Constraints on the Hot Circumgalactic Medium around Nearby L* Galaxies from SRG/eROSITA All Sky Survey

TL;DR

This study presents the first systematic search for a hot circumgalactic medium around nearby L* galaxies using the SRG/eROSITA all-sky survey and image stacking. The diffuse soft X-ray emission is detected out to ~50 kpc and is best described by a hot thermal component with a radial distribution modeled by a PSF-convolved β-profile, giving $R_c ≈ 8.2$ kpc and $β ≈ 0.50$, corresponding to a hot gas mass of $M_{hot} ≈ 2×10^{10} M_sun$ and a 0.5–2 keV luminosity of $L_{0.5-2} ≈ 6×10^{39}$ erg s^-1 per galaxy within 10–200 kpc. Spectral analysis favors a predominantly thermal origin with a best-fit temperature around $T ≈ 0.23$ keV and a log-normal dispersion, arguing against a non-thermal, CR-dominated halo. The observed profile agrees with IllustrisTNG50 MW analog predictions, supporting current feedback implementations, and the results provide empirical benchmarks to calibrate hot CGM physics in next-generation cosmological simulations. The hot CGM content appears to correlate with stellar mass, star formation activity, and AGN presence, offering insights into how feedback shapes halo gas in the local universe.

Abstract

The circumgalactic medium (CGM) is a multi-phase, dynamic interface between galaxy and the intergalactic medium, providing crucial diagnostics of galaxy evolution. However, direct evidence for a hot (million-Kelvin) CGM around present-day L* galaxies remains elusive. Here, we present the first systematic search of the hot CGM around nearby (< 50 Mpc) L* galaxies, by stacking their X-ray images and spectra from the SRG/eROSITA all-sky survey. Significant diffuse X-ray emission is detected out to ~ 50 kpc, with spectral signatures consistent with a hot gas but arguing against a predominantly non-thermal origin. The radial distribution and total amount of the hot gas are in agreement with prediction by IllustrisTNG simulations. The constraints on the hot CGM derived in this study hold promise for calibrating key physical processes in next-generation cosmological simulations.

Figures (11)

• Figure 1: Spatial distribution and galaxy properties of our sample galaxies. (a): Projected sky position (in Galactic coordinates) of the 474 $L^*$ galaxies (magenta cicles), with the symbol size scaling inversely with the galaxy's distance. (b): Histogram of the B-band absolute magnitude of the parent sample, i.e. the 50 Mpc Galaxy Catalog (grey) and our final sample of $L^*$ galaxies (magenta). (c)--(e): Histograms of the distance, star formation rate and stellar mass of the sample galaxies. The high-SFR and low-SFR hosts are shown in blue and red, respectively, and the the full sample in black. The vertical dashed lines in panels (c) and (e) indicate the median values of the corresponding samples, while in panel (d) the vertical dashed line separates the high-SFR and low-SFR hosts.
• Figure 2: Background-subtracted 0.5--2 keV radial intensity profiles of the full sample and various subsamples of $L^*$ galaxies. (a): Radial intensity profiles of the full sample (black circles with error bars), adaptively binned to achieve a signal-to-noise ratio greater than 2 in each bin. The angular scale has been converted to a physical scale according to a nominal distance of 50 Mpc, while the observed 0.2--2.3 keV net count rate has been converted into unabsorbed energy flux by adopting the best-fit spectral model to the stacked halo X-ray spectrum (see supplementary_methods). The last radial bin with the arrow denotes 2 $\sigma$ upper limit. The blue dashed and red solid curves show the normalized WISE near-infrared starlight profile (for the disk component; supplementary_methods) and PSF-convolved $\beta$-model (for the halo component), respectively, while their sum is shown by the cyan dash-dotted curve. For comparison, the eROSITA 0.5--2 keV intensity profile obtained by stacking $\sim 3\times10^4$ Milky Way-sized galaxies mainly drawn from the Sloan Digital Sky Survey, taken from Zhang_2024, is shown with purple symbols. Also shown as the green solid curve is the mean intensity profile calculated for 152 Milky Way analogs in the Illustris-TNG50 simulations, as defined by Pillepich_2024. This simulation-predicted profile has been convolved with the eROSITA PSF, and its 1 $\sigma$ scatter is shown by the light green strip. (b): Background-subtracted 0.5--2 keV intensity profiles of the high-mass (blue) and low-mass (red) galaxies. The TNG50-predicted profile in panel (a) is also overlaid, which compares better with the high-mass subgroup with stellar masses more comparable to the MW. (c): Similar to panel (b), but for the high-SFR (blue) and low-SFR (red) galaxies. (d): Similar to panel (b), but for the AGN hosts (blue) and non-AGN hosts (red).
• Figure 3: Background-subtracted stacked eROSITA X-ray spectrum, extracted from the radial range of 10--50 kpc around the full sample of $L^*$ galaxies. The spectrum has been adaptively grouped to achieve a signal-to-noise ratio greater than 2 per bin. From left to right shows the fitted model: an absorbed power-law, an absorbed single-temperature thermal plasma plus a fixed power-law, and an absorbed thermal plasma with a log-normal temperature distribution plus a fixed power-law. In all panels, the total model is represented by a solid black curve. In panels (b) and (c), the hot plasma component and the power-law component are indicated by red dashed and blue dotted lines, respectively. See details in supplementary_methods.
• Figure 4: Stacked surface brightness image and azimuthal intensity profiles of a subsample of highly inclined $L^*$ galaxies.Left: Stacked background-subtracted 0.2--2.3 keV surface brightness image of 114 galaxies with a highly inclined disk, defined as having $b/a<0.5$, where $a$ and $b$ are the semi-major and semi-minor axes of the $I_{22,W1}$ isophote derived from WISE W1 images (see supplementary_methods). The X-ray images of individual galaxies have been rotated such that their major-axis is aligned with the horizontal (X) direction. The white contours are derived from the WISE image aligned and stacked in the same way, illustrating the extent of the inclined stellar disk. Right: Azimuthal 0.2--2.3 keV intensity profiles extracted from the stacked X-ray image of the highly inclined galaxies, over a radial range of 10--50 kpc (black) and 10--30 kpc (blue), which avoids the disk region. A position angle of $0^{\circ}$ corresponds to the positive X-axis (i.e. the galactic plane) and $90^{\circ}$ indicates the minor-axis. Each bin is $36^{\circ}$ wide, with counts from regions above and below the galactic plane combined to improve the the $S/N$. The horizontal lines represent the average intensity. Both profiles are consistent with no significant azimuthal dependence. Synthetic intensity profiles of the TNG50 MW analogs are plotted as the black (10--50 kpc) and blue (10--30 kpc) dashed lines with 1$\sigma$ scatters represented by the strips, which are also consistent with no significant azimuthal dependence.
• Figure S1: Distribution of the eRASS1 detected point-like sources within the analysis footprint.Left: 0.2--2.3 keV flux distribution of 33695 eRASS1 point-like sources falling within the footprint of our sample galaxies. The vertical line indicates the median value. Middle: Azimuthally averaged surface density profile of the eRASS1 point-like sources as a function of the galactocentric radius. The profile can be roughly described by the combination of two components (red curve): a $\rm S\acute{e}rsic$ profile accounting for sources associated with the host galaxies, and a constant accounting for the cosmic X-ray background sources. Here the $\rm S\acute{e}rsic$ index $n$ = 2.1 is derived from the stacked WISE $W1$ image of the sample galaxies (Section \ref{['subsec:WISE']}). Right: 0.2--2.3 keV luminosity distribution of sources associated with the nuclei of central (green) and satellite (blue) galaxies.