SubhaloFlag for central subhalos is 1?

Rui Lan Zhang
  • 1 Jul '22

From the following excerpt of the Data Specifications:

If zero, this subhalo should generally be excluded, and is not thought to be of cosmological origin.
If one, this subhalo should be considered a 'galaxy' or 'satellite' of cosmological origin.
(Note: always true for centrals).

Am I correct in interpreting this as: "SubhaloFlag" is always equal to 1 for central/primary subhalos?
And therefore subhalos whose indices are returned by "GroupFirstSub" (which are central/primary?) should have 1 as their "SubhaloFlag"?

Using the indices given by "GroupFirstSub", I find some instances where "SubhaloFlag" is 0.
I am using TNG300-1 and I give one example below.

halos["GroupFirstSub"][841]  # --> 207049
subhalos["SubhaloFlag"][207049] # --> 0
Dylan Nelson
  • 4 Jul '22

Yes you are correct, this does seem strange. I cannot remember immediately how this could happen.

What snapshot of TNG300-1? For snapshot 99, I see GroupFirstSub[841] = 648064.

Rui Lan Zhang
  • 4 Jul '22

Sorry, I forgot to mention the snapshot. It's snapshot 43.

Dylan Nelson
  • 6 Jul '22

Hi Rui,

You're right, there are a small number of central subhalos with SubhaloFlag==1.

The actual definition of this field is:

# construct SubhaloFlag (i.e. define)
# flag == True if satisfies all three conditions, otherwise False

SubhaloFlag = data['isSat_atForm'] * \
             (data['rad_rvir_atForm'] < 1.0) * \
             (data['dmFrac_atForm'] < 0.8)

SubhaloFlag = np.invert( SubhaloFlag.astype('bool') ) # 0=bad, 1=good

In particular, a subhalo can be flagged bad so long as it is a satellite subhalo at its 'time of formation'. I suspect, then, that any central subhalos which are flagged were previously satellites, i.e. these are backsplash-like galaxies (and/or structures which were then ejected out of the halo due to gravitational dynamics).

Rui Lan Zhang
  • 6 Jul '22

Ah I see. Thank you for looking into this for me!

  • Page 1 of 1