An Improved Statistical Point-source Foreground Model for the Epoch of Reionization
© 2017. The American Astronomical Society. All rights reserved.. We present a sophisticated statistical point-source foreground model for low-frequency radio Epoch of Reionization (EoR) experiments using the 21 cm neutral hydrogen emission line. Motivated by our understanding of the low-frequency ra...
| Main Authors: | , , |
|---|---|
| Format: | Journal Article |
| Published: |
Institute of Physics Publishing
2017
|
| Online Access: | http://purl.org/au-research/grants/arc/DE140100316 http://hdl.handle.net/20.500.11937/57761 |
| _version_ | 1848760091190231040 |
|---|---|
| author | Murray, Steven Trott, Cathryn Jordan, C. |
| author_facet | Murray, Steven Trott, Cathryn Jordan, C. |
| author_sort | Murray, Steven |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | © 2017. The American Astronomical Society. All rights reserved.. We present a sophisticated statistical point-source foreground model for low-frequency radio Epoch of Reionization (EoR) experiments using the 21 cm neutral hydrogen emission line. Motivated by our understanding of the low-frequency radio sky, we enhance the realism of two model components compared with existing models: the source count distributions as a function of flux density and spatial position (source clustering), extending current formalisms for the foreground covariance of 2D power-spectral modes in 21 cm EoR experiments. The former we generalize to an arbitrarily broken power law, and the latter to an arbitrary isotropically correlated field. This paper presents expressions for the modified covariance under these extensions, and shows that for a more realistic source spatial distribution, extra covariance arises in the EoR window that was previously unaccounted for. Failure to include this contribution can yield bias in the final power-spectrum and under-estimate uncertainties, potentially leading to a false detection of signal. The extent of this effect is uncertain, owing to ignorance of physical model parameters, but we show that it is dependent on the relative abundance of faint sources, to the effect that our extension will become more important for future deep surveys. Finally, we show that under some parameter choices, ignoring source clustering can lead to false detections on large scales, due to both the induced bias and an artificial reduction in the estimated measurement uncertainty. |
| first_indexed | 2025-11-14T10:10:16Z |
| format | Journal Article |
| id | curtin-20.500.11937-57761 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T10:10:16Z |
| publishDate | 2017 |
| publisher | Institute of Physics Publishing |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-577612022-09-06T04:28:48Z An Improved Statistical Point-source Foreground Model for the Epoch of Reionization Murray, Steven Trott, Cathryn Jordan, C. © 2017. The American Astronomical Society. All rights reserved.. We present a sophisticated statistical point-source foreground model for low-frequency radio Epoch of Reionization (EoR) experiments using the 21 cm neutral hydrogen emission line. Motivated by our understanding of the low-frequency radio sky, we enhance the realism of two model components compared with existing models: the source count distributions as a function of flux density and spatial position (source clustering), extending current formalisms for the foreground covariance of 2D power-spectral modes in 21 cm EoR experiments. The former we generalize to an arbitrarily broken power law, and the latter to an arbitrary isotropically correlated field. This paper presents expressions for the modified covariance under these extensions, and shows that for a more realistic source spatial distribution, extra covariance arises in the EoR window that was previously unaccounted for. Failure to include this contribution can yield bias in the final power-spectrum and under-estimate uncertainties, potentially leading to a false detection of signal. The extent of this effect is uncertain, owing to ignorance of physical model parameters, but we show that it is dependent on the relative abundance of faint sources, to the effect that our extension will become more important for future deep surveys. Finally, we show that under some parameter choices, ignoring source clustering can lead to false detections on large scales, due to both the induced bias and an artificial reduction in the estimated measurement uncertainty. 2017 Journal Article http://hdl.handle.net/20.500.11937/57761 10.3847/1538-4357/aa7d0a http://purl.org/au-research/grants/arc/DE140100316 Institute of Physics Publishing restricted |
| spellingShingle | Murray, Steven Trott, Cathryn Jordan, C. An Improved Statistical Point-source Foreground Model for the Epoch of Reionization |
| title | An Improved Statistical Point-source Foreground Model for the Epoch of Reionization |
| title_full | An Improved Statistical Point-source Foreground Model for the Epoch of Reionization |
| title_fullStr | An Improved Statistical Point-source Foreground Model for the Epoch of Reionization |
| title_full_unstemmed | An Improved Statistical Point-source Foreground Model for the Epoch of Reionization |
| title_short | An Improved Statistical Point-source Foreground Model for the Epoch of Reionization |
| title_sort | improved statistical point-source foreground model for the epoch of reionization |
| url | http://purl.org/au-research/grants/arc/DE140100316 http://hdl.handle.net/20.500.11937/57761 |