If, HaloID is not a field, it is implicit ie: is the index of any Group* dataset, i.e. the row number at any given redshift, then how can we trace back the accretion history of any given Halo (considering the number of Halos differs across redshifts)? Reading through the forums it seems that amongst the options, we can either:
(1) identify the most significant (central) subhalo within any FoF/Group and then use their Merger-Tree information (which brings the question of whether any given central sub-halo at z=N will always be central at higher red-shifts z>N), another method,
(2) proposed by Rodriguez-Gomez* et al, in section 3.3, “Merger trees of halos” where a splitting algorithm is used across the sub-halos merger trees of any given Halo to build the Halo level merger-tree, and finally, also on this forum,
(3) and specifically for sub-Halos, is suggested one loads the "main progenitor branch" of the SubLink merger tree - but how does this work at the Halo level?
Reference*: “The merger rate of galaxies in the Illustris Simulation: a comparison with observations and semi-empirical models”, Rodriguez-Gomez et al in Monthly Notices of the Royal Astronomical Society, Volume 449, Issue 1, 1 May 2015
May you please provide some guidance on how to gather the accretion info / history for any given halo, whether (1),(2),(3) or if there are other methods, more tailored for Halos etc...
Regards (and Merry Xmas)
André
Dylan Nelson
28 Dec '22
Hello Andre,
(2) While one could construct a halo-based merger tree, we don't have any such structure available. All our merger trees are subhalo based.
(1,3) As you suggest, the best way is to simply go to the central of the halo of interest (via GroupFirstSub), then look at its main progenitor branch. For each subhalo along the MPB, you can look at the parent halo (via SubhaloGrNr). This is essentially a "halo MPB". You are not guaranteed that subhalos along a MPB are always the central of their halos, but I expect it would be rare otherwise, and this procedure work for most cases.
Hi All,
If, HaloID is not a field, it is implicit ie: is the index of any Group* dataset, i.e. the row number at any given redshift, then how can we trace back the accretion history of any given Halo (considering the number of Halos differs across redshifts)? Reading through the forums it seems that amongst the options, we can either:
(1) identify the most significant (central) subhalo within any FoF/Group and then use their Merger-Tree information (which brings the question of whether any given central sub-halo at z=N will always be central at higher red-shifts z>N), another method,
(2) proposed by Rodriguez-Gomez* et al, in section 3.3, “Merger trees of halos” where a splitting algorithm is used across the sub-halos merger trees of any given Halo to build the Halo level merger-tree, and finally, also on this forum,
(3) and specifically for sub-Halos, is suggested one loads the "main progenitor branch" of the SubLink merger tree - but how does this work at the Halo level?
Reference*: “The merger rate of galaxies in the Illustris Simulation: a comparison with observations and semi-empirical models”, Rodriguez-Gomez et al in Monthly Notices of the Royal Astronomical Society, Volume 449, Issue 1, 1 May 2015
May you please provide some guidance on how to gather the accretion info / history for any given halo, whether (1),(2),(3) or if there are other methods, more tailored for Halos etc...
Regards (and Merry Xmas)
André
Hello Andre,
(2) While one could construct a halo-based merger tree, we don't have any such structure available. All our merger trees are subhalo based.
(1,3) As you suggest, the best way is to simply go to the central of the halo of interest (via GroupFirstSub), then look at its main progenitor branch. For each subhalo along the MPB, you can look at the parent halo (via SubhaloGrNr). This is essentially a "halo MPB". You are not guaranteed that subhalos along a MPB are always the central of their halos, but I expect it would be rare otherwise, and this procedure work for most cases.
Thanks Dylan,
will try this approach.
Regards