A Sample of Nearby Isolated Dwarf Galaxies: A First Look at the Mass Function of Field Dwarfs

TL;DR

ELVES-Field creates a volume-limited census of isolated Local Volume dwarfs (D < $10$ Mpc) over ~ $3{,}000$ deg$^2$ to robustly measure the field-dwarf SMF and test environmental effects. It deploys a semi-automated, LSB-focused detection pipeline on Legacy Surveys data, supplemented by deep Subaru/HSC imaging for SBF-based distances, achieving distance confirmation for a large sample and a probabilistic framework for unconfirmed dwarfs. The resulting SMF agrees with redshift-based surveys and IllustrisTNG predictions at the high-mass end, and forward-modeling demonstrates consistency across environments amid substantial cosmic variance. The study provides a scalable blueprint for Rubin/Euclid/Roman-era work to build larger, distance-confirmed field-dwarf samples and probe small-scale structure and quenching in isolated systems.

Abstract

We present the results of the Exploration of Local VolumE Survey - Field (ELVES-Field), a survey of the dwarf galaxies in the Local Volume (LV; $D<10$ Mpc) over roughly $3,000$ square degrees, focusing on the field dwarf population. Candidates are detected using a semi-automated algorithm tailored for low surface brightness dwarfs. Using tests with injected galaxies, we show the detection is $50\%$ complete to $m_g\sim20$ mag and $M_\star \sim 10^6$ $M_\odot$. Candidates are confirmed to be true nearby dwarfs through distance measurements including redshift, tip of the red giant branch, and surface brightness fluctuations. We identify isolated, field dwarfs using various environmental criteria. Over the survey footprint, we detect and confirm 95 LV dwarfs, 44 of which we consider isolated. Using this sample, we infer the field dwarf mass function and find good agreement at the high-mass end with previous redshift surveys and with the predictions of the IllustrisTNG simulation. This sample of isolated, field dwarfs represents a powerful dataset to investigate aspects of small-scale structure and the effect of environment on dwarf galaxy evolution.

Figures (9)

• Figure 1: The area surveyed for LV field dwarfs in ELVES-Field. The area shown is specifically the Legacy Survey footprint that is used in the object detection. These footprints are chosen because they also have existing deeper archival HSC imaging which is used for the SBF measurements. HSC data coming from the HSC-SSP versus other archival sources are shown in different colors. The two large, gray excluded areas correspond to the Virgo Cluster and the NGC 5846 group and are not searched for dwarfs. The points show the actual dwarf candidates that we detect with the different markers indicating the results of the distance confirmation step in §\ref{['sec:distances']}.
• Figure 2: A demonstration of the dwarf detection process for a portion of the LSDR10 brick '1849p062'. The left panel shows a $grz$ composite of the original LSDR10 imaging. Detections coming from the Sienna Galaxy Atlas moustakas2023 and tractor that pass our photometric cuts are shown in blue solid and pink dashed ellipses, respectively. The right panel shows the image after the tractor models for stars, high surface brightness sources, and very red sources are subtracted out. The background in the right panel appears less flat due to a more aggressive image stretch to bring out faint and low surface brightness objects. On this cleaned image, a separate low surface brightness-focused detection step is run. The resulting detections that pass our photometric cuts are shown in the green dotted ellipses.
• Figure 3: An overview of the photometric cuts applied to detected objects to isolate possible LV dwarfs. Each panel shows a different projection of photometric space and the region used to select dwarf candidates. The titles of each panel give the actual selection cut used. Effective radii are measured in arcseconds. In the bottom left panel, $\epsilon$ refers to the ellipticity of the source. For reference, we show known LV dwarfs, including ELVES satellites from carlsten2022 and field LV dwarfs from karachentsev2013. The small black points show the distribution of all sources from Legacy Survey photometry for a representative sample of LSDR10 bricks.
• Figure 4: The completeness of the survey as shown through injection of artificial dwarfs (left) and recovery of known LV dwarfs (right). For the injection tests, we inject artificial dwarfs across a range in size and magnitude and quantify the recovery fraction. The left panel shows that the recovery is generally $\gtrsim 90\%$ above the size and luminosity cuts applied and brighter than $\mu_{\mathrm{eff}}<27$ mag arcsec$^{-2}$ in surface brightness. The dot-dashed green line shows the mass-size relation of carlsten2021a at a distance of 8 Mpc for reference. The right panel confirms this completeness level for actual known LV dwarfs. Filled green circles are LV dwarfs that are recovered by the detection algorithm while empty red circles are missed.
• Figure 5: Assuming dwarfs are uniformly distributed in space and follow the mass-size relation of carlsten2021a, we collapse the completeness estimate shown in Figure \ref{['fig:completeness']} to an estimate of the completeness as just a function of apparent $g$ band magnitude (left) and stellar mass (right). We do this separately for dwarfs with $g-r$ color characteristic of a star-forming dwarf (blue) and that characteristic of a quenched dwarf (red). The green points show the completeness of the detection algorithm to the same reference sample of known LV dwarfs shown in Figure \ref{['fig:completeness']}.