Validating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity Function
Although neural-network-based emulators enable efficient parameter estimation in 21 cm cosmology, the accuracy of such constraints is poorly understood. We employ nested sampling to fit mock data of the global 21 cm signal and high- z galaxy ultraviolet luminosity function (UVLF) and compare for the...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2023-01-01
|
| Series: | The Astrophysical Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-4357/ad003e |
| _version_ | 1797404121539543040 |
|---|---|
| author | J. Dorigo Jones D. Rapetti J. Mirocha J. J. Hibbard J. O. Burns N. Bassett |
| author_facet | J. Dorigo Jones D. Rapetti J. Mirocha J. J. Hibbard J. O. Burns N. Bassett |
| author_sort | J. Dorigo Jones |
| collection | DOAJ |
| description | Although neural-network-based emulators enable efficient parameter estimation in 21 cm cosmology, the accuracy of such constraints is poorly understood. We employ nested sampling to fit mock data of the global 21 cm signal and high- z galaxy ultraviolet luminosity function (UVLF) and compare for the first time the emulated posteriors obtained using the global signal emulator globalemu to the “true” posteriors obtained using the full model on which the emulator is trained using ARES . Of the eight model parameters we employ, four control the star formation efficiency (SFE) and thus can be constrained by UVLF data, while the remaining four control UV and X-ray photon production and the minimum virial temperature of star-forming halos ( ${T}_{\min }$ ) and thus are uniquely probed by reionization and 21 cm measurements. For noise levels of 50 and 250 mK in the 21 cm data being jointly fit, the emulated and “true” posteriors are consistent to within 1 σ . However, at lower noise levels of 10 and 25 mK, globalemu overpredicts ${T}_{\min }$ and underpredicts γ _lo , an SFE parameter, by ≈3 σ –4 σ , while the “true” ARES posteriors capture their fiducial values within 1 σ . We find that jointly fitting the mock UVLF and 21 cm data significantly improves constraints on the SFE parameters by breaking degeneracies in the ARES parameter space. Our results demonstrate the astrophysical constraints that can be expected for global 21 cm experiments for a range of noise levels from pessimistic to optimistic, as well as the potential for probing redshift evolution of SFE parameters by including UVLF data. |
| first_indexed | 2024-03-09T02:50:24Z |
| format | Article |
| id | doaj.art-7fdfd042f8114ca7bf55d8281b6c549e |
| institution | Directory Open Access Journal |
| issn | 1538-4357 |
| language | English |
| last_indexed | 2024-03-09T02:50:24Z |
| publishDate | 2023-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astrophysical Journal |
| spelling | doaj.art-7fdfd042f8114ca7bf55d8281b6c549e2023-12-05T13:24:11ZengIOP PublishingThe Astrophysical Journal1538-43572023-01-0195914910.3847/1538-4357/ad003eValidating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity FunctionJ. Dorigo Jones0https://orcid.org/0000-0002-3292-9784D. Rapetti1https://orcid.org/0000-0003-2196-6675J. Mirocha2https://orcid.org/0000-0002-8802-5581J. J. Hibbard3https://orcid.org/0000-0002-9377-5133J. O. Burns4https://orcid.org/0000-0002-4468-2117N. Bassett5https://orcid.org/0000-0001-7051-6385Center for Astrophysics and Space Astronomy, Department of Astrophysical and Planetary Sciences, University of Colorado Boulder , Boulder, CO 80309, USA johnny.dorigojones@colorado.eduCenter for Astrophysics and Space Astronomy, Department of Astrophysical and Planetary Sciences, University of Colorado Boulder , Boulder, CO 80309, USA johnny.dorigojones@colorado.edu; NASA Ames Research Center , Moffett Field, CA 94035, USA; Research Institute for Advanced Computer Science, Universities Space Research Association , Washington, DC 20024, USAJet Propulsion Laboratory, California Institute of Technology , 4800 Oak Grove Drive, Pasadena, CA 91109, USA; California Institute of Technology , 1200 E. California Boulevard, Pasadena, CA 91125, USACenter for Astrophysics and Space Astronomy, Department of Astrophysical and Planetary Sciences, University of Colorado Boulder , Boulder, CO 80309, USA johnny.dorigojones@colorado.eduCenter for Astrophysics and Space Astronomy, Department of Astrophysical and Planetary Sciences, University of Colorado Boulder , Boulder, CO 80309, USA johnny.dorigojones@colorado.eduCenter for Astrophysics and Space Astronomy, Department of Astrophysical and Planetary Sciences, University of Colorado Boulder , Boulder, CO 80309, USA johnny.dorigojones@colorado.eduAlthough neural-network-based emulators enable efficient parameter estimation in 21 cm cosmology, the accuracy of such constraints is poorly understood. We employ nested sampling to fit mock data of the global 21 cm signal and high- z galaxy ultraviolet luminosity function (UVLF) and compare for the first time the emulated posteriors obtained using the global signal emulator globalemu to the “true” posteriors obtained using the full model on which the emulator is trained using ARES . Of the eight model parameters we employ, four control the star formation efficiency (SFE) and thus can be constrained by UVLF data, while the remaining four control UV and X-ray photon production and the minimum virial temperature of star-forming halos ( ${T}_{\min }$ ) and thus are uniquely probed by reionization and 21 cm measurements. For noise levels of 50 and 250 mK in the 21 cm data being jointly fit, the emulated and “true” posteriors are consistent to within 1 σ . However, at lower noise levels of 10 and 25 mK, globalemu overpredicts ${T}_{\min }$ and underpredicts γ _lo , an SFE parameter, by ≈3 σ –4 σ , while the “true” ARES posteriors capture their fiducial values within 1 σ . We find that jointly fitting the mock UVLF and 21 cm data significantly improves constraints on the SFE parameters by breaking degeneracies in the ARES parameter space. Our results demonstrate the astrophysical constraints that can be expected for global 21 cm experiments for a range of noise levels from pessimistic to optimistic, as well as the potential for probing redshift evolution of SFE parameters by including UVLF data.https://doi.org/10.3847/1538-4357/ad003eNested samplingReionizationLuminosity functionNeural networksPosterior distributionRadio astronomy |
| spellingShingle | J. Dorigo Jones D. Rapetti J. Mirocha J. J. Hibbard J. O. Burns N. Bassett Validating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity Function The Astrophysical Journal Nested sampling Reionization Luminosity function Neural networks Posterior distribution Radio astronomy |
| title | Validating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity Function |
| title_full | Validating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity Function |
| title_fullStr | Validating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity Function |
| title_full_unstemmed | Validating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity Function |
| title_short | Validating Posteriors Obtained by an Emulator When Jointly Fitting Mock Data of the Global 21 cm Signal and High-z Galaxy UV Luminosity Function |
| title_sort | validating posteriors obtained by an emulator when jointly fitting mock data of the global 21 cm signal and high z galaxy uv luminosity function |
| topic | Nested sampling Reionization Luminosity function Neural networks Posterior distribution Radio astronomy |
| url | https://doi.org/10.3847/1538-4357/ad003e |
| work_keys_str_mv | AT jdorigojones validatingposteriorsobtainedbyanemulatorwhenjointlyfittingmockdataoftheglobal21cmsignalandhighzgalaxyuvluminosityfunction AT drapetti validatingposteriorsobtainedbyanemulatorwhenjointlyfittingmockdataoftheglobal21cmsignalandhighzgalaxyuvluminosityfunction AT jmirocha validatingposteriorsobtainedbyanemulatorwhenjointlyfittingmockdataoftheglobal21cmsignalandhighzgalaxyuvluminosityfunction AT jjhibbard validatingposteriorsobtainedbyanemulatorwhenjointlyfittingmockdataoftheglobal21cmsignalandhighzgalaxyuvluminosityfunction AT joburns validatingposteriorsobtainedbyanemulatorwhenjointlyfittingmockdataoftheglobal21cmsignalandhighzgalaxyuvluminosityfunction AT nbassett validatingposteriorsobtainedbyanemulatorwhenjointlyfittingmockdataoftheglobal21cmsignalandhighzgalaxyuvluminosityfunction |