Skip to content

Latest commit

 

History

History
133 lines (107 loc) · 4.53 KB

README.md

File metadata and controls

133 lines (107 loc) · 4.53 KB
Raw

HMcode2020 Emulator

Emulated HMcode2020 non-linear power spectra for fast weak lensing analysis.

Installation

To install it, just clone the repository, go to the folder and do

pip install . [--user]

Requirements

Required python packages:

  • numpy
  • scipy

For tutorials:

  • matplotlib
  • emcee
  • getdist

Usage

import HMcode2020Emu as hmcodeemu


params = {
    'omega_cdm'     :  [0.315],
    'As'            :  [np.exp(3.07)*1.e-10],
    'omega_baryon'  :  [0.05],
    'ns'            :  [0.96],
    'hubble'        :  [0.67],
    'neutrino_mass' :  [0.0],
    'w0'            :  [-1.0],
    'wa'            :  [0.0],
    'log10TAGN'     :  [7.8],
    'z'             :  [0.]
}
emulator = hmcodeemu.Matter_powerspectrum()

k_lin, pk_lin_total = emulator.get_linear_pk(nonu=False, **params)
k_lin, pk_lin_nonu = emulator.get_linear_pk(nonu=True, **params)

k_nonlin, pk_nonlin_total = emulator.get_nonlinear_pk(baryonic_boost=True, **params)

Note that for neutrino calculations we assume 2 massless neutrinos and 1 massive neutrino with the mass equal to 'neutrino_mass'. You can also see an example in the tutorials-folder as well as an initialising CAMB file.

Parameter ranges

parameter limits
omega_cdm [0.1, 0.8]
omega_baryon [0.01, 0.1]
hubble [0.4, 1.]
As [0.495e-9, 5.459e-9]
ns [0.6, 1.2]
neutrino_mass [eV] [0., 0.5]
w0 [-3., -0.3]
wa [-3., 3.]
log10TAGN [7.6, 8.3]
z [0.0, 4.]
:---: :---:
k_lin [h/Mpc] [3.7e-4, 50]
k_nonlin [h/Mpc] [0.01, 50]

Note that $w = w_0 + (1-a)w_a$ must be negative at all redshifts, hence we impose the following condition: $w_0 + w_a \leq 0$. One could add it on the level of priors into analysis too.

Citation

If you use HMcode2020Emu at any point in your work please cite the HMcode2020 paper:

@article{Mead2020,
        author = {Mead, Alexander and Brieden, Samuel and Tr\"oster, Tilman and Heymans, Catherine},
        title = {HMcode-2020: Improved modelling of non-linear cosmological power spectra with baryonic feedback},
        journal={Monthly Notices of the Royal Astronomical Society},
        publisher={Oxford University Press (OUP)},
        year = {2021},
        month = {Mar},
        volume = {502},
        number = {1},
        ISSN = {0035-8711},
        url = {https://doi.org/10.1093/mnras/stab082},
        pages = {1401-1422},
        DOI = {10.1093/mnras/stab082},
        archivePrefix = {arXiv},
        eprint = {2009.01858},
        primaryClass = {astro-ph.CO},
}

the CosmoPower paper:

@article{SpurioMancini2022,
         title={CosmoPower: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys},
         volume={511},
         ISSN={1365-2966},
         url={http://dx.doi.org/10.1093/mnras/stac064},
         DOI={10.1093/mnras/stac064},
         number={2},
         journal={Monthly Notices of the Royal Astronomical Society},
         publisher={Oxford University Press (OUP)},
         author={Spurio Mancini, Alessio and Piras, Davide and Alsing, Justin and Joachimi, Benjamin and Hobson, Michael P},
         year={2022},
         month={Jan},
         pages={1771–1788}
         }

as well as this paper where the similar data production and emulation pipelines have been used:

@article{Tsedrik2024,
    author = Tsedrik, Maria and Bose, Benjamin and Carrilho, Pedro and Pourtsidou, Alkistis and Pamuk, Sefa and Casas, Santiago and Lesgourgues, Julien,
    title = {Stage-IV Cosmic Shear with Modified Gravity and Model-independent Screening},
    journal = {JCAP},
    year = {2024},
    month = {Oct},
    volume = {2024},
    number = {10},
    eid = {099},
    pages = {099},
    doi = {10.1088/1475-7516/2024/10/099},
    eprint = {2404.11508},
    archivePrefix = {arXiv}
}

License

MIT