Tracking all subhalos of specific halo at different snapshots
Neda Stojkovic
12 Mar '20
Hello,
I'm trying to find all of the subhalos (their masses and radii) of one specific halo at different snapshots. Is there a way to do this without downloading all of the group catalogs, with only complete merger tree? It's easy to do this with all of the group catalogs and you don't even need merger tree, but I have limited space and that data is somewhere inside the tree.
Another question: at snapshot 135 there is a subhalo with subfind id 2. Beginning with snapshot 112 there is two strange continual decreasing of the subhalo's mass (dark matter mass mostly) - between 112. and 117. snapshot its mass dips from order of magnitude 10^13 to order of magnitude 10^12 (with lowest mass at 114. snapshot), and at 117. snapshot its mass returns to order of magnitude of 10^13 Msun. The second dip is at 131. snapshot, and mass continually decreases even at 135. snapshot, going from order of magnitude 10^13 to 10^12 Msun. The dips are continuous, not discrete, so it's not subhalo switching problem, and the quick check of SubhaloParent told me that the subhalo is central in its halo in all of the relevant snapshots (110-135) (excluding one discrete dip at 120. snapshot, so a subhalo switching problem here I guess), so it's not decreasing of the mass because of the halo finder assigning the most of the mass to the primary subhalo (central). What could be the cause of this decreasing?
Thank you so much in advance!
Neda
Neda Stojkovic
13 Mar '20
Also, related to the second question, the subhalo does not have merger with mass ratio greater then 0.1 in period after snapshot 110. Also, there is halo merger between snapshots 119 and 120 (between haloes 18 and 0) through which subhalo becomes part of the halo 0.
Dylan Nelson
13 Mar '20
Hi Neda,
You can download subsets of the group catalogs, one field at a time. So I'd suggest to do four of these, to get: GroupFirstSub, GroupNsubs, SubhaloMassInRadType, SubhaloHalfmassRadType (for example). Use the first two, given the FoF halo ID at each snapshot, to get its list of subhalo indices. Then use the second two to get their masses and radii.
If you see a smooth decrease of a satellite subhalo total/dark matter mass with time, I would think this indicates (tidal) stripping as it accretes towards the center of its host halo. Note that it also becomes very hard to "find" the outskirts (weakly bound particles) of a satellite when it is deep inside a massive host, so its total mass can also decrease somewhat due to this numerical limitation (in Subfind).
Neda Stojkovic
13 Mar '20
Thank you for ideas for downloading subsets of group catalogs!
Related to the second question: I'm familiar with the phenomenons of tidal stripping of satellite mass, and also with limitations of the subfind algorithm, but this is not the case, because the subhalo I'm analyzing is central of the halo, and the most massive. Is there another mechanism that would strip mass from the central subhalo continuously?
Dylan Nelson
13 Mar '20
I presume it must be environmental - it could be interacting with a nearby halo, but remain a central.
Hello,
I'm trying to find all of the subhalos (their masses and radii) of one specific halo at different snapshots. Is there a way to do this without downloading all of the group catalogs, with only complete merger tree? It's easy to do this with all of the group catalogs and you don't even need merger tree, but I have limited space and that data is somewhere inside the tree.
Another question: at snapshot 135 there is a subhalo with subfind id 2. Beginning with snapshot 112 there is two strange continual decreasing of the subhalo's mass (dark matter mass mostly) - between 112. and 117. snapshot its mass dips from order of magnitude 10^13 to order of magnitude 10^12 (with lowest mass at 114. snapshot), and at 117. snapshot its mass returns to order of magnitude of 10^13 Msun. The second dip is at 131. snapshot, and mass continually decreases even at 135. snapshot, going from order of magnitude 10^13 to 10^12 Msun. The dips are continuous, not discrete, so it's not subhalo switching problem, and the quick check of SubhaloParent told me that the subhalo is central in its halo in all of the relevant snapshots (110-135) (excluding one discrete dip at 120. snapshot, so a subhalo switching problem here I guess), so it's not decreasing of the mass because of the halo finder assigning the most of the mass to the primary subhalo (central). What could be the cause of this decreasing?
Thank you so much in advance!
Neda
Also, related to the second question, the subhalo does not have merger with mass ratio greater then 0.1 in period after snapshot 110. Also, there is halo merger between snapshots 119 and 120 (between haloes 18 and 0) through which subhalo becomes part of the halo 0.
Hi Neda,
You can download subsets of the group catalogs, one field at a time. So I'd suggest to do four of these, to get:
GroupFirstSub,GroupNsubs,SubhaloMassInRadType,SubhaloHalfmassRadType(for example). Use the first two, given the FoF halo ID at each snapshot, to get its list of subhalo indices. Then use the second two to get their masses and radii.For the details, search for
{subset_query}on the API documentation page.If you see a smooth decrease of a satellite subhalo total/dark matter mass with time, I would think this indicates (tidal) stripping as it accretes towards the center of its host halo. Note that it also becomes very hard to "find" the outskirts (weakly bound particles) of a satellite when it is deep inside a massive host, so its total mass can also decrease somewhat due to this numerical limitation (in Subfind).
Thank you for ideas for downloading subsets of group catalogs!
Related to the second question: I'm familiar with the phenomenons of tidal stripping of satellite mass, and also with limitations of the subfind algorithm, but this is not the case, because the subhalo I'm analyzing is central of the halo, and the most massive. Is there another mechanism that would strip mass from the central subhalo continuously?
I presume it must be environmental - it could be interacting with a nearby halo, but remain a central.
Thank you so much for your help!