search

CMB-S4 / s4mapbasedsims

CMB-S4 map based simulations
3 stars 1 forks source link

polarized dust doesn't show up correctly in SAT output foreground maps #4

Closed cbischoff closed 4 years ago

cbischoff commented 4 years ago
keskitalo commented 4 years ago

There seems to be something funny about the input maps. The dust input map is /global/cscratch1/sd/zonca/cmbs4/map_based_simulations/202006_foregrounds_extragalactic_cmb_tophat/512/dust/0000/cmbs4_dust_uKCMB_SAT-HFS2_nside512_0000.fits and, like you said, it should form the bulk of the combined foreground map at /global/cscratch1/sd/zonca/cmbs4/map_based_simulations/202006_foregrounds_extragalactic_cmb_tophat/512/combined_foregrounds/0000/cmbs4_combined_foregrounds_uKCMB_SAT-HFS2_nside512_0000.fits.

Here are some standard deviations measured in dust and combined_foregrounds:

                            |  full sky                     |   SAT HFS2 patch
                            |   I     |     Q    |     U    |   I     |     Q    |     U    |
dust                        |  3241   |   59.3   |   28.0   |  131.1  |    6.3   |    5.4   |
combined_foregrounds        |  3309   |   61.1   |   31.5   |  138.1  |   15.7   |   15.4   |
combined_foregrounds - dust |  93.9   |   14.4   |   14.4   |   17.7  |   14.4   |   14.4   |

So the combined foreground polarization in the SAT patch at HFS2 is dominated by something entirely different than dust. What do you think it is @zonca ?

keskitalo commented 4 years ago

Found it! The CIB component in /global/cscratch1/sd/zonca/cmbs4/map_based_simulations/202006_foregrounds_extragalactic_cmb_tophat/512/cib/0000/cmbs4_cib_uKCMB_SAT-HFS2_nside512_0000.fits has exactly same map in the I, Q and U columns of the FITS file. The standard deviation also matches the 14.4uK excess we saw above.

zonca commented 4 years ago

sorry, I noticed something in the CIB, see https://github.com/CMB-S4/s4mapbasedsims/issues/3, but haven't investigate yet.

zonca commented 4 years ago

there was a wrong broadcasting, so CIB which is T only was replicated on polarization as well.

I have rerun the SAT and generated CIB and the combined map again, can you please rerun your checks on these? Currently are on my scratch, if they are fine I'll run on LAT and copy to the project folder.

/global/cscratch1/sd/zonca/cmbs4/map_based_simulations/202006_foregrounds_extragalactic_cmb_tophat/512/cib/0000
/global/cscratch1/sd/zonca/cmbs4/map_based_simulations/202006_foregrounds_extragalactic_cmb_tophat/512/combined_foregrounds/0000
cbischoff commented 4 years ago

Yes, this looks much improved. Just to check, the CIB component is now totally unpolarized?

zonca commented 4 years ago

Yes, Q and U are 0 arrays

keskitalo commented 4 years ago

The new Pole SAT maps are here: /global/cscratch1/sd/keskital/s4sim/reference_tool/out_v2.

zonca commented 4 years ago

ok, I regenerated the foregrounds outputs at: /global/cscratch1/sd/zonca/cmbs4/s4refsimtool/s4_reference_design_foregrounds

if they are fine I'll copy to project

keskitalo commented 4 years ago

@zonca When will you have the LAT foreground maps prepared?

zonca commented 4 years ago

I'm trying on Jupyter, if that works, tonight, if not, Monday night

zonca commented 4 years ago

@keskitalo ok, LAT maps are available at /global/cscratch1/sd/zonca/cmbs4/map_based_simulations/202006_foregrounds_extragalactic_cmb_tophat/4096/combined_foregrounds/0000

keskitalo commented 4 years ago

All of the filtered foreground maps can be found here: /global/cscratch1/sd/keskital/s4sim/reference_tool/out_v2

zonca commented 4 years ago

thanks @keskitalo

LAT output maps are at /global/cscratch1/sd/zonca/cmbs4/s4refsimtool/s4_reference_design_foregrounds

I will run some validation and then copy everything to project, please write here if you have any more feedback.

jdborrill commented 4 years ago

Just to get us all in the right habits, we're actually running verification (did we do the thing right) on the maps now, not validation (did we do the right thing).

Julian

On Mon, Aug 10, 2020 at 9:40 AM Andrea Zonca notifications@github.com wrote:

thanks @keskitalo https://github.com/keskitalo

LAT output maps are at /global/cscratch1/sd/zonca/cmbs4/s4refsimtool/s4_reference_design_foregrounds

I will run some validation and then copy everything to project, please write here if you have any more feedback.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/CMB-S4/s4mapbasedsims/issues/4#issuecomment-671462945, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAC4LSUWJSF5QELNWUUEX23SAAPIZANCNFSM4PTX7XRQ .

zonca commented 4 years ago

@jdborrill correct, thanks

zonca commented 4 years ago
zonca commented 4 years ago

added gnomview plots to check consistency of input and output foregrounds: https://github.com/CMB-S4/s4mapbasedsims/blob/master/202006_reference_design/plot_gnomview/README.md

zonca commented 4 years ago

ok, this is completed, I posted link to the DM reference sims docs on the sim and analysis logbook: https://cmb-s4.org/wiki/index.php/Simulation_and_Forecasting_Logbook#Logbook_Entries_.28reverse_chronological.29