Cosmologies#
Cosmology Specifications#
This page describes the specification of the Cosmologies and the CLASS parameters that they define. The CLASS parameter files and resulting power spectra and transfer functions are available in the AbacusSummit/Cosmologies directory.
We have given the cosmologies numbers, so that the simulations will refer to c123.
There are various sets of cosmologies in this list:
The primary cosmology (c000) is a Planck2018 LCDM version, specifically the mean of base_plikHM_TTTEEE_lowl_lowE_lensing. This cosmology has 25 base boxes, plus other mass resolution options. Planck, and AbacusSummit, include a massive neutrino in this cosmology.
c009 is the same cosmology, but with massless neutrinos, at fixed omega_cb, H0, and sigma_cb. c019 and c020 have two and zero 60 meV neutrino species, at fixed omega_cb, theta_CMB, and sigma_cb.
There are 4 secondary cosmologies (c001-4), each with 6 base boxes. There is a low-omega_c choice based on WMAP7, a wCDM choice, a high-Neff choice, and a low-sigma8 choice.
There are ten reference cosmologies (c010, c012-c018, c021-022) that match the choices employed in flagship runs from other groups, so as to ease comparisons between code bases. Note that most of these have massless neutrinos. c011 was not run.
There is a linear derivative grid (c100-c116) that provides 8 matched pairs, with small positive and negative steps in an 8-dimensional parameter space, plus one additional simulation that is the high-sigma8 partner to the low-sigma8 secondary cosmology. Another set of ten simulations (c117-c126) are pairs with smaller spacing.
There is a larger unstructured emulator grid (c130-c181) that provides a wider coverage of this 8-dimensional space.
Further details are below the table.
All cosmologies use tau=0.0544. Most use 60 meV neutrinos (also the choice of the Planck 2018 baseline cosmology), omega_nu = 0.00064420, scaling from z = 1. We use HyRec, rather than RecFast.
CLASS is run with the pk_ref.pre precision choices, unless the name ends with _fast, in which case we use the defaults. There was one case where CLASS underflowed an integration tolerance with the pk_ref precisions; we reverted to pk_permille.pre for this.
Remember that Omega_m = (omega_b+omega_cdm+omega_ncdm)/h^2.
We output five redshifts from CLASS, z = 0.0, 1.0, 3.0, 7.0, and 49, which are called z1,z2,z3,z4,z5.
We use the CDM+Baryon power spectrum at z = 1 (z2_pk_cb) and scale back by the ratio of growth factors \(D(z_\mathrm{init})/D(1)\) to define our matter-dominated CDM-only simulation IC. The growth function includes the neutrinos as a smooth component.
Primordial Non-Gaussianity#
The cosmologies numbered 300-303 contain local-type primordial non-Gaussianity variations (f_NL).
Cosmologies Table#
The following is a listing of all cosmologies used in AbacusSummit simulations. You can also download the cosmologies table as a CSV here.
In analysis applications, users are encouraged to use the cosmological parameters stored as in the header field
of the ASDF data product files (which is loaded into the meta field of Astropy tables, or the header field of
CompaSOHaloCatalog objects) rather than referencing the cosmologies table.
Another way to access the cosmology of a particular AbacusSummit simulation is
through the abacusnbody.metadata module: Simulation Metadata
- Column Names
The names of the parameter columns in the following table correspond to CLASS parameters. A good place to look for CLASS parameter definitions is this well-commented file: abacus_base.pre. It is derived from the example CLASS input file, so most of the comments are CLASS’s. Note that this file is also the base AbacusSummit parameter file used to run CLASS. For the full CLASS documentation, see here.
root |
notes |
omega_b |
omega_cdm |
h |
A_s |
n_s |
alpha_s |
N_ur |
N_ncdm |
omega_ncdm |
w0_fld |
wa_fld |
sigma8_m |
sigma8_cb |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
abacus_cosm000 |
Baseline LCDM, Planck 2018 base_plikHM_TTTEEE_lowl_lowE_lensing mean |
0.02237 |
0.1200 |
0.6736 |
2.0830e-9 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.807952 |
0.811355 |
abacus_cosm001 |
WMAP9+ACT+SPT LCDM, Calabrese++ 2017 |
0.02242 |
0.1134 |
0.7030 |
2.0376e-9 |
0.9638 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.776779 |
0.780222 |
abacus_cosm002 |
wCDM with thawing model w0 = -0.7, wa = -0.5 |
0.02237 |
0.1200 |
0.6278 |
2.3140e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-0.7 |
-0.5 |
0.808189 |
0.811577 |
abacus_cosm003 |
Neff=3.70, from base_nnu_plikHM_TT_lowl_lowE_Riess18_post_BAO |
0.02260 |
0.1291 |
0.7160 |
2.2438e-9 |
0.9876 |
0.0 |
2.6868 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.855190 |
0.858583 |
abacus_cosm004 |
Low sigma8_matter = 0.75, otherwise Baseline LCDM |
0.02237 |
0.1200 |
0.6736 |
1.7949e-9 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.749999 |
0.753159 |
abacus_cosm009 |
Baseline LCDM with massless neutrinos matching omega_cb & sigma8_cb |
0.02237 |
0.1200 |
0.6736 |
2.0417e-9 |
0.9649 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.811362 |
0.811362 |
abacus_cosm010 |
AbacusCosmos Planck LCDM cosmology |
0.02222 |
0.1199 |
0.6726 |
2.100e-9 |
0.9652 |
0.0 |
3.04 |
0 |
0.0 |
-1.0 |
0.0 |
0.823630 |
0.823630 |
abacus_cosm011 |
AbacusCosmos Planck LCDM cosmology +10% in sigma8 |
0.02222 |
0.1199 |
0.6726 |
2.541e-9 |
0.9652 |
0.0 |
3.04 |
0 |
0.0 |
-1.0 |
0.0 |
0.905993 |
0.905993 |
abacus_cosm012 |
Euclid Flagship1 LCDM, sigma8_m=0.8279 |
0.02200 |
0.1212 |
0.6700 |
2.1000e-9 |
0.9600 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.827899 |
0.827899 |
abacus_cosm013 |
Euclid Flagship2 LCDM, sigma8_m=0.813715, sigma8_cb=0.817135 |
0.02200 |
0.1206 |
0.6700 |
2.1000e-9 |
0.9600 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.813715 |
0.817135 |
abacus_cosm014 |
OuterRim LCDM, sigma8=0.80 |
0.02258 |
0.1109 |
0.7100 |
2.1591e-9 |
0.9630 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.800000 |
0.800000 |
abacus_cosm015 |
DarkSky LCDM, sigma8=0.835 |
0.02215 |
0.1175 |
0.6880 |
2.1852e-9 |
0.9688 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.835005 |
0.835005 |
abacus_cosm016 |
Horizon Run 4 LCDM, sigma8=0.7937 |
0.02281 |
0.1120 |
0.7200 |
2.0996e-9 |
0.9600 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.793693 |
0.793693 |
abacus_cosm017 |
IllustrisTNG LCDM, sigma8=0.8159 |
0.02230 |
0.1194 |
0.6774 |
2.0671e-9 |
0.9667 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.815903 |
0.815903 |
abacus_cosm018 |
MultiDark Planck LCDM, sigma8=0.8228 |
0.02214 |
0.1189 |
0.6777 |
2.1022e-9 |
0.9600 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.819708 |
0.819708 |
abacus_cosm019 |
Baseline LCDM with two 0.06 eV neutrinos |
0.02237 |
0.1200 |
0.6683 |
2.1353e-09 |
0.9649 |
0.0 |
1.0196 |
2 |
0.00064420,0.00064420 |
-1.0 |
0.0 |
0.805050 |
0.811826 |
abacus_cosm020 |
Baseline LCDM w/ massless neutrinos matching theta_star & sigma8_cb |
0.02237 |
0.1200 |
0.6790 |
2.0342e-09 |
0.9649 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.811350 |
0.811350 |
abacus_cosm021 |
MassiveNUs massless base |
0.02254 |
0.12446 |
0.700 |
2.1000e-09 |
0.9700 |
0.0 |
3.046 |
0 |
0.0 |
-1.0 |
0.0 |
0.849842 |
0.849842 |
abacus_cosm022 |
MassiveNUs with one 0.1 eV massive neutrino species |
0.02254 |
0.12339 |
0.700 |
2.1000e-09 |
0.9700 |
0.0 |
2.0328 |
1 |
0.001074 |
-1.0 |
0.0 |
0.824961 |
0.830296 |
abacus_cosm100 |
Baseline +2% ln(omega_b) |
0.02282 |
0.1200 |
0.6777 |
2.0934e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808181 |
0.811575 |
abacus_cosm101 |
Baseline -2% ln(omega_b) |
0.02193 |
0.1200 |
0.6696 |
2.0751e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808156 |
0.811570 |
abacus_cosm102 |
Baseline +3.3% ln(omega_c) |
0.02237 |
0.1240 |
0.6597 |
2.0205e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808270 |
0.811574 |
abacus_cosm103 |
Baseline -3.3% ln(omega_c) |
0.02237 |
0.1161 |
0.6877 |
2.1541e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808075 |
0.811582 |
abacus_cosm104 |
Baseline +0.01 n_s |
0.02237 |
0.1200 |
0.6736 |
2.0684e-09 |
0.9749 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808166 |
0.811572 |
abacus_cosm105 |
Baseline -0.01 n_s |
0.02237 |
0.1200 |
0.6736 |
2.0999e-09 |
0.9549 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808177 |
0.811580 |
abacus_cosm106 |
Baseline +0.02 nrun |
0.02237 |
0.1200 |
0.6736 |
2.0638e-09 |
0.9649 |
0.02 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808181 |
0.811586 |
abacus_cosm107 |
Baseline -0.02 nrun |
0.02237 |
0.1200 |
0.6736 |
2.1045e-09 |
0.9649 |
-0.02 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808168 |
0.811571 |
abacus_cosm108 |
Baseline +0.1 w0 |
0.02237 |
0.1200 |
0.6444 |
2.2390e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-0.9 |
0.0 |
0.808177 |
0.811570 |
abacus_cosm109 |
Baseline -0.1 w0 |
0.02237 |
0.1200 |
0.7037 |
1.9465e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.1 |
0.0 |
0.808161 |
0.811576 |
abacus_cosm110 |
Baseline +0.4 wa, -0.1 w0 |
0.02237 |
0.1200 |
0.6698 |
2.1219e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.1 |
0.4 |
0.808170 |
0.811572 |
abacus_cosm111 |
Baseline -0.4 wa, +0.1 w0 |
0.02237 |
0.1200 |
0.6752 |
2.0639e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-0.9 |
-0.4 |
0.808179 |
0.811584 |
abacus_cosm112 |
Baseline +2% sigma8 |
0.02237 |
0.1200 |
0.6736 |
2.1672e-9 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.824120 |
0.827591 |
abacus_cosm113 |
Baseline -2% sigma8 |
0.02237 |
0.1200 |
0.6736 |
2.0021e-9 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.792107 |
0.795443 |
abacus_cosm114 |
Baseline +0.3 nnu, +3.3% ln(omega_c), +0.01 n_s |
0.02237 |
0.1240 |
0.6947 |
2.0463e-09 |
0.9749 |
0.0 |
2.3328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808245 |
0.811563 |
abacus_cosm115 |
Baseline -0.3 nnu, -3.3% ln(omega_c), -0.01 n_s |
0.02237 |
0.1161 |
0.6517 |
2.1211e-09 |
0.9549 |
0.0 |
1.7328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808089 |
0.811582 |
abacus_cosm116 |
Emulator grid around baseline cosmology |
0.02237 |
0.1200 |
0.6736 |
2.3938e-09 |
0.9649 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.866133 |
0.869782 |
abacus_cosm117 |
Baseline +0.83% ln(omega_c) |
0.02237 |
0.1210 |
0.6701 |
2.0675e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808205 |
0.811584 |
abacus_cosm118 |
Baseline -0.83% ln(omega_c) |
0.02237 |
0.1190 |
0.6771 |
2.1014e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808152 |
0.811582 |
abacus_cosm119 |
Baseline +0.003 n_s |
0.02237 |
0.1200 |
0.6736 |
2.0794e-09 |
0.9679 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808168 |
0.811573 |
abacus_cosm120 |
Baseline -0.003 n_s |
0.02237 |
0.1200 |
0.6736 |
2.0889e-09 |
0.9619 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.808181 |
0.811585 |
abacus_cosm121 |
Baseline +0.025 w0 |
0.02237 |
0.1200 |
0.6662 |
2.1211e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-0.975 |
0.0 |
0.808169 |
0.811571 |
abacus_cosm122 |
Baseline -0.025 w0 |
0.02237 |
0.1200 |
0.6810 |
2.0483e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.025 |
0.0 |
0.808170 |
0.811577 |
abacus_cosm123 |
Baseline +0.1 wa, -0.025 w0 |
0.02237 |
0.1200 |
0.6729 |
2.0915e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.025 |
0.1 |
0.808176 |
0.811580 |
abacus_cosm124 |
Baseline -0.1 wa, +0.025 w0 |
0.02237 |
0.1200 |
0.6742 |
2.0778e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-0.975 |
-0.1 |
0.808175 |
0.811580 |
abacus_cosm125 |
Baseline +0.5% sigma8 |
0.02237 |
0.1200 |
0.6736 |
2.1039e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.811995 |
0.815415 |
abacus_cosm126 |
Baseline -0.5% sigma8 |
0.02237 |
0.1200 |
0.6736 |
2.0622e-09 |
0.9649 |
0.0 |
2.0328 |
1 |
0.00064420 |
-1.0 |
0.0 |
0.803908 |
0.807294 |
abacus_cosm130 |
Emulator grid around baseline cosmology |
0.02237 |
0.1200 |
0.6736 |
1.6140e-09 |
0.9649 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.711201 |
0.714197 |
abacus_cosm131 |
Emulator grid around baseline cosmology |
0.02237 |
0.1086 |
0.7165 |
2.3146e-09 |
0.9649 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.807866 |
0.811587 |
abacus_cosm132 |
Emulator grid around baseline cosmology |
0.02237 |
0.1200 |
0.6736 |
2.1791e-09 |
0.9049 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.808189 |
0.811584 |
abacus_cosm133 |
Emulator grid around baseline cosmology |
0.02237 |
0.1200 |
0.6736 |
2.6144e-09 |
0.9649 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.905163 |
0.908976 |
abacus_cosm134 |
Emulator grid around baseline cosmology |
0.02237 |
0.1326 |
0.6319 |
1.9066e-09 |
0.9649 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.808458 |
0.811566 |
abacus_cosm135 |
Emulator grid around baseline cosmology |
0.02237 |
0.1200 |
0.6736 |
1.9904e-09 |
1.0249 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.808160 |
0.811573 |
abacus_cosm136 |
Emulator grid around baseline cosmology |
0.02073 |
0.1192 |
0.6618 |
2.5239e-09 |
0.9303 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.882475 |
0.886249 |
abacus_cosm137 |
Emulator grid around baseline cosmology |
0.02212 |
0.1271 |
0.6472 |
1.6540e-09 |
0.9252 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.729693 |
0.732609 |
abacus_cosm138 |
Emulator grid around baseline cosmology |
0.02108 |
0.1138 |
0.6847 |
2.2282e-09 |
0.9723 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.821432 |
0.825095 |
abacus_cosm139 |
Emulator grid around baseline cosmology |
0.02416 |
0.1128 |
0.7164 |
2.1681e-09 |
0.9732 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.793003 |
0.796493 |
abacus_cosm140 |
Emulator grid around baseline cosmology |
0.02096 |
0.1221 |
0.6536 |
1.8126e-09 |
0.9893 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.772033 |
0.775267 |
abacus_cosm141 |
Emulator grid around baseline cosmology |
0.02381 |
0.1272 |
0.6623 |
1.9945e-09 |
0.9384 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.799575 |
0.802737 |
abacus_cosm142 |
Emulator grid around baseline cosmology |
0.02287 |
0.1130 |
0.7038 |
1.6701e-09 |
0.9927 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.707082 |
0.710219 |
abacus_cosm143 |
Emulator grid around baseline cosmology |
0.02206 |
0.1278 |
0.6443 |
2.1084e-09 |
0.9952 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.847522 |
0.850903 |
abacus_cosm144 |
Emulator grid around baseline cosmology |
0.02210 |
0.1130 |
0.6968 |
2.7653e-09 |
0.9279 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.891360 |
0.895324 |
abacus_cosm145 |
Emulator grid around baseline cosmology |
0.02428 |
0.1186 |
0.6961 |
2.1634e-09 |
0.9347 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.801404 |
0.804769 |
abacus_cosm146 |
Emulator grid around baseline cosmology |
0.02097 |
0.1180 |
0.6682 |
1.9020e-09 |
0.9351 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.762696 |
0.765984 |
abacus_cosm147 |
Emulator grid around baseline cosmology |
0.02113 |
0.1215 |
0.6498 |
1.9081e-09 |
0.9587 |
0.000 |
2.0328 |
1 |
0.00064420 |
-0.809 |
-0.628 |
0.777157 |
0.780416 |
abacus_cosm148 |
Emulator grid around baseline cosmology |
0.02289 |
0.1201 |
0.6199 |
1.9629e-09 |
0.9380 |
0.000 |
2.0328 |
1 |
0.00064420 |
-0.925 |
0.393 |
0.714913 |
0.717888 |
abacus_cosm149 |
Emulator grid around baseline cosmology |
0.02188 |
0.1200 |
0.6688 |
2.1521e-09 |
0.9913 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.174 |
0.613 |
0.824854 |
0.828343 |
abacus_cosm150 |
Emulator grid around baseline cosmology |
0.02248 |
0.1216 |
0.7224 |
2.5099e-09 |
0.9355 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.090 |
-0.373 |
0.937655 |
0.941570 |
abacus_cosm151 |
Emulator grid around baseline cosmology |
0.02315 |
0.1254 |
0.6582 |
2.2686e-09 |
0.9757 |
0.000 |
2.0328 |
1 |
0.00064420 |
-0.826 |
-0.615 |
0.860820 |
0.864288 |
abacus_cosm152 |
Emulator grid around baseline cosmology |
0.02165 |
0.1148 |
0.6204 |
1.7965e-09 |
0.9541 |
0.000 |
2.0328 |
1 |
0.00064420 |
-0.732 |
-0.187 |
0.677885 |
0.680849 |
abacus_cosm153 |
Emulator grid around baseline cosmology |
0.02192 |
0.1199 |
0.6092 |
2.2946e-09 |
0.9917 |
0.000 |
2.0328 |
1 |
0.00064420 |
-0.735 |
-0.172 |
0.794389 |
0.797729 |
abacus_cosm154 |
Emulator grid around baseline cosmology |
0.02158 |
0.1148 |
0.7426 |
2.0689e-09 |
0.9538 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.076 |
-0.395 |
0.838698 |
0.842413 |
abacus_cosm155 |
Emulator grid around baseline cosmology |
0.02369 |
0.1184 |
0.6247 |
2.0723e-09 |
0.9713 |
0.000 |
2.0328 |
1 |
0.00064420 |
-0.727 |
-0.184 |
0.735302 |
0.738388 |
abacus_cosm156 |
Emulator grid around baseline cosmology |
0.02202 |
0.1261 |
0.6499 |
1.9738e-09 |
0.9678 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.176 |
0.621 |
0.801974 |
0.805210 |
abacus_cosm157 |
Emulator grid around baseline cosmology |
0.02247 |
0.1214 |
0.6668 |
2.5108e-09 |
0.9356 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.167 |
0.616 |
0.872315 |
0.875941 |
abacus_cosm158 |
Emulator grid around baseline cosmology |
0.02201 |
0.1262 |
0.6958 |
1.8626e-09 |
0.9667 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.082 |
-0.392 |
0.829816 |
0.833179 |
abacus_cosm159 |
Emulator grid around baseline cosmology |
0.02206 |
0.1261 |
0.5967 |
1.7977e-09 |
0.9673 |
0.000 |
2.0328 |
1 |
0.00064420 |
-0.726 |
-0.220 |
0.718521 |
0.721402 |
abacus_cosm160 |
Emulator grid around baseline cosmology |
0.02223 |
0.1185 |
0.7456 |
2.0963e-09 |
0.9942 |
0.000 |
2.0328 |
1 |
0.00064420 |
-1.271 |
0.217 |
0.876756 |
0.880523 |
abacus_cosm161 |
Emulator grid around baseline cosmology |
0.02282 |
0.1274 |
0.7330 |
1.9688e-09 |
0.9649 |
0.026 |
2.7283 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.793066 |
0.796247 |
abacus_cosm162 |
Emulator grid around baseline cosmology |
0.02136 |
0.1048 |
0.6383 |
2.1592e-09 |
0.9297 |
-0.026 |
1.3521 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.779589 |
0.783285 |
abacus_cosm163 |
Emulator grid around baseline cosmology |
0.02135 |
0.1290 |
0.7280 |
2.1689e-09 |
0.9922 |
-0.018 |
2.8575 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.838824 |
0.842187 |
abacus_cosm164 |
Emulator grid around baseline cosmology |
0.02173 |
0.1091 |
0.6042 |
2.0003e-09 |
0.9211 |
0.016 |
1.1769 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.774159 |
0.777681 |
abacus_cosm165 |
Emulator grid around baseline cosmology |
0.02225 |
0.1278 |
0.7293 |
2.0676e-09 |
1.0241 |
0.026 |
2.7558 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.835954 |
0.839319 |
abacus_cosm166 |
Emulator grid around baseline cosmology |
0.02306 |
0.1151 |
0.6826 |
2.2779e-09 |
0.9691 |
0.038 |
1.9089 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.837463 |
0.841106 |
abacus_cosm167 |
Emulator grid around baseline cosmology |
0.02248 |
0.1400 |
0.7050 |
1.7321e-09 |
0.9715 |
-0.017 |
2.8889 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.768419 |
0.771255 |
abacus_cosm168 |
Emulator grid around baseline cosmology |
0.02157 |
0.1084 |
0.6070 |
2.5378e-09 |
0.9275 |
0.016 |
1.1911 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.871407 |
0.875401 |
abacus_cosm169 |
Emulator grid around baseline cosmology |
0.02189 |
0.1222 |
0.6474 |
1.8067e-09 |
0.9898 |
0.038 |
1.9109 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.777925 |
0.781158 |
abacus_cosm170 |
Emulator grid around baseline cosmology |
0.02319 |
0.1076 |
0.6418 |
1.7994e-09 |
0.9387 |
-0.026 |
1.3411 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.716059 |
0.719332 |
abacus_cosm171 |
Emulator grid around baseline cosmology |
0.02348 |
0.1272 |
0.6733 |
2.2585e-09 |
0.9762 |
-0.038 |
2.1543 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.852878 |
0.856267 |
abacus_cosm172 |
Emulator grid around baseline cosmology |
0.02282 |
0.1032 |
0.6369 |
2.1259e-09 |
0.9012 |
0.017 |
1.1954 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.765650 |
0.769278 |
abacus_cosm173 |
Emulator grid around baseline cosmology |
0.02322 |
0.1161 |
0.6800 |
1.8904e-09 |
0.9628 |
0.038 |
1.9049 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.763962 |
0.767255 |
abacus_cosm174 |
Emulator grid around baseline cosmology |
0.02302 |
0.1064 |
0.6427 |
2.4889e-09 |
0.9437 |
-0.026 |
1.3242 |
1 |
0.00064420 |
-1.000 |
0.000 |
0.840113 |
0.843996 |
abacus_cosm175 |
Emulator grid around baseline cosmology |
0.02173 |
0.1336 |
0.6432 |
1.7429e-09 |
0.9898 |
-0.003 |
2.7456 |
1 |
0.00064420 |
-0.911 |
0.350 |
0.708760 |
0.711484 |
abacus_cosm176 |
Emulator grid around baseline cosmology |
0.02238 |
0.1199 |
0.6972 |
2.3120e-09 |
0.9459 |
0.036 |
1.9043 |
1 |
0.00064420 |
-1.222 |
0.339 |
0.892483 |
0.896318 |
abacus_cosm177 |
Emulator grid around baseline cosmology |
0.02239 |
0.1344 |
0.6820 |
2.0641e-09 |
1.0002 |
0.007 |
2.8643 |
1 |
0.00064420 |
-0.757 |
-0.443 |
0.806026 |
0.809110 |
abacus_cosm178 |
Emulator grid around baseline cosmology |
0.02234 |
0.1204 |
0.6542 |
1.9665e-09 |
0.9424 |
0.037 |
1.8987 |
1 |
0.00064420 |
-0.874 |
-0.455 |
0.791239 |
0.794555 |
abacus_cosm179 |
Emulator grid around baseline cosmology |
0.02240 |
0.1067 |
0.5881 |
2.2365e-09 |
0.9308 |
-0.003 |
1.1884 |
1 |
0.00064420 |
-0.755 |
-0.435 |
0.775969 |
0.779541 |
abacus_cosm180 |
Emulator grid around baseline cosmology |
0.02258 |
0.1209 |
0.7451 |
2.2790e-09 |
0.9796 |
-0.034 |
2.1722 |
1 |
0.00064420 |
-1.108 |
-0.274 |
0.894071 |
0.897833 |
abacus_cosm181 |
Emulator grid around baseline cosmology |
0.02169 |
0.1112 |
0.5745 |
2.0651e-09 |
0.9336 |
-0.013 |
1.4059 |
1 |
0.00064420 |
-0.910 |
0.350 |
0.730036 |
0.733292 |
Running CLASS#
We used Willem Elber’s fork of CLASS with N-body gauge support. Specifically, we used git hash 6cf8e3.
The CLASS outputs stored in AbacusSummit/Cosmologies can be reproduced by running this code passing both the CLASS.ini file for a given cosmology and the abacus_base.pre or abacus_base_fast.pre file. For example:
$ cd AbacusSummit/Cosmologies/abacus_cosm000
$ <path to class executable> CLASS.ini ../abacus_base.pre
Additional Details#
Beyond the Planck2018 LCDM primary cosmology, we chose 4 other secondary cosmologies. One was WMAP7, to have a large change in omega_m, H0, and sigma8. Others were one wCDM, one high Neff, and one low S8.
wCDM: Chose w0=-0.7, wa=-0.5 to be an extreme thawing model.
Neff=3.70 cosmology: Took the chains from base_nnu_plikHM_TT_lowl_lowE_Riess18_post_BAO and averaged those in 3.595 < nnu < 3.90, chosen so that the weighted mean was 3.70. Also standardized As to tau=0.0544.
Low sigma8: Opted to drop the amplitude by about 7.7%, to make sigma8(matter)=0.75. This is a sizeable shift, but there’s lots of ways to damp power.
Then we are doing a large grid of cosmologies to provide control of first and second derivatives around the primary LCDM model.
For the grid of positive/negative excursions for linear derivatives around the baseline LCDM, we opted for the simplicity of rectalinear derivatives in ln(omega_b), ln(omega_c), ns, nrun, sigma8_m, w0. Note that we treat sigma8_m, not As, as the independent variable, in the expectation that this will keep large-scale structure closer to constant. For wa, we opt to hold w(z=0.333)=w0+0.25*wa fixed, close to the mirage model. For Neff, the Planck chains suggested substantial degeneracies with omegac and ns, so we opt to move these two along with Neff to stay close to the CMB degeneracy direction.
We added one extra simulation to be the paired opposite to the low-sigma8 secondary cosmology.
For the broader emulator set, we construct the unstructured grid as follows: We place points on the surface of an 8-dimensional unit sphere, denoting these v0..v7, then map them into the 8-dimensional parameter space by:
sigma8cb = 0.811355 (1 + 0.12 v0 - 0.125 v4 + 0.06 u0), where u0 is another random number, uniform in [-1,1].
omega_c = 0.1200 exp(0.100 v1 + 0.165 v6)
ns = 0.9649 + 0.06 v2 + 0.05 v6
omegab = 0.02237 exp(0.10 v3)
w0 = -1.0 + 0.3 v4 -0.2 v5
wa = 0.8 v5
Nur = 2.0328 + 1.2 v6
alpha_s = 0.05 v7
These parameter ranges were chosen to be relatively large (5-8 sigma) beyond today’s CMB+LSS constraints, but it is important to note that most of an 8-d sphere is not close to the extreme in any one parameter, but rather 1/sqrt(8) of that extreme.
We have continued to have omega_c and ns vary with Nur, and w0 to vary with wa (so that variations in wa hold w(z=0.333) constant). In addition, we opted to have sigma8 vary with w(0.333), not as much as a pure wCDM fit to the CMB would imply, but to partially track that behavior.
Finally, we add extra +-6% scatter to sigma8. Note that if we were holding the amplitude of the CMB anisotropies fixed (and fixed tau), then our parameter variations would vary sigma8 quite a lot. But we have not varied tau or neutrino mass, so we want to allow some scatter in sigma8.
Next, we have to specify the distribution of points on the 8-d unit sphere. We want to keep the points well separated, but also impose some constraints. We seek to have some of the points sit in subspaces, so that one doesn’t have to be using the entire 8-dimensional space in all fits. And we want to avoid most antipodal points, as these provide only redundant information about second derivatives (since we already have the linear derivative set). We also want to mildly exceed the number of simulations needed to constrain the second derivatives, so that there is some ability to drop simulations for cross-validation.
We use 49 antipodal pairs of points on the sphere. These are subject to the constraints below, but otherwise are evolved from their random start to an electrostatic glass, resulting in a well dispersed set of points. The constraints:
a) The first 3 pairs are forced to be at the unit vectors in the v0, v1, and v2 directions, which will map to individual extremal excursions in sigma8cb, omega_c, and ns. We retain both points of each pair in the grid, as these are particularly important directions. In all cases below, we keep only the first point of each pair.
b) The next 11 pairs sample only the v0..v3 directions and are constrained to have v4..v7 = 0, so that they will only sample sigma8cb, omega_c, ns, and omega_b. We note that the 4-dimensional space has 10 second derivatives, for which we’re 17 simulations (and 14 non-antipodal).
c) The next 14 pairs sample the v0..v5 directions, holding v6..v7 = 0. These will add w0, wa to the space. This introduces 11 new second derivatives.
d) The next 14 pairs sample the v0..v3 + v6..7 directions, holding v4..v5 = 0. These will add Nur and alpha_s to the LCDM space. Again, this introduces 11 new second derivatives.
e) The last 7 pairs sample the full space, and hence have excursions in w0, wa, Nur, and alpha_s. This last subspace has 4 new second derivatives to measure.
The randomness of the starting point was subjected to some patterns on the sign of certain coordinates in order to encourage a glass with better balance in 2-d projections. This was judged simply by eye.
Matching sigma_8#
For deviations from the c000 cosmology where we want to hold sigma8 fixed instead of A_s, we employed the following procedure.
Run CLASS with the
abacus_base_fast.preparameters, modifying some parameters but holding A_s fixed;look at the CLASS-reported sigma8 value;
compute a new A_s = A_s_c000 * (sigma8_c104 / sigma8_c000)^2;
record that new A_s in the cosmologies table;
run CLASS again with the new A_s, using the full-precision
abacus_base.preparameters;look at the new sigma8 and record it in the cosmologies table (in theory, it will be the same as c000).
The final sigma8 in step (6) may not match that of c000 exactly because the initial calibration run was
done using the faster, less accurate abacus_base_fast.pre settings, while the final run used full
precision. However, the sigma8 match is still generally very good, as can be verified by inspection
of the cosmologies table.