Review Committee Meeting Monday 22 January 2007 12:00 Pacific / 15:00 Eastern
Minutes: Monday 22 January 2007 12:00 Pacific / 15:00 Eastern
Agenda and Contact Info
- Discussion of the GRB Search Paper
- Jolien's comments [ TXT ]
- Brian's comments [ TXT ]
- Keith's BH-NS research [ TXT ]
- Soumya's response to Jolien's comments [ PDF ] and revised paper (Version 7) [ PDF ]
- Isabel's analysis of calibration phase errors [ HTML ]
InterCall telecon service:
Phone: 1-866-380-5536, participant code: 435 672 9587 #
International callers ++1-816-249-4731 with same code
International access numbers (~10 cents/min surcharge for LIGO Lab):
Paris 017 080 7156 ;
Lyon 042 603 0036 ;
Germany toll-free 0800 182 1591 ;
Berlin 030 726 167 371 ;
Rome 00645 217 080 ;
UK toll-free 0808 234 7914 ;
London 0203 107 0293
Handy participant command: Press *6 ["*M"] to mute, #6 to unmute
- Discussion of propogation of phase errors in calibration
- Derived by determining a magnitude (1-sigma) for time delay
in terms of samples. Injections then done with shifted signals.
(An alternative would be to simply adjust the phases of the response
functions used in conditioning vs. the response function used for
injection. This approach was not used.)
- Primarily affect S2 results for SG around 100 or 150 Hz because
these are where the phase errors are largest. Primarily affect
circular polarization results.
- These errors will need to be included. Isabel will compute them
for all GRB results. Need to change from 1-sigma to 2-sigma
error estimates (though these are probably pretty conservative already).
- Discussion of errors in fitting procedure
- Isabel reminded us that such errors are accounted for in the
following way: in comparing ULs from circular and linear polarizations
for H1-L1, the ratio of the results should be a constant factor. Measure
the spread in the distribution of ratios and use this as an estimate
of both the uncertainty in the fit method and Monte Carlo errors. See
Polarization document [ PDF ] for details.
- Discussion of use of hrss90%.
- Sergey is concerned that hrss90% is not the best choice for a
sensitivity estimate. Better to use hrss50%, for one reason because
the hrss50% for linear and circular polarization waveforms should be
nearly the same.
- Sergey is also concerned that hrss90% might not be a valid upper
limit. Sergey will send an email note explaining his concern
but the issue was tabled for now so that we could discuss other issues.
- Ray: other substantive changes seem OK. Inclusion of explicit
inclination angle factors in equations will be done.
- NS-NS, BH-BH, NS-BH discussion:
- NS-NS might not be irrelivant. Here, not targeting the actual
chirp but instead the modes of an meta-stable hypermassive neutron
star product. Characteristic time for emission from modes is less
than 100 ms at frequencies ~ 1.5 kHz. Problem is that the inspiral
and merger waveforms may affect data conditioning procedure. Too
uncertain to discuss in such detail in paper.
- NS-BH: Keith Thorne discussed recent results that are unpublished.
How to handle this? Should have as many references as possible since
we're throwing a wide net. But don't have too much detail since our
uncertainty doesn't warrent. Jolien is concerned that expressing
energy emission in terms of total mass seems misleading (since it ought
to scale with reduced mass, and perhaps inversely with total mass
for large mass ratios).
- BH-BH: used here to estimate energy emission, not to be interpreted
as a viable GRB model. Need a phrase to express this.
- Soumya discussed the changes to the population study.
- Most changes adopted. Note: scaling of PSD is done to the actual
sensitivity at the time of each GRB, not just the sensitivity of the
run at the time of each GRB. This was a change that was introduced.
- An accurate implementation would need to take into account the
changing spectral characteristics in Eq. (17), redshifting of the
standard candle, error propogation, etc. This analysis is clearly
labeled as exploratory.
- Discussion of content of paper.
- Brian thinks that paper loses focus at beginning of Section VII.
- Joel thinks that Section VIII is too long given uncertainty in GRB
progenitors (essentially the same equation repeated with different factors,
which are irrelevant).
- Ray comments that input was sought on this issue. Aware that this
is not a pure results paper. Feeling of LSC was divided on the issue.
- Jolien agrees with Brian and Joel that the tone of the paper takes
an abrupt shift, but that Sections VII and VIII do add useful content,
and it is probably best to keep them at this stage rather than trying
to change the content significantly (e.g., putting them in an appendix,
putting them into a separate publication).
- Brian's question: "Why 180s". Ray will send an email detailing
the justification for this amount of time and will add a sentence in the
paper better describing why this amount was chosen.
$Id: minutes-2007-01-22.html,v 1.5 2007/01/23 17:30:06 jolien Exp $