> so I would think you would want to use something more physical, like a
> common alpha.
I agree - at least as a plot option I would like to see this.
Clem
--
**********************************************************************
Clem Pryke - Associate Professor - Physics
University of Minnesota,
Room 313 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-624-7578 Fax: 612-624-4578 email: pryke(a)physics.umn.edu
**********************************************************************
Jon,
This is cool stuff.
I worry a bit about the inconsistent best fit angle for TB and EB - TB more
negative for both B2 and Keck - and TB+EB smaller offset from zero than either
alone. This last seems weird to me - can you explain it.
I also worry that there is perhaps interaction with diff ellip deproj which as
we noted on telecon seems to do weird things to supfac.
Please try your code again on dp1100 - perhaps the diff ellip is taking out
some TB which is actually real making TB and EB angles inconsistent.
Also does the BB curve in fig 3 include r=0.1?!? Why does the amplitude of the
ell=80 bump not change when I flick B2/Keck?
Clem
--
**********************************************************************
Clem Pryke - Associate Professor - Physics
University of Minnesota,
Room 313 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-624-7578 Fax: 612-624-4578 email: pryke(a)physics.umn.edu
**********************************************************************
> so you could imagine elements in the TB+EB covariance matrix might
> have some stronger effect than just the TB or EB alone (?)
You aren't taking the sim derived cov mat at face value are you? All but the
first off diag should be zeroed - for 50 sim set like this there is loads of
noise.
> I believe in the parlance of our times that pureB means r=0.1
Huh? pureB means Kendrick style estimator - nothing to do with sim input
spectrum which is of course r=0?!
Clem
--
**********************************************************************
Clem Pryke - Associate Professor - Physics
University of Minnesota,
Room 313 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-624-7578 Fax: 612-624-4578 email: pryke(a)physics.umn.edu
**********************************************************************
> So that tells us something isn't right. If Planck was measuring
> dust and it was scaled properly, the cross-spectra should agree with the auto.
So to be sure I understand: you are complaining that the orange line in
Jamie's fig 2 is approx 0.003 uK^2 but the red line in the BfBb panel of the
second figure 2 is much lower?
This seems to imply that either:
1) the extrapolation is wrong in an amplitude sense and the pol dust power is
much less than r=0.05 equivalent.
2) The extrapolation is wrong in a spatial sense and the pol angles are
getting scrambled or similar.
Jamie T: What does the temperature equivalent of fig 2 look like? Can the TfTb
panel of the second fig 2 possibly be right? Can the noise of a cross spectrum
taken against the B2 T map really only be a few uK^2?
Clem
--
**********************************************************************
Clem Pryke - Associate Professor - Physics
University of Minnesota,
Room 313 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-624-7578 Fax: 612-624-4578 email: pryke(a)physics.umn.edu
**********************************************************************
I've added a posting which shows how BICEP2's TB and EB are consistent with
birefringence (or alternatively, a miscalibration of detector polarization
orientation).
http://bicep.caltech.edu/~spuder/analysis_logbook/analysis/20130902_B2_TB_E…
I think it would be worth looking at my figure 2 tomorrow although we don't
need to spend much time on this. I certainly don't want to get bogged down
in a discussion of whether or not we're actually seeing cosmological
birefringence. The summary is this:
BICEP2's TB and EB are consistent with birefringence with an angle of -1 to
~4 sigma. It is not, however, enough to explain the power in BB.
-Jon
Hi Jamie,
I looked at your new post with the suppression factor included, and the
conclusion is pretty similar. The scaled Planck auto-spectra give an
appreciable level of contamination at 150 GHz (~r = 0.05 at 150 GHz) at ell
= 100. But there is no detection in the cross-spectra. So that tells us
something isn't right. If Planck was measuring dust and it was scaled
properly, the cross-spectra should agree with the auto. Maybe there is a
noise bias in the Planck map that is not accounted in the auto-spectrum.
There is noise (and bias) in the Planck estimate of polarization fraction
for example.
I also want to check if you are sure that the Planck polarization fraction
maps and angles are based on 353 GHz temperature map and not 353 GHz dust
map? I looked again at the Bernard presentation and I think that is the
case, though the presentation is not entirely explicit.
I suspect the noise bias but I just wanted to see if Sergi has any insights.
Jamie
++++++++++++++++++++++++++++++++++++++++++++++++++++
Caltech (MTTh) JPL (WF)
M/S 367-17 M/S 169-327
California Inst. of Technology Jet Propulsion Laboratory
Pasadena, CA 91125 Pasadena, CA 91109
Voice: (626)-395-2017 (818)-354-0715
Fax: (626)-395-2366 (818)-354-8895
++++++++++++++++++++++++++++++++++++++++++++++++++++
12:00 Eastern, 11:00 Central, 9:00 am Pacific
1-866-890-3820 (toll: 1-334-323-7229) Passcode: 59702175#
Agenda is taking shape in the google doc
http://goo.gl/LNvpx
--
**********************************************************************
Clem Pryke - Associate Professor - Physics
University of Minnesota,
Room 313 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-624-7578 Fax: 612-624-4578 email: pryke(a)physics.umn.edu
**********************************************************************
> the second-nearest-neighbors in either direction are centered slightly
> above zero
Are you claiming significant detection of next-but-one?
> Cosmic ray hits can be used as a calibration sample for estimating
> crosstalk. This measures crosstalk on the Al transition
Don't you mean Ti?
Clem
--
**********************************************************************
Clem Pryke - Associate Professor - Physics
University of Minnesota,
Room 313 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-624-7578 Fax: 612-624-4578 email: pryke(a)physics.umn.edu
**********************************************************************