RAD@home Citizen Science Discovery of Two Spiral Galaxies Where the 30–220 kpc Radio Lobes Are Possibly Shaped by Ram Pressure Stripping

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors,

The article titled "RAD@home citizen science discovery of two galaxies where the 30-220 kpc radio lobes are possibly shaped by ram pressure stripping" describes the discovery of two lobed radio sources that are hosted by young spiral galaxies. We commend the authors on a well-written manuscript and their discoveries and only recommend a moderate amount of revision prior to submission.

I would like to commend the RAD@home project on their efforts and the fantastic long term impact such a project can have on both the astronomy community and that of the greater general public.  Therefore it is in this positive light that I provide my comments below that I hope are constructive and improve the accuracy of this paper.

Here are my comments in order of appearance in the manuscript:
1) Section 2: please provide more information (and references) on where the UV, optical and IR images originated.

2) Section 3.1: can the authors provide further justification for the host galaxy cross match for NGC 3898?  What is the probability of a chance coincidence?

3) Section 3.1, 1st paragraph, 2nd sentence: there is some inconsistency with the claim that this galaxy is close to face-on. Vega Beltran et al 2001 (https://ui.adsabs.harvard.edu/abs/2001A%26A...374..394V/abstract) found the inclination to be ~54 degs. They also concluded from NGC3898's HI and Halpha kinematics that the ionised and neutral gas are in a smooth distribution and a regular velocity field.

Pignatelli et al 2001 also analysed the kinematics of NGC 3898.

4) Section 3.1, 3rd paragraph: perhaps clarify that the angular separation and distance are projected angular separation and distance

5) Section 3.1, 3rd paragraph, 2nd last sentence: Can you mark this in a figure for further clarification on which compact source that you are describing?

6) Section 3.1, 5th paragraph: The outer star forming arms are also visible on the eastern side (of the UV image) even if it is more pronounced on the western side.  The diffuse star forming arms of NGC 3898 resemble those from the extended UV (XUV) galaxies  (eg. Thilker et al 2007). Given the diffuse ISM of the outer disks, the observed asymmetry is not unusual and does not require additional external interaction to cause such asymmetries.  The regular kinematics observed in the gas provides a hint about the relative stability of this galaxy as a process such as ram pressure is likely to affect the HI distribution and kinematics.  Please amend your statements to reflect such previous evidence.

7) Section 3.1, 6th paragraph, 3rd last sentence:  Please revise or remove this statement.  This claim of similarity needs to be moderated, the similarity of the "ionised tail" has been overstated between Fig 4b and that of the NGC 4438 (Kenney et al 2008;  https://ui.adsabs.harvard.edu/abs/2008ApJ...687L..69K/abstract).  From Figs 4a and 4b, and assuming that the cyan contours are consistent between the 2 panels, NGC3898's ionised gas does not appear to even extend beyond the stellar body of the galaxy, much less demonstrate the extent of extraplanar ionised gas that is observed in NGC 4438.  Also the ionised emission of NGC 4438 comes from shock ionisation rather than ionisation by its radio AGN.

Furthermore the stellar disk of NGC 3898 isn't disturbed and irregular such as that of NGC 4438.  This galaxy is likely a better analogue of the "Thumbs Up" galaxy that you discuss below.

8) Section 3.2, 1st paragraph, 3rd sentence & Fig 5: The northern most radio "peak" appears more like sidelobes in both NVSS and TGSS. What is the signal-to-noise ratio of this northernmost "peak" ? Is this peak confirmed in newer observations from the ASKAP RACS surveys?  I know that you do mention later in this paper that these observations will be the subject of a future paper but it may be worth saying if this is confirmed in higher sensitivity observations as well at this point.

This Thumbs-Up galaxy resides within a cluster, WHL J221654.9-042302 (Wen & Han 2015; https://ui.adsabs.harvard.edu/abs/2015ApJ...807..178W/abstract). The cluster BCG appears to be ~2arcmin North-east of Thumbs-Up's host spiral galaxy. 

There is also a supercluster in the background at z~0.38 (Sankhyayan et al 2023 https://ui.adsabs.harvard.edu/abs/2023ApJ...958...62S/abstract; Wen et al 2012 https://ui.adsabs.harvard.edu/abs/2012ApJS..199...34W/abstract), projected 3.5 arcmin to the south-east of the Thumbs-Up spiral.

The fist component of the radio emission is projected in coincidence with a more distant galaxy at 22:16:52 -04:24:59.5 which has a DESI photo-z ~0.38. This galaxy is observed by the PanStarrs PS1 survey and the Legacy Survey DR10. Is it possible that the association of this radio component to Spiral Host galaxy is a case of mistaken identity?  If not, please provide further reasons against the possibility of a chance coincidence.

9) Section 3.2, 2nd paragraph, 3rd sentence: the "resolving out" of the diffuse emission is not convincing evidence that the western radio component to the spiral host galaxy is not a chance coincidence with an unassociated source or cluster in the background.

If the diffuse emission to the west of the spiral is indeed older lobe emission, is this lobe emission steeper in radio spectral index than that of the purported core emission from the central regions of the spiral galaxy?

10) Fig 7: It would be useful for this figure to include the western component as well (as per the point above). Also, the green part of the colour scale in Fig 7 is too coarse for the reader to distinguish between a flat vs a steep spectral index. Green appears to be able to represent a spectral index ranging between -0.6 to -1.1.

11) Section 3.2, 6th paragraph, 3rd last sentence: I agree with this conclusion.

On the other hand as noted earlier, there is a WHL cluster BCG 1.5 arcmin north of the spiral that is at the same redshift.  Therefore, while the elliptical that appears closest in projection may be in the foreground, it does seem that this galaxy could be residing in a cluster.

Other recent discoveries of spirals/late-type galaxies hosting double radio AGN from Gao et al (2023;  https://ui.adsabs.harvard.edu/abs/2023RAA....23c5005G/abstract) and Yuan  et al (2024; https://ui.adsabs.harvard.edu/abs/2024RAA....24d5007Y/abstract) suggest a connection between such galaxies and a denser local environment such as that of a group and cluster. Perhaps a comparison between the "Thumbs-up" and these recent results would be useful.

12) Section 4.1: What does the spectral index map look like for NGC 3898?

13) Section 4.1, last sentence: There is no evidence for the compression of gas (eg Kantharia et al 2005) along the leading edge of interaction. The highly ordered star-forming outer arms of NGC 3898 would also suggest a dearth of strong external influence that is either depleting the gas or suppressing the star formation process within this galaxy. Hence, more substantial evidence is needed from the multiwavelength observations of NGC 3898, to justify the authors' assertions for ram pressure stripping.

14) Section 4.3: Perhaps a reference to the more recent paper by Ignesti et al (2022; https://ui.adsabs.harvard.edu/abs/2022ApJ...937...58I/abstract) which show some very nice example of ram pressure-stripped radio tails, would also be very relevant to this paper. 

Author Response

Referee 1

Dear authors,

The article titled "RAD@home citizen science discovery of two galaxies where the 30-220 kpc radio lobes are possibly shaped by ram pressure stripping" describes the discovery of two lobed radio sources that are hosted by young spiral galaxies. We commend the authors on a well-written manuscript and their discoveries and only recommend a moderate amount of revision prior to submission.

I would like to commend the RAD@home project on their efforts and the fantastic long term impact such a project can have on both the astronomy community and that of the greater general public.  Therefore it is in this positive light that I provide my comments below that I hope are constructive and improve the accuracy of this paper.

Authors: We sincerely thank the referee for their encouraging and supportive comments on our manuscript and for recognising the scientific value of the discoveries reported. We are especially grateful for the kind words regarding the RAD@home citizen science initiative and its broader impact. We also thank the referee for carefully going through the manuscript and for the critical comments, which have significantly improved the quality and clarity of the paper. The constructive suggestions provided have been thoroughly considered and incorporated into the revised version. We trust that the changes made have strengthened the manuscript as intended.

Referee: Here are my comments in order of appearance in the manuscript:
1) Section 2: Please provide more information (and references) on where the UV, optical and IR images originated.
Authors: Thank you. We have now given references to the GALEX and WISE surveys used in our RGB images.

Referee: 2) Section 3.1: Can the authors provide further justification for the host galaxy cross-match for NGC 3898?  What is the probability of a chance coincidence?

Authors: We thank the referee for raising this important point. The association of large radio sources with spiral galaxies, while once considered rare, is now well documented, with over three dozen such cases known in the literature. In the specific case of NGC 3898, the FR II-like morphology of the radio source naturally confines the possible host candidates to a small spatial region along the radio axis.

To address the possibility of a chance alignment, we have added a new figure (revised Fig. 3) showing the LOFAR high-resolution radio contours overlaid on the DESI Legacy optical image. This clearly reveals a compact radio source near the midpoint of the lobes, offset by more than 2 arcseconds from the nearest optical galaxy, which has a photometric redshift of z~0.5. Given this offset and the lack of radio emission associated with the optical counterpart at this location in other surveys (e.g. FIRST, VLASS), we find no compelling evidence that this background galaxy is responsible for the large-scale radio emission.

In contrast, the spiral NGC 3898 is coincident with the radio core component and lies well within the radio axis. Its location within a cluster-like environment further supports its role as the host galaxy. As also noted in the revised manuscript, establishing definitive radio-optical associations generally requires imaging of connecting jets. In their absence, we rely on a combination of positional alignment, multi-frequency morphology, and contextual environment to justify the association. We believe the newly added data and discussion clarify and strengthen this identification.

Referee: 3) Section 3.1, 1st paragraph, 2nd sentence: there is some inconsistency with the claim that this galaxy is close to face-on. Vega Beltran et al 2001 (https://ui.adsabs.harvard.edu/abs/2001A%26A...374..394V/abstract) found the inclination to be ~54 degs. They also concluded from NGC3898's HI and Halpha kinematics that the ionised and neutral gas are in a smooth distribution and a regular velocity field. Pignatelli et al 2001 also analysed the kinematics of NGC 3898.

Authors: Thank you. We removed the face-on word. We have also cited both of these kinematics studies in the paragraph describing HI and H-alpha studies.

Referee: 4) Section 3.1, 3rd paragraph: perhaps clarify that the angular separation and distance are projected angular separation and distance

Authors: Thank you. We have now expressed it as projected angular separation. Similarly, we have made corrections in other places in the manuscript.

Referee: 5) Section 3.1, 3rd paragraph, 2nd last sentence: Can you mark this in a figure for further clarification on which compact source that you are describing?

Authors: A separate image (new Figure 3) has been created for presenting detailed radio-optical positional mismatch. The offset, not along the jet/lobe but orthogonal, is a clear sign that it is unlikely the radio core and only a background galaxy at photo-z around 0.5. A small paragraph of write-up on this point source, not being the distant host, is now added to the manuscript.

Referee: 6) Section 3.1, 5th paragraph: The outer star-forming arms are also visible on the eastern side (of the UV image), even if it is more pronounced on the western side.  The diffuse star-forming arms of NGC 3898 resemble those from the extended UV (XUV) galaxies  (e.g., Thilker et al 2007). Given the diffuse ISM of the outer disks, the observed asymmetry is not unusual and does not require additional external interaction to cause such asymmetries.  The regular kinematics observed in the gas provides a hint about the relative stability of this galaxy as a process such as ram pressure is likely to affect the HI distribution and kinematics.  Please amend your statements to reflect such previous evidence.

Authors: Thank you. We have given reference to Thilker et al. for the extended UV disk. At this point in the manuscript, we have not presented enough asymmetries to claim ram pressure stripping. Hence, arguments of the inner disk showing regular velocity fields are presented in a later section 4.1. 

Referee: 7) Section 3.1, 6th paragraph, 3rd last sentence:  Please revise or remove this statement.  This claim of similarity needs to be moderated, the similarity of the "ionised tail" has been overstated between Fig 4b and that of the NGC 4438 (Kenney et al 2008;  https://ui.adsabs.harvard.edu/abs/2008ApJ...687L..69K/abstract).  From Figs 4a and 4b, and assuming that the cyan contours are consistent between the 2 panels, NGC3898's ionised gas does not appear to even extend beyond the stellar body of the galaxy, much less demonstrate the extent of extraplanar ionised gas that is observed in NGC 4438.  Also the ionised emission of NGC 4438 comes from shock ionisation rather than ionisation by its radio AGN. Furthermore the stellar disk of NGC 3898 isn't disturbed and irregular such as that of NGC 4438.  This galaxy is likely a better analogue of the "Thumbs Up" galaxy that you discuss below.

Authors: We agree that the similarity with NGC4438 is an overstatement. We have now removed that statement. We agree that neither the stellar disk is disturbed nor the velocity field of the intensity-weighted inner ionised gas disk. However, ram pressure stripping works well where the stellar disk gravity is weak, that is, on the outer parts of the disk. It may be noted that gas in the stellar disk may remain well-defined, but only deep imaging can show the long ram pressure stripped tails, which are currently unavailable. A good example is NGC4569, an undisturbed disk with a ~230 kpc H-alpha tail. Similarly, the 100 kpc long HI tail is seen from the undisturbed disk galaxy NGC4388.  

Referee: 8) Section 3.2, 1st paragraph, 3rd sentence & Fig 5: The northern most radio "peak" appears more like sidelobes in both NVSS and TGSS. What is the signal-to-noise ratio of this northernmost "peak" ? Is this peak confirmed in newer observations from the ASKAP RACS surveys?  I know that you do mention later in this paper that these observations will be the subject of a future paper but it may be worth saying if this is confirmed in higher sensitivity observations as well at this point.

Authors: To strengthen the case, we have incorporated both RACS and EMU radio data into the manuscript (Fig. 9). The RACS image clearly shows the connection between the diffuse emission and the radio core, while the EMU image distinctly reveals the faint thumb-like component along with its filamentary connections. RMS noise levels have been added to the figure captions, ensuring that the radio contours accurately represent the significance of the detections. Specifically, the northern peak of the thumb feature in the EMU image appears at the 3rd contour (1.5 mJy/b) level (Fig. 9 top right panel), corresponding to a significance of over 32*sigma. Furthermore, the peak at the thumb is 1.8 mJy/b, corresponding to a  signal-to-noise ratio of ∼39. These measurements confirm that the fundamental features of the Thumbs Up Galaxy are robust and well-detected.

This Thumbs-Up galaxy resides within a cluster, WHL J221654.9-042302 (Wen & Han 2015; https://ui.adsabs.harvard.edu/abs/2015ApJ...807..178W/abstract). The cluster BCG appears to be ~2arcmin North-east of Thumbs-Up's host spiral galaxy. There is also a supercluster in the background at z~0.38 (Sankhyayan et al 2023 https://ui.adsabs.harvard.edu/abs/2023ApJ...958...62S/abstract; Wen et al 2012 https://ui.adsabs.harvard.edu/abs/2012ApJS..199...34W/abstract), projected 3.5 arcmin to the south-east of the Thumbs-Up spiral.

Authors:  We have cited all three references for the clusters and superclusters. We have described it in a couple of sentences at the end of the paragraph describing EMU images and corresponding optical images.  We have now clearly mentioned that the target galaxy is a member of the cluster. 

Referee: The fist component of the radio emission is projected in coincidence with a more distant galaxy at 22:16:52 -04:24:59.5 which has a DESI photo-z ~0.38. This galaxy is observed by the PanStarrs PS1 survey and the Legacy Survey DR10. Is it possible that the association of this radio component to Spiral Host galaxy is a case of mistaken identity?  If not, please provide further reasons against the possibility of a chance coincidence.

 

Authors: We thank the referee for raising this important point. Indeed, the line-of-sight projection of a background galaxy (DESI photo-z∼0.38z \sim 0.38z∼0.38) near the "fist" component of the Thumbs Up radio source warrants careful consideration. However, we believe the association of the radio emission with the nearby disturbed spiral galaxy is robust, and not a case of mistaken identity, for the following reasons:

1. Radio–Optical Association of the Core: The central "wrist" or radio core component is spatially coincident with the spiral galaxy, a member of the cluster under discussion. This is well supported by the FIRST, RACS and EMU data.
   
   
2. Morphological Continuity in RACS and EMU: As shown in Fig. 9 of the revised manuscript, the newly included RACS data reveal a plume-like extension connecting the "wrist" (radio core) to the "fist" component (top left panel). Additionally, the deeper EMU image (top right) clearly shows filamentary structures linking the "wrist," "fist" and "thumb" components, indicating a coherent radio morphology consistent with a common origin, the disturbed spiral.
   
   
3. Lack of Radio Emission from Background Galaxy: The background galaxy near the "fist" position does not exhibit any radio emission in either the EMU or FIRST data. If it were the true origin of the radio lobe, some compact or extended radio counterpart would be expected, particularly in the high-sensitivity EMU image. This absence of emission argues against any physical association.
   
   
4. Foreground Cluster Membership and Interpretation: The disturbed spiral is a confirmed member of a cluster, where ram pressure stripping is likely shaping the large-scale radio morphology. This interpretation is supported by the UV/optical asymmetries, as discussed in the manuscript (Sections 3.2 and 4.1). The radio lobes—particularly the one forming the "fist"—are plausibly shaped by environmental effects acting on the foreground galaxy, not by a background source.

In conclusion, the consistent alignment of all three radio components with the spiral galaxy, the absence of any radio detection from the background galaxy, and the morphological coherence in the RACS and EMU images strongly favour the association of the Thumbs Up radio source with the foreground disturbed spiral. Therefore, we consider the possibility of a chance projection to be highly unlikely.

Referee: 9) Section 3.2, 2nd paragraph, 3rd sentence: The "resolving out" of the diffuse emission is not convincing evidence that the western radio component to the spiral host galaxy is not a chance coincidence with an unassociated source or cluster in the background.

Authors: Indeed, the resolving out of the diffuse emission does not prove a connection with the spiral galaxy but can be anything else. However, the newly presented RACS image shows the connection between (plume) the radio core component and the diffuse fist part. Other than the Thumbs Up source, we do not detect any diffuse emission source to suggest association with the background cluster.

Referee: If the diffuse emission to the west of the spiral is indeed older lobe emission, is this lobe emission steeper in radio spectral index than that of the purported core emission from the central regions of the spiral galaxy?

Authors: Yes. As presented in Figure 8, the spectral index values are steeper as we move westward from the radio emission associated with the spiral galaxy. Furthermore, the flat emission is not just overlapping with the spiral galaxy, but the plume seen in the RACS image is also a relatively flat component, suggesting that the spiral galaxy, through the plume, has likely supplied diffuse plasma to the fist in the past.

Referee: 10) Fig 7: It would be useful for this figure to include the western component as well (as per the point above). Also, the green part of the colour scale in Fig 7 is too coarse for the reader to distinguish between a flat vs a steep spectral index. Green appears to be able to represent a spectral index ranging between -0.6 to -1.1.

Authors: We thank the referee for this constructive suggestion. In response, we have updated the TGSS–NVSS spectral index map (previously Fig. 7, now presented as Fig. 8 in the revised manuscript) to include the full extent of the radio source, including the western ("fist") component. This addition ensures a complete view of the source morphology and spectral variations across both lobes. We have also carefully revised the colour scale of the spectral index map to enhance interpretability. Specifically, we adjusted the dynamic range and improved the contrast to allow for a more discernible distinction between relatively flat-spectrum regions (near the core) and steep-spectrum regions (towards the edges). As a result, the plume visible in the RACS image is now clearly identifiable in the spectral index map as a feature connecting the core to the western edge, where the plasma exhibits the steepest spectral index. This updated figure not only improves visual clarity but also strengthens our interpretation that the radio emission becomes progressively steeper away from the core, consistent with aged synchrotron plasma and a  pattern expected from ram pressure stripping.

Referee: 11) Section 3.2, 6th paragraph, 3rd last sentence: I agree with this conclusion. On the other hand as noted earlier, there is a WHL cluster BCG 1.5 arcmin north of the spiral that is at the same redshift.  Therefore, while the elliptical that appears closest in projection may be in the foreground, it does seem that this galaxy could be residing in a cluster.

Authors: Yes, indeed. In reference to the WHL cluster, we have now expressed that the Thumbs Up galaxy is part of that cluster, whose BCG is seen here.  

Referee: Other recent discoveries of spirals/late-type galaxies hosting double radio AGN from Gao et al (2023;  https://ui.adsabs.harvard.edu/abs/2023RAA....23c5005G/abstract) and Yuan  et al (2024; https://ui.adsabs.harvard.edu/abs/2024RAA....24d5007Y/abstract) suggest a connection between such galaxies and a denser local environment such as that of a group and cluster. Perhaps a comparison between the "Thumbs-up" and these recent results would be useful.

Authors: We have now cited both Gao et al and Yuan et al in the manuscript. Indeed, spiral-host radio galaxies reside in similar group/cluster environments as standard radio galaxies. We have expressed the consistency of our findings with those of these authors. 

Referee: 12) Section 4.1: What does the spectral index map look like for NGC 3898?

Authors: Currently, no high-frequency GHz data is available to complement the LoTSS 144 MHz data, and hence, the spectral index map information is beyond the scope of this paper.

Referee: 13) Section 4.1, last sentence: There is no evidence for the compression of gas (eg Kantharia et al 2005) along the leading edge of interaction. The highly ordered star-forming outer arms of NGC 3898 would also suggest a dearth of strong external influence that is either depleting the gas or suppressing the star formation process within this galaxy. Hence, more substantial evidence is needed from the multiwavelength observations of NGC 3898, to justify the authors' assertions for ram pressure stripping.

Authors: Typically, the radio lobes are far from the stellar/gaseous disks of the host galaxies.
In Fig. 5, we have added the LOFAR radio contours to the optical and HI-contours of NGC3898. This summarised the extents of stars, gas and radio lobes (although seen in projection, specially the radio lobes). Notice that the HI on the east is already shrunken (compared to that in the west) and the short radio lobe is seen (in projection) confined to the HI-gas disk while the longer lobe is seen beyond the HI gas disk.  In this study, the radio lobe asymmetry has been associated with asymmetries in the UV/optical asymmetries in the stellar/gaseous disks, which is also seen in other cases of proven ram pressure strippings. Hence, finding a direct proof is probably not possible. Deep observations in the future may find convincing evidence like a faint H-alpha tail in NGC4569 or HI tail in NGC4388.

Referee: 14) Section 4.3: Perhaps a reference to the more recent paper by Ignesti et al (2022; https://ui.adsabs.harvard.edu/abs/2022ApJ...937...58I/abstract) which show some very nice example of ram pressure-stripped radio tails, would also be very relevant to this paper. 

Authors:  We have added a reference to this paper. Thank you.

Reviewer 2 Report

Comments and Suggestions for Authors

Context: The Authors present the study of two peculiar radio sources that emerged from the citizen science project RAD@home. The effort to involve young students in the study of astrophysics is commendable and deserves recognition. The two targets have been studied by making use of multiple archival observations to conduct a detailed multiwavelength analysis of their morphology. 
However, the proposed scenarios for both sources are not convincing, and more detailed work is requested before this work deserves publication:

Major Comments:

-NGC3898: The Authors claim that the galaxy host double-jetted radio emission which has been shaped by the ram pressure. The LOFAR images show that the eastern lobe is brighter than the western one, which is a strong indication of Doppler boosting resulting from the jet orientation, i.e. indicates that the eastern jet is pointing toward the observer and the western one points in the opposite direction (https://ui.adsabs.harvard.edu/abs/2009A%2526A...494..471O ). Combined with the misaligned of the lobes with respect to the spiral galaxy, the geometry of the system appears unclear. The Authors should address this issue, possibly by adding a new figure depicting the system geometry they propose. Additionally, the comparison of Halpha, HI, and UV images suggests a weak ram pressure, which has not been able to strip the galaxy of its gas. On the other hand, bending and stripping radio galaxies requires strong ram pressure of the order of ~1e-11 erg cm^-3 (https://ui.adsabs.harvard.edu/abs/2017A%26A...608A..58T/abstract), which is typically verified in clusters. To conclude, The Authors should discuss also other possible scenarios, such as the possibility that the two radio lobes are part of a background/foreground source aligned with NGC3898.  In this regard, Figure 1 (left panel) shows a point-like radio source aligned with the two radio lobes, thus being a potential candidate to be the real radio galaxy host. The Authors should quantitatively rule out this possibility before discussing more complicated scenarios involving environmental processes;

WISEA J221656.57-042434.1 : Similarly to the previous case, the Authors propose a complicated scenario without having ruled out other, simpler possibilities. First, before invoking the effect of ram pressure, the Authors should verify and report that the galaxy is part of a larger structure hosting environmental plasma. Secondly, the asymmetrical optical and FIRST images suggest that the galaxy is indeed under the effect of ram pressure. Specifically, the high-resolution radio morphology is indeed similar to the one observed at radio frequencies for these galaxies, with the radio continuum emission extending in the galaxy wake (https://ui.adsabs.harvard.edu/abs/2021A%26A...650A.111R/abstract , https://ui.adsabs.harvard.edu/abs/2004AJ....127.3375V/abstract).  In this interpretation, the ram pressure appears to be acting from South to North. Thus its radio emission does not necessarily imply the presence of current AGN radio-loud activity. Concerning the extended radio emission, the Authors should discuss also other scenarios for its origin, e.g. remnant AGN activity from bright elliptical galaxies (https://ui.adsabs.harvard.edu/abs/2020A%26A...634A...4M/abstract). 

Minor comments and suggestions:

-Scientific images should report coordinates and scale bar to ease the interpretation;
-Surface brightness levels in the radio images should be reported in the conventional format, i.e. levels at 3,6,12,24...x rms level, where rms=...;
-Report the observed radio frequencies;
-Add labels on figures to identify the different sources and components discussed in the text;
-Add a figure depicting the proposed scenarios with a simple diagram.

Author Response

Referee 2

Referee: Context: The Authors present the study of two peculiar radio sources that emerged from the citizen science project RAD@home. The effort to involve young students in the study of astrophysics is commendable and deserves recognition. The two targets have been studied by making use of multiple archival observations to conduct a detailed multiwavelength analysis of their morphology.
However, the proposed scenarios for both sources are not convincing, and more detailed work is requested before this work deserves publication.

Authors: We thank the referee for taking the time to review our manuscript and appreciate the acknowledgement of our citizen science initiative. We have carefully addressed the concerns raised and have revised the manuscript significantly to strengthen the interpretation of both sources. We hope the updated analysis and additional data now present a more convincing case.

Major Comments:

Referee: -NGC3898: The Authors claim that the galaxy host double-jetted radio emission which has been shaped by the ram pressure. The LOFAR images show that the eastern lobe is brighter than the western one, which is a strong indication of Doppler boosting resulting from the jet orientation, i.e. indicates that the eastern jet is pointing toward the observer and the western one points in the opposite direction (https://ui.adsabs.harvard.edu/abs/2009A%2526A...494..471O ). Combined with the misaligned of the lobes with respect to the spiral galaxy, the geometry of the system appears unclear. The Authors should address this issue, possibly by adding a new figure depicting the system geometry they propose. 

Authors: We thank the referee for raising this important point. We agree that Doppler boosting is a plausible contributor to the observed brightness asymmetry between the radio lobes and have now explicitly acknowledged this in the revised manuscript. In the updated Discussion section, we have added a dedicated paragraph near the beginning to outline the standard explanations for radio lobe asymmetries, including Doppler boosting, light travel time effects, and differences in ambient medium properties, citing O’Dea 2009 as suggested. However, our analysis indicates that in the case of NGC3898, Doppler boosting alone is insufficient to explain the full set of observed features, particularly the detailed backflow structures and environmental asymmetries visible in the optical/UV data. We now highlight this with a more nuanced discussion, supported by additional multiwavelength evidence. To clarify the proposed system geometry, we have added two schematic diagrams illustrating the orientation and impact of ram pressure stripping on the radio morphology for both galaxies (now presented in Figure 11). We trust that these additions make the underlying physical picture and our interpretation more transparent and accessible to readers.

 

Referee: Additionally, the comparison of Halpha, HI, and UV images suggests a weak ram pressure, which has not been able to strip the galaxy of its gas. On the other hand, bending and stripping radio galaxies requires strong ram pressure of the order of ~1e-11 erg cm^-3 (https://ui.adsabs.harvard.edu/abs/2017A%26A...608A..58T/abstract), which is typically verified in clusters. To conclude, The Authors should discuss also other possible scenarios, such as the possibility that the two radio lobes are part of a background/foreground source aligned with NGC3898.  In this regard, Figure 1 (left panel) shows a point-like radio source aligned with the two radio lobes, thus being a potential candidate to be the real radio galaxy host. The Authors should quantitatively rule out this possibility before discussing more complicated scenarios involving environmental processes.

 

Authors: We have added a detailed discussion addressing the possibility that the observed radio lobes might belong to a background radio galaxy. Specifically, in the case of NGC3898, we examined the candidate point-like radio source aligned along the radio axis. In the revised manuscript, we present a high-resolution LOFAR image overlaid on the DESI Legacy optical image (now Figure 3), clearly showing a significant spatial offset (>2″) between the radio peak and the nearby optical galaxy. The radio source is faint (0.8 mJy at 144 MHz) and undetected in FIRST and VLASS, suggesting that it is unlikely to be a true radio core. Moreover, we have verified the astrometric accuracy of the LOFAR image using field sources, further reinforcing this conclusion.

Additionally, we now include deeper imaging from RACS and EMU (new Figure 9), which reveals a coherent morphological structure in the Thumbs Up galaxy: a well-detected radio core (coincident with the disturbed spiral), a connecting plume (seen in RACS), and extended lobes with filamentary structure (in EMU). The morphology and alignment strongly support a two-episode radio outflow scenario originating from the disturbed spiral host, rather than from a background or unrelated source.

We have also incorporated a discussion on alternative mechanisms for the observed asymmetries, including Doppler boosting and projection effects, at the start of the Discussion section. 

While we appreciate the referee's reference to Gregory et al. (2017), we believe the comparison is not directly applicable to our case. That study focuses on classical head-tail radio galaxies in dense cluster environments, typically hosted by massive elliptical galaxies moving at high velocities through a dense intracluster medium (ICM). In such scenarios, ram pressure values on the order of ~1e-11 erg cm^-3  are indeed required to cause significant bending of the jets or displacement of the radio lobes. In contrast, our case involves a spiral galaxy, NGC 3898, which is not located in the core of a rich cluster but rather in a moderately rich group environment (with M200~10^13 Msun​). Spiral hosts are structurally and dynamically different from ellipticals: they are gas-rich with shallower central potential wells, particularly in their outer discs, and may respond to even modest ram pressure. The observed asymmetry in the radio lobes and the alignment of UV/optical gas features are consistent with weak ram pressure acting preferentially on the more tenuous backflow plasma of the radio lobes rather than on the host's stellar or gaseous disc.

Referee: WISEA J221656.57-042434.1 : Similarly to the previous case, the Authors propose a complicated scenario without having ruled out other, simpler possibilities. First, before invoking the effect of ram pressure, the Authors should verify and report that the galaxy is part of a larger structure hosting environmental plasma. Secondly, the asymmetrical optical and FIRST images suggest that the galaxy is indeed under the effect of ram pressure. Specifically, the high-resolution radio morphology is indeed similar to the one observed at radio frequencies for these galaxies, with the radio continuum emission extending in the galaxy wake (https://ui.adsabs.harvard.edu/abs/2021A%26A...650A.111R/abstract , https://ui.adsabs.harvard.edu/abs/2004AJ....127.3375V/abstract).  In this interpretation, the ram pressure appears to be acting from South to North. Thus its radio emission does not necessarily imply the presence of current AGN radio-loud activity. Concerning the extended radio emission, the Authors should discuss also other scenarios for its origin, e.g. remnant AGN activity from bright elliptical galaxies (https://ui.adsabs.harvard.edu/abs/2020A%26A...634A...4M/abstract). 

Authors: We thank the referee for their comments and for highlighting alternative interpretations that merit discussion. However, we would like to clarify that the central aim of our manuscript is not to establish whether the galaxy is undergoing ram pressure stripping of its ISM in the traditional jellyfish sense. Instead, we present observational evidence that the extended radio morphology,  particularly the asymmetric, non-aligned lobes well beyond the stellar disk, may be influenced by environmental effects such as ram pressure acting on the radio plasma, not just the ISM. There are many known examples of wide-angle-tailed radio galaxies found in group environments, which demonstrate that even the relatively modest ram pressure present in such settings can be sufficient to bend or distort radio lobes and plumes. This is well illustrated in the study by Freeland & Wilcots (2011), where intra-group gas densities were quantified and shown to be capable of shaping radio structures. These cases establish that a dense cluster environment is not a prerequisite for observing ram pressure effects on radio lobes, supporting our interpretation that the features seen in the Thumbs Up galaxy may arise from similar environmental interactions.

As highlighted by the referee, a recent study using LoTSS data (Ignesti et al. 2022) finds that many jellyfish galaxies exhibit radio continuum tails even in the absence of diffuse H-alpha emission. However, such radio-tailed galaxies lacking double-lobed AGNs likely represent a physically distinct class, differing in both origin and morphology from classical radio galaxies.

An excellent example is the spiral galaxy NGC4569, which exhibits a 230 kpc H-alpha tail due to ram pressure stripping (Sun et al.). The associated asymmetric radio lobes are only ~24 kpc across and lie embedded within the ionised gas tail (Chyży et al.). The magnetic field vectors in the radio lobes are well aligned with the direction of the inferred ram pressure, while those within the host galaxy are nearly orthogonal. This alignment suggests that once the radio lobes extend beyond the stellar and gaseous boundaries of the galaxy, they can be more easily bent and displaced by ram pressure. In our case, the spectral index map of the Thumbs Up galaxy supports the interpretation that the more distant, diffuse emission is a remnant of an earlier episode of AGN activity, visible now only as a steep-spectrum, low-surface-brightness feature.

Minor comments and suggestions:

Referee: -Scientific images should report coordinates and scale bar to ease the interpretation;

Authors: New images presented have coordinate and scale bars. The reason some radio images do not contain coordinates is because we have directly presented the RAD@home -RGB-maker web-tool outputs describing the discovery process. By design, these images do not contain source identity because citizen scientists directly share these images for group discussion. Discussions facilitate their intellectual growth but at the same time images without RA-Dec protect their intellectual property (identity of a potential discovery). Trust that is acceptable.

Referee: -Surface brightness levels in the radio images should be reported in the conventional format, i.e. levels at 3,6,12,24...x rms level, where rms=...;

Authors: We have now expressed the rms noise levels in each figure caption to better understand statistical significance from contour levels. With the presentation of the EMU image, the weak feature Thumb is now beyond doubt a real component.

Referee: -Report the observed radio frequencies;
Authors:  We have now reported the frequency of observation in each figure.
Referee: -Add labels on figures to identify the different sources and components discussed in the text.
Authors:  We have now added labels to the images for better mapping between images and text descriptions.
Referee: -Add a figure depicting the proposed scenarios with a simple diagram.

Authors:  We have now added two schematic figures to explain the ram pressure stripping scenario proposed here for each of these two targets. Trust that will make the explanations clear. Thank you very much.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

-NGC3898: I am still not convinced by the proposed scenario. Spiral galaxies with radio jets are rare, hence, this discovery would be very valuable, but the Authors should provide solid evidence to support their claim.
1) Regarding the LOFAR point source reported in Figure 3, the Authors state that: 
"The compact radio source has a flux density of 0.8 mJy at 144 MHz but is not detected in higher frequency radio surveys, including FIRST (3σ upper limit: 0.5 mJy) and Very Large Array Sky Survey(VLASS; 3σ upper limit: 0.4 mJy). This nondetection at higher frequencies suggests that the source does not exhibit the flat-spectrum characteristics typical of radio cores." 
However, I note that the non-detection sets an upper limit for its spectral index: alpha < log(0.8/0.5)/log(144/1400) = -0.2, which would be consistent with a radio core. I further add that, due to the 6" resolution, a 2" shift is not enough to rule out an overlap between the radio and optical emission.
2) Regarding the HI emission, the Authors report about a truncation in the eastern part, which would indicate that the side is facing the ram pressure. However, the surface brightness contours are more closely-spaced on the western side than the eastern one, which would suggest that the ram pressure is compressing the western side (https://ui.adsabs.harvard.edu/abs/2023A%26A...676A.118D/abstract).
3) I suggest that a test to rule out the possibility that the radio lobes are hosted by the background galaxy. According to the images, the lobes' angular size is ~1 arcmin, which at NGC3898 redshift would correspond to ~5 kpc. On the contrary, if they were at the putative radio core redshift (z=0.469), their linear size would be ~350 kpc. In those two scenarios, the lobes would have different synchrotron self-absorption frequency, which typically decreases with the linear size (e.g., https://academic.oup.com/mnras/article/458/4/3786/2613792). Specifically, if the lobes are at the same redshift as NGC3898, the peak frequency would be between 100 and 1000 MHz. On the contrary, in the other scenario, the peak frequency would be below LOFAR frequency. I checked that the lobes are observed both at 144 and 1400 MHz (NVSS), thus I suggest the Authors to compute an integrated spectral index between these frequencies for the two lobes. If the lobes are actually associated with the spiral galaxy, then the peak frequency should be between these two observed frequencies, and, hence, the spectral index would be close to 0. On the other hand, if the spectral index is steeper than -0.5, then it would be evidence that the peak frequency is lower than 144 MHz and, thus, that the radio lobes are much larger than 5 kpc. 

-Thumb-up galaxy: The inclusion of the new data really highlights how complex this source is. I agree that the interpretation is not trivial; however, in the proposed scenario, it is still unclear why the emission between the thumbs and the wrist is undetected, whereas the bridge between the fist and the thumb is. I add that "This elliptical galaxy was also undetected in the FIRST
radio survey image (Fig. 7), which supports the absence of any phoenix- or halo-like diffuse emission from the cluster BCG." could be misleading. Not observing the emission at higher frequencies is a signature of a steep spectral index, which is a key characteristic of diffuse or relic sources. Regarding the interpretation, I wonder if the Authors could consider, or rule out, the possibility that the spiral galaxy could have re-ignited the emission of a pre-existing fossil plasma cloud by crossing it (https://ui.adsabs.harvard.edu/abs/2025A%26A...697A..45R/abstract). The spectral index map shows a tentative gradient along the SE-NW direction, which may support this scenario. 

Minor comments: 
-Scale-bar in figures should report physical scales (e.g. kpc), not angular ones;
-I appreciate that the Authors have included a new figure depicting their proposed scenarios, but in those pictures, they report the exact structure shown in the real images. I suggest either attempt presenting an alternative point of view, or a time-sequence of the events that would have resulted in the observed morphologies.
-

Author Response

Dear Editor, 

We thank the referee for his/her critical comments on the manuscript. It has helped us present the interpretation in an unbiased and clear way. We have modified figures and added several paragraphs (red font) in our revised version. These modifications are primarily to express that the association of radio lobes with NGC3898, and thumb-fist feature with the host of the mini-thumbs up are debatable. However, given data and arguments we have put our interpretation as they are associated and suggest the effect of ram pressure stripping on the asymmetric radio lobes. We bring attention to an earlier work which also had proposed effects of ram pressure stripping on asymmetric radio galaxies Gopal-Krishna & Wiita 1996 (https://ui.adsabs.harvard.edu/abs/1996ApJ...467..191G/abstract )
Our work brings observational evidence to this from multi-wavelength imaging studies. We hope these modifications are acceptable to the referee. Please find below detailed reply to the referee comments. 

Referee: -NGC3898: I am still not convinced by the proposed scenario. Spiral galaxies with radio jets are rare, hence, this discovery would be very valuable, but the Authors should provide solid evidence to support their claim.

Authors: We agree that radio lobe association with spiral galaxies is rare and in our case has become difficult to prove. To express this ambiguity prominently, along with additions in the text (in red) we have changed the titles of the subsections in the result as follows. 1. Radio lobes of NGC3898 or a background radio galaxy ? 2. RAD-Thumbs up Galaxy: Speca or radio Phoenix? Additionally, 

Referee: 1) Regarding the LOFAR point source reported in Figure 3, the Authors state that:
"The compact radio source has a flux density of 0.8 mJy at 144 MHz but is not detected in higher frequency radio surveys, including FIRST (3σ upper limit: 0.5 mJy) and Very Large Array Sky Survey(VLASS; 3σ upper limit: 0.4 mJy). This nondetection at higher frequencies suggests that the source does not exhibit the flat-spectrum characteristics typical of radio cores."
However, I note that the non-detection sets an upper limit for its spectral index: alpha < log(0.8/0.5)/log(144/1400) = -0.2, which would be consistent with a radio core. I further add that, due to the 6" resolution, a 2" shift is not enough to rule out an overlap between the radio and optical emission.

Authors: We have now mentioned in the paper the consistency of the spectral index value with a flat-spectrum radio core. Since this is an upper limit, it does not rule out the possibility of it being a steep-spectrum high-z radio galaxy whose host galaxy is even fainter than the BASS detection limit. We have also added a discussion about the 0.2” accuracy of LOFAR high-resolution images. We have added a figure (Fig. 3, right panel) where we show four point sources, both in radio and optical, seen in four directions around NGC3898. The LoTSS high-resolution image contours are seen perfectly aligned with the centre of the optical sources. We have now clearly expressed this ambiguity in the title of the Result sub-section (Radio lobes of NGC3898 or a background radio galaxy ?) The significant positional offset of the compact source and the absence of any connecting jet lead us to favour an association of the asymmetric lobes with NGC 3898.  

Referee: 2) Regarding the HI emission, the Authors report about a truncation in the eastern part, which would indicate that the side is facing the ram pressure. However, the surface brightness contours are more closely spaced on the western side than the eastern one, which would suggest that the ram pressure is compressing the western side (https://ui.adsabs.harvard.edu/abs/2023A%26A...676A.118D/abstract).

Authors:  The best evidence for the direction of ram pressure stripping is long tails of HI or H-alpha (e.g. galaxies in A1367 https://ui.adsabs.harvard.edu/abs/2001ApJ...563L..23G/abstract ). HI study of three tailed galaxies of the same system shows truncation/compression features that can be compared with what is seen in NGC3898 (https://ui.adsabs.harvard.edu/abs/2007BASI...35..121H/abstract ). HI contours show truncation in the side facing the wind, despite some contour compression seen in the side of the tail (e.g. CGCG 097073 and CGCG 097079). Similar truncation on the side facing the wind and accumulation of HI on the side of the tail can also be seen in the edge on system CGCG 097087.  Taken together, the observational evidence indicates that any ram‑pressure force acting on NGC 3898 is most plausibly directed from east to west.

Referee: 3) I suggest that a test to rule out the possibility that the radio lobes are hosted by the background galaxy. According to the images, the lobes' angular size is ~1 arcmin, which at NGC3898 redshift would correspond to ~5 kpc. On the contrary, if they were at the putative radio core redshift (z=0.469), their linear size would be ~350 kpc. In those two scenarios, the lobes would have different synchrotron self-absorption frequency, which typically decreases with the linear size (e.g., https://academic.oup.com/mnras/article/458/4/3786/2613792 ). Specifically, if the lobes are at the same redshift as NGC3898, the peak frequency would be between 100 and 1000 MHz. On the contrary, in the other scenario, the peak frequency would be below LOFAR frequency. I checked that the lobes are observed both at 144 and 1400 MHz (NVSS), thus I suggest the Authors to compute an integrated spectral index between these frequencies for the two lobes. If the lobes are actually associated with the spiral galaxy, then the peak frequency should be between these two observed frequencies, and, hence, the spectral index would be close to 0. On the other hand, if the spectral index is steeper than -0.5, then it would be evidence that the peak frequency is lower than 144 MHz and, thus, that the radio lobes are much larger than 5 kpc. 

Authors: We have measured integrated TGSS–NVSS spectral indices for the eastern lobe, central component, and western lobe of the source. The values are –0.79, –1.00, and –0.58, respectively. The eastern lobe index is likely contaminated by core emission, whereas the western lobe provides a clean measurement. Nearby Seyfert and ram-pressure-stripped spirals with lobes only a few kiloparsecs across can show similarly steep two-point indices (Hota & Saikia 2006 https://ui.adsabs.harvard.edu/abs/2006MNRAS.371..945H/abstract); examples include NGC 3079 (–0.66 for a 3.4 kpc bubble) and NGC 5548 (–0.79 for a 3.4 kpc jet). Hence, a slope of –0.6 to –0.8 is not, by itself, evidence that the source is hundreds of kiloparsecs in size.  The v_peak​–size relation cited by the referee (Jeyakumar 2016) was derived for GPS/CSS sources, which are less than ~20 kpc in size. Large (>30kpc), classical FR-II lobes hundreds of kiloparsecs across generally show their spectral turnovers well below 100 MHz and are not synchrotron self-absorbed at 144 MHz or 1.4 GHz, so the relation is not directly applicable.

Referee:-Thumb-up galaxy: The inclusion of the new data really highlights how complex this source is. I agree that the interpretation is not trivial; however, in the proposed scenario, it is still unclear why the emission between the thumbs and the wrist is undetected, whereas the bridge between the fist and the thumb is. I add that "This elliptical galaxy was also undetected in the FIRST
radio survey image (Fig. 7), which supports the absence of any phoenix- or halo-like diffuse emission from the cluster BCG." could be misleading. Not observing the emission at higher frequencies is a signature of a steep spectral index, which is a key characteristic of diffuse or relic sources. Regarding the interpretation, I wonder if the Authors could consider, or rule out, the possibility that the spiral galaxy could have re-ignited the emission of a pre-existing fossil plasma cloud by crossing it (https://ui.adsabs.harvard.edu/abs/2025A%26A...697A..45R/abstract). The spectral index map shows a tentative gradient along the SE-NW direction, which may support this scenario. 

Authors: The EMU image (Fig. 9 right hand top panel) does show the wrist-to-thumb filament but very faint. We expressed it in the text.  We corrected the misleading statement, the FIRST/EMU non-detection of the BCG is now mentioned as a fact only, without interpretation. We have now clearly discussed that the thumb and fist components could be phoenix. We have further added that given the absence of a possible host galaxy for these two radio phoenixes the best candidate is the same host galaxy on the east. Hence, WAT lobes (mini thumbs up) as well as the phoenixes constitute a spiral-host episodic radio galaxy (Speca). Since we do not know about the possibility of mergers in this cluster and even no sign of interaction between two galaxies, shock-revival of remnant lobes to show up as  a phoenix has become doubtful.  We could not feel strongly about passing a galaxy through its own remnant radio lobes to re-ignite it. 

Minor comments:
-Scale-bar in figures should report physical scales (e.g. kpc), not angular ones;

Authors: We have now expressed the scale bar in both angular and linear values in every figure caption.
Referee: -I appreciate that the Authors have included a new figure depicting their proposed scenarios, but in those pictures, they report the exact structure shown in the real images. I suggest either attempt presenting an alternative point of view, or a time-sequence of the events that would have resulted in the observed morphologies.

Authors: We have now added schematics for the time evolution (in two stages) of both sources as advised. The eastward movement of the host galaxy NGC3898 from a symmetric FR II structure to an asymmetric lobe phase is now depicted clearly. Similarly the two episodes of radio lobes along with movement of the host or remnant lobes is clearly seen in the schematic for Thumbs up galaxy. Hope that is clear. Thank you very much.

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

I appreciate the Authors efforts that further improve the manuscript presentation. I am still not convinced by the Authors interpretation of the first source, but at this point, it is my personal opinion so it should not delay the paper acceptance any longer. 

Just one last minor comment: on lines 285-286, they report, "This is also unlike cluster radio halos, which are relatively flat spectral and seen around cluster BCGs."  Radio halos, and diffuse radio sources in general, are typically characterized by a spectral index steeper than -1 and thus they are referred to as 'steep-spectrum sources'.

Congratulations on your work.

Author Response

Dear Editor,
We thank the reviewer for his/her prompt action on our revised manuscript. 

The minor comment was "Just one last minor comment: on lines 285-286, they report, "This is also unlike cluster radio halos, which are relatively flat spectral and seen around cluster BCGs."  Radio halos, and diffuse radio sources in general, are typically characterized by a spectral index steeper than -1 and thus they are referred to as 'steep-spectrum sources'."

We have modified the particular line to the following " This is also unlike cluster radio halos, which are steep spectral diffuse sources and are seen around cluster BCGs without links to any individual galaxy." We say this  primarily because of the thumb to core filamentary connection in EMU map. 

Trust our paper can now be formally accepted.  We thank the referee again for his/her stance on agreeing to a difference of opinion which future observations can only resolve.

with best wishes

Dr Ananda Hota for all authors