3.1. Radio: EMU

The radio continuum morphology of NGC 5938 as observed with the EMU survey (Hotan et al. Reference Hotan2021; Norris et al. Reference Norris2021; Hopkins et al. Reference Hopkins2025) is presented in Figure 1. The observation was conducted on 21 May 2024 using the full complement of 36 ASKAP antennas, at the central frequency of 943 MHz with a 288 MHz bandwidth (scheduling blocks SB62461). The data were processed using the ASKAPsoft pipelines, including multi-frequency synthesis imaging and multi-scale cleaning. For a complementary view and quantitative analysis of the overall emission, the data were also convolved to a common beam size of 15 $^{\prime\prime}$ $\times$ 15 $^{\prime\prime}$ with a position angle of PA = 0 degrees, from which flux densities were calculated. The higher resolution EMU-HiRES image, as shown in Figure 1, was utilised to effectively illustrate the detailed radio structure of NGC 5938 and to study the centre of the galaxy for the existence of a potential counter jet. Hi-Res processing enhances the visibility of fine details within galaxies, AGN, and other radio sources, as demonstrated by Guzman et al. (Reference Guzman2019).

Additionally, we detected NGC 5938 in the Rapid ASKAP Continuum Survey (RACS; McConnell et al. Reference McConnell2020; Hale et al. Reference Hale2021; Duchesne et al. Reference Duchesne2023, Reference Duchesne2025) across its Low ( $\nu \approx 888\textrm{MHz}$ ), Mid ( $\nu \approx 1\,367\,\textrm{MHz}$ ), and High ( $\nu \approx 1\,655\,\textrm{MHz}$ ) frequency bands. RACS provides wide-field continuum imaging with angular resolutions of approximately $15{^{\prime\prime}}$ (Low), $12{^{\prime\prime}}$ (Mid), and $10{^{\prime\prime}}$ (High), and typical sensitivities of $\sim250$ – $300\,\mu\textrm{Jy beam}^{-1}$ . Measured integrated flux densities for each band are listed in Table 1. In all three RACS bands, NGC 5938 is detected as a bright, compact radio core coincident with the optical nucleus, accompanied by extended emission associated with the eastern jet region (spans $\sim33{^{\prime\prime}}$ ). This extended component has integrated flux densities of $4.3\pm0.46$ mJy (888 MHz/Low band), $3.0\pm0.60$ mJy (1 368 MHz/Mid band), and $2.0\,\pm\,0.69$ mJy (1 655 MHz/High band). While the Low band exhibits the highest integrated flux density, consistent with a non-thermal spectrum, the emission morphology is most clearly resolved and detected in the RACS-Mid band (Figure 1). This difference in visual clarity is attributed to the varying angular resolution across the bands. The Low band has the largest beam, which spreads the emission, resulting in lower peak surface brightness and a less distinct morphology, despite having the highest total flux. The detected surface-brightness levels are low, corresponding to only 1.2–1.6 times the RACS rms noise ( $\sim0.25$ mJy beam $^{-1}$ ). It is important to note that the RACS sensitivity of $\sim 0.25\,\textrm{mJy beam}^{-1}$ (rms) is roughly five times lower than the EMU survey sensitivity of $\sim 0.05\,\textrm{mJy beam}^{-1}$ . This difference may lead to an underestimation of flux density in RACS, particularly in regions of low surface brightness where faint emission approaches the detection threshold.

Table 1.

EMU and RACS radio survey flux density and spectral index measurements of regions within NGC 5938.

$^\textrm{a}$ RMS and beam values for each survey: $^{\textrm{L}}$ RACS Low (888 MHz) rms = 0.25 mJy beam $^{-1}$ , 15 $^{\prime\prime}$ beam; $^{\textrm{E}}$ EMU (944 MHz) rms = 0.05 mJy beam $^{-1}$ , 15 $^{\prime\prime}$ beam; $^{\textrm{M}}$ RACS Mid (1 368 MHz) rms = 0.25 mJy beam $^{-1}$ , 12 $^{\prime\prime}$ beam; $^{\textrm{H}}$ RACS High (1 655 MHz) rms = 0.25 mJy beam $^{-1}$ , 10 $^{\prime\prime}$ beam. Greyed-out values mark RACS flux densities that may be underestimated due to the higher rms noise and lower angular resolution of the RACS data. Flux uncertainties are calculated as $\delta S = \textrm{rms} \sqrt{N_{\textrm{beam}}}$ .

Figure 1.

Radio and NIR observations of NGC 5938 (Araish). The top row (from left to right) shows the EMU Hi-Res image, the EMU Convolved Image, and the DECaPS2 optical z-band image. The bottom row shows the RACS Low, RACS Mid, and RACS High band images with the EMU Hi-Res contours overlaid at 3, 6, 9, 12, 15, 25, 50, and 75 $\sigma$ on all panels, where $\sigma$ = $65\,\mu$ Jy beam $^{-1}$ .

Flux densities for the EMU convolved image (SB62461) and the three RACS bands were measured for the jet region, the galaxy centre, and the integrated flux region using Cube Analysis and Rendering Tool for Astronomy (CARTA; Comrie et al. Reference Comrie2021). We used CARTA’s statistical tools to measure the integrated flux density within the elliptical apertures shown in the EMU Convolved Image Figure 1. These apertures were manually defined to enclose the contiguous emission above the local $6\sigma$ level for the east jet and the central region, while the galaxy-integrated region encompasses all emission above $3\sigma$ . The semi-major axes of these apertures correspond to the characteristic angular extents listed in Table 1.

• • Central region: an ellipse with semi-major axis $\sim32{^{\prime\prime}}$ centred on the optical nucleus

• • East Jet region: an ellipse with semi-major axis $\sim 33{^{\prime\prime}}$ , centred on the peak of the eastern extension, encloses the jet-like emission

• • Galaxy Integrated region: a larger ellipse with semi-major axis $\sim 176{^{\prime\prime}}$ encloses the full galaxy-integrated emission above $3\sigma$ level.

Flux density values were extracted using the image statistics panel, and background noise was estimated locally to subtract any baseline offset. No primary beam correction was applied. Flux uncertainties were estimated using the rms noise expressed in mJy beam $^{-1}$ , following the standard relation for uncorrelated noise per synthesised beam:

\begin{align*}\delta S = \textrm{rms}\times\sqrt{N_{\textrm{beam}}}\end{align*}

where $N_\textrm{beam}$ is the number of synthesised beams within the flux–extraction region,

\begin{align*}N_{\textrm{beam}} = A_{\textrm{flux}} / A_{\textrm{beam}}\end{align*}

and $A_\textrm{beam}$ is the solid angle area of the synthesised beam.

\begin{align*} A_{\textrm{beam}}=\pi\theta_{\textrm{maj}}\theta_{\textrm{min}}/(4\ln 2)\end{align*}

This provides the correct noise propagation when integrating flux across extended emission. The measured flux densities and corresponding uncertainties are reported in Table 1, where greyed-out values represent RACS measurements for which the flux density may be underestimated due to the lower angular resolution and higher rms noise of the RACS data, particularly in regions of low surface brightness where emission approaches the detection threshold. We used these flux density measurements to calculate the spectral index ( $\alpha$ )Footnote ^a and the radio luminosity given in (Table 1).

The EMU in-band spectral index of the galaxy’s nucleus were also measured using multi-frequency synthesis imaging with Taylor terms $T_0$ and $T_1$ . The EMU Taylor images are outputs of the multi-frequency synthesis process, representing the sky brightness as a Taylor series expansion as a function of frequency. The $T_0$ term corresponds to the intensity image at the reference frequency, while $T_1$ represents the first-order spectral slope component. The spectral index $\alpha$ for the centre region was derived from the ratio

\begin{align*}\alpha = \frac{T_1}{T_0},\end{align*}

following the approach outlined by Rau & Cornwell (Reference Rau and Cornwell2011). The flux density of the centre region $T_0$ was measured as $3.72 \times 10^{-2}$  Jy with a standard deviation of $7.71 \times 10^{-3}$  Jy beam $^{-1}$ , while $T_1$ was $-2.61 \times 10^{-2}$  Jy with a standard deviation of $5.64 \times 10^{-3}$  Jy beam $^{-1}$ . This results in a spectral index $\alpha = -0.70 \pm 0.21$ , indicating a steep spectral slope characteristic of non-thermal emission processes. This result is consistent with the value derived from EMU and RACS flux densities (Table 1). The spectral index is further discussed in Section 4.1. All data are available through the ASKAP Science Data Archive (CASDAFootnote ^b ).

3.2. Optical/NIR: DECaPS2

Figure 2(b) shows NGC 5938 in optical and near-IR images from DECaPS2 (Saydjari et al. Reference Saydjari2023).Footnote ^c The single-exposure limiting magnitudes reach $(23.5, 22.6, 21.6)$ mag in the GRZ (475, 644, and 926 nm) bands.

Figure 2.

Multi-wavelength observations of the spiral galaxy NGC 5938 (Araish) (a) Radio continuum image from EMU 944 MHz tile SB62461, Convolved Image has the beam size of $15{^{\prime\prime}}$ , at PA = 0 $^{\circ}$ , with an EMU Hi-Res inset image showing finer structures of the nucleus. (b) DECaPS2 optical/NIR GRZ (475, 644, and 926 nm) bands composite image represented by Blue, Green and Red colours, respectively, with inset showing Hi-Res radio contours overlaid on the DECaPS2 z-band image, highlighting jet–ISM cavity overlap. (c) WISE Band-3. (d) eROSITA X-ray Observation: 0.2–5.0 keV, Gaussian smoothing at Full Width at Half Maximum (FWHM)= $3{^{\prime\prime}}$ , with inset showing Hi-Res radio contours overlaid on X-ray image, highlighting radio jet and X-ray adjoining. The EMU convolved radio contours are applied on all base images at 3, 9, 15, 25, and 50 $\sigma$ , where $\sigma$ = $75 \mu$ Jy beam $^{-1}$ . The radio emission from the jet region is encompassed by the $25\sigma$ radio contour.

3.3. IR: WISE

Infrared observations (Figures 2c and 3) were obtained from WISE (Meisner et al. Reference Meisner, Lang, Schlafly and Schlegel2022) where band W1, W2, W3, and W4 correspond to 3.4, 4.6, 12, and 22 $\mu$ m, respectively. Image cutouts were retrieved from Caltech datasets available through the Infrared Science Archive (IRSA) at NASA/IPAC.Footnote ^d The WISE AllWISE release provides typical $5\sigma$ point-source sensitivities of 16.5, 15.5, 11.2 and 7.9 mag (Vega) in W1–W4, which correspond to approximate flux density limits of 0.08, 0.11, 1.0 and 6.0 mJy. These sensitivity levels are sufficient to detect faint stellar populations and dust emission in nearby galaxies, although sensitivity degrades in regions of high background noise such as the Galactic plane (Wright et al. Reference Wright2010). We also calculated the infrared Vega magnitudes across W1–W4 images as listed in Table 2.

Figure 3.

Infrared observations of NGC 5938 (a) WISE Band W1, (b) Band W2, (c) Band W3, (d) Band W4 images. 3, 9, 15, 25, and 50 $\sigma$ contours applied from EMU convolved radio observations, where $\sigma$ = $75\,\mu$ Jy beam $^{-1}$ .

3.4. X-Ray: eROSITA

Figure 2(d) shows the X-ray image constructed from the combined data from the first four All-Sky Surveys eRASS (Merloni et al. Reference Merloni2024), eRASS1, eRASS2, eRASS3, and eRASS4, obtained with the eROSITA (Predehl et al. Reference Predehl2021) onboard the Russian-German SRG mission (Sunyaev et al. Reference Sunyaev2021). The eROSITA data were calibrated and cleaned using the pipeline version 020 of the eROSITA Science Analysis Software (eSASS, Brunner et al. Reference Brunner2022, version 20211004). We merged the photon events from all seven telescopes into one event list file for each eRASS. To increase the signal-to-noise ratio, we also generated an event list combining all the data from the four eRASS epochs (hereafter eRASS:4). We applied eRASS:4 (0.2–5.0 keV) image smoothing using a Gaussian kernel with a FWHM of 3 $^{\prime\prime}$ using the Python Astropy 2D convolution kernel functions. This smoothing scale optimally reveals the features of interest, specifically the X-ray cocoon surrounding the radio emission.

4.1. Radio observations

As shown in Figure 1, EMU Hi-Res and Convolved observations, there is an extended region of radio emission perpendicular to the galaxy’s major axis, resembling possible AGN jets originating from the central SMBH. Taking the redshift independent distance of $\approx 26.6$ Mpc to the galaxy, angular measurements indicate the core to tip length of eastern jet $\approx 40{^{\prime\prime}}$ and the total jet extent of the AGN $ \approx 64{^{\prime\prime}}$ , which converts into physical sizes $ \approx 5.15$ kpc and $ \approx 8.2$ kpc, respectively. Notably, Figure 2(a) (Hi-Res Inset image) tentatively labels a feature to the north-west of the nucleus as a ‘counter jet’. This feature appears to be approximately aligned with the East jet axis but displaced from a direct line of sight, potentially indicating that it is angled out of the plane. Interpretation of this feature include a counter (west) jet, or possibly backflow.

Similar to the well-studied and more energetic galaxy NGC 1068 (Young et al. Reference Young, Wilson and Shopbell2001), which hosts prominent AGN jets, our observations indicate that the jet is oriented roughly perpendicular to the galactic disk and its major axis. However, this does not definitively establish the jet’s 3D orientation relative to the galaxy’s actual disk. The jet could be truly perpendicular to the disk, extending directly out of the plane, or it could be inclined at some angle, like in NGC 1068. It is worth noting that in NGC 1068, despite the active AGN, the star formation contributes significantly to the radio emission and the AGN itself accounts for only about half the total radio luminosity (Cecil et al. Reference Cecil, Bland-Hawthorn, Veilleux and Filippenko2002). If the jet in NGC 5938 is similarly inclined, it may still be propagating through the outer regions of the ISM, rather than through the main disk itself.

Evidence for the jet-ISM interaction, which could help distinguish between these scenarios (Morganti Reference Morganti2017), is limited in our current data. The optical image (Figure 2b) shows morphological features that are consistent with jet-ISM interaction, though the infrared image (Figure 2c) show no sign of obscuration of the dust lanes, nor any obvious heating of gas or dust along the jet’s path. These observations suggest that the jet is likely propagating through relatively low-density ISM, producing subtle structural effects (e.g. a shallow cavity) but without strong radiative or thermal signatures typical of jet impact on dense molecular gas.

The central AGN region has a spectral index of $\alpha = - 0.7\pm0.4$ , while the eastern jet region shows a steeper index of $\alpha = -1.2\pm0.2$ . This steeper index suggests a non-thermal emission mechanism, consistent with synchrotron radiation (Filipović & Tothill Reference Filipović and Tothill2021). The spectral index of NGC 5938’s centre was also calculated using the ratio of EMU Taylor T0 and T1 images; This method gives us the spectral index $\alpha = -0.7\pm0.2$ , which is consistent with the spectral index calculated using the flux densities from EMU and RACS surveys (Table 1). The observation of radio emission in NGC 5938 extending perpendicular to the galaxy’s major axis (Figure 2a), combined with the non-thermal spectral index, strongly supports the interpretation of an AGN jet.

4.2. Optical observations

The optical observations from DECaPS2 (Figure 2b) show the galaxy in visible light, highlighting the luminous nucleus, distribution of stars and the overall morphology. NGC 5938 is classified as a barred spiral, with a smooth disk and a less prominent central bulge compared to most spiral galaxies, following Moorthy & Holtzman (Reference Moorthy and Holtzman2006). The dust lanes are apparent in the optical, partially obscuring the galaxy’s optical emissions from the nucleus. The DECaPS2 z-band image (Figure 2b) reveals a diffuse stellar and ISM distribution, within which a pronounced under-luminous cavity (highlighted by the yellow ellipse) is identifiable. This cavity appears to be caused by a low density of stars or dust absorption in the region. Although the origin of this cavity is uncertain, it is spatially aligned with the radio-jet axis ( $f_{\textrm{overlap}} \approx 50\%$ , white contours). This geometric association suggests a possible imprint of the jet on the surrounding ISM (Morganti Reference Morganti2017). Such a marginal structural depression would be consistent with a jet propagating through comparatively low-density medium, producing subtle clearing without the strong heating or dust-lane disruption typically associated with jet–ISM interaction in denser regions of galactic disks.

4.3. Infrared observations

Unlike visible light, infrared emission can pass through dust, making it invaluable for studying obscured star formation and AGN. The longer wavelengths of infrared radiation enable us to detect star formation and AGN that are invisible due to dust blockage in the optical spectrum. It also traces the distribution and properties of the dust itself, a key component of the ISM (Saydjari et al. Reference Saydjari2023). NGC 5938 observations from WISE (Wright et al. Reference Wright2010) W1–W4 bands (Figure 3, Table 2) show prominent infrared emission.

In the W1 band, the galaxy exhibits an apparent magnitude of $8.10 \pm 0.02$ . For context, Nikutta et al. (Reference Nikutta, Hunt-Walker, Nenkova, Ivezić and Elitzur2014) shows that WISE sources with $W1 \lt 11$ are typically dominated by stars, and extragalactic objects become more prevalent at fainter magnitudes, i.e. $W1 \gt 11$ .Footnote ^e Therefore, our target’s W1 magnitude of $8.10$ places it firmly among the brightest extragalactic objects detected in W1, suggesting a substantial near-infrared luminosity.

Moreover, mid-infrared studies of AGN populations, such as those utilising WISE often focus on sources with W1 magnitudes in the range $10\!-\!15\,\textrm{mag}$ i.e. AGN selection criteria by Stern (Reference Stern2012) and O’Connor et al. (Reference O’Connor, Rosenberg, Satyapal and Secrest2016). Within these samples, an object with $W1 = 8.10$ would be considered exceptionally bright, implying that its infrared emission is significant relative to typical AGN hosts.

Regarding the colours:

• • The W1–W2 colour of $ - 0.01 \pm 0.04$ lies near zero, indicative of a source whose infrared SED is characteristic of a cool stellar population with minimal hot dust or AGN torus contribution (Wright et al. Reference Wright2010).

• • The W2–W3 colour of $2.81 \pm 0.04$ and W3–W4 colour of $2.01 \pm 0.05$ show significant brightening at longer mid-IR wavelengths, consistent with warm dust emission from star-forming regions and/or lower-luminosity AGN systems (Cluver et al. Reference Cluver2014, Reference Cluver2017).

Table 2.

Infrared magnitude – WISE.

The position of our target galaxy NGC 5938 (black circle in Figure 4-top panel) on the WISE colour–colour diagram indicates W1–W2 = $-0.01 \pm 0.04$ and W2–W3 = $2.81 \pm 0.04$ . These values place the source below the AGN-dominated region but firmly within the dusty, star-forming locus, consistent with the regime of galaxies exhibiting significant warm dust emission (Lisenfeld et al. Reference Lisenfeld, Ogle, Appleton, Jarrett and Moncada-Cuadri2023). The location also lies above the stellar-dominated sequence, suggesting that mid-infrared emission from heated dust makes a substantial contribution. While its location on the Star Formation Main Sequence (SFMS) in Figure 4-bottom panel confirms its classification as a star forming galaxy (Brinchmann et al. Reference Brinchmann2004), where its star formation rate $\log \textrm{SFR}\;[M_\odot\,\textrm{yr}^{-1}] = 0.58 \pm 0.08$ is obtained from Buzzo et al. (Reference Buzzo2024) and the stellar mass $\log {M}^* = 11.24 \pm 0.10$ is taken from Parkash et al. (Reference Parkash, Brown, Jarrett and Bonne2018).

The presence of cool dust around the centre of the NGC 5938, as indicated by W2–W3 colour, could be due to interactions between high-velocity particles in AGN jets and the interstellar dust grains (Cielo et al. Reference Cielo, Antonuccio-Delogu, Silk and Romeo2017; Cluver et al. Reference Cluver2014). However, it is important to note that in the W1–W3–W4 RGB image (Figure 5), we do not observe any increase in infrared emission, disturbance in dust distribution, or other signs of jet-ISM interaction coincident with the radio jet’s location. No other prominent infrared sources are detected within $6{^{\prime\prime}}$ of the radio jet. The native angular resolution of WISE is $\sim 6{^{\prime\prime}}$ in W1–W3 and $\sim12{^{\prime\prime}}$ in W4 (Wright et al. Reference Wright2010), which is sufficient to trace global dust emission but not to resolve small-scale interactions between the radio jet and the interstellar dust. Thus, the observed WISE colours provide evidence for cool dust in the galaxy, although higher-resolution infrared data would be required to directly probe any jet–ISM interaction.

Figure 4.

Top: WISE colour-colour diagram showing NGC 5938’s position relative to stellar-dominated, dusty star-forming, SFMS, warm/AGN, and Quasar/AGN regions. The dashed curves mark template tracks from Lisenfeld et al. (Reference Lisenfeld, Ogle, Appleton, Jarrett and Moncada-Cuadri2023). The location of NGC 5938 (black dot) near the dusty star-forming/warm–AGN boundary indicates the presence of significant dust. Bottom: Star Formation Main Sequence diagram plotting mid-infrared–corrected star formation rate (SFR $_{\textrm{MIRcor}}$ ) derived from WISE 12 $\mu$ m luminosity following Buzzo et al. (Reference Buzzo2024) versus stellar mass. The contours represent the distribution of nearby galaxies from the combined 2MRS+S4G sample from Lisenfeld et al. (Reference Lisenfeld, Ogle, Appleton, Jarrett and Moncada-Cuadri2023), while the solid magenta line shows the SFMS fit from Buzzo et al. (Reference Buzzo2024). The dashed grey line corresponds to a specific star formation rate of log(sSFR) = –11 yr $^{-1}$ . NGC 5938 located on the SFMS, consistent with active star formation.

Figure 5.

RGB WISE Red-W4, Green-W3, and Blue-W1. W1 colour indicates excess hot dust emission from an AGN while W3 and W4 luminosity is sensitive to star formation. Radio Contours 3, 9, 15, 25, and 50 $\sigma$ applied from EMU radio observations.

4.3.1. Infrared and radio emission analysis

To further investigate AGN activity in NGC 5938, we calculated the radio-to-infrared flux density ratio, a valuable diagnostic tool commonly used in identifying AGN jets (Weston et al. Reference Weston2018). This ratio was computed using the integrated radio flux density at 1.4 GHz (Table 1) and infrared flux density at 3.6 $\mu$ m to generate the ratio. We used the integrated radio flux density at 944 MHz, $S_{944} = 90.6 \pm 0.49$ mJy (Table 1), and the measured spectral index $\alpha = -1.5 \pm 0.2$ to extrapolate flux density at 1.4 GHz using the power-law relation (Equation 1), yielding $S_{1.4} = 50.16 \pm 4.27$ mJy.

(1) \begin{equation} S_2 = S_1 \left( \frac{\nu_2}{\nu_1} \right)^{\alpha}\end{equation}

For infrared measurements, the WISE W1 Vega magnitude of $M_{W1} = 8.10 \pm 0.02$ was converted to a flux density (Equation 2; Wright et al. Reference Wright2010), where $S_{\nu_0} = 306.7$ Jy is the W1 zero-magnitude flux density (Wright et al. Reference Wright2010). This yields $S_{3.4} = 176.48 \pm 4.0$  mJy. Extrapolating to $3.6\,\mu$ m using the measured infrared spectral index ( $\alpha_{\textrm{IR}} \approx -1.0$ ) gives $S_{3.6} \approx 209.5 \pm 4.0$ mJy.

(2) \begin{equation}S_{W1} = S_{\nu_0} \times 10^{-0.4 \times M_{W1}}\end{equation}

(3) \begin{equation}R_{1.4/3.6} = \frac{S_{1.4,\textrm{GHz}}}{S_{3.6,\mu\textrm{m}}}\end{equation}

The radio-to-infrared ratio (Equation 3) is $0.24 \pm 0.02$ , which lies well below the canonical radio-loud quasar threshold ( $R \gt 10$ ; Elvis et al. Reference Elvis1994), indicating that radio emission is not the dominant power source. In Figure 6, NGC 5938 (green dot) lies marginally above the median star-forming galaxy locus but well below the radio-loud AGN regime.

Figure 6.

The ratio between the radio 1.4 GHz and extrapolated 3.6 $\mu$ m integrated flux density plotted as a function of redshift. Black points represent galaxies from Weston et al. (Reference Weston2018). The green circled dot marks the flux density ratio of NGC 5938 (this work). The grey shaded area denotes the loci of radio-loud AGN as per Weston et al. (Reference Weston2018). Template tracks are shown for the starburst galaxy M82 (black dashed line), and the spiral galaxy NGC 4429 (black solid line) from Seymour et al. (Reference Seymour2007), radio-loud quasars (red dotted line), and radio-quiet quasars (red dash-dotted line) from Elvis et al. (Reference Elvis1994).

To place this ratio in a broader context, we compared NGC 5938’s stellar mass ( $\log M_* / M_\odot = 11.02 \pm 0.05$ ) and radio luminosity at 888 MHz ( $\log L_{888\,\textrm{MHz}} = 21.50 \pm 0.03$ W Hz $^{-1}$ ) to the distributions of AGN hosts identified using optical Baldwin, Phillips & Terlevich (BPT) diagnostics, WISE colours, the Mass–Excitation diagram, and ProSpect(0010) modelling from Prathap et al. (Reference Prathap2024). While the radio-to-infrared ratio alone would not classify NGC 5938 as strongly radio-loud, its radio luminosity is consistent with the median values for AGN selected by ProSpect(0010), a method designed to detect lower-luminosity AGN in massive hosts.

This quantitative cross-comparison indicates that NGC 5938 likely harbours a low luminosity AGN where radio output is modest compared to its infrared emission, though consistent with multi-wavelength AGN diagnostics. The results emphasise that in composite systems, where star formation and AGN activity both contribute to the observed flux, a single diagnostic can under-represent the AGN component, and multi-method analyses are required for reliable classification.

4.4. X-ray observations

4.4.1. X-ray emission from the radio jet

In the X-ray regime, we investigated the galaxy NGC 5938 utilising the eROSITA-eRASS:4 X-ray 0.2–5.0 keV processed image (Figure 2d). Using the ermldet task, no strong point-like X-ray source was detected at the radio jet’s position. Instead, the observed X-ray morphology reveals extended, diffuse emission broadly following the radio contours (Figures 2d and 7), consistent with a cocoon of shocked gas surrounding the jet. To quantify this, we extracted photons within the $3\sigma$ radio contour, detecting 29 counts (4 background) in the 0.2–2.3 keV band, corresponding to a significance $\gt 3\sigma$ .

Figure 7.

RGB composite image of NGC 5938 illustrating the distribution of radio emission (Red)-EMU Hi-Res, infrared W3 emission (Green)-WISE and X-ray emission (Blue) – eROSITA. The radio jet seems to be collimated by the surrounding X-ray gas.

To estimate the X-ray upper limit of the jet region, we extracted X-ray spectra using srctool with a $10 {^{\prime\prime}}$ circular region centred at the position of the east radio jet (EEF $=0.4$ ; Tubín-Arenas et al. Reference Tubín-Arenas2024) and an annulus ( $70{{^{\prime\prime}}}$ – $130{{^{\prime\prime}}}$ ) for background subtraction. Following Kraft, Burrows, & Nousek (Reference Kraft, Burrows and Nousek1991), assuming a power-law spectrum with galactic absorption ( $N_{\textrm{H}} = 1.4 \times 10^{21}\ \textrm{cm}^{-2}$ ) and photon index $\Gamma = 1.8$ , we obtained a $3\sigma$ flux upper limit of $S_{0.2 - 2.0\,\textrm{keV}} \approx 0.7 \times 10^{-13}\, \textrm{erg cm}^{-2}\, \textrm{s}^{-1}$ , resulting in a soft X-ray luminosity of $L_{\textrm{X}} \approx 6.0 \times 10^{39}\,\textrm{erg s}^{-1}$ ( $\log L_{\textrm{X}} \approx 39.8$ ).

To assess the possibility of a central AGN, we extracted the X-ray events from a $10{^{\prime\prime}}$ circular region centred at the core (Figure 7). This yielded 7 photons (0.7 background), significant at $\gt 3\sigma$ . However, all photons fall below (2 keV), with no hard X-rays ( $\geq 2\,\textrm{keV}$ ) detected. As a result, the inferred luminosity pertains only to the soft X-ray band. Due to the low photon counts, we estimated the soft X-ray flux assuming an absorbed power-law model ( $N_{\textrm{H}} = 1.4 \times 10^{21}\ \textrm{cm}^{-2}$ and $\Gamma = 1.8$ ). The resulting soft X-ray luminosity ( $3.4\pm3 \times 10^{39}\,\textrm{erg s}^{-1}$ ) is consistent with LINER galaxies ( $\log L_{\textrm{X}} \approx 39.9$ ) and higher than Hii galaxies ( $\log L_{\textrm{X}} \approx 39.3$ ), while remaining well below the typical Seyfert regime ( $\log L_{\textrm{X}} \approx 41.0$ , Georgantopoulos, Basilakos, & Plionis Reference Georgantopoulos, Basilakos and Plionis1999). Notably, this luminosity is also broadly consistent with empirical relations for X-ray binary (XRB) populations and shock-heated plasma in a galaxy with an SFR $\approx 0.58$ M $_{\odot}$ /yr (Mineo, Gilfanov, & Sunyaev Reference Mineo, Gilfanov and Sunyaev2012; Ranalli, Comastri, & Setti Reference Ranalli, Comastri and Setti2003). This suggests that the observed soft emission may be dominated by stellar processes and shock-heated gas associated with the radio-jet interaction rather than the AGN itself.

Although no hard-band emission is detected, we evaluated whether NGC 5938 could host a heavily obscured near-Compton-thick AGN (Ricci et al. Reference Ricci2015). We derived $3\sigma$ upper limits from the eROSITA data, assuming a heavily obscured power-law model with $N_{\textrm{H}} \gtrsim 1.0 \times 10^{24}\,\textrm{cm}^{-2}$ and $\Gamma = 1.8$ . This observed hard-band upper limit corresponds to an intrinsic luminosity limit of $L_{2-10\,\textrm{keV}} \lesssim 6.8 \times 10^{42}\,\textrm{erg s}^{-1}$ , with a corresponding soft-band intrinsic limit of $L_{0.2-2\,\textrm{keV}} \lesssim 6.6 \times 10^{42}\,\textrm{erg s}^{-1}$ . These limits, along with the estimated soft X-ray luminosity under low absorption scenario, imply that NGC 5938 does not host a luminous Seyfert-like AGN, and that any nuclear activity is either intrinsically weak or heavily absorbed (Gilli, Comastri, & Hasinger Reference Gilli, Comastri and Hasinger2007; Brightman & Nandra Reference Brightman and Nandra2012; Ricci et al. Reference Ricci2015). Comparison with the BASS survey (Tokayer et al. Reference Tokayer2025) further shows that NGC 5938 is significantly less luminous and lacks the detectable hard X-ray emissions ( $\log L_{14-195\,\textrm{keV}} \geq 43$ ), characteristics of the powerful AGN probed in the study (Tokayer et al. Reference Tokayer2025).

Although more complex absorption scenarios, such as highly ionised absorbers, could in principle suppress the hard band while allowing some soft emission (Done, Gierliński, & Kubota Reference Done, Gierliński and Kubota2007); however the limited signal-to-noise ratio of the available data prevents any meaningful constraints on these models. These results are consistent with a scenario in which the AGN in NGC 5938 may be undergoing a low-accretion-rate, pre-blowout phase, where the nuclear feedback is currently insufficient to clear the surrounding obscuring material. We therefore conclude that the current data cannot rule out a deeply buried, low-luminosity nucleus, although the observed emission is more naturally explained by non-AGN processes. Deeper hard X-ray observations (e.g. with NuSTAR; Harrison et al. Reference Harrison2013) would be required to detect or place meaningful constraint on any obscured nuclear emission.