What factors shape the radio luminosity of star-forming galaxies? A new calibration from LoTSS-DR2 2025 What factors shape the radio luminosity of star-forming galaxies? A new calibration from LoTSS-DR2 – 11

Shravya Shenoy, Daniel J. B. Smith, Sarah K. Biddle, Gülay Gürkan, Martin J. Hardcastle, Marina I. Arnaudova, Soumyadeep Das, Luke R. Holden, Gaoxiang Jin, Leah K. Morabito, Huub J. A. Röttgering · 2026

Read What factors shape the radio luminosity of star-forming galaxies? A new calibration from LoTSS-DR2 2025 What factors shape the radio luminosity of star-forming galaxies? A new calibration from LoTSS-DR2 – 11 by Shravya Shenoy, Daniel J. B. Smith, Sarah K. Biddle, Gülay Gürkan, Martin J. Hardcastle, Marina I. Arnaudova, Soumyadeep Das, Luke R. Holden, Gaoxiang Jin, Leah K. Morabito, Huub J. A. Röttgering (2026) 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-2601-15374.txt.

Introduction

Shravya Shenoy, Daniel J. B. Smith, and co-authors ask what sets the radio brightness of a star-forming galaxy. Their 2026 arXiv paper deals with the radio continuum to star formation rate relation. Radio light offers a view of star birth that dust cannot hide.

New evidence hints that stellar mass bends that relation. The authors ask which other galaxy properties shape it too. They build a new calibration from the LoTSS-DR2 survey data. A cleaner relation means radio surveys can read star formation rates with far more trust than ever before.

Catalog record

CREATORShravya Shenoy, Daniel J. B. Smith, Sarah K. Biddle, Gülay Gürkan, Martin J. Hardcastle, Marina I. Arnaudova, Soumyadeep Das, Luke R. Holden, Gaoxiang Jin, Leah K. Morabito, Huub J. A. Röttgering
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE71 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-15374

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

What factors shape the radio luminosity of star-forming galaxies? A new calibration from LoTSS-DR2 2025 What factors shape the radio luminosity of star-forming galaxies? A new calibration from LoTSS-DR2 – 11 Shravya Shenoy, Daniel J. B. Smith, Sarah K. Biddle, Gülay Gürkan, Martin J. Hardcastle, Marina I. Arnaudova, Soumyadeep Das, Luke R. Holden, Gaoxiang Jin, Leah K. Morabito, Huub J. A. Röttgering Abstract Radio observations offer a dust-unobscured view of galaxy star formation via the radio continuum-star formation rate (RC–SFR) relation. Emerging evidence of a stellar mass dependence in the RC–SFR relation raises the broader question of how other galaxy properties may influence this relation. In this work, we study the dependence of the global RC–SFR relation on galaxy properties in local ( z ≤ z\,\leq 0.3) star-forming galaxies (SFGs) using the second data release of the LOFAR Two-Metre Sky Survey (LoTSS-DR2). Employing a non-parametric decision-tree regression algorithm, we identify the most important galaxy properties for estimating the radio luminosity using a sample of 18,828 emission-line-classified SFGs based on spectroscopic data from the SDSS-DR8. Along with the spectroscopically obtained SFRs and stellar mass values, we also use SFRs and stellar masses derived using photometric SED-fitting from the GALEX –SDSS– WISE Legacy Catalogue (GSWLC) for the same sample. We find that a galaxy’s SFR is most important for predicting the radio luminosity, followed by the stellar mass, at > 5 ​ σ >5\sigma significance. Complementing the LoTSS catalogue 150 MHz flux densities with aperture photometry for the rest of the emission-line classified sample (35,099 galaxies in total), we obtain a new calibration of the RC–SFR relation, which does not change significantly whether we use spectroscopic or photometrically derived SFRs and stellar masses, despite the fact that the methods probe star formation on different characteristic timescales. Our results highlight the utility of decision-tree algorithms for handling censored radio-selected galaxy samples, which will be useful for future spectroscopic surveys of radio sources. 1 Introduction Exploring star formation in galaxies is important for understanding the evolution of baryonic matter. Traditionally, star formation rate (SFR) calibrations relied on ultraviolet and optical observations of ionising radiation produced by young and massive stars, because they dominate the total stellar luminosities of star-forming galaxies (see 69 ; 28 and 77 for reviews). However, reliably using optical and UV data to determine the intrinsic SFRs of galaxies is challenging, especially in the distant universe, due to the large and uncertain dust corrections required (e.g. 68 ). Dust absorbs a fraction of the UV/optical radiation arising from short-lived luminous stars, and the energy is then re-radiated at far-infrared (FIR) wavelengths, where the luminosity is therefore a generally reliable means of estimating SFRs in dusty galaxies (e.g. 101 ; 42 ; 68 ; 69 ). However, along with the practical difficulties associated with obtaining FIR observations in the post- Herschel era, there are also issues arising due to the low spatial resolution of FIR data ( 116 ; 82 , e.g.) , as well as the possible impact of unobscured sight-lines ( 100 ; 120 ; 119 ; 56 , e.g.) and additional heating from evolved stellar populations ( 101 , e.g.) . The dust that affects short-wavelength star formation indicators is irrelevant at radio frequencies, where a tight correlation between radio and far-infrared luminosities has been observed (e.g. 129 ; 41 ; 60 ; 64 ; 63 ; 118 ; 104 ; 84 ). Non-thermal radio emission in SFGs is attributed to electrons accelerated by supernova shocks moving in the ambient magnetic field ( 13 ) . Massive stars (\gtrsim 8\,{M}_{\odot} ) are responsible for core-collapse supernovae ( 35 ) , and so the observed non-thermal radio emission can be used as an indicator of the recent ( \simeq 100 Myr averaged) star formation rate of galaxies. Multiple observational works have studied the relationship between galaxies’ SFRs and their non-thermal synchrotron radio emission at 150 MHz including 50 ; 133 ; 117 ; 20 ; 58 ; 15 and 38 . These works assumed a power law relation between SFR and radio luminosity (at various frequencies), with the gradient often deviating from linearity. A gradient equal to one would mean that the galaxy is a perfect electron calorimeter and that the cosmic ray electrons lose all their energy before being transported away from the sites at which radiation is observed. However, the radio emission may underestimate the SFRs of galaxies compared to far-infrared tracers when a fraction of cosmic rays accelerated by supernovae escape the host galaxy ( 34 ) . At the same time, massive galaxies could retain cosmic ray electrons more efficiently due to deeper potentials, overestimating star-formation rates in galaxies and so giving rise to a super-linear slope of the RC–SFR power-law relation as observed e.g. by 39 . 133 report values as steep as \sim 1.35 using an infrared (60 \mu m) selected subsample of LOw-Frequency ARray Two-metre Sky Survey (LoTSS; 113 ) galaxies at 150 MHz. On spatially-resolved scales using integral field spectroscopy (IFS)-derived SFR, 65 report a superlinear slope of 1.16. Works including 50 and 117 have explored the dependence of the RC–SFR relation on stellar mass, demonstrating that including a mass-dependence reduces the scatter on the RC–SFR relation. Possible interpretations for stellar mass dependence of the RC–SFR relation are that more massive galaxies are better cosmic ray calorimeters or that stellar mass dependence comes from undiagnosed contamination by active galactic nuclei (AGN), since it is well-known that AGN are also associated with non-thermal GHz emission ( 14 , e.g.) and that the AGN fraction is dependent on stellar mass ( 106 ) . However, the stellar mass dependence in 50 and 117 extends to relatively low-mass galaxies ( <10^{10}\,{M}_{\odot} ) where the AGN fraction is small ( 135 ; 79 ; 15 ; 44 ; 38 ), consistent with results from simulations like The Tiered Radio Extragalactic Continuum Simulation or T-RECS ( 19 ) . Therefore, AGN contamination is unlikely to be the only explanation for stellar mass dependence in the RC–SFR relation. The stellar mass dependence of the RC–SFR relation prompts a broader question: which additional galaxy properties might influence the relation? To address this question, we generate a non-parametric random forest (RF) regression model that predicts the radio luminosity of SFGs using galaxy properties (which can be derived using photometric, spectroscopic, or a mixture of the two types of observations). The application of RF regressors ( 23 ) to explore fundamental scaling relations in extragalactic studies is becoming increasingly popular (see, e.g. 110 ; 18 ; 17 ; 103 ; 10 ; 11 ; 78 ; 111 among others), since ensemble learning methods with decision trees like RF regression can identify highly non-linear relationships in multi-dimensional datasets. While RF regressors cannot predict analytical relations between the input and target variables directly, the relative importance of features that are included in the model can be estimated, leading to a better intuitive understanding of the model. In this work, we use a sample of 35,099 emission-line classified SFGs, of which 18,828 are detected at \geq 3\sigma at 150 MHz. We apply RF regression to the radio-detected sources (18,828 sources) to investigate the dependence of radio luminosity on galaxy features including SFR, stellar mass,

Read the complete text