20 May
2013
20 May
'13
11:03 p.m.
Hi Walt and Sergi,
I'm expecting on tomorrow's telecon we will discuss Sergi's posting
(deferred now twice):
2013 May 6: Planck 143 nominal maps versus BICEP2 revisited (SRH)
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20130506_planck_…
Sergi added "NOTE Added 20130514: Do not check out this posting..."
Ignoring this warning, I see the posting claims a detection of the 2.2%
calibration power scaling between Planck and WMAP, similar to the level
Planck has reported at degree scales.
In talking with Kris Gorski on Friday we reviewed the spectral shape of
this discrepancy that Planck and WMAP have been fretting over--it
appears roughly consistent with linear slope which ranges from 0 at low
ell to 2-2.5% at ell=200, and roughly constant above that.
I'm eager to see a side-by-side abscal analysis of B2 derived from WMAP
vs from Planck, including the slope parameter introduced by Walt in this
posting:
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20121101_abscal_…
Perhaps it would be worth running this slope specifically for our
ell=30-200 bins (rather than our usual 100-300). I'd like to see the
average difference in slope parameter for individual detectors on Planck
vs WMAP, and the same for the coadded abscal as well. It is not clear
to me whether our beam measurements can reach high enough precision to
be able to say whether it is Planck or WMAP that has the "right"
slope--that would be quite newsworthy--but as a first step it would be
interesting to see whether we can detect this ell-dependent discrepancy
between the two experiments.
Walt, is this abscal analysis already in the works?
thanks,
John
--
___________________________________________________________________
John Kovac jmkovac(a)cfa.harvard.edu
Assistant Professor, Astronomy and Physics, Harvard University
160 Concord Ave rm 310, Cambridge MA 02138, 617-496-0611
21 May
21 May
2:55 a.m.
Hi John,
A coadded-abscal analysis is currently in the works. I'm planning to
use a selection of "ref" and "cal" including Planck 100, Planck 143,
and WMAP 9-year. This can easily include the slope parameter.
I haven't done per-detector abscal using Planck, but I can work on that
next.
- Walt
4:20 p.m.
Near the end of the telecon just now I asked Randol whether he had taken
his BICEP2 beam profile measurements from BBNS data, e.g. Fig 7 here
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20130124_zemax_c…
and translated them into B_l's which can bracket uncertainties across
the ell=30-200 range. Randol had dropped off so didn't answer.
But I am surprised Chris didn't speak up since (as I only remembered
after the telecon) he had indeed made B_l's from similar profiles in
this posting
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20130227_BBNS_be…
Check out Chris' third figure of this post. The lower right panel
titled "beam window function differences" shows that the difference
between the BBNS-measured beam and our usual assumed gaussian grows from
0 to 2% up to ell=100, and then turns over--a similar amplitude to the
WMAP/Planck discrepancy. The difference between A and B profiles (which
was the main focus of our attention) is much smaller--perhaps indicating
that we've measured our true effective B_l across this range to decent
precision.
Walt, are the WMAP and Planck input maps that you are using for the
abscal analysis smoothed with gaussians? Or has the BBNS profile become
our new standard?
John
On 5/20/13 11:03 PM, John Kovac wrote:
Hi Walt and Sergi,
I'm expecting on tomorrow's telecon we will discuss Sergi's posting
(deferred now twice):
2013 May 6: Planck 143 nominal maps versus BICEP2 revisited (SRH)
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20130506_planck_…
Sergi added "NOTE Added 20130514: Do not check out this posting..."
Ignoring this warning, I see the posting claims a detection of the 2.2%
calibration power scaling between Planck and WMAP, similar to the level
Planck has reported at degree scales.
In talking with Kris Gorski on Friday we reviewed the spectral shape of
this discrepancy that Planck and WMAP have been fretting over--it
appears roughly consistent with linear slope which ranges from 0 at low
ell to 2-2.5% at ell=200, and roughly constant above that.
I'm eager to see a side-by-side abscal analysis of B2 derived from WMAP
vs from Planck, including the slope parameter introduced by Walt in this
posting:
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20121101_abscal_…
Perhaps it would be worth running this slope specifically for our
ell=30-200 bins (rather than our usual 100-300). I'd like to see the
average difference in slope parameter for individual detectors on Planck
vs WMAP, and the same for the coadded abscal as well. It is not clear
to me whether our beam measurements can reach high enough precision to
be able to say whether it is Planck or WMAP that has the "right"
slope--that would be quite newsworthy--but as a first step it would be
interesting to see whether we can detect this ell-dependent discrepancy
between the two experiments.
Walt, is this abscal analysis already in the works?
thanks,
John
--
___________________________________________________________________
John Kovac jmkovac(a)cfa.harvard.edu
Assistant Professor, Astronomy and Physics, Harvard University
160 Concord Ave rm 310, Cambridge MA 02138, 617-496-0611
24 May
24 May
4:06 p.m.
Hi John,
Walt, are the WMAP and Planck input maps that you are
using for the
abscal analysis smoothed with gaussians? Or has the BBNS profile
become our new standard?
As I documented in my posting, I've been using circular Gaussians of
31.22 arcmin. How would I go about using the BBNS profile? I imagine
it's possible to prepare a set of B_l's that represent a beam that's
radially symmetric but non-Gaussian, with wings similar to the BBNS
beams, and then feed that to alteralms. Is that what you have in mind?
- Walt
4:10 p.m.
Walt,
In my posting that John points out, the crux of it was that I prepared a
fits file for the B_ls of the azimuthally averaged BBNS beam profile. It is
circular and non Gaussian.
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20130227_BBNS_be…
If you want this fits file so you can use it with synfast and alteralm, let
me know and I'll point you to it.
-Chris
On Fri, May 24, 2013 at 3:06 PM, Walt Ogburn <ogburn(a)stanford.edu> wrote:
Hi John,
--
**********************************************************************
Christopher Sheehy - Ph.D. candidate - University of Chicago
Research Specialist, University of Minnesota, Department of Physics
Room 220 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-625-1802 Fax: 612-624-4578 email: csheehy(a)uchicago.edu
**********************************************************************
Walt, are the WMAP and Planck input maps that you
are using for the
abscal analysis smoothed with gaussians? Or has the BBNS profile
become our new standard?
As I documented in my posting, I've been using circular Gaussians of
31.22 arcmin. How would I go about using the BBNS profile? I imagine
it's possible to prepare a set of B_l's that represent a beam that's
radially symmetric but non-Gaussian, with wings similar to the BBNS
beams, and then feed that to alteralms. Is that what you have in mind?
- Walt
_______________________________________________
Bicep2-list mailing list
Bicep2-list(a)lists.fas.harvard.edu
https://lists.fas.harvard.edu/mailman/listinfo/bicep2-list
4:18 p.m.
Hi Chris,
Yes, if you can point me to the fits file for the B_l's, that would be
great. I have only prepared input maps following a pretty limited set
of recipes, with circular Gaussian beams, so I'm not very familiar with
the other options that are available in alteralm. I take it that the
fits file you mention is a file with beam properties only, which can be
applied to any healpix map as a way of performing the beam smoothing?
If we get this working, it sounds as though it would be preferable to
(as John says) make it the standard for abscal. In that case I will
just drop the abscal sims I've been running with circular Gaussian
beams.
- Walt
On Fri, 24 May 2013 15:10:39 -0500
Chris Sheehy <csheehy(a)uchicago.edu> wrote:
Walt,
In my posting that John points out, the crux of it was that I
prepared a fits file for the B_ls of the azimuthally averaged BBNS
beam profile. It is circular and non Gaussian.
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20130227_BBNS_be…
If you want this fits file so you can use it with synfast and
alteralm, let me know and I'll point you to it.
-Chris
On Fri, May 24, 2013 at 3:06 PM, Walt Ogburn <ogburn(a)stanford.edu>
wrote:
Hi John,
Walt, are the WMAP and Planck input maps that you
are using for
the abscal analysis smoothed with gaussians? Or has the BBNS
profile become our new standard?
As I documented in my posting, I've been using circular Gaussians of
31.22 arcmin. How would I go about using the BBNS profile? I
imagine it's possible to prepare a set of B_l's that represent a
beam that's radially symmetric but non-Gaussian, with wings similar
to the BBNS beams, and then feed that to alteralms. Is that what
you have in mind?
- Walt
_______________________________________________
Bicep2-list mailing list
Bicep2-list(a)lists.fas.harvard.edu
https://lists.fas.harvard.edu/mailman/listinfo/bicep2-list
4:27 p.m.
Hi Walt,
I think it won't make much difference, but I'd be curious to see.
The beams files are located in:
~csheehy/bicep2_analysis/
beamfile_20130222_A.fits
beamfile_20130222_B.fits
beamfile_BBNS.fits
The A and B files are for A and B detectors. The BBNS file should be for
the A and B stacked beams. A and B I know are good. Please check that BBNS
is not much different than either A or B, though I think it should be fine.
In alteralm, you can just specify beam_file_out = ... instead of
fwhm_arcmin_out = ...
This is my input file for alteralm to make a WMAP map smoothed by the BBNS
A beam.
infile_alms =
/n/home03/rwa/wmap7yr/alms_wmap_band_iqumap_r9_7yr_V_v4_gal.fits
outfile_alms = alms_wmap_7yr_V_cel_BBNS_20130222_A.fits
beam_file_in = /n/home03/rwa/wmap7yr/bl_v_wmap_7yr.fits
#fwhm_arcmin_out = 31.22
beam_file_out = /n/home06/csheehy/bicep2_analysis/beamfile_20130222_A.fits
coord_in = G
coord_out = C
nsmax_in = 512
nsmax_out = 512
nlmax_out = 1000
-Chris
On Fri, May 24, 2013 at 3:18 PM, Walt Ogburn <ogburn(a)stanford.edu> wrote:
Hi Chris,
Yes, if you can point me to the fits file for the B_l's, that would be
great. I have only prepared input maps following a pretty limited set
of recipes, with circular Gaussian beams, so I'm not very familiar with
the other options that are available in alteralm. I take it that the
fits file you mention is a file with beam properties only, which can be
applied to any healpix map as a way of performing the beam smoothing?
If we get this working, it sounds as though it would be preferable to
(as John says) make it the standard for abscal. In that case I will
just drop the abscal sims I've been running with circular Gaussian
beams.
- Walt
On Fri, 24 May 2013 15:10:39 -0500
Chris Sheehy <csheehy(a)uchicago.edu> wrote:
--
**********************************************************************
Christopher Sheehy - Ph.D. candidate - University of Chicago
Research Specialist, University of Minnesota, Department of Physics
Room 220 Tate, 116 Church Street S.E. Minneapolis, MN, 55455
Tel: 612-625-1802 Fax: 612-624-4578 email: csheehy(a)uchicago.edu
**********************************************************************
Walt,
In my posting that John points out, the crux of it was that I
prepared a fits file for the B_ls of the azimuthally averaged BBNS
beam profile. It is circular and non Gaussian.
http://bmode.caltech.edu/~spuder/analysis_logbook/analysis/20130227_BBNS_be…
If you want this fits file so you can use it with synfast and
alteralm, let me know and I'll point you to it.
-Chris
On Fri, May 24, 2013 at 3:06 PM, Walt Ogburn <ogburn(a)stanford.edu>
wrote:
Hi John,
Walt, are the WMAP and Planck input maps that you
are using for
the abscal analysis smoothed with gaussians? Or has the BBNS
profile become our new standard?
As I documented in my posting, I've been using circular Gaussians of
31.22 arcmin. How would I go about using the BBNS profile? I
imagine it's possible to prepare a set of B_l's that represent a
beam that's radially symmetric but non-Gaussian, with wings similar
to the BBNS beams, and then feed that to alteralms. Is that what
you have in mind?
- Walt
_______________________________________________
Bicep2-list mailing list
Bicep2-list(a)lists.fas.harvard.edu
https://lists.fas.harvard.edu/mailman/listinfo/bicep2-list
5 Jun
5 Jun
3:32 a.m.
Hi John,
I've now put up a finer abscal comparison for ell=30-200.
Interestingly, it looks like Planck 143 (or 100) GHz and WMAP W band
are quite well matched within this range. The WMAP V band is the odd
man out, with a different slope at about the level we're interested in.
http://bicep0.caltech.edu/~spuder/analysis_logbook/analysis/20130604_abscal…
- Walt
4000
days inactive
4015
days old
bicep2-list@lists.fas.harvard.edu
7 comments
3 participants
participants (3)
-
Chris Sheehy -
John Kovac -
Walt Ogburn