Public Data Access Overview / Discussion Forum

My name is Jose Benavides, a PhD student at the Universidad Nacional de Córdoba (Argentina). I notice that within a FoF group their SubhaloParent index can take a value from 0 to any number, but not necessarily all numbers exist in this range (for example: there are subhalos with Parent 0,1,2,6,10,12, .. ., but there is no 3,4,5,7,8,9,11, ...) why? What does it mean that the members of a group have associated a SubhaloParent 0 or 1 or 2, ...? There may even be a single subhalo with a unique parent number. Initially we think that these parent indexes refer to substructures within the FoF groups (giving them a more physical criterion), something like:

Subhalos with SubhaloParent = 0 to a "subgroup 0" within the "groupFoF 0" (for example: SubhaloGrNr == 0 in a specific snap)),

Subhalos with SubhaloParent = 1 to a "subgroup 1" within "groupFoF 0" (for example: SubhaloGrNr == 0 in a specific snap)),

and so on.

I am not sure if we are going the right way, I would appreciate it if you can give me some help about the meaning (physical or computational for example) of this SubhaloParent.

Cordial greetings and thanks,

Jose.

Hi Jose,

The description of SubhaloParent which I suspect you've already seen is:

Index into the Subhalo table of the unique SUBF host/parent of this Subhalo. This index is local to the group (i.e. 2 indicates the third most massive subhalo of the parent halo of this subhalo, not the third most massive of the whole snapshot). The values are often zero for all subhalos of a group, indicating that there is no resolved hierarchical structure in that group, beyond the primary subhalo having as direct children all of the secondary subhalos.

Note that we very rarely use this field - it would only be needed if you are searching for subhalos of subhalos (i.e. hierarchical substructure). Perhaps if you can say more what you are trying to do, I can suggest something.

Hi Dr. Nelson,

Thanks for the last message. We have been working analyzing clusters of galaxies that are accreted by clusters (Groups FoF), we have used the variable SubhaloParent to identify the subhalos associated to a host galaxy (substructures within the groups). But I would like to know, what is the form or criterion to associate a subhalo with another?

Thank you,

Hi Jose,

The determination of this field comes from the Subfind algorithm, for details you can read Springel+ (2001), section 4.2 in particular.

Your approach is interesting and might work, but it isn't clear to me that all the member galaxies of the group, which then falls into a larger cluster, are going to remain associated (according to subfind).

An alternative would be to identify the group prior to infall, e.g. when its distance is 3rvir from the future cluster parent, then track it (and all its member subhalos) forward in time using the merger trees, to see where they end up and how they evolve.

Hi Dr. Nelson,

Thank you very much, together with my director we defined a criterion for the time of infall (doing tests with 2R_vir, 3R_vir, ...), we made the identification of the FoF group of these subhalos and followed their "satellites" in the time as you mention, We have found interesting things, but we had doubts about this variable (SubhaloParent). Already in the past I had reviewed this article by Springel (it is very good), I will revise it to better understand this variable.
Thank you.

Hi Dylan,

I would like to identify the splashback halos in the TNG300-1. I was thinking that 'SubhaloParent' might be useful for this (although I have the same problem than several users, and I do not fully understand the meaning: Index into the Subhalo table of the unique Subfind parent of this Subhalo).

So, I'm using the merger trees for the main branch of the central galaxies of each halo. A part of the code:

**
GroupFirstSub = il.groupcat.loadHalos(basePath,99,fields=['GroupFirstSub','Group_M_Crit200','GroupNsubs'])
FirstSub = GroupFirstSub['GroupFirstSub']

fields = ['SubhaloParent']
for i in range(0, Ntot):
tree = il.sublink.loadTree(basePath2,99,FirstSub[i],fields=fields,onlyMPB=True)
**

I've printed some of the SubhaloParent values together with the snapshot number:

SubhaloParent1 = [2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
Snapshotnumber1 = [99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 80 79 78 77 76 75
74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50
49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25
24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2]

SubhaloParent2=[0 0 0 2 2 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
Snapshotnumber2 = [99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 80 79 78 77 76 75
74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50
49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25
24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8]

SubhaloParent3 = [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]
Snapshotnumber3 = [99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 80 79 78 77 76 75
74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50
49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25
24 23 22 21 20 19 18 17 16 15 14 13 12 11 9]

So, what does it mean?
For the first one: was the parent subhalo a satellite in the two previous snapshots? (when SubhaloParent = 2)
For the second one: had it also a period in which it was a satellite? (when SubhaloParent = 2)
For the third one: it was always a central, right?
Do you think that SubhaloParent can be used to track the splashback halos?

Thanks!
Celeste

Hello,

You shouldn't use SubhaloParent, this is a hierarchical index, so it is meant to identify a subhalo of a subhalo (this is a rare occurrence in these types of simulations).

Better to just use SubhaloGrNr, which tells you the parent FoF, then of this group use GroupFirstSub. If this index points back to the subhalo you started from, it is a central. if not, then not.

Hi Maria,
To identify "backsplash" subhalos I would recommend tracing the orbits and seeing if they are "recent infalls" or satellites that have already had at least a first entry into the Rvir of the halo (true "backsplash"). Do not use the "SubhaloParent", because this variable only associates subhalos with other subhalos by proximity criteria but they do not necessarily have an important interaction along the time.

Thanks Dylan and Jose!!