Intensity Mapping: A New Window into the Cosmos
461
Fig. 1 Angular power
spectrum C HI
at redshift 0.8,
using the best-fitting HI
astrophysical parameters
from the halo model and a
redshift binning of
Δz = 0.05. The error bars
shown in red represent the
standard deviation ΔC HI
l ,
calculated for a SKA 1
MID-like configuration
10
0
10
1
10
2
10
3
10
−8
10
−7
10
−6
C
z = 0.8
h Ω m n s Ω b σ 8
0.0
0.1
0.2
0.3
0.4
σ
A /A
Fixed astrophysics
With astrophysical prior
Marginalized
v c,0
β
Fixed cosmology
Without cosmological prior
Fig. 2 Left panel: Constraints on the cosmological parameters from a SKA I MID-like configuration, (1) marginalized over astrophysics, (2) with fixed astrophysics and (3) with the astrophysical
prior coming from the present data. Right panel: Astrophysical forecasts (1) without cosmological
priors and (2) with fixed cosmology. Figure adapted from Ref. [13]
Cross-correlating 21 cm intensity maps with galaxy surveys (both photometric
and spectroscopic) offer exciting prospects for constraining astrophysical and
cosmological parameters. It can be shown [12] that cross-correlating such surveys
covering similar redshift ranges and sky areas significantly improves astrophysical
constraints (see Fig. 3 for an example of a CHIME-like and DESI-like survey crosscorrelated in the northern hemisphere). Further, cross-correlation is a valuable tool
to mitigate the effects of contaminants and foregrounds, which are expected to be
significantly uncorrelated between the two surveys (e.g., Ref. [2]).
The Fisher formalism can also be used to quantify how a complementary effect,
the accuracy of cosmological constraints, is affected by the astrophysical prior
information. This can be done by calculating the relative biases on cosmological
parameters, induced by a wrong assumption on the astrophysical ones. Such biases
can be naturally quantified using the ‘nested likelihoods’ (e.g., Ref. [3]) framework
in which the parameter space is split into two subsets: one containing all the
parameters of interest and the other, all those deemed ‘nuisance’ or systematic
461
Fig. 1 Angular power
spectrum C HI
at redshift 0.8,
using the best-fitting HI
astrophysical parameters
from the halo model and a
redshift binning of
Δz = 0.05. The error bars
shown in red represent the
standard deviation ΔC HI
l ,
calculated for a SKA 1
MID-like configuration
10
0
10
1
10
2
10
3
10
−8
10
−7
10
−6
C
z = 0.8
h Ω m n s Ω b σ 8
0.0
0.1
0.2
0.3
0.4
σ
A /A
Fixed astrophysics
With astrophysical prior
Marginalized
v c,0
β
Fixed cosmology
Without cosmological prior
Fig. 2 Left panel: Constraints on the cosmological parameters from a SKA I MID-like configuration, (1) marginalized over astrophysics, (2) with fixed astrophysics and (3) with the astrophysical
prior coming from the present data. Right panel: Astrophysical forecasts (1) without cosmological
priors and (2) with fixed cosmology. Figure adapted from Ref. [13]
Cross-correlating 21 cm intensity maps with galaxy surveys (both photometric
and spectroscopic) offer exciting prospects for constraining astrophysical and
cosmological parameters. It can be shown [12] that cross-correlating such surveys
covering similar redshift ranges and sky areas significantly improves astrophysical
constraints (see Fig. 3 for an example of a CHIME-like and DESI-like survey crosscorrelated in the northern hemisphere). Further, cross-correlation is a valuable tool
to mitigate the effects of contaminants and foregrounds, which are expected to be
significantly uncorrelated between the two surveys (e.g., Ref. [2]).
The Fisher formalism can also be used to quantify how a complementary effect,
the accuracy of cosmological constraints, is affected by the astrophysical prior
information. This can be done by calculating the relative biases on cosmological
parameters, induced by a wrong assumption on the astrophysical ones. Such biases
can be naturally quantified using the ‘nested likelihoods’ (e.g., Ref. [3]) framework
in which the parameter space is split into two subsets: one containing all the
parameters of interest and the other, all those deemed ‘nuisance’ or systematic