Clustering Confuses Spectro-photometry: An Investigation of 2D and 3D Forced Profile Matching for Stacking Line-intensity Mapping Data on Source Catalogues numpy ( 20 ) ; scipy ( 34 ) ; Matplotlib ( 21 ) . This research made use of Photutils, an Astropy package for detection and photometry of astron
Ella M. Mansfield, Delaney A. Dunne, Dongwoo T. Chung · 2025
Read Clustering Confuses Spectro-photometry: An Investigation of 2D and 3D Forced Profile Matching for Stacking Line-intensity Mapping Data on Source Catalogues numpy ( 20 ) ; scipy ( 34 ) ; Matplotlib ( 21 ) . This research made use of Photutils, an Astropy package for detection and photometry of astron by Ella M. Mansfield, Delaney A. Dunne, Dongwoo T. Chung (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-22356.txt.
Introduction
Ella M. Mansfield, Delaney A. Dunne, and Dongwoo T. Chung sharpen a technique for finding faint cosmic signals by stacking. Line-intensity mapping charts the universe at low resolution through spectral line glow, and stacking co-adds small map cutouts at known emitter positions to lift signal out of noise. Their upgrade fits an assumed signal shape rather than simply averaging, in two spatial dimensions or in three including the spectral axis.
They also stack at source coordinates more precise than the map's resolution. Tested on simulations mimicking the COMAP Pathfinder, the fitting methods gained up to twenty-five percent in detection significance over the plain stack. The best profile was wider than the telescope beam and Lorentzian in frequency. Clustering effects, including fingers-of-God smearing, shape what works best.
Catalog record
| CREATOR | Ella M. Mansfield, Delaney A. Dunne, Dongwoo T. Chung |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-22356
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Clustering Confuses Spectro-photometry: An Investigation of 2D and 3D Forced Profile Matching for Stacking Line-intensity Mapping Data on Source Catalogues numpy ( 20 ) ; scipy ( 34 ) ; Matplotlib ( 21 ) . This research made use of Photutils, an Astropy package for detection and photometry of astronomical sources ( 6 ) . Ella M. Mansfield, Delaney A. Dunne, Dongwoo T. Chung Abstract Line-intensity mapping (LIM) is an emerging observational technique that is used to observe the universe on large scales at low resolution through spectral line emission. Stacking analyses coadd cutouts of LIM data on positions of known signal emitters, robustly detecting signal otherwise hidden in a noisy map. In this article, we present two augmentations of a stacking pipeline, both aiming to refine the sensitivity of the stack by assuming a specific observed signal shape in 2D spatial axes or 3D spatial and spectral axes, as well as stacking on source coordinates more precise than the resolution of the LIM data cube. We test these methods on a series of simplistic and complex simulations mimicking observations with the CO Mapping Array Project (COMAP) Pathfinder. We find that these fitting methods provide up to a 25% advantage in detection significance over the original stack method in realistic COMAP-like simulations. We also find that the optimal fitting profile, given our CO line model and galaxy model, is larger than the 5 ′ 5^{\prime} width of the COMAP beam and takes on a Lorentzian shape in the spectral dimension. Our findings suggest a nuanced dependence of the optimal profile size and shape on the LIM signal itself, including redshift-space clustering and fingers-of-God effects that depend on the tracer luminosity function and bias. I Introduction Average L^{\prime}_{\rm CO} Average L^{\prime}_{\rm CO} Average L^{\prime}_{\rm CO} Average L^{\prime}_{\rm CO} L^{\prime}_{\rm CO} (1) L^{\prime}_{\rm CO} (2) L^{\prime}_{\rm CO} (3) Figure 1: A depiction of the stacking analysis and its three different stacking methods. Line intensity mapping (LIM) is an observational technique that considers the aggregate spectral line emission of all sources in an observation field across large patches of the sky ( 22 ; 4 , see reviews by, e.g.:) . LIM is primarily used to obtain low-resolution maps of the universe at different redshifts to understand its structure on very large scales. The Carbon monOxide Mapping Array Project (COMAP) Pathfinder ( 13 ) is an example of an experiment using this technique to trace molecular carbon monoxide (CO) transitions. The rotational CO(1–0) transition at rest-frame emission frequency 115.27 GHz is a valuable tracer of star formation. It can trace cool molecular clouds in star-forming regions most directly ( 7 , for a review, see) compared to other typical LIM target lines (e.g. H \alpha , [C ii ], 21 cm H i ), which only trace them indirectly or not at all. The COMAP Pathfinder focuses on mapping this transition between redshifts 2.4 to 3.4 , corresponding to a frequency range of 26-34 GHz. For a more complete description of the instrumentation of the COMAP Pathfinder, see 23 . COMAP is an advanced, dedicated facility for star-formation LIM, a subfield where similar experiments are still in an era of commissioning—e.g., CONCERTO ( 16 ) and TIME ( 35 ) for [C ii ]—and targeting first detections. Early detections, even of the brightest peaks of the line-intensity signal, present a significant challenge for these experiments. The low resolution instrumentation used in LIM leads to signal from unresolved sources being attenuated across a single pixel, producing a lower signal to noise ratio (SNR) and making detections more difficult. These reasons strongly motivate the optimization of analysis methods to accelerate the initial detection of a signal or provide more accurate upper limits on the luminosities of undetected signals. In this article, we thoroughly analyze two augmentations of a COMAP data analysis pipeline first presented in 19 used to ‘stack’ the COMAP data cube on sources in the eBOSS quasar sample ( 15 ) . Combining LIM data with catalogues of known emitters is a robust way to make more confident detections, as shown repeatedly in the context of low-redshift 21 cm H i LIM ( 8 ; 27 ; 1 ; 9 ) and most recently in z\sim 3 Lyman- \alpha ( \mathrm{Ly}\alpha ) LIM using the Hobby–Eberly Telescope Dark Energy eXperiment ( 25 , HETDEX;) . The augmentations to the stacking pipeline presented in this work assume specific signal profiles in 2D and 3D with the goal of lowering the uncertainty around recovered line luminosity values, thus increasing the SNR with which we recover said line luminosity. In addition to incorporating prior information about the stack signal, the assumption of such profiles also allows for sub-voxel positioning of those profiles in fitting. This may provide an advantage over simply coadding cutouts at the voxel level, as the coordinates of the CO sources stacked on are typically known to an accuracy far greater than the resolution of COMAP. For instance, source coordinates in the HETDEX catalogue of \rm{Ly}\alpha emitters—a primary cross-correlation target for COMAP ( 11 ) —are known to within a few arcseconds ( 28 ) . This is about 40 \times smaller than COMAP pixels, which are 2^{\prime} wide to Nyquist-sample the beam FWHM of 4.5 ′ at 30 GHz ( 23 ) . In the spectral axis, the COMAP data are binned to a 31.25 MHz resolution for scientific analysis, for a resolving power of \nu/\delta\nu\sim 1000 much better matched to the HETDEX redshift accuracy. By supplying the stacking analysis of LIM data with information such as signal profile models in the form of point response functions (PRF; thus weighting most heavily the voxels which contain the most signal), and sub-pixel/sub-voxel coordinates (thus concentrating the signal with respect to the noise), we aim to achieve more accurate and confident signal recovery in contexts like the COMAP Pathfinder. The work is extremely timely as the latest results from COMAP Pathfinder stacks ( 18 ) on external catalogs are beginning to disfavor an increasing volume of the parameter space for the luminosity function of CO emission. We devise and test our augmented profile-based stacking pipelines in the limit of uniformly distributed sources with completely non-overlapping profiles, and on realistically clustered simulations of CO emitters. In doing so, this work aims to answer the following questions: • Can the forced fitting of 2D and 3D profiles at the sub-pixel level enhance the accuracy and significance of existing LIM stacking methods? • Do the optimal parameters for these profiles depend solely on the instrument and survey design, or do they also depend on the LIM signal itself? We structure the paper as follows: Section 2 provides a description of each of the three stacking methods presented in this article and how they derive a final luminosity and associated uncertainty. Section 3 and Section 4 discuss simulated tests discerning the behavior of the two forced stacking methods against the traditional stacking method. Section 5 discusses the findings before we consider future directions in the concluding Section VI . Where relevant, we assume a flat \Lambda CDM cosmology with \Omega_{m}=0.286 , \Omega_{b}=0.047 , H_{0}=100h km s -1 Mpc -1 with h=0.7 , and \sigma_{8}=0.82 , for consistency with