The Millennium Simulation data, exactly as plotted in
Figures 1,2,4,5,6 and 7 of astro-ph/07081367
(Cole, Helly, Frenk and Parkinson 2008 MNRAS 383,546),
is tabulated in the following data files.
Figure 1: The conditional mass function histograms
Figure 2: The scaled conditional mass functions histograms
Figure 4: Mass distributions
of first and second most massive progenitors.
Figure 5: The final mass distribution
histograms
Figure 6: Redshift distribution histograms
of the most recent major mergers.
Figure 7: The distribution
of normalized accretion rates as
a function of redshift.
Acknowledgement: Work
that makes use of these data should reference the paper in which they
are
presented,
Cole, Helly, Frenk and Parkinson 2007 MNRAS 383, 546.
The friends-of friends N-body halo merger trees used in Cole, Helly,
Frenk and Parkinson
(2008) are available on the web from MyMillennium
.
Through this site it is possible to query an SQL database which
contains halo and galaxy catalogues derived from the Millennium
simulation. Access to the full database requires an account, which can
be obtained by sending an email to j.c.helly AT durham.ac.uk.
There is also a version of the database, based on the smaller
"milli-millennium" simulation, which doesn't require an account
at: milli-millennium.
The only difference between this and the full database is that the
simulation box is only 62.5Mpc/h across rather than 500Mpc/h.
Database layout
It is possible to download merger trees, calculate statistics or
extract subsets by running SQL queries on the appropriate database
tables. The friends of friends (FoF) merger trees used in Cole et al
(2008) are stored in the FoFHalo table in the FoFTrees database.
Each row in the table represents one FoF group. The columns store
various properties of the groups. The columns which define the merger
tree structure are:
- FoFHaloID - This is a
unique identifier for the halo. They are unique
across the whole simulation and over all snapshots, so the 'same' halo
does not retain its ID number between snapshots. These IDs are assigned
by walking the merger tree in depth first order, which ensures that the
IDs of the progenitors of a particular halo form a continuous sequence.
- DescendantID - The
FoFHaloID of the descendant of this halo
- LastProgenitorID -
The highest FoFHaloID of any progenitor of this halo
- EndMainBranchID -
The highest FoFHaloID of any progenitor on the main
branch of the merger tree of this halo.
- SnapNum
- the snapshot this halo exists at.
The progenitors of a halo can be found by locating all haloes with FoFHaloID between
the FoFHaloID and the LastProgenitorID of the descendant halo. The main
progenitor branch for a particular halo can be found in the
same way, if EndMainBranchID is used instead of LastProgenitorID.
Example queries
These queries access the milli-millennium database so they can be run
without an account.
The following query could be used to locate all progenitors at z=1
(snapshot 41) of haloes at z=0 (snapshot 63) with masses greater than
1.0e14 Msolar/h:
select *
from
millimil..FoFHalo as d,
millimil..FoFHalo as p
where
d.snapnum =63 and p.snapnum = 41
and p.fofhaloid between d.fofhaloid and d.lastprogenitorid
and (d.np*8.61e8) > 1.0e14
The constant 8.61e8 is the particle mass in Msolar/h.
The SQL 'group
by' command can be used to make the query return a histogram of
progenitor masses:
select
floor(log10(p.np*8.61e8)/0.2)*0.2 as logm,
count(*) as n
from
millimil..FoFHalo as d,
millimil..FoFHalo as p
where
d.snapnum =63 and p.snapnum = 41
and p.fofhaloid between d.fofhaloid and d.lastprogenitorid
and (d.np*8.61e8) > 1.0e14
group by
floor(log10(p.np*8.61e8)/0.2)*0.2
order by
floor(log10(p.np*8.61e8)/0.2)*0.2
Here, the progenitors are grouped according to log(mass) rounded
down to the nearest 0.2, and the query returns the number of progenitors
in each group. The 'order by' is just to ensure that the rows returned
are sorted in ascending order of mass.