostatistics in 1996 and created and maintain a Web site that provides links to statistical codes and services on the Web. They also have played host to two international conferences at Penn State with a third, titled "Statistical Challenges in Modern Astronomy III," scheduled at the University Park campus this summer. Approximately 120 scientists, some astronomers and some statisticians, are expected to visit campus in late July. The conferences are designed to facilitate a dialog between astronomers and statisticians regarding important research issues. While large datasets provide many challenges, Babu points to the assistance the interdisciplinary approach has provided in recent years. Adapting or utilizing advanced statistical methods to serve the research needs of observational astronomy provides the primary goal.

"What Eric and I are doing is almost like match making," Babu said. "We're helping to put together groups of like-minded statisticians and astronomers across the country for collaborations on interdisciplinary astrostatistics research."

As more astronomers and statisticians become aware of the possibilities and become comfortable with the methods, and as graduate students and postdoctoral students grow and learn in an atmosphere that encourages the use of interdisciplinary work, Babu believes the approach will become a vital tool throughout observational astronomy.

<S W S>