next up previous
Next: 13. Why is the Up: Einstein@Home S3 Analysis Summary Previous: 11. Typical S3 results


12. The search is not equally sensitive over the entire sky.

As described earlier, the signal from a pulsar is modulated by the Earth's rotation about its axis and its motion around the Sun. The amount and type of this modulation depends upon sky location. Because the detector's output contains instrument noise artifacts which are purely sinusoidal (for example, 60 Hz harmonics and violin modes of the optical suspension wires) this means that our search is more sensitive in some regions of the sky than in others.

Shown in the following two figures are plots which are similar to the previous figures but in the band of frequencies around the violin modes of the optical suspensions (340-350 Hz). Many LIGO optical elements are suspended from wires, and thermal fluctuations and other noise sources cause these wires to vibrate in the same way as a plucked violin string, hence the name ``violin modes''. The violin modes of the optical suspensions produce quite prominent false candidates. (For example, just compare the number counts in these two figures with the figures in the previous section for the 300-325 Hz frequency band.)

The all-sky map (a) shows that the violin modes produce a "noisy belt" on the sky. Along this belt, spectral line artifacts appear like a pulsar signal (See the next section for the reason why!). Essentially, we do not have much sensitivity to a signal in this region.

The frequency - declination plot shows that although the violin modes make a moderately large region of the sky insensitive to gravitational waves from a pulsar, the obscured region only covers a narrow range of frequency and still leaves large sensitive areas in the parameter space intact. In fact, when preparing the (final, second time through) S3 data set for analysis by Einstein@Home, the violin modes were removed to reduce spurious candidates.

Figure 12.1: Color map of the number of coincidences among sixty 10-hour data segments in the frequency range 340-350 Hz. The first harmonics of the violin modes (340-350 Hz) produce quite prominent candidates. Since they are not produced by pulsars, these misleading events are called "false candidates" or "false alarms". (a) All-sky map. (b) frequency-declination plane.
(a) All-sky map (b) declination - frequency map
\includegraphics[height=8cm,clip]{dirty_sky_blue.eps} \includegraphics[height=8cm,clip]{dirty_freq_blue.eps}

In the next section, we explain why sinusoidal noise sources in the instrument such as the violin modes of the suspension wires produce these characteristic patterns (great circles) on the celestial sphere.


next up previous
Next: 13. Why is the Up: Einstein@Home S3 Analysis Summary Previous: 11. Typical S3 results
LSC LOGO Contents[1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]Bibliography
Einstein@Home S3 Analysis Summary
Last Revised: 2007.03.28 08:59:23 UTC
Copyright © 2005 Bruce Allen for the LIGO Scientific Collaboration
Document version: 1.132