GWDAW 2002

Peter Saulson, Syracuse University

The seventh installment of the Gravitational Wave Data Analysis Workshop was held in Keihanna Science City in Kyoto Prefecture, Japan, on 17-19 December 2002. This meeting marked the rapid progress of the various gravitational wave detection efforts around the globe. LIGO Lab Director Barry Barish summarized the mood of most attendees in his remarks to the workshop banquet, when he said that the meeting program represented a mature field, with rapid progress on many instruments, discussions of new analyses of unprecedentedly sensitive data, and ongoing research on data analysis.

The first day's talks were devoted to instrument progress reports. Interferometer reports all described substantial progress in commissioning and initial data taking runs. TAMA, GEO, and LIGO have all collected some science data in runs interspersed with commissioning aimed at achieving full design sensitivity. VIRGO has had a very successful commissioning exercise of its Central Interferometer, and will soon start commissioning its 3 kilometer arms. Attendees also learned of the vigorous prototyping program preparing for Japan's Large Cryogenic Gravitational Telescope (LCGT), slated for installation in the Kamioka mine. A 7-meter single-arm cryogenic test facility, CLIK, has been built at ICRR, while the LISM 20-meter room temperature interferometer is in operation at Kamioka. Construction of the 100-meter cryogenic CLIO interferometer has begun in the Kamioka mine; the tunnel has been prepared, and infrastructure work is now in progress. LCGT aims to take advantage of the low seismic noise of its underground site, and the low thermal noise of 20 K cryogenic test masses, to reach sensitivities sufficient to see neutron star binary inspirals at a distance of 200 Mpc. Advanced LIGO and others are also aiming at similar goals, although via different technological means.

The growing excitement about the prospects of space-based detectors were the subject of the first afternoon's talks. LISA is at the center of the world's planning (it is a joint ESA-NASA project), but attendees also heard about a Japanese initiative called DECIGO, promoted by Seiji Kawamura, which is aimed at the 0.1 Hz band between LISA's most sensitive band and that of the ground-based interferometers. Among the bars, EXPLORER and NAUTILUS are operating well (more about them later), and ALLEGRO and AURIGA are about to come on line after major transducer upgrades.

The second day's talks began the discussion of data analysis per se. The discussion was organized around signal character (burst, inspiral, sinusoidal, and stochastic), and moved fluently between bar and interferometer analyses. There were a number of talks from LIGO authors on the methods used to analyze data from the recent (late August to early September 2002) S1 run. Preliminary results are still embargoed to allow for revision during discussions internal to the LIGO Scientific Collaboration; most members of the LSC had only heard the first results of the analyses a week or two before the meeting, and an active review process is now under way. (The first big announcement of still-preliminary results is expected in mid-February at a meeting of the AAAS.)

The most-anticipated session of the meeting was the late-afternoon section devoted to recent coincident analyses of data from bars. Giovanni Prodi opened with a very clear discussion of the methods used by the IGEC collaboration to set upper limits using several years worth of data from the entire worldwide bar network. The rest of the talks had as their subject the recent result from the Rome group on 2001 data from EXPLORER and NAUTILUS (Astone et al., Class. Quant. Gravity 19, 5449-63 (2002) and gr-qc/0210053.) In that paper, "indications" were reported of the emission of gravitational waves from sources scattered throughout the Galactic disk. Pia Astone led off with a discussion of a Bayesian interpretation of the results, (Eugenio Coccia having summarized the frequentist analysis of the paper, during his status report the previous day.) Then came two strong critiques of the statistical significance of the result, presented in turn by Sam Finn and Warren Johnson. Finn demonstrated that results as significant as those reported would be expected due to chance alone 25% of the time, hardly dramatic evidence of a discovery. Astone, in remarks interspersed during Johnson's presentation, emphasized that no discovery was claimed. Coccia closed the session by defending the significance of the result, on the grounds that detections at sidereal time 4 hours were special because of the link to the Galaxy, but also reiterating that only further observations could promote the claimed "indications" into a discovery (or, of course, rule them out.)

Discussion spilled over into the evening and on to the workshop banquet, aided by freely flowing sake and Asahi beer. Warren Johnson and Eugenio Coccia mugged for the cameras, pretending to throw roundhouse punches at one another. Then, Johnson and Pia Astone kissed and made up, literally and repeatedly, until everyone's camera had recorded the public reconciliation.

The third and final day was filled with talks on new data analysis methods, discussions of sources, and accounts of detector characterization techniques. While less easily summarized than the previous two days' talks, these in some way were the most future-oriented heart of the meeting, laying the foundations for the new results to be expected in time for the next year's workshop.

On the sidelines of the meeting, another important discussion was taking place. After the first day, representatives of TAMA and LIGO met in the International Institute of Advanced Studies' beautifully appointed seminar lounge to sign a Memorandum of Understanding for joint analysis of data from the upcoming S2 data run. This brought to fruition a series of negotiations spearheaded by Nobuyuki Kanda for TAMA and Albert Lazzarini for LIGO. After the third and final day's talks, a first working meeting of the two teams was held. Plans were sketched for joint searches for burst signals and for chirps from binary inspirals, intended to be completed six months after S2 concluded in mid-April 2003.

Intensely-focused discussions of other ongoing or soon-to-be-initiated analyses were held during coffee breaks and in the evenings. Combined with the work formally presented in talks, these indicate a field truly mature, and ready for a steady stream of new scientific results over the next few years.