Generating lensed hosts for all the Twinkle systems in "sprinkled_agn_230.txt" (HackWeek at Fermilab)

[linan7788626]

Questions for all hackers in this project.

• How to organize the catalogs to make the workflow more clear?
  • The current format of catalogs seems reasonably straightforward. If there are not more updates, we will use the ones from @jbkalmbach yesterday.
• How to calculate the intrinsic magnitudes of the source AGNs? mag_src = mag_lensed_AGN - 2.5 log10(1/magnification)?
  • Bryce created a new catalog for the host galaxies without magnifications, and this issue should be fixed.
• Whether there would be other ways of matching the Twinkles systems with their lens galaxies and host galaxies beyond using the 0th lensed AGN?

[linan7788626]

Hackers : @linan7788626, @mpwiesner, @jbkalmbach, @drphilmarshall, @bnord

Introduction

This project is following the hackday project in the DESC meeting at SLAC. We are going to produce lensed host galaxies based one the catalogs we created in the hackday. The goal of this work is generating lensed host galaxies for lensed AGNs, which is a preparation for the next stage of the Twinkles simulation.

The basic idea is:

Create a one-to-one relation between the 0th lensed AGN and the other components of the Twinkle systems.

Inputs

• sprinkled_agn_230.txt

  • AGN_IDs, Twinkle_IDs, and other properties of the lensed AGNs...

• sprinkled_lens_galaxies_230.txt

  • All the parameters of the images of lens galaxies.

• sprinkled_hosts_230.txt

  • All the parameters of the host galaxies, including multiple components.

• twinkles_lenses_v2.fits

  • Details of the mass model of the twinkle systems, the positions and magnifications of source AGNs.

Output

• host_cat = (x1, x2, mag, Reff_a, Reff_b, zl, ndex, SED, Twinkle_ID, Lens_ID)
• lens_cat = (x1, x2, vd, ql, phl, zl, ext_shears, phiG, Twinkle_ID, Lens_ID)
• lens_gal_cat = (y1, y2, mag, Reff_a, Reff_b, zs, ndex, SED, Twinkle_ID, Lens_ID)
• lensed_hosts_cat
  • Fits file as a 2d histogram for shooting photos in PhoSim
  • [Unique_lensID] [lensra] [lensdec] [mag] [SED] [zs] _ [PixelSize] . fits

The Goal by the end of this Hackweek

• [x] Create all the parameters of creating instance catalogs for PhoSim

  • [x] Sprinkled AGNs.
  • [x] Images of lens galaxies.
  • [x] Stamps and parameters of lensed host galaxies.
    • [x] Create the images of lensed host galaxies.
    • [x] Extract all the parameters we need for the stamps above.

• [x] If we have time, we will feed the instance catalogs of lensed AGNs, lens galaxies, and lensed host galaxies into PhoSim to Create a more realistic mock image for Twinkles. The field of view would be the same as that covered by the sprinkled AGNs in the catalog above.

Here is an image of a lensing system at (53.065667, -27.479433), output by PhoSim. Left is images of AGN and lensing galaxy. Right also includes the lensed host. We now have many of these, this is just a sample. This is a gri image and represents a 30 second exposure time.

Future Work

• Double check the relative positions of lensed AGNs to the lensed host galaxies, to make sure the lensing calculation in the OM10 and this project are exactly the same.
• Merge all the lensed host galaxies and into the current Twinkles Simulation. A Sample Image of Twinkles
• Catalogrize the lensed host galaxies and make an updated OM10?

[drphilmarshall]

Rock. ON. This is going to be great! Let's try hard to get the PhoSim outputs in place: I really want to see how visible these host galaxies are, by eye, preferably by blinking back and forth between two gri color composite cutout image PNGs :-) Even if we don't get that far, a key aspect of this project is visualization: let's display the different components of the cutout image as we make them, to check that we are building up sensible things. This will also work best if we use gri composites, always with the same cutout field of view and the same color scales and nonlinearity. We have code to make these, HumVI, but it will need upgrading to work with the LSST/PhoSim/whatever format headers that our FITS files have.

[linan7788626]

Hi @drphilmarshall

Thank you for your suggestion. Yes, you are right, let's try to get the PhoSim outputs first. We all are interested in the visibility of the lensed host galaxies, but we may go further if there is enough time. :-).

Regardless, this is just the first version of the task list, if you would like to update it with your ideas and suggestions, I really appreciate it. Looking forward to meeting you at Fermilab next week. :-)

Best, Nan

[jbkalmbach]

Hi @linan7788626 here is an updated version of sprinkled_hosts_230.txt with no magnification in the AGN hosts. It was created using the notebook I made and am merging shortly.

sprinkled_hosts_230.txt

[linan7788626]

Hi, @jbkalmbach, thank you for the updated catalog, it is very helpful. I am working on wrapping up all the codes for this project and will post the updates here tomorrow.

[jbkalmbach]

sprinkled_agn_230.txt sprinkled_lens_galaxies_230.txt

Here are updated agn and lens catalogs as well. You can match the AGN and the AGN hosts to the lens using lens_galaxy_uID column.

[linan7788626]

Fantastic, Thanks a lot.

[linan7788626]

Hi @mpwiesner,

Please check out these files from Bryce:

sprinkled_agn_230.txt sprinkled_lens_galaxies_230.txt

0. We plan to feed the instance catalogs into PhoSim in three steps:

   • Lensed AGNs
   • image of lens galaxies
   • lensed host galaxies

1. You can find all the parameters you need for creating the instance catalogs of lensed AGNs and lens galaxies for PhoSim. Let's create these instance catalogs first.

2. I have obtained all the fits files and needed parameters for creating the instance catalogs of lensed host galaxies. When you are ready for this, please download the data below. It includes all the stamps of the lensed components of the host galaxies. And the parameters are shown in the file name as the following:

lensed host galaxies

[Unique_lensID] [lensra] [lensdec] [mag] [SED] [zs] _ [PixelSize] . fits

Thanks, Nan

[mpwiesner]

@linan7788626 and @jbkalmbach this looks great. I can easily take these all and make instance files and run PhoSim. Is it correct that our end goal is 199 individual FITS images, one for each lensing system?

[linan7788626]

Hi, @mpwiesner , in my opinion, the end goal is not 199 individual FITS images, but a patch of the sky including all the Twinkle systems. Probably, it would be similar to the sample picture in the description of Twinkles.

Hi @jbkalmbach and @drphilmarshall , could you please confirm the end products of this hackweek is 199 individual FITS images or a patch of the sky including 199 twinkles system? Thank you so much!

Best, Nan

[drphilmarshall]

I'd say go for the sensor image - we can make cutout images later. But either works - do whatever is fastest and easiest! :-)

On Apr 5, 2017 3:25 PM, "Nan Li" notifications@github.com wrote:

Hi, @mpwiesner https://github.com/mpwiesner , in my opinion, the end goal is not 199 individual FITS images, but a patch of the sky including all the Twinkle systems. Probably, it would be similar to the sample picture in the description of Twinkles.

Hi @jbkalmbach https://github.com/jbkalmbach and @drphilmarshall https://github.com/drphilmarshall , could you please confirm the end products of this hackweek is 199 individual FITS images or a patch of the sky including 199 twinkles system? Thank you so much!

Best, Nan

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/LSSTDESC/Twinkles/issues/443#issuecomment-291986190, or mute the thread https://github.com/notifications/unsubscribe-auth/AArY9ykfZjSAv-hyYT3CZ8y3ku4Uz2o1ks5rs_ixgaJpZM4Mo0X8 .

[mpwiesner]

Here is an image of a lensing system at (53.065667, -27.479433), output by PhoSim. Left is images of AGN and lensing galaxy. Right also includes the lensed host. We now have many of these, this is just a sample. This is a gri image and represents a 30 second exposure time.

[linan7788626]

Hi @mpwiesner , the new images are good, but we need to double check something first.

• Left Image:

  • Is blue one lensed AGN?
  • Is there only one lensed AGN?
  • Is the yellow ball the image of lens galaxy?
  • Is another lensed AGN obstructed by the lens galaxy?

• Right Image:

  • The relative positions of the lensed hosts to the lens galaxy are reasonable to me.
  • However, there is a considerable offset between the lensed AGNs and lensed hosts, and this might be the issue mentioned by @drphilmarshall yesterday. But It could be because of the incorrect position of the lens galaxy. Did you obtain the position of lens galaxy from this catalog sprinkled_lens_galaxies_230.txt? I did not found a lens galaxy at (53.065667, -27.479433).

In my opionion, the critical issue in current results is that Making sure the locations of lensed AGNs and lens galaxies are correct. If you could give me any details, it should be awesome.

Thanks, Nan

[mpwiesner]

One more of several hundred lensing systems we simulated, this one at ( 52.920242, -27.427762).