[yt-users] HaloProfiler ActualOverdensity
Britton Smith
brittonsmith at gmail.com
Wed Dec 2 15:08:49 PST 2009
Shankar,
The overdensity in the profile is referring to dark matter + baryons. The
field called "Density" is just baryon density.
The nans come from spherical shells of the radial profile that had nothing
in them. If you look at the TotalMassMsun field, you'll notice that it is
flat over the points where you get the nans and the value of myweight is 0.
Britton
On Wed, Dec 2, 2009 at 3:49 PM, Eric Hallman <Eric.Hallman at colorado.edu>wrote:
> For question 1, you would use that only if you believe the *correct* OD for
> virialization is 200 only with respect to either mean matter density or
> critical density. The number 200 is somewhat arbitrary. But for your
> purposes, your conversion is correct. As to which number you should use for
> virial overdensity, I would recommend looking through linear collapse theory
> in an expanding universe if you are concerned at this level which is correct
> (e.g. Peebles or the Press-Schechter papers, or any of the new ones)
>
>
> On Dec 2, 2009, at 3:40 PM, Agarwal, Shankar wrote:
>
> Britton and Eric,
>
> So I guess if the convention in halo_profiler is (Baryon + Dark Matter
> Density) / (Mean density) , then I should use
>
> virial_overdensity = 200/Omega_matter = 200/0.258 = 775
>
> instead of 200.
>
>
>
> A separate stupid question :
>
> I have run a Box=200Mpc/h with 512^3 dark matter particle simulation.
> http://drop.io/slice
> The most dense point in the box is only 1.3e-27 gm/cm^3, which is about
> 520*matter_mean_density (for omega_matter=0.26).
>
> Then I ran halo_profiler.py on it. Here are the first few lines...
>
>
> ActualOverdensity CellVolume Density RadiusMpc
> Temperature TotalMassMsun myweight
> 1.411083e+03 1.884580e+73 5.022445e-28 1.086579e+00
> 3.470612e+07 3.350352e+13 4.667973e+12
> 1.349560e+03 2.826870e+73 2.846061e-28 1.108807e+00
> 2.697442e+07 4.806416e+13 1.054647e+12
> 1.199229e+03 3.298015e+73 3.425752e-28 1.131491e+00
> 2.577368e+07 4.982850e+13 8.115086e+11
> 1.029234e+03 5.653740e+73 2.792583e-28 1.154639e+00
> 3.303925e+07 7.331169e+13 2.847611e+12
> 1.135776e+03 7.538320e+73 3.070320e-28 1.178260e+00
> 3.784802e+07 1.078674e+14 2.675091e+12
> 1.135776e+03 7.538320e+73 nan 1.202365e+00 nan
> 1.078674e+14 0.000000e+00
> 1.135776e+03 7.538320e+73 nan 1.226962e+00 nan
> 1.078674e+14 0.000000e+00
> 1.018853e+03 8.951755e+73 3.040255e-28 1.252063e+00
> 3.025227e+07 1.149060e+14 1.974464e+12
> 9.159449e+02 1.083633e+74 1.950294e-28 1.277677e+00
> 3.573580e+07 1.250475e+14 1.774217e+12
> 9.158806e+02 1.177862e+74 1.275998e-28 1.303816e+00
> 3.568055e+07 1.359117e+14 5.672694e+11
> 8.432311e+02 1.413435e+74 1.915018e-28 1.330489e+00
> 2.360579e+07 1.501570e+14 1.880053e+12
> 7.995894e+02 1.649007e+74 2.735319e-28 1.357708e+00
> 2.651133e+07 1.661165e+14 2.674878e+12
>
>
> Looking at ActualOverdensity column, Is it possible to have a halo with
> overdensities in 1000's when the peak overdensity is 520.
>
> And do the "nan" correspond to some kind of profile maxima ?
>
>
> regards
> shankar
>
>
>
>
> -----Original Message-----
> From: yt-users-bounces at lists.spacepope.org on behalf of Eric Hallman
> Sent: Wed 12/2/2009 12:56 PM
> To: Discussion of the yt analysis package
> Subject: Re: [yt-users] HaloProfiler ActualOverdensity
>
> Shankar and Britton,
> I have gone through this argument extensively, and discovered that
> both definitions are used, with roughly equal frequency in the
> literature. Lately, it has been trending toward OD = density/mean
> density (matter). I think the main reason for the density with respect
> to critical in the past has been due to an historical preference for
> omega_m=1 universes (SCDM).
>
> cheers,
>
>
> On Dec 2, 2009, at 11:50 AM, Britton Smith wrote:
>
> Shankar,
>
>
> As I stated in my previous email, you will find both definitions
>
> used in the literature. I'm not going to spend time in a citation
>
> battle. The good news is that both definitions differ only by a
>
> factor of Omega_matter, so you can change the value of
>
> virial_overdensity accordingly.
>
>
> Britton
>
>
> On Wed, Dec 2, 2009 at 11:27 AM, Agarwal, Shankar <sagarwal at ku.edu>
>
> wrote:
>
> Hi Britton,
>
>
> I just wanted a clarification on the definition of
>
> ActualOverdensity. In your mail, you said...
>
>
> "ActualOverdensity has the same physical meaning as the regular
>
> Overdensity
>
> field. It is (Baryon Density + Dark Matter Density) / (Mean
>
> density of the
>
> universe). If you search the literature, you will find alternate
>
> definitions that use critical density instead of mean density. The
>
> one used
>
> here is a little more common."
>
>
>
> Are you sure you did not mean (Baryon Density + Dark Matter
>
> Density) / (critical density) ?
>
>
>
> Because, in halo_profiler.py, there is a virial filter...
>
>
> hp.add_halo_filter(HP.VirialFilter,must_be_virialized=True,
>
> overdensity_field='ActualOverdensity',
>
> virial_overdensity=200,
>
> virial_filters=[['TotalMassMsun','>=','1e14']],
>
> virial_quantities=['TotalMassMsun','RadiusMpc'])
>
>
>
> Isn't 200 w.r.t rho_ciritcal (not rho_mean_matter)? Look
> http://arxiv.org/abs/astro-ph/0011495
>
>
>
> shankar
>
>
>
>
>
> -----Original Message-----
>
> From: yt-users-bounces at lists.spacepope.org on behalf of Britton Smith
>
> Sent: Wed 11/25/2009 1:30 PM
>
> To: Discussion of the yt analysis package
>
> Subject: Re: [yt-users] Hop vs HaloProfiler
>
>
> Hi Shankar,
>
>
> I will answer the HaloProfiler related questions and leave the Hop
>
> questions
>
> for Stephen. However, could you post the failure output for what
>
> you tried
>
> to do with Hop. That will probably be helpful for answering your
>
> question.
>
>
> Is Halo_0000_profile.dat related to the first Halo listed in
>
> HopAnalysis.out
>
> file ?
>
>
>
> Yes, the HaloProfiler will always use the same indices for halos as
>
> all of
>
> the halo finders in yt. So, the file Halo_N_profile.dat will always
>
> refer
>
> to halo N on the halo list.
>
>
> What is # ?
>
>
>
> This is just a comment character so that plotting programs don't try
>
> to read
>
> in what's on that line. The tab between it and the first field name
>
> is
>
> there to accommodate the routines that read those files back in to the
>
> HaloProfiler.
>
>
>
> What is the meaning of ActualOverdensity ? And units ?
>
>
>
> ActualOverdensity has the same physical meaning as the regular
>
> Overdensity
>
> field. It is (Baryon Density + Dark Matter Density) / (Mean density
>
> of the
>
> universe). If you search the literature, you will find alternate
>
> definitions that use critical density instead of mean density. The
>
> one used
>
> here is a little more common (I think, but I'm not sure). As per
>
> these
>
> definitions, overdensity is a unitless quantity.
>
>
> In yt, there is an Overdensity field that is calculated on a cell-by-
>
> cell
>
> basis. For calculation of virial quantities for halos, the
>
> overdensity you
>
> calculate should be explicitly the total mass (baryon + dm) / total
>
> volume /
>
> mean density, where total refers to all cells enclosed within the
>
> sphere of
>
> the radial profile, not just the shell from r_(i-1) to r_i.
>
> Technically,
>
> you could get this by doing a profile of the Overdensity field,
>
> weighted by
>
> CellVolume, with accumulation set to True. However, if for some
>
> reason, the
>
> user wanted to do profiles of the overdensity field in a different
>
> way, say
>
> weighted by CellMass, or just counting the material shell-by-shell
>
> (accumulation=False), this number would not be correct for the
>
> calculation
>
> of virial quantities. For that reason, the HaloProfiler automatically
>
> generates this ActualOverdensity field which is assured to be
>
> calculated in
>
> the correct way. That way the user doesn't accidentally override a
>
> proper
>
> calculation of the overdensity used for the virial quantities.
>
>
> Other than the ActualOverdensity field (which is automatically
>
> generated by
>
> the HaloProfiler, and thus does not exist outside that context) you
>
> can
>
> always get the units of any field with:
>
> lagos.fieldInfo[some_field].units
>
> For more information, see here:
>
>
> http://yt.enzotools.org/doc/faq.html#how-do-i-know-what-the-units-returned-are
>
>
> What is the meaning of CellVolume? And units ?
>
>
> What is Density referring to ? And units ?
>
>
> CellVolume and Density are the volume of a cell and the baryon
>
> density. In
>
> this context, they are the values of those fields in the radial
>
> profile. If
>
> you did it as per the recipe, the CellVolume is the total
>
> accumulated volume
>
> for all cells within the sphere radius. Density SHOULD be the
>
> mass-weighted, mean baryon density for cells within spherical shells.
>
> However, after looking at the recipe on the website, I see that, in
>
> error, I
>
> set the weight_field of the Density profile to None, instead of
>
> CellMassMsun. Therefore, the Density profile may be meaningless as
>
> is. I
>
> will change the recipe on the website ASAP.
>
>
>
> What is mywieght ?
>
>
>
> The myweight field is a temporary field for keeping track of the
>
> weight
>
> field for a weight radial profile. It should be ignored and we
>
> might just
>
> want to not write it out in the future.
>
>
>
> I also got the projection/ directory. But I got errors...
>
>
>
> It looks like the HaloProfiler tried to do a projection of
>
> Metallicity, but
>
> your simulation data did not have the MetalDensity field. You need
>
> to set
>
> CosmologyUseMetallicityField (or something like that) to 1 in your
>
> enzo
>
> parameter file in order to get this field. You can also just remove
>
> the
>
> line in the HaloProfiler recipe that adds the metallicity field to
>
> the list
>
> of projections.
>
>
> Cheers,
>
>
> Britton
>
>
>
> _______________________________________________
>
> yt-users mailing list
>
> yt-users at lists.spacepope.org
>
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>
>
>
> _______________________________________________
>
> yt-users mailing list
>
> yt-users at lists.spacepope.org
>
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>
>
> Dr. Eric J. Hallman
> NSF Astronomy and Astrophysics Postdoctoral Fellow
> Center for Astrophysics and Space Astronomy
> University of Colorado at Boulder
> hallman (at) casa.colorado.edu
> Phone: (312) 725-4626
> http://solo.colorado.edu/~hallman/ <http://solo.colorado.edu/%7Ehallman/>
>
>
> <winmail.dat>_______________________________________________
>
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>
>
> Dr. Eric J. Hallman
> NSF Astronomy and Astrophysics Postdoctoral Fellow
> Center for Astrophysics and Space Astronomy
> University of Colorado at Boulder
> hallman (at) casa.colorado.edu
> Phone: (312) 725-4626
> http://solo.colorado.edu/~hallman/ <http://solo.colorado.edu/%7Ehallman/>
>
>
> _______________________________________________
> yt-users mailing list
> yt-users at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-users-spacepope.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.spacepope.org/pipermail/yt-users-spacepope.org/attachments/20091202/85473cb1/attachment.html>
More information about the yt-users
mailing list