iCosmo: Galaxy redshift distribution

Use of Healpix, camb, CLASS, cosmomc, compilers, etc.
Post Reply
Charles Shapiro
Posts: 24
Joined: February 05 2005
Affiliation: University of Portsmouth

iCosmo: Galaxy redshift distribution

Post by Charles Shapiro » February 22 2009

Hello,

Can someone please explain how to generate lensing Cls using my own galaxy redshift distribution, p(z) ? It's not clear to me from the help messages.

Thanks a lot,
Chaz

Charles Shapiro
Posts: 24
Joined: February 05 2005
Affiliation: University of Portsmouth

iCosmo: Galaxy redshift distribution

Post by Charles Shapiro » February 22 2009

Actually I'm having a little trouble using the preset parameters, too. If I specify

Code: Select all

fid=set_fiducial(expt_in={sv1_n_zbin:1,sv1_dndzp:[1.,4.]})
sv=mk_survey(fid,'sv1')
then I get a whole bunch of these:

Code: Select all

% Program caused arithmetic error: Floating underflow
So it would be helpful to see some examples of the right way to set the galaxy distribution.

Thanks,
Chaz

Charles Shapiro
Posts: 24
Joined: February 05 2005
Affiliation: University of Portsmouth

iCosmo: Galaxy redshift distribution

Post by Charles Shapiro » February 23 2009

Ok, despite the (many!) warning messages, the Smail parameter inputs do seem to work.

However, when I specify a median redshift e.g.

Code: Select all

expt_in={sv1_z_med:1.1}
this input gets ignored. I've traced the problem to the "expt_" functions such as expt_generic.pro and expt_snap.pro. In these files, I find

Code: Select all

if tag_check(expt_in,'sv1_z_med',val=val) then sv1_z_m=val
where the variable sv1_z\_m seems like it should be sv1_z\_med. Likewise for sv2, sv3.

Chaz

Ivan Debono
Posts: 25
Joined: April 17 2007
Affiliation: APC, Paris 7 / OBSPM

iCosmo: Galaxy redshift distribution

Post by Ivan Debono » February 24 2009

There's a bug in the expt routines. The output variable should be sv1_z_med, sv2_z_med, etc.

Anais Rassat
Posts: 24
Joined: October 20 2004
Affiliation: LASTRO (EPFL)
Contact:

iCosmo: Galaxy redshift distribution

Post by Anais Rassat » March 04 2009

Hi,

Here is a link from djfanning's IDL website about getting rid of over/under flow errors in IDL.

http://www.dfanning.com/math_tips/underflow.html

There is indeed a bug in the expt_generic.pro routine, which will be corrected in the next version of iCosmo. In the meantime you can replace the following lines:

Code: Select all

if tag_check(expt_in, 'sv1_z_med', val=val) then sv1_z_m=val
by

Code: Select all

if tag_check(expt_in, 'sv1_z_med', val=val) then sv1_z_med=val 
and do this for lines corresponding to survey1, survey2 and survey3.
Anais.

Charles Shapiro
Posts: 24
Joined: February 05 2005
Affiliation: University of Portsmouth

iCosmo: Galaxy redshift distribution

Post by Charles Shapiro » May 10 2010

Looks like this never got answered...

What is the correct way to call mk_survey() in order to specify my own tabulated galaxy distribution, p(z)?

Thanks,
Chaz

Post Reply