Public Data Access Overview / Discussion Forum

Hi,

I am trying to identify a disk galaxy using CircAbove07Frac, however I dont understand how to actually call it?

I am trying to do something along the lines of:

fields=['CircAbove07Frac']

diskiness=il.snapshot.loadSuhalo(basePath, int(input("SnapNum: ")),int(input("ID: ")),'stars',fields = fields)

however this obviously doesnt work. Basically how can I access the CircAbove07Frac for a given input galaxy?

Thanks

Hi Leo,

This data about circularities is in the "supplementary data catalog", not part of the original group catalog (so loadSubhalo won't work). Have you downloaded it for the simulation you want already? If so, e.g.

with h5py.File('stellar_circs.hdf5','r') as f:
    SubfindID = f['Snapshot_99/SubfindID'][()]
    CircAbove07Frac = f['Snapshot_99/CircAbove07Frac'][()]

target_subhalo_id = 12345
w = np.where(SubfindID == target_subhalo_id)
print('circ for this subhalo: ', CircAbove07Frac[w])

Hi,
Thanks for your response.

I am getting the error:

UnicodeDecodeError: 'utf-8' codec can't decode byte 0x89 in position 0: invalid start byte

I tried changing 'r' to 'rb' but it gave the same error. If you (or anyone) have any ideas it would be greatly appreciated.

Thanks

Hi Leo,

Not sure, that doesn't make much sense. What is the full output of:

f = h5py.File('stellar_circs.hdf5','r')
x = f['Snapshot_99/SubfindID'][()]

If it makes an error, either the file is wrong/corrupt (try redownloading), or your install of python/hdf5/h5py is wrong/corrupt.

Hi Dylan,

Late reply - I got all that stellar_circs stuff working so thank you for that, I just have a question on the Subfind/subhaloIDs.

If I have a subfindID of say 1000 in snapshot 50, and I want to extract the CircAbove07Frac for that galaxy throughout all the snapshots, I believe I need to:
1) Identify the subhaloID of the galaxy
2) use the subhaloID and the snapshot number for each point in time to find the corresponding subfindID for that galaxy at each snapshot
3) use the subfindID to extract the CircAbove07Frac for the galaxy at each snapshot.

Is this correct? I am just a bit confused as to how the different IDs work.

Thanks

Hi Leo,

"subfindID" and "subhaloID" are going to be the same thing. So you would follow the first code snippet, replacing 12345 by 1000 and snapshot 99 by 50. If you want to follow this object through time, you need to load it's merger tree. Once you load the main progenitor branch (backwards in time), you have two arrays SnapNum and SubfindID. At each snapshot, you re-do the same process to locate where, in the stellar circs catalog, that subfind ID is located.

Hi Dylan,

Many thanks for your help.

I am trying to identify how many Galaxies in our sample of Fof/Groups (MW analogues, based on MC200) are Disk Galaxies.

However, our sample was based on the Illustris-TNG-1-Dark and the CircAbove07Frac field found under (c) Stellar Circularities, Angular Momenta, Axis Ratios is calculated from the baryonic simulations.

2 questions please:

(1) is there a 1to1 mapping between Illustris-TNG-1Dark and Illustris-TNG-1-baryonic?

(2) in case there is what is the criteria applied to CircAbove07Frac in order to define a Disk Galaxy?

Regards,
André

(1) The initial conditions for TNG100-1 and TNG100-1-Dark are identical, so yes all (well resolved) halos will exist in both simulations, i.e. with about the same mass, and in the same location. The easiest way to match (for FoF halos) is simply to search in the other simulation for the closest halo within a reasonable mass range (e.g. +/- 0.2 dex). This should work for the vast majority of halos.

(2) You can apply whatever criterion to CircAbove07Frac that you wish, depending on how conservative/aggressive (i.e. false positives vs false negatives) you want to be. As a starting point, you could plot the histogram of all such values, and see if there is a natural choice (i.e. halfway between two peaks representing disks and spheroids).

Thanks Dylan,

(1) by "closest halo" should we use GroupPos ([x,y,x]] ?; and which Mass value should we use?

(2) since we already have a sample of DM FoFs, can we use their Sub-Halo DM (centrals) and then try to map these to their corresponding baryonic Sub-Halos (using (1)) -> so that we can then get CircAbove07Frac?

I am struggling to understand why we need to follow:
[path A] DM FoF sample <-[map]-> Baryonic FoF -> get SubHalos -> getCircAbove07Frac
instead of
[path B] DM FoF sample -> DM Centrals/SubHalos <-[map]-> Baryonic SubHalos ->getCircAbove07Frac

Otherwise, could we just use the (d) Subhalo Matching Between Runs Catalog?

Yes I would try to use "(d) Subhalo Matching Between Runs Catalog" and see if it suitable for your purpose. Otherwise, you can try either of the approaches above, and see how they may differ.

Thanks Dylan, using (d) seems to work (since all IDs are found). The Original distro of getCircAbove07Frac has right tails with a max at 0.15 whereas the MW Sample shows a symmetric distro with max at 0.5 and left tails. However the latter is not showing any bimodality, discerning disks and spheroids.