EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 685 (May 2024) A&A, 685 (2024) L13 Full HTML
Open Access
Issue
A&A
Volume 685, May 2024
Article Number L13
Number of page(s) 15
Section Letters to the Editor
DOI https://doi.org/10.1051/0004-6361/202449982
Published online 20 May 2024

© The Authors 2024

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1. Introduction

Feedback from active galactic nuclei (AGN) is an important element for the evolution of galaxies (e.g., Weinberger et al. 2018). However, detecting deeply buried AGN is still difficult due to dust obscuration. This is particularly challenging in local luminous and ultra-luminous infrared (IR) galaxies (U/LIRGs; LIR > 1011 L⊙). Local U/LIRGs are mostly interacting or merging systems, so they represent a key phase in the evolution of galaxies (e.g., Patton et al. 2020), and their activity (AGN and star formation) occurs in extremely dust-embedded environments.

Activity diagnostics in the optical fail in many U/LIRGs since they host the most obscured nuclei in the local Universe (NH > 1025 cm−2; e.g., Falstad et al. 2021; García-Bernete et al. 2022a). Indirect methods using mid-IR spectroscopy pointed out the presence of AGN in a large fraction, ∼70%, of ULIRGs (e.g., Nardini et al. 2009; Veilleux et al. 2009). These mid-IR methods suggested a modest AGN contribution to the total luminosity. However, radio and sub-millimeter observations, which are insensitive to dust extinction, revealed compact nuclei exceeding the surface brightness density of a maximal starburst, hence likely AGN, dominating the bolometric luminosity of the majority of ULIRGs (Barcos-Muñoz et al. 2017; Pereira-Santaella et al. 2021; Hayashi et al. 2021). Therefore, the direct detection of these AGN remains elusive and its exact contribution to the total luminosity is still debated.

The reduced dust extinction in the near- and mid-IR spectral ranges, along with the superb sensitivity and angular resolution of the James Webb Space Telescope (JWST), for the first time allows the search for high-ionization emission lines, which may eventually provide direct evidence of these elusive AGN in U/LIRGs. Specifically, the [Mg IV] 4.487 μm line, with a high ionization-potential (IP; 80 eV), is not expected in gas photoionized by young stars since they do not emit significant quantities of photons with the energies beyond 54 eV (He+ ionization energy). Therefore, the detection of the [Mg IV] line, which lies close to the minimum of the dust opacity (e.g., Corrales et al. 2016), has been proposed as an excellent tracer of deeply embedded AGN (e.g., Satyapal et al. 2021). However, fast ionizing-shocks (> 80 km s−1; Sutherland & Dopita 2017) can also produce highly ionized gas that can emit high-ionization lines. For instance, the [O IV] 25.9 μm line (IP 55 eV), which is well correlated with the AGN power (e.g., Diamond-Stanic et al. 2009; Pereira-Santaella et al. 2010b; García-Bernete et al. 2016), is also associated with shocks in starbursts (e.g., Lutz et al. 1998; Petric et al. 2011; Alonso-Herrero et al. 2012).

In this Letter we investigate the origin (AGN, star formation, or shocks) of the extended [Mg IV] emission detected in a sample of four local LIRGs observed with the JWST/Near Infrared Spectrograph (NIRSpec; Jakobsen et al. 2022). We first analyze the spatial distribution of the [Mg IV] emission and compare the [Mg IV] line profiles with those of other low-ionization and H recombination lines. Then, we use new grids of photoionization and shock models to determine the physical and dynamical conditions of the gas producing this line.

2. Sample and data reduction

Four local (39 < DL/Mpc < 161) LIRGs, with LIR = 1011.6 − 11.9 L⊙, were observed with JWST/NIRSpec using the high spectral resolution grating G395H (R ∼1900–3600; 2.87–5.27 μm) and the integral field unit (IFU) mode (Böker et al. 2022), and with JWST/Mid-Infrared Instrument (MIRI) using all the bands of the Medium Resolution Spectrograph (MRS; Wright et al. 2023; Argyriou et al. 2023). These observations were part of the Director’s Discretionary Early Release Science (DD-ERS) Program #1328 (PI: L. Armus and A. Evans). These four objects cover a wide range of properties, from pre-mergers to late merger stages and from heavily obscured AGN to Type 1 AGN (see Table 1).

Table 1.

Sample of local LIRGs.

For the data reduction, we used the JWST calibration pipeline (version 1.12.4; Bushouse et al. 2023) and the context 1197. We followed the standard reduction recipe complemented by a number of custom steps to reduce the effect of bad pixels and cosmic rays. Further details on the data reduction of the NIRSpec and MIRI/MRS data observations can be found in Pereira-Santaella et al. (2022, 2024) and García-Bernete et al. (2024a).

3. Analysis and results

3.1. Spatially extended [Mg IV] emission

Figure 1 shows the rest-frame 4.31–4.57 μm spectra from 23 regions selected in the four systems. To extract these spectra we used apertures with a diameter of 054 (100–400 pc depending on the target). We did not apply any aperture correction as the emission appears spatially resolved. Nevertheless, in a worst-case scenario of a point source, the error in the line ratios discussed below (Sect. 4) would be small, < 5%. The regions were selected to cover the extended [Mg IV] emission detected in the line maps (see Figs. 2 and A.1–A.4). For simplicity, for this work we excluded regions with a strong CO v = 1–0 band since the [Mg IV] transition is almost coincident with the CO v = 1–0 R(24) 4.489 μm line (Fig. 1). We note, however, that after modeling the band, it is also possible to detect [Mg IV] emission in these regions (see González-Alfonso et al. 2024c; García-Bernete et al. 2024b).

thumbnail Fig. 1.

Scaled and shifted JWST/NIRSpec 4.31–4.57 μm spectra of the selected regions (see Sect. 3.1 and Figs. 2 and A.1–A.4). The wavelength of the transitions tracing ionized gas (Hu-12), warm molecular gas (H2 0–0 S(10) 4.410 μm and 1–1 S(11) 4.417 μm), highly ionized gas ([Mg IV] and [Ar VI]), and the CO v = 1–0 band are indicated by the green, blue, red, and purple vertical lines, respectively.

thumbnail Fig. 2.

Line maps and velocity fields for NGC 3256 S. The remaining objects are in Figs. A.1–A.4. Top row from left to right: Line maps of Hu-12, H2 S(8), [Mg IV], and [Fe II]. The areas filled with purple crosses in the [Mg IV] panel correspond to regions where the CO v = 1–0 band is strong and complicates the modeling of the [Mg IV] line. The black cross indicates the position of the nucleus. Bottom row from left to right: Velocity fields from single Gaussian fits of Pf-γ, H2 S(8), [Mg IV], and [Fe II]. The circles in the first and third panel give the location and size of the selected regions. The units of the color scale are 10−15 erg cm−2 s−1 arcsec−1 for the line maps and km s−1 for the velocity fields.

The spectra show clear [Mg IV] detections, two H2 lines, the Humphreys-12 4.376 μm (Hu-12) recombination line, and in some cases weak absorptions from the R-branch of the CO v = 1–0 band. The [Ar VI] 4.530 μm transition, with an IP of 75 eV slightly lower than [Mg IV], is only detected in the nucleus and outflow of the Type 1 AGN NGC 7469 with the two lines having comparable intensities (see also Bianchin et al. 2024). The latter is consistent with the Infrared Space Observatory (ISO) detections of the [Mg IV] and [Ar VI] line pair in two AGN (Circinus and NGC 1068), where the [Ar VI] line is approximately two times brighter (Sturm et al. 2002). Based on these previous results for AGN, the absence of the [Ar VI] transition in the remaining regions is surprising and suggests that distinct physical conditions, perhaps not related to an AGN, can lead to this [Mg IV] emission.

To determine the origin of the [Mg IV] line, we created 2D emission maps of tracers of the ionized gas phase (Hu-12 and Pf-γ 3.741 μm), the warm molecular gas phase (H2 0–0 S(8) 5.053 μm), and fast shocks ([Fe II] 5.340 μm). Warm molecular gas excited by shocks has been identified in interacting systems and U/LIRGs (e.g., Guillard et al. 2009; Pereira-Santaella et al. 2014), but the shock velocities are typically lower than those of the ionizing shocks traced by [Fe II]. We extracted the spectra using a running 2x2 spaxels box and fitted a local linear continuum level and a Gaussian profile to determine the flux and velocity at each position (Figs. 2 and A.1–A.4). The [Mg IV] morphology is spatially extended, from a few hundred pc to ∼1 kpc, and its spatial distribution is more similar to that of the ionized and shocked gas (Hu-12 and [Fe II]) than to the warm molecular H2 emission. Actually, for the regions selected above, there is a very good linear correlation between the Hu-12 and the [Mg IV] luminosities (excluding the nucleus and outflow of NGC 7469) which spans almost 2 dex and has a Spearman correlation coefficient of 0.94 (Fig. 3 top). This correlation also holds when we consider the individual spaxels (Fig. 3 bottom). The correlation between the shock tracer [Fe II] and [Mg IV] is also good (Fig. C.1 bottom). For the warm H2, the correlation coefficient is lower, 0.72, and the correlation, if real, would not be linear (Fig. C.1 top).

thumbnail Fig. 3.

[Mg IV] vs. Hu-12 luminosities of the selected regions (top) and for every spaxel (bottom). Different colors and symbols are used to identify the regions of each object. In the top panel, the labels of the points are the names of the regions. The Spearman correlation coefficient, rs, is indicated. The solid and dashed red lines are the best fit (log y = 0.30 + 0.98log x) and best linear fit (log y = ( − 0.29 ± 0.18)+log x), respectively, to the luminosities of the regions.

3.2. Broad [Mg IV] line profiles

We compared the kinematics of the gas emitting the [Mg IV] line with that of the ionized gas traced by other transitions by modeling their profiles. In particular, we analyzed the following lines in the spectra of the selected regions: the Hu-12 line that is close in wavelength to [Mg IV] and the effects of differential extinction are minimized; Pf-γ, which can be more affected by extinction in these dusty galaxies (APf−γ/A[Mg IV]= 1.14; Chiar & Tielens 2006), but is intrinsically brighter than Hu-12 (Pf-γ/Hu-12 = 5.53 at 10 000 K; Storey & Hummer 1995); [Fe II] 5.340 μm, which is a good tracer of shocks (Koo et al. 2016); and [Ar II] 6.985 μm, which also traces ionized gas, but has significantly higher signal-to-noise ratios (S/N).

The observed line profiles and best-fit models are presented in Figs. 4 and B.1. A single Gaussian component reproduces the [Mg IV] and Hu-12 profiles well. We find that the [Mg IV] profile is systematically broader than the Hu-12 profile (top panel of Fig. 5). The average widths, corrected for the instrumental resolution, are σcorr([Mg IV]) = 90 ± 25 km s−1 and σcorr(Hu − 12) = 57 ± 15 km s−1. In addition, the [Mg IV] emission presents velocity shifts, up to ±50 km s−1, with respect to Pf-γ (the NIRSpec absolute wavelength accuracy for the G395H grating is ∼12 km s−1; priv. comm.). In the higher S/N profiles of Pf-γ, a second broader component is clearly detected through higher velocity wings in all the profiles. Wings are also detected in the [Fe II] and [Ar II] lines (Fig. 4, two rightmost panels). Broad profiles (> 200 km s−1) have been detected in the mid-IR lines of local U/LIRGs typically associated with the narrow-line region (NLR) of AGN and ionized outflows (Spoon et al. 2009; Dasyra et al. 2011; Alonso-Herrero et al. 2013; Vivian et al. 2022; García-Bernete et al. 2022b; Armus et al. 2023). However, we note that the regions selected here are widespread over the entire extent of these objects and were not specifically selected to encompass outflows or the NLR of AGN.

thumbnail Fig. 4.

Observed profiles of the [Mg IV], Hu-12, Pf-γ, [Fe II], and [Ar II] emission lines (solid black line) for two selected regions. The remaining regions are shown in Fig. B.1. The filled gray Gaussian represents the instrumental spectral resolution. The blue line in the first panel is the best single Gaussian fit to the [Mg IV] profile. The [Mg IV] model is also plotted as a blue dashed line in the second and third panels, normalized to the peak of Hu-12 and the broad Pf-γ component, respectively, for reference. The two component fits for Pf-γ, [Fe II], and [Ar II], are represented by the pink (narrow component) and green (broad component) lines and the total model in red. For reference, the dashed lines indicate the rest velocity for each region derived from the narrow component of Pf-γ and the zero flux level. The observed (not corrected for instrumental resolution) σ and velocity shift derived from the Gaussian fits are indicated in each panel in km s−1.

We show the best-fit two-Gaussian model for these lines in the third, fourth, and fifth columns of Figs. 4 and B.1. We find a remarkable similarity between the widths and velocity shifts of the [Mg IV] and the broad Pf-γ profiles. The (Pf-γ) = 105 ± 15 km s−1 is comparable to σcorr([Mg IV]) and the great majority (85%) of the velocity shifts have the same sign (i.e., lie in the blue or red shaded areas in the bottom panel of Fig. 5).

thumbnail Fig. 5.

Comparison between the width, corrected for the instrumental resolution, of the [Mg IV] and Hu-12 lines (top) and between the velocity shifts of [Mg IV] and the broad Pf-γ component relative to the narrow Pf-γ component (bottom). The symbols are as in Fig. 3. The large blue star in the top panel represents the mean σcorr in each axis. The solid black line is the one-to-one relation. The shaded blue and red areas in the bottom panel indicate blue and red velocity shifts, respectively.

In most of the cases (∼75%) the [Mg IV] line is blueshifted relative to the narrow component in Pf-γ. Although extinction is greatly reduced at ∼4.5 μm, it might affect the receding side of the [Mg IV] emission and explain this uneven distribution of the sign of the shifts. A detailed study of the extinction is beyond the scope of this Letter. We also note that the sign of the shifts of the broad component of [Fe II] and [Ar II] follow the same trend (Fig. C.2) and the widths are also comparable to that of the [Mg IV] line: ([Fe II]) = 112 ± 27 km s−1 and ([Ar II]) = 102 ± 20 km s−1.

4. Origin of the [Mg IV] emission: Star formation, AGN, and shock models

Despite its high IP, ther are three reasons why the results described in Sect. 3 are not fully consistent with an AGN origin of the [Mg IV] emission: (1) it is extended and correlated with the Hu-12 emission, a H recombination line tracing star-forming regions based on its morphology; (2) the [Ar VI] line, with a slightly lower IP and typically found in other AGN, is undetected, except in the outflow and nucleus of the Type 1 AGN; and (3) the [Mg IV] profile is broader and shifted compared to the other transitions. Therefore, we created three grids of models (star formation, AGN, and shocks) to investigate the origin of the [Mg IV] emission. We briefly describe the models in this section and provide further details in Appendix D.

We modeled the emission from gas photoionized by AGN or star formation using the spectral synthesis code CLOUDY (Chatzikos et al. 2023). We mostly followed Pereira-Santaella et al. (2017) to create these grids, but updated the spectral energy distributions (SEDs) of the incident radiation field. For the star formation grids, we used the Binary Population and Spectral Synthesis (BPASS) library (Eldridge et al. 2017), which includes stripped-envelope stars that have harder ionizing spectra (e.g., Götberg et al. 2019) and could contribute to the [Mg IV] emission. For the AGN models, we used the three SEDs for low, medium, and high Eddington ratios derived by Jin et al. (2012), which provide a more realistic approximation to the intrinsic SED of AGN than a power law. Based on these grids, we can reject that the [Mg IV] line originates in gas photoionized by stars since the observed [Mg IV]/Hu-12 ratio (∼0.5; Fig. 3) is > 2.3 dex larger than the predicted ratio (Fig. D.1).

The AGN photoionization predictions for the [Mg IV]/Hu-12 and [Ar VI]/[Mg IV] ratios are shown in the left panel of Fig. 6. The outflow and nucleus of the Sy1 NGC 7469 lie close to the grid with ionization parameters, log U, between −2.00 and −2.25 in the range of Seyfert AGN (e.g., Pérez-Díaz et al. 2022). For the remaining regions, the lower [Mg IV]/Hu-12 ratio and the nondetection of [Ar VI] implies log U < −3.25, which is low, but is also found in some AGN and low-ionization nuclear emission regions (LINERs; e.g., Thomas et al. 2018).

thumbnail Fig. 6.

[Ar VI]/[Mg IV] vs. [Mg IV]/Hu-12 predicted by AGN (left) and shock (right) models. For the shock models we only consider the flux of the broad component of Hu-12 estimated from the broad to narrow flux ratio of Pf-γ. Left: Reference AGN models (Z⊙ and nH = 103 cm−3) for the low (blue), mid (orange), and high (red) Eddington ratio SEDs. The shaded areas highlight the range of ratios predicted varying the metallicities (0.4–2 Z⊙) and nH (102–105 cm−3). Models with the same ionization parameter, U, are connected by dashed lines. Right: Reference shock models (ram pressure R = 106) for the 0.5 Z⊙ (green), Z⊙ (black), and 2 Z⊙ (purple) metallicities. The shaded areas highlight the range of ratios predicted varying R (104–108). Models with the same shock velocity, vs, are connected by dashed lines. The observed ratios use the same symbols as in Fig. 3. The models are described in Sect. 4 and Appendix D.

Finally, we created a grid of shock models using MAPPINGS V following Sutherland & Dopita (2017), which includes several improvements compared to the models in Allen et al. (2008; see Appendix D.3). The [Mg IV]/Hu-12broad ratio and the [Ar VI] 3σ upper limit are compatible with the emission produced by shocks with shock velocities vs = 100 − 130 km s−1 (Fig. 6 right). The vs is not directly equivalent to σcorr (e.g., Ho et al. 2014; Perna et al. 2020); however, the fact that they are similar, σcorr([Mg IV]) = 90 ± 25 km s−1 (Sect. 3.2), supports the idea that shocks are a possible origin of the [Mg IV] emission.

In summary, while we cannot completely reject the AGN origin in a very low log U environment based on these line ratios, the morphology and kinematic properties of the [Mg IV] emission (broader profile and velocity shifts) favor the shock origin. Under this assumption, the good correlation between the extended [Mg IV] emission and the recombination line Hu-12, which traces star-forming clumps, is naturally explained if the shocks producing the [Mg IV] emissions are powered by supernove (SNe). To estimate whether the energy released by SNe can produce the observed [Mg IV] luminosity relative to the star formation rate (i.e., [Mg IV]/Hu-12 ratio), we combined stellar population evolution models and the shock models. Shock models predict efficient [Mg IV] emission at vs ∼ 110–160 km s−1, so observing [Mg IV] emission at these velocities is also favored (Fig. D.3). Using these relations (see Appendix D.4), the predicted log [Mg IV]/Hu-12 ratio from a star-forming region with shocks produced by SNe is between −1.4 and −0.5. The observed log [Mg IV]/Hu-12 ratio, −0.29 ± 0.18, lies at the upper end of the range, but is consistent with it considering the relatively large uncertainties of the shock models (e.g., vs, Mg abundance) and the star formation models (e.g., initial mass function, energy per SN, electron temperature). Therefore, it is plausible that SNe can provide enough mechanical energy to generate the observed [Mg IV] emission.

5. Summary and conclusions

Using JWST/NIRSpec integral field spectroscopy, we investigated the extended high-ionization [Mg IV] 4.487 μm (IP 80 eV) emission in a sample of four local LIRGs, only one of them hosting a Seyfert-like AGN. Excluding the nucleus and outflow of this AGN, we find two reasons why shocks related to star formation are the most likely origin of the extended [Mg IV] emission. First, the [Mg IV] luminosity is well correlated with the recombination line Hu-12 4.376 μm, which traces star-forming clumps in these objects; instead, the [Ar VI] 4.530 μm line (IP 75 eV), which is common in the spectra of AGN, remains undetected. The second reason is that the [Mg IV] profile is broader (σcorr([Mg IV]) = 90 ± 25 km s−1) and shifted up to ±50 km s−1 relative to the recombination lines (σcorr(Hu-12) = 57 ± 15 km s−1). The [Mg IV] line kinematics actually track the faint wings of the lower ionization lines (Pf-γ 3.741 μm, [Fe II] 5.340 μm, and [Ar II] 6.985 μm) and resembles the large-scale rotating velocity field of the galaxies.

Supporting this interpretation, shock models with vs ∼ 100–130 km s−1 (i.e., similar to the σ of [Mg IV] and the broad component in other species) are consistent with the observed [Mg IV]/Hu-12 ratio and the [Ar VI] upper limit. In addition, the [Mg IV] luminosity is comparable to that expected from shocks associated with the mechanical energy released by SNe in these regions. Based on these results, the detection of [Mg IV] emission is expected in pure star-forming galaxies, at least in objects with high gas and star formation surface densities similar to those of local LIRGs (e.g., Sánchez-García et al. 2022).

Due to its high IP, [Mg IV] offers a unique view of the shocked gas by tracing the highly ionized phase, contrasting with classic shock tracers such as the [Fe II] transitions, which have a lower IP of 7.9 eV and are more easily ionized. Compared to the [O IV] 25.9 μm line (IP 55 eV), the higher sensitivity and angular and spectral resolutions of JWST at the shorter wavelength of the [Mg IV] line will be beneficial for future studies of stellar feedback caused by these widespread shocks in strong starbursts.

Acknowledgments

We thank the referee for their useful comments and suggestions. The authors acknowledge the GOALS ERS team for developing their observing program. MPS acknowledges support from grant RYC2021-033094-I funded by MICIU/AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR. IGB and DR acknowledge support from STFC through grants ST/S000488/1 and ST/W000903/1. AAH acknowledges support from grant PID2021-124665NB-I00 funded by the Spanish Ministry of Science and Innovation and the State Agency of Research MCIN/AEI/10.13039/501100011033 and ERDF A way of making Europe. This work is based on observations made with the NASA/ESA/CSA James Webb Space Telescope. The data were obtained from the Mikulski Archive for Space Telescopes at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS 5-03127 for JWST; and from the European JWST archive (eJWST) operated by the ESAC Science Data Centre (ESDC) of the European Space Agency. These observations are associated with program #1328.

References

  1. Allen, M. G., Groves, B. A., Dopita, M. A., Sutherland, R. S., & Kewley, L. J. 2008, ApJS, 178, 20 [Google Scholar]
  2. Alonso-Herrero, A., Pereira-Santaella, M., Rieke, G. H., & Rigopoulou, D. 2012, ApJ, 744, 2 [Google Scholar]
  3. Alonso-Herrero, A., Pereira-Santaella, M., Rieke, G. H., et al. 2013, ApJ, 765, 78 [NASA ADS] [CrossRef] [Google Scholar]
  4. Argyriou, I., Glasse, A., Law, D. R., et al. 2023, A&A, 675, A111 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Armus, L., Mazzarella, J. M., Evans, A. S., et al. 2009, PASP, 121, 559 [Google Scholar]
  6. Armus, L., Lai, T., Vivian, U., et al. 2023, ApJ, 942, L37 [NASA ADS] [CrossRef] [Google Scholar]
  7. Barcos-Muñoz, L., Leroy, A. K., Evans, A. S., et al. 2017, ApJ, 843, 117 [CrossRef] [Google Scholar]
  8. Bianchin, M., U, V., Song, Y., et al. 2024, ApJ, 965, 103 [NASA ADS] [CrossRef] [Google Scholar]
  9. Böker, T., Arribas, S., Lützgendorf, N., et al. 2022, A&A, 661, A82 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  10. Bushouse, H., Eisenhamer, J., Dencheva, N., et al. 2023, https://doi.org/10.5281/zenodo.7577320 [Google Scholar]
  11. Chatzikos, M., Bianchi, S., Camilloni, F., et al. 2023, Rev. Mex. Astron. Astrofis., 59, 327 [Google Scholar]
  12. Chiar, J. E., & Tielens, A. G. G. M. 2006, ApJ, 637, 774 [NASA ADS] [CrossRef] [Google Scholar]
  13. Corrales, L. R., García, J., Wilms, J., & Baganoff, F. 2016, MNRAS, 458, 1345 [Google Scholar]
  14. Dasyra, K. M., Ho, L. C., Netzer, H., et al. 2011, ApJ, 740, 94 [NASA ADS] [CrossRef] [Google Scholar]
  15. Del Zanna, G., Dere, K. P., Young, P. R., & Landi, E. 2021, ApJ, 909, 38 [NASA ADS] [CrossRef] [Google Scholar]
  16. Diamond-Stanic, A. M., Rieke, G. H., & Rigby, J. R. 2009, ApJ, 698, 623 [Google Scholar]
  17. Donnan, F. R., García-Bernete, I., Rigopoulou, D., et al. 2023, MNRAS, 519, 3691 [CrossRef] [Google Scholar]
  18. Eldridge, J. J., Stanway, E. R., Xiao, L., et al. 2017, PASA, 34, e058 [Google Scholar]
  19. Falstad, N., Aalto, S., König, S., et al. 2021, A&A, 649, A105 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Ferland, G. J., Done, C., Jin, C., Landt, H., & Ward, M. J. 2020, MNRAS, 494, 5917 [Google Scholar]
  21. García-Bernete, I., Ramos Almeida, C., Acosta-Pulido, J. A., et al. 2016, MNRAS, 463, 3531 [CrossRef] [Google Scholar]
  22. García-Bernete, I., Rigopoulou, D., Aalto, S., et al. 2022a, A&A, 663, A46 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  23. García-Bernete, I., Rigopoulou, D., Alonso-Herrero, A., et al. 2022b, MNRAS, 509, 4256 [Google Scholar]
  24. García-Bernete, I., Alonso-Herrero, A., Rigopoulou, D., et al. 2024a, A&A, 681, L7 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  25. García-Bernete, I., Pereira-Santaella, M., González-Alfonso, E., et al. 2024b, A&A, 682, L5 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  26. González-Alfonso, E., García-Bernete, I., Pereira-Santaella, M., et al. 2024c, A&A, 682, A182 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  27. Götberg, Y., de Mink, S. E., Groh, J. H., Leitherer, C., & Norman, C. 2019, A&A, 629, A134 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  28. Guillard, P., Boulanger, F., Pineau Des Forêts, G., & Appleton, P. N. 2009, A&A, 502, 515 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  29. Hayashi, T. J., Hagiwara, Y., & Imanishi, M. 2021, MNRAS, 504, 2675 [NASA ADS] [CrossRef] [Google Scholar]
  30. Ho, I. T., Kewley, L. J., Dopita, M. A., et al. 2014, MNRAS, 444, 3894 [Google Scholar]
  31. Inami, H., Surace, J., Armus, L., et al. 2022, ApJ, 940, L6 [NASA ADS] [CrossRef] [Google Scholar]
  32. Jakobsen, P., Ferruit, P., Alves de Oliveira, C., et al. 2022, A&A, 661, A80 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  33. Jin, C., Ward, M., & Done, C. 2012, MNRAS, 425, 907 [NASA ADS] [CrossRef] [Google Scholar]
  34. Koo, B.-C., Raymond, J. C., & Kim, H.-J. 2016, J. Kor. Astron. Soc., 49, 109 [NASA ADS] [CrossRef] [Google Scholar]
  35. Kroupa, P. 2001, MNRAS, 322, 231 [NASA ADS] [CrossRef] [Google Scholar]
  36. Lutz, D., Kunze, D., Spoon, H. W. W., & Thornley, M. D. 1998, A&A, 333, L75 [NASA ADS] [Google Scholar]
  37. Lykins, M. L., Ferland, G. J., Kisielius, R., et al. 2015, ApJ, 807, 118 [NASA ADS] [CrossRef] [Google Scholar]
  38. Mouri, H., Kawara, K., & Taniguchi, Y. 2000, ApJ, 528, 186 [Google Scholar]
  39. Nardini, E., Risaliti, G., Salvati, M., et al. 2009, MNRAS, 399, 1373 [Google Scholar]
  40. Ohyama, Y., Terashima, Y., & Sakamoto, K. 2015, ApJ, 805, 162 [Google Scholar]
  41. Osterbrock, D. E., & Martel, A. 1993, ApJ, 414, 552 [Google Scholar]
  42. Patton, D. R., Wilson, K. D., Metrow, C. J., et al. 2020, MNRAS, 494, 4969 [NASA ADS] [CrossRef] [Google Scholar]
  43. Pereira-Santaella, M., Alonso-Herrero, A., Rieke, G. H., et al. 2010a, ApJS, 188, 447 [NASA ADS] [CrossRef] [Google Scholar]
  44. Pereira-Santaella, M., Diamond-Stanic, A. M., Alonso-Herrero, A., & Rieke, G. H. 2010b, ApJ, 725, 2270 [Google Scholar]
  45. Pereira-Santaella, M., Spinoglio, L., van der Werf, P. P., & Piqueras López, J. 2014, A&A, 566, A49 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  46. Pereira-Santaella, M., Rigopoulou, D., Farrah, D., Lebouteiller, V., & Li, J. 2017, MNRAS, 470, 1218 [NASA ADS] [CrossRef] [Google Scholar]
  47. Pereira-Santaella, M., Colina, L., García-Burillo, S., et al. 2021, A&A, 651, A42 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  48. Pereira-Santaella, M., Álvarez-Márquez, J., García-Bernete, I., et al. 2022, A&A, 665, L11 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  49. Pereira-Santaella, M., González-Alfonso, E., García-Bernete, I., García-Burillo, S., & Rigopoulou, D. 2024, A&A, 681, A117 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  50. Pérez-Díaz, B., Pérez-Montero, E., Fernández-Ontiveros, J. A., & Vílchez, J. M. 2022, A&A, 666, A115 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  51. Perna, M., Arribas, S., Catalán-Torrecilla, C., et al. 2020, A&A, 643, A139 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  52. Petric, A. O., Armus, L., Howell, J., et al. 2011, ApJ, 730, 28 [Google Scholar]
  53. Rich, J., Aalto, S., Evans, A. S., et al. 2023, ApJ, 944, L50 [CrossRef] [Google Scholar]
  54. Rigby, J. R., & Rieke, G. H. 2004, ApJ, 606, 237 [NASA ADS] [CrossRef] [Google Scholar]
  55. Rogantini, D., Costantini, E., Zeegers, S. T., et al. 2020, A&A, 641, A149 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  56. Sánchez-García, M., Pereira-Santaella, M., García-Burillo, S., et al. 2022, A&A, 659, A102 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  57. Satyapal, S., Kamal, L., Cann, J. M., Secrest, N. J., & Abel, N. P. 2021, ApJ, 906, 35 [NASA ADS] [CrossRef] [Google Scholar]
  58. Spoon, H. W. W., Armus, L., Marshall, J. A., et al. 2009, ApJ, 693, 1223 [NASA ADS] [CrossRef] [Google Scholar]
  59. Stierwalt, S., Armus, L., Surace, J. A., et al. 2013, ApJS, 206, 1 [Google Scholar]
  60. Storey, P. J., & Hummer, D. G. 1995, MNRAS, 272, 41 [NASA ADS] [CrossRef] [Google Scholar]
  61. Sturm, E., Lutz, D., Verma, A., et al. 2002, A&A, 393, 821 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  62. Sutherland, R. S., & Dopita, M. A. 2017, ApJS, 229, 34 [Google Scholar]
  63. Thomas, A. D., Dopita, M. A., Kewley, L. J., et al. 2018, ApJ, 856, 89 [NASA ADS] [CrossRef] [Google Scholar]
  64. Veilleux, S., Rupke, D. S. N., Kim, D.-C., et al. 2009, ApJS, 182, 628 [Google Scholar]
  65. Vivian, U., Lai, T., Bianchin, M., et al. 2022, ApJ, 940, L5 [NASA ADS] [CrossRef] [Google Scholar]
  66. Weinberger, R., Springel, V., Pakmor, R., et al. 2018, MNRAS, 479, 4056 [Google Scholar]
  67. Wright, G. S., Rieke, G. H., Glasse, A., et al. 2023, PASP, 135, 048003 [NASA ADS] [CrossRef] [Google Scholar]
  68. Zapartas, E., de Mink, S. E., Izzard, R. G., et al. 2017, A&A, 601, A29 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]

Appendix A: Emission line maps

Figures A.1–A.4 show the emission maps for the complete sample.

thumbnail Fig. A.1.

Same as Fig. 2, but for VV 114 E.

thumbnail Fig. A.2.

Same as Fig. 2, but for NGC 3256 N.

thumbnail Fig. A.3.

Same as Fig. 2, but for II Zw 096.

thumbnail Fig. A.4.

Same as Fig. 2, but for NGC 7469.

Appendix B: Emission line profiles

Figure B.1 shows the emission line profiles for all the selected regions.

thumbnail Fig. B.1.

continued.

thumbnail Fig. B.1.

continued.

thumbnail Fig. B.1.

continued.

Appendix C: Relation between the luminosity and kinematics of [Mg IV] and H2 S(8), [Fe II], and [Ar II]

Figure C.1 shows the relation between the [Mg IV] and the H2 S(8) and [Fe II] 5.34 μm luminosities. We scaled the MRS spectra to match the NIRSpec continuum at ∼5 μm. The scaling factors are between 0.7 and 1.3. We did not correct for the differential extinction affecting these lines. Figure C.2 presents the comparison between the [Mg IV] line kinematics observed with JWST/NIRSpec and the longer wavelength lines ([Fe II] and [Ar II]) observed with JWST/MRS.

thumbnail Fig. C.1.

Luminosities of [Mg IV] vs. H2 S(8) (top) and vs. [Fe II] (bottom). The symbols are as in Fig. 3. The solid red line is the best fit (log y = −8.20 + 1.17log x) and the dashed red line the best linear fit (log y = ( − 1.56 ± 0.25)+log x).

thumbnail Fig. C.2.

Same as the bottom panel of Fig. 5, but for [Fe II] (top) and [Ar II] (bottom).

Appendix D: Star formation, AGN, and shock models

In this section we present a more detailed description of the grids of models used in Sect. 4. For the star formation and AGN photoionization models we used CLOUDY version 23.01 (Chatzikos et al. 2023) and mostly followed Pereira-Santaella et al. (2017) to create the grids. The main differences are the updated SEDs used for the incident radiation field (see below). For the shock models we used the MAPPINGS V code version 5.2.0 following Sutherland & Dopita (2017).

D.1. Star formation models

We used a constant density slab model. The incident radiation field is from the BPASS library (version 2.2; Eldridge et al. 2017). This library includes the effect of stripped-envelope stars in binary systems, which increase the hardness of the spectrum. We assumed an instantaneous burst of star formation with ages between 106 to 108 with 0.1 dex steps. We considered three stellar metallicities (0.4 Z⊙, Z⊙, and 2 Z⊙), and two initial mass functions (IMFs), both with an upper stellar mass limit of 300 M⊙: a standard Kroupa (2001) IMF and a top-heavy IMF with a power-law exponent of −2 between 0.5 and 300 M⊙. The initial gas phase abundance was matched to the stellar metallicity and the metals depleted using the ISM_CloudyClassic.dpl table. The gas-to-dust mass ratio were adjusted as in Pereira-Santaella et al. (2017). We varied the gas volume density, nH, between 102 and 104 cm−3 in 1 dex steps, and the ionization parameters between log U=−6.75 and −1.0 in 0.25 dex steps. The models were stopped when the temperature dropped below 1000 K or the H+ abundance was below 0.1%.

In order to trace the temporal evolution of the line ratios, we needed to define how log U varies with the age of the stellar population. log U is proportional to the rate of ionizing photons, Q(H). Thus, as in Rigby & Rieke (2004), we defined, for every combination of IMF and stellar metallicity, log U0 as the ionization parameter when Q(H) is maximum (Q(H)0; typically at ages of ∼1–2 Myr), and scaled log Uage relative to log U0 using the ratio Q(H)(age)/Q(H)0. Then, the line emission at each stellar age is interpolated from the grid using log Uage.

Figure D.1 shows the [Mg IV]/Hu-12 ratio as a function of stellar age for log U0 between −1.0 and −3.0. Only the nH = 102 cm−3 ratios are presented since the variation of this ratio in the explored nH range (102–104 cm−3) is small, < 30 %.

thumbnail Fig. D.1.

[Mg IV]/Hu-12 ratio for an instantaneous burst of star formation as a function of stellar age. The circles, diamonds, and stars indicate the ratios for 0.4 Z⊙, Z⊙, and 2 Z⊙. The filled (empty) symbols connected by a solid (dashed) line corresponds to a standard (top heavy) IMF. The grid with nH = 102 cm−3 is plotted. The color of the lines indicates the ionization parameter, U0.

D.2. AGN photoionization models

As we did for the star formation models (Sect. D.1), we assumed a constant density slab. For the incident radiation, we used the three SEDs derived by Jin et al. (2012) for low, medium, and high Eddington ratios. These SEDs represent a more realistic characterization of the SED of AGN than a power law (e.g., Ferland et al. 2020). We considered three gas phase abundances (0.4 Z⊙, Z⊙, and 2 Z⊙), which were depleted using the ISM_CloudyClassic.dpl table, and their gas-to-dust mass ratios were adjusted (see Sect. D.1). We varied nH between 102 and 105 cm−3 in 0.5 dex steps, and log U between −4.0 and −1.75 in 0.25 dex steps. Line ratios derived from these grids are shown in the left panel of Fig. 6.

D.3. Shock models

We used MAPPINGS V code version 5.2.0 to create a grid of shock models Sutherland & Dopita (2017). This grid includes several improvements compared to the models presented by Allen et al. (2008). For instance, the fully time-dependent solution for the photoionized precursor is computed for the first time (see Sutherland & Dopita 2017 for details).

The effect of dust is not taken into account by the MAPPINGS V shock code. This is justified for the high-temperature post-shock gas where dust grains can be effectively destroyed (see Allen et al. 2008). However, dust will be present in the photoionized precursor. This is particularly important since Mg is known to condense on silicate grains (e.g., Rogantini et al. 2020). As a first-order correction, we reduced the gas phase abundances of metals in the precursor using the default CLOUDY depletion factors (0.2 for Mg). We also updated the atomic parameters relevant for the emission lines of interest (e.g., [Mg IV] and [Ar VI]) to match those used in CLOUDY version 23.01. We used atomic data from the CHIANTI database version 10 (Del Zanna et al. 2021) and the Stout database from CLOUDY (Lykins et al. 2015).

We followed Sutherland & Dopita (2017) to create the grids varying four input parameters: the gas metallicity, Zgas; the shock velocity, vs; the ram pressure parameter, R=(nH/1 cm−3) × (vs/km s−1)2; and the magnetic to ram pressure ratio, , where B is the magnetic field and ρ the density. For Zgas we used 0.5 Z⊙, Z⊙, and 2 Z⊙, and for vs we used a logarithmic grid between 80 and 500 km s−1 with ∼0.04 dex steps. We used three pressures, R = 104, 106, and 108, which are equivalent to densities of nH = 1, 100, and 104 cm−3 at 100 km s−1. For the magnetic field, we used ηM = 10−4, 10−2, and 10−1, which correspond to the “standard”, “moderate”, and “strong” magnetic cases identified in Sutherland & Dopita (2017) and are equivalent to B = 5.4, 54, and 170 μG for R = 106. The standard magnetic case is presented in the right panel of Fig. 6 and the moderate and strong cases in Fig. D.2. The main effect of the magnetic field on the [Mg IV]/Hu-12 ratios is that that a higher vs is needed to obtain the same ratio because more energy is needed to compress the gas when the magnetic field is stronger.

thumbnail Fig. D.2.

Same as the right panel of Fig. 6, but for the moderate (left) and strong (right) magnetic field cases.

D.4. [Mg IV] luminosity from SNe

We estimated the expected [Mg IV] luminosity produced by SNe shocks for a constant star formation rate, traced by H recombination lines, by combining the shock models and the stellar population evolution models from the BPASS library. Assuming Case B conditions, the H recombination lines can be used to measure Q(H). For temperatures between 5 000 and 10 000 K, the Hα/Hu-12 ratio is 1250–1820 (Storey & Hummer 1995) and Q(H)(s−1) = (0.85 − 1.41)×1015L(Hu-12)(erg s−1).

According to the BPASS stellar population models, for metallicities between 0.4 Z⊙ and 2 Z⊙, and assuming a constant star formation rate, log(Q(H)(s−1)/ĖSN(erg s−1)) = 11.72–12.10, where ĖSN is the power released by SNe. This ratio is sensitive to the assumed energy released per SN event (1051 erg), and also to the IMF (e.g., Zapartas et al. 2017). For this calculation, we used the Kroupa (2001) IMF and a top-heavy IMF (see Sect. D.1).

For these metallicities, shock models (Sect. D.3) predict that [Mg IV] is emitted more efficiently at vs ∼110–160 km s−1. These vs are also comparable to the observed dispersion of the [Mg IV] line profiles. Therefore, in this vs range, log L([Mg IV])/Ėshock, where is the rate of mechanical energy flux across the shock, is between –4.25 and –4.05 (Fig. D.3). We note that this fraction is very small compared to the ∼0.2% shock energy that is released by some [Fe II] transitions (e.g., Mouri et al. 2000).

thumbnail Fig. D.3.

Ratio of the [Mg IV] luminosity to the rate of mechanical energy flux across the shock, Ėshock. The symbols are as in the right panel of Fig. 6.

Combining these three relations, matching the abundances of the stellar population and the shock models, and taking into account the Hu-12 emission from the shock (right panel of Fig.6), the predicted log L([Mg IV])/L(Hu-12) ratio in a region with constant star formation rate and shocks produced by SNe would be –1.4 and –0.5.

All Tables

Table 1.

Sample of local LIRGs.

In the text

All Figures

thumbnail Fig. 1.

Scaled and shifted JWST/NIRSpec 4.31–4.57 μm spectra of the selected regions (see Sect. 3.1 and Figs. 2 and A.1–A.4). The wavelength of the transitions tracing ionized gas (Hu-12), warm molecular gas (H2 0–0 S(10) 4.410 μm and 1–1 S(11) 4.417 μm), highly ionized gas ([Mg IV] and [Ar VI]), and the CO v = 1–0 band are indicated by the green, blue, red, and purple vertical lines, respectively.
In the text
thumbnail Fig. 2.

Line maps and velocity fields for NGC 3256 S. The remaining objects are in Figs. A.1–A.4. Top row from left to right: Line maps of Hu-12, H2 S(8), [Mg IV], and [Fe II]. The areas filled with purple crosses in the [Mg IV] panel correspond to regions where the CO v = 1–0 band is strong and complicates the modeling of the [Mg IV] line. The black cross indicates the position of the nucleus. Bottom row from left to right: Velocity fields from single Gaussian fits of Pf-γ, H2 S(8), [Mg IV], and [Fe II]. The circles in the first and third panel give the location and size of the selected regions. The units of the color scale are 10−15 erg cm−2 s−1 arcsec−1 for the line maps and km s−1 for the velocity fields.
In the text
thumbnail Fig. 3.

[Mg IV] vs. Hu-12 luminosities of the selected regions (top) and for every spaxel (bottom). Different colors and symbols are used to identify the regions of each object. In the top panel, the labels of the points are the names of the regions. The Spearman correlation coefficient, rs, is indicated. The solid and dashed red lines are the best fit (log y = 0.30 + 0.98log x) and best linear fit (log y = ( − 0.29 ± 0.18)+log x), respectively, to the luminosities of the regions.
In the text
thumbnail Fig. 4.

Observed profiles of the [Mg IV], Hu-12, Pf-γ, [Fe II], and [Ar II] emission lines (solid black line) for two selected regions. The remaining regions are shown in Fig. B.1. The filled gray Gaussian represents the instrumental spectral resolution. The blue line in the first panel is the best single Gaussian fit to the [Mg IV] profile. The [Mg IV] model is also plotted as a blue dashed line in the second and third panels, normalized to the peak of Hu-12 and the broad Pf-γ component, respectively, for reference. The two component fits for Pf-γ, [Fe II], and [Ar II], are represented by the pink (narrow component) and green (broad component) lines and the total model in red. For reference, the dashed lines indicate the rest velocity for each region derived from the narrow component of Pf-γ and the zero flux level. The observed (not corrected for instrumental resolution) σ and velocity shift derived from the Gaussian fits are indicated in each panel in km s−1.
In the text
thumbnail Fig. 5.

Comparison between the width, corrected for the instrumental resolution, of the [Mg IV] and Hu-12 lines (top) and between the velocity shifts of [Mg IV] and the broad Pf-γ component relative to the narrow Pf-γ component (bottom). The symbols are as in Fig. 3. The large blue star in the top panel represents the mean σcorr in each axis. The solid black line is the one-to-one relation. The shaded blue and red areas in the bottom panel indicate blue and red velocity shifts, respectively.
In the text
thumbnail Fig. 6.

[Ar VI]/[Mg IV] vs. [Mg IV]/Hu-12 predicted by AGN (left) and shock (right) models. For the shock models we only consider the flux of the broad component of Hu-12 estimated from the broad to narrow flux ratio of Pf-γ. Left: Reference AGN models (Z⊙ and nH = 103 cm−3) for the low (blue), mid (orange), and high (red) Eddington ratio SEDs. The shaded areas highlight the range of ratios predicted varying the metallicities (0.4–2 Z⊙) and nH (102–105 cm−3). Models with the same ionization parameter, U, are connected by dashed lines. Right: Reference shock models (ram pressure R = 106) for the 0.5 Z⊙ (green), Z⊙ (black), and 2 Z⊙ (purple) metallicities. The shaded areas highlight the range of ratios predicted varying R (104–108). Models with the same shock velocity, vs, are connected by dashed lines. The observed ratios use the same symbols as in Fig. 3. The models are described in Sect. 4 and Appendix D.
In the text
thumbnail Fig. A.1.

Same as Fig. 2, but for VV 114 E.
In the text
thumbnail Fig. A.2.

Same as Fig. 2, but for NGC 3256 N.
In the text
thumbnail Fig. A.3.

Same as Fig. 2, but for II Zw 096.
In the text
thumbnail Fig. A.4.

Same as Fig. 2, but for NGC 7469.
In the text
thumbnail Fig. B.1.

continued.
In the text
thumbnail Fig. B.1.

continued.
In the text
thumbnail Fig. B.1.

continued.
In the text
thumbnail Fig. C.1.

Luminosities of [Mg IV] vs. H2 S(8) (top) and vs. [Fe II] (bottom). The symbols are as in Fig. 3. The solid red line is the best fit (log y = −8.20 + 1.17log x) and the dashed red line the best linear fit (log y = ( − 1.56 ± 0.25)+log x).
In the text
thumbnail Fig. C.2.

Same as the bottom panel of Fig. 5, but for [Fe II] (top) and [Ar II] (bottom).
In the text
thumbnail Fig. D.1.

[Mg IV]/Hu-12 ratio for an instantaneous burst of star formation as a function of stellar age. The circles, diamonds, and stars indicate the ratios for 0.4 Z⊙, Z⊙, and 2 Z⊙. The filled (empty) symbols connected by a solid (dashed) line corresponds to a standard (top heavy) IMF. The grid with nH = 102 cm−3 is plotted. The color of the lines indicates the ionization parameter, U0.
In the text
thumbnail Fig. D.2.

Same as the right panel of Fig. 6, but for the moderate (left) and strong (right) magnetic field cases.
In the text
thumbnail Fig. D.3.

Ratio of the [Mg IV] luminosity to the rate of mechanical energy flux across the shock, Ėshock. The symbols are as in the right panel of Fig. 6.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.