ODIN: A New Lyman Alpha Blob Selection Method, Sample, and Statistical Analysis at $z\sim3.1$

TL;DR

ODIN targets Ly$\alpha$ blobs (LABs) at $z\sim3.1$ with a large, uniform narrowband survey to overcome small-number statistics and cosmic variance. It introduces two complementary LAB selection pipelines—the traditional extended-LAE method and a new extended-beyond-continuum (Tractor-based) approach—and reports $112$ LABs in the E-COSMOS field, including $23$ LABs uniquely detected by the Tractor method. Spectroscopic confirmations from Gemini/GMOS and DESI/literature, along with environment-focused analyses, show that proto-cluster regions host higher LAB densities and brighter Ly$\alpha$ halos, indicating environment dependence of LAB statistics and luminosities. ODIN will expand to $\sim100$ deg$^{2}$ across six additional fields and two redshift windows, enabling robust tests of LAB formation mechanisms and their connection to large-scale structure.

Abstract

Ly$α$ blobs (LABs) are large, spatially extended Ly$α$-emitting objects whose nature remains unclear. Their statistical properties such as number densities and luminosity functions are still uncertain because of small sample sizes and large cosmic variance. The One-hundred-deg$^2$ DECam Imaging in Narrowbands (ODIN) survey, with its large volume, offers an opportunity to overcome these limitations. We describe our LAB selection method and present 112 new LABs in the 9 deg$^2$ E-COSMOS field. We begin with the conventional LAB selection approach, cross-matching LAEs with extended Ly$α$ sources, yielding 89 LAB candidates. To obtain a more complete LAB sample, we introduce a new selection pipeline that models all galaxies detected in deep broadband imaging, subtracts them from the narrowband image, and then directly detects extended Ly$α$ emission. This method successfully identifies 23 additional low-surface-brightness LABs which could otherwise be missed by the conventional method. The number density of ODIN LABs near an ODIN protocluster ($n=7.5\times10^{-5}$ cMpc$^{-3}$) is comparable to that found in the SSA22 proto-cluster and is four times higher than the average across the field. The cumulative Ly$α$ luminosity function within the protocluster regions is similar to that measured for the LABs in the SSA22 proto-cluster, suggesting a large excess of luminous LABs relative to the average field. These findings suggest the Ly$α$ luminosities and number densities of LABs are environment-dependent. ODIN will provide an expansive LAB and protocluster samples across six additional fields and two more redshifts, allowing us to investigate the nature of LABs in relation to their environments.

Figures (14)

• Figure 1: Size-luminosity relation for LAB candidates selected using the extended-LAE selection method. Blue circles represent the LAB candidates. The ODIN LAE sample Firestone24 shows two sequences. One extends to LABs implying that they could be LAH candidates, while the other follows the simulated point sources, indicating that our LAB candidates are distinct from bright and compact point sources. Black dotted and solid lines represent 1.5$\sigma_{\rm SB_1}$ and 2$\sigma_{\rm SB_1}$, respectively, where $\sigma_{\rm SB_1}$ represents the detection threshold per 1 arcsec$^2$. The blue solid lines outline the LAB selection criteria: size over 20 arcsec$^{2}$ and 3$\sigma$ above the size-luminosity relation defined by the simulated point sources.
• Figure 2: Postage stamp ($20\times20$ arcsec$^{2}$) images for examples of diffuse and extended Ly$\alpha$ halos identified by our visual inspection. These objects are not selected as LAB in extended-LAE method, requiring additional LAB selection is necessary for assembling a complete LAB samples. Left: Color image constructed with $N501$, $g$, and $r$ bands. Green shows the diffuse and extended signal in the $N501$ image. Middle: High contrast $N501$ image revealing the diffuse and extended source. Right: Filtered Ly$\alpha$ image. Yellow contours ($1.5\sigma_{\rm SB1}=3.3\times10^{-18}$ erg s$^{-1}$ cm$^{-2}$ arcsec$^{-2}$) highlighting the diffuse and extended Ly$\alpha$ emission.
• Figure 3: Example $192\times192$ pixel$^2$ (52$\times$52) HSC SSP $g$ band image that illustrates our tiling strategy. We use $64\times64$ pixel$^2$ sub-tile images to perform model fitting for $g$-band galaxies. To ensure accurate model fitting and avoid fragmentation near the sub-tile edges, We offset each tile by half a tile-size from its nearest adjoining tile. This pattern is illustrated by the solid blue and red dashed squares, whose centers are defined by the blue circles and red crosses, respectively.
• Figure 4: Postage stamp images ($64\times64$ pixel$^2$) demonstrating extended-beyond-continuum method. Top row: HSC SSP $g$-band image (left), the Tractor model image of galaxies (center), and the residuals from the fits (right). The Tractor models do an excellent job reproducing the continuum emission from the galaxies. Middle row: the same set of images for the $N501$ frame. The galaxy models have the exact same shape parameters as for the $g$-band data. Note the excess Ly$\alpha$ flux on the residual image. Bottom row: The residual Ly$\alpha$ image which is used for the detection of LABs in the extended-beyond-continuum method (right). The image is constructed by only the residual $N501$ image. The flux and sizes are measured using the Ly$\alpha$ image from extended-LAE method (left).
• Figure 5: Cutout images ($3\arcmin\times3\arcmin$) illustrating the robustness of the Tractor-based residual Ly$\alpha$ detection image. Top: $N501$ and $g$ band images. A bright internal reflection (dark stripe) is clearly visible across the $g$-band image, introducing large-scale artifacts. Bottom: Ly$\alpha$ detection images from the extended-LAE (left) and the extended-beyond-continuum methods, respectively. Yellow contours indicate regions exceeding the detection threshold for Ly$\alpha$ blob candidates identified by SExtractor. In the extended-LAE method (Section \ref{['sec:traditional_LAB_selection']}), the artifact is propagated into the Ly$\alpha$ detection image, and the over-subtracted background near this artifact leads to numerous false detections (yellow contours). In contrast, the residual Ly$\alpha$ image from Tractor is effectively free of this artifact, significantly reducing spurious detections.