Constraints on the minor merging and star formation history of the Wolf–Rayet galaxy NGC 5430 through observations

Abstract

We used multi-wavelength analysis of the newly observed molecular gas [¹²CO and ¹³CO(1–0)] with interferometer CARMA and archival star formation tracers to constrain the interaction, merging, and star formation history of an off-center minor merger, a three-spiral barred galaxy NGC 5430 and its satellite embedded in the bar. Morphology of the molecular gas in the bar of NGC 5430 shows minimal signs of recent interactions with our resolution. The apparent morphological remnant of the past galaxy interaction is an asymmetric spiral arm, containing more molecular gas and exhibiting higher star formation rate (SFR) surface density than the two primary arms. Rotation curve analysis suggests that NGC 5430 collided with its satellite several Gyr ago. History of star formation was constrained by using SFRs that trace different timescales (infrared, radio continuum, and Hα). The collision occurred 5–10 Myr ago, triggering a transient off-center starburst of Wolf–Rayet stars at the eastern bar end. In the past, the global SFR during the Wolf–Rayet starburst peaked at 35 M_⊙ yr⁻¹. At present, the merger-driven starburst is rapidly decaying and the current global SFR has decreased to the Galactic value. The SFR will continue to decay as suggested by the present amount of dense gas [traced by HCN(1–0)]. Nonetheless, the global SFR is still dominated by the Wolf–Rayet region rather than the circumnuclear region. Compared with other barred galaxies, the circumnuclear region exhibits a particularly low dense gas fraction, low star formation activity, and high concentration of gas. Physical properties of the molecular gas are inferred by using the large velocity gradient calculations. The initial mass ratio of NGC 5430 to its satellite is suggested to be in an intermediate ratio range of 7:1–20:1.

1 Introduction

Hierarchical formation models suggest that galaxy–galaxy interactions play a key role in the formation, evolution, and morphology of galaxies (e.g., White & Rees 1978; Kazantzidis et al. 2008). Minor mergers, i.e., mergers of galaxies with large mass ratios, take place frequently in z < 1, and their rate is significantly higher than that of major mergers (mergers between galaxies of equal mass). While major mergers tend to form featureless massive elliptical galaxies, minor mergers are more relevant to disk galaxies, because they exert a relatively small effect on the primary disks (e.g., Mastropietro et al. 2005; Casteels et al. 2014). Minor mergers help to form a bulge and increase disk thickness in disk galaxies through deposition of satellite materials (Shioya et al. 1998; Aguerri et al. 2001; Eliche-Moral et al. 2006; Hopkins et al. 2010). These mergers also help to form galactic bars by triggering disk instability (Hernquist & Mihos 1995; Skibba et al. 2012), and probably trigger some low or intermediate luminosity active galactic nuclei (AGNs; Hopkins et al. 2008; Kazantzidis et al. 2008; Villalobos & Helmi 2008; Mapelli et al. 2015).

Galaxy–galaxy interactions and mergers trigger star formation. Observations suggest that galaxy–galaxy interactions increase star formation activity through violent compression of molecular gas (Lambas et al. 2003; Kauffmann et al. 2004; Lin et al. 2007; Smith et al. 2007; Woods & Geller 2007; Kaviraj 2014). Kaviraj (2014) suggested that about half of the star formation events in the local universe are triggered by minor merger processes. The notion of merger-induced star formation is also supported by theoretical studies (e.g., Cox et al. 2008; Lotz et al. 2010).

Observations suggest that off-center mergers are common and can occur repeatedly in a primary disk (e.g., Mandelker et al. 2014; Guo et al. 2015). The collision sites are shown as discrete star-forming regions in the disks of primary galaxies and contribute significant amounts of UV and optical light to the host galaxies (Wuyts et al. 2012; Guo et al. 2015).

In this work, we diagnosed the history of mergers and star formation events in a nearby off-center minor-merger system: NGC 5430 and its satellite galaxy HOLM 569 B.¹ The results can help to constrain the star formation properties and history of the high-redshift counterparts that often exhibit a higher rate of mergers but cannot be resolved in detail.

A Sloan digital sky survey (SDSS) gri composite image of the primary disk galaxy NGC 5430 is shown in figure 1. NGC 5430 (14^h00^m45|${^{\rm s}_{.}}$|7, +59°19^′42^″) is classified as SB(s)b, located at 42 Mpc, where 1^″ = 205 pc. NGC 5430 is a member of a galaxy group (Geller & Huchra 1983). The satellite dwarf galaxy HOLM 569 B (14^h00^m47|${^{\rm s}_{.}}$|3, +59°19^′27^″) is embedded in the eastern bar end, and is shown as a blue spot in figure 1, around 4.5 kpc (22^″) from the galaxy center (Keel 1982). A 2D Gaussian fit to its i band suggests that the diameter of HOLM 569 B is ∼ 1 kpc.

Fig. 1.

SDSS (Baillard et al. 2011) gri composite of NGC 5430 system taken from the NASA/IPAC Extragalactic Database. The white circle indicates the area of CARMA observations. The diameter of the circle is 80^″ or ∼ 16 kpc. (Color online)

The minor-merger system is manifested by the asymmetric spiral arms and the unusual star formation activity at the eastern bar end. NGC 5340 contains three spiral arms, including two primary arms and one asymmetric arm located in the north of the galactic nucleus, implying a morphology distortion caused by an external effect. The global infrared luminosity of NGC 5430 is not particularly high (∼ 7.5 × 10¹⁰ L_⊙; Sanders et al. 2003) and is inferior to the luminous infrared galaxy (LIRG; ≥10¹¹ L_⊙). However, the giant H ii region at the eastern bar end is unusually bright, with star formation rate equal to that of the entire Milky Way. Moreover, the giant H ii region deviates from the global H ii region luminosity function of the galaxy, suggesting that the bright core is a separate object (Keel 1982). The most massive stars are forming in the giant H ii region, including 10⁴–10⁵ OB stars and Wolf–Rayet stars (> 20 M_⊙) (Keel 1982; Fernandes et al. 2004). In the rest of this work, we refer to the eastern bar end (the position of HOLM 569 B and the surrounding 1 kpc-diameter area) as a Wolf–Rayet (W–R) region. The contribution of UV light and the size of the W–R region are similar to those of the giant star-forming clumps that are formed through minor mergers in the local and high-redshift universe (Elmegreen et al. 2007; Livermore et al. 2012; Guo et al. 2015).

The spatial distribution of molecular gas in NGC 5430 was unknown prior to this work, while single-point observations were performed in a few studies. Kruegel, Chini, and Steppe (1990) observed low density gas in ¹²CO(1–0), ¹²CO(2–1), and ¹³CO(2–1) in the galactic nucleus by using the IRAM 30 m telescope. Later, Contini et al. (1997) used the same telescope to measure dense gas tracers HCN(1–0) and CS(3–2), as well as to perform ¹²CO(2–1) observations toward the nucleus and the W–R region. No CS(3–2) emission was detected in either region, while HCN(1–0) was detected in the nucleus. The HCN–far infrared (FIR) luminosity relation for this nucleus is the same as for other nearby galaxies; however, the ¹²CO(2–1) and ¹²CO(1–0) intensities are unusually high relative to the FIR fluxes. The unusual nuclear CO intensities were reported by Contini et al. (1997) for two other Wolf–Rayet galaxies. ¹²CO(2–1) was detected in the W–R region. The intensity of ¹²CO(2–1) in the W–R region was 13 times weaker than in the nucleus.

This paper is structured as follows. The mapping observations of ¹²CO(1–0) and ¹³CO(1–0) are described in section 2. The results of these observations, including the images and morphological and kinematic analyses, are shown in section 3. In section 4, we present the physical properties of molecular gas and the history of star formation. In section 5, we discuss the possible initial mass ratio of NGC 5430 to its satellite galaxy. In section 6, we summarize our work.

2 Observation and data reduction

Molecular gas of NGC 5430 was mapped by using the combined array for research in millimeter-wave astronomy (CARMA). CARMA consists of six 10.4 m-diameter antennas and nine 6.1 m-diameter antennas. Observations were performed on 2013 July 30, August 1, and August 3 by using E configuration, with baselines in the 8–66 m range. Observations were performed toward the central 80^″ (diameter of ∼ 16 kpc) of the galaxy (white circle in figure 1). Two narrow bands (250 MHz) were centered at 115.27 GHz [¹²CO(1–0)] and 110.20 GHz [¹³CO(1–0)]. The total velocity width of a narrow band was ∼ 740 km⁻¹, with spectral resolution of ∼ 9 km s⁻¹. Wide bands (500 MHz) were placed between the narrow bands for the calibration of the continuum. Flux, passband, and phase calibrators were MWC 349, 1635+381, and 1642+689, respectively. The total observation time was ∼ 15 hr, including the target and the calibrators.

Calibration, imaging, and deconvolution were performed by using the standard procedures of the MIRIAD package (Sault et al. 1995). The final ¹²CO(1–0) and ¹³CO(1–0) clean maps have spatial resolutions of 7|${^{\prime\prime}_{.}}$|6 (1.5 kpc) × 5|${^{\prime\prime}_{.}}$|3 (1.1 kpc) with P.A. = −71|$_{.}^{\circ}$|8, and 8|${^{\prime\prime}_{.}}$|1 (1.7 kpc) × 5|${^{\prime\prime}_{.}}$|5 (1.1 kpc) with P.A. = −68|$_{.}^{\circ}$|0, respectively. With the velocity resolution binned to 12 km s⁻¹, the final spectral cube of ¹²CO(1–0) has the sensitivity of σ_RMS ≈ 11 mJy beam⁻¹. Because ¹³CO(1–0) is significantly weaker than ¹²CO(1–0), it has been binned to the lower spectral resolution of 25 km s⁻¹ to improve the signal-to-noise ratio (S/N). The final σ_RMS of ¹³CO(1–0) is 6 mJy beam⁻¹. The largest structure that can be detected in these observations is ∼ 82^″ or ∼ 17 kpc. Therefore, missing flux is negligible for this observation.

3 Results

3.1 Galactic morphology

3.1.1 Channel maps of CO

The channel map of ¹²CO(1–0) (hereafter ¹²CO) is delineated in figure 2 with white contours. The SDSS i-band image is also shown on the grayscale with black contours. The galactic center and the W–R region are marked with a plus sign. The ¹²CO exhibits a wide range of velocities, from 2730 to 3170 km s⁻¹. The strongest emission was generated at the galactic center (∼ 40σ_RMS or ∼ 0.44 Jy beam⁻¹). Two elongated structures emerge from the center, and were detected with a significance of ∼ 10 σ_RMS or ∼ 0.11 Jy beam⁻¹. The elongated structures correspond to the bar regions of the optical image. The W–R region was detected with a significance of ∼ 10 σ_RMS or ∼ 0.11 Jy beam⁻¹.

Fig. 2.

Channel map of ¹²CO(1–0) in white contours superimposed on the SDSS i-band image (Baillard et al. 2011) in black contours and on a grayscale. Velocity resolution of the data cube is 12 km s⁻¹. The figure presents every other channel to save the space. The contours of ¹²CO are 2, 3, 4, 5, 7, 10, 15, 20, and 30 σ_RMS, where 1 σ_RMS is 11 mJy. The plus signs denote the galactic center (upper one) and the W–R region (lower), respectively. Beam size of ¹²CO(1–0) is 7|${^{\prime\prime}_{.}}$|6 × 5|${^{\prime\prime}_{.}}$|3 (1.5 × 1.1 kpc) with P.A. = −71|$_{.}^{\circ}$|8, showing in the right-hand lower corner of each channel. The velocity of each channel is displayed in units of km s⁻¹ at the upper left.

The channel map of ¹²CO reveals features that are typical of the barred galaxy. The ¹²CO emission from the bar shifts toward the downstream of galactic rotation, known as an offset ridge of the bar. The offset ridges trace the shocks that are associated with the crowded gas orbits, as suggested by theoretical studies (Athanassoula 1992). The two offset ridges appear to be spatially and kinematically symmetric with respect to the galactic center. After passing through the offset ridge shocks, the gas moves inward and accumulates in the central region, yielding high gas concentration at the center (e.g., Kuno et al. 2007). This phenomenon is well known as bar-driven gas inflow (Sakamoto et al. 1999). The widths of velocity ranges of the galactic center, the eastern-, and western-offset ridges were ∼ 440, 180, and 160 km s⁻¹, respectively. ¹²CO was also detected in the asymmetric arm, emerging in the 2780 to 2860 km s⁻¹ range. The ¹²CO was observed in a spot of the northern primary arm with velocity in the 2854–2830 km s⁻¹ range and with a significance of ∼ 4 σ_RMS or 0.044 Jy beam⁻¹.

Figure 3 shows the channel map of ¹³CO(1–0) (hereafter ¹³CO). The keys are the same as in figure 2. ¹³CO emission was generated in the same velocity range as that of ¹²CO. The galactic center was solidly detected (∼ 14 σ_RMS or 0.084 Jy beam⁻¹). The eastern bar was detected with a significance of ∼ 8 σ_RMS or 0.048 Jy beam⁻¹, while the other side was barely detected with ∼ 3 σ_RMS or 0.018 Jy beam⁻¹. The ¹³CO of the asymmetric arm was observed in the velocity of ∼ 2800 km s⁻¹, consistent with the velocity of ¹²CO emission. We did not detect any ¹³CO from the two primary arms.

Fig. 3.

Channel map of ¹³CO(1–0) in white contours superimposed on the SDSS i-band image in black contours and on a grayscale. The symbols are the same as in figure 2. The contours are 2, 3, 4, 5, 7, and 10 σ_RMS, where 1 σ_RMS is 6 mJy. Beam size of ¹³CO(1–0) is 8|${^{\prime\prime}_{.}}$|1 × 5|${^{\prime\prime}_{.}}$|5 (1.7 × 1.1 kpc) with P.A. = −68|$_{.}^{\circ}$|0. Note that the channel velocity width of ¹³CO(1–0) is larger than that of ¹²CO(1–0). Velocity resolution of the ¹³CO(1–0) data cube is 25 km s⁻¹.

3.1.2 Integrated intensity maps of CO

Integrated intensity maps of ¹²CO and ¹³CO are shown in figure 4 with white contours, superimposed on the SDSS i-band image. The morphology of molecular gas at the center and throughout the bar appears not to be severely disturbed by the galactic mergers. Both ¹²CO and ¹³CO intensity maps contain a bright central component with a radius of ∼ 1.5 kpc. There is an offset between the strongest CO intensity and the galactic center. A reason for this is the enclosure of the galactic nucleus within a ringlike structure (García-Barreto et al. 1996); thus, the CO peak intensity position is determined from the convolution of the ring with our CO beam. Therefore, the CO peak is not necessarily at the galactic center if the gas distribution in the ring is not uniform. Offset ridges at the two sides of the bar are seen in ¹²CO, and are symmetric with respect to the galactic center. The offset ridges extend from the center to a distance of ∼ 5.6 kpc, and have a width of ∼ 1.2 kpc. The eastern ridge exhibits stronger ¹²CO intensity than the western ridge. Within this resolution, it cannot be determined whether the eastern ridge is, in general, stronger throughout the entire ridge or whether the W–R region, containing extra molecular gas contributed by the satellite, has been convolved with a large beam. ¹³CO is only detected in the eastern ridge. The length and width are consistent with those observed for ¹²CO. Because the two sides of the bar exhibit different star formation activity (stronger at the eastern ridge and weaker at the western ridge), their intrinsic ¹²C, ¹³C, ¹²CO, and ¹³CO abundances can be different. Both ¹²C and ¹³C can be enriched by the rapid cycling of interstellar medium (ISM) through stars. Once they are used to form ¹²CO and ¹³CO, the molecules can be destroyed by photo-dissociation through the UV photons from stars. The intrinsic ¹²CO/¹³CO abundance ratio then depends on the net result of these processes (e.g., Milam et al. 2005; Henkel et al. 2014; Szücs et al. 2014 and references therein). The observed line intensity ratio of ¹²CO/¹³CO also depends on the radiative transfer and on the excitation of molecular lines. Therefore, star formation activity and the ISM environment play a critical role in determining the observed ¹²CO/¹³CO ratio. The undetected ¹³CO of the western ridge may imply nonsimilar physical properties of molecular gas across the two sides of the bar.

Fig. 4.

(a) Integrated intensity map of ¹²CO(1–0) (white contours) superimposed on the SDSS i-band image (black contours and grayscale). Contour of ¹²CO(1–0) is plotted with intensity levels of 0.6, 0.9, 1.5, 1.9, 3, 4, 6, 8, 13, 20, 30, 40, 60, 80, and 100 Jy beam⁻¹ km s⁻¹. The galactic center and the W–R region are marked with a plus sign. The noise level of the intensity map is ∼ 0.2 Jy beam⁻¹ km s⁻¹. (b) Integrated intensity map of ¹³CO(1–0) (white contours) superimposed on the SDSS i-band image. Contour of ¹³CO is displayed with intensity levels of 0.3, 0.4, 0.9, 1.7, 2.0, 3.0, 4.5, 6.0, 8.0, 10, and 13 Jy beam⁻¹ km s⁻¹. The noise level is ∼ 0.1 Jy beam⁻¹ km s⁻¹.

The asymmetric spiral arm is connected to the eastern ridge in ¹²CO. The ¹²CO emission of the asymmetric spiral arm is spatially correlated with the stellar light traced in the SDSS image. The asymmetric spiral arm is more fragmented in ¹³CO, and is not connected to the eastern ridge. The detected ¹³CO is found in the regions with high ¹²CO intensity and stellar light.

3.1.3 Central concentration of molecular gas and bar strength

The total gas mass (H₂ + He) in the central ∼ 1 kpc of NGC 5430 is considerably higher than other strong bar (SB and SAB) galaxies. We estimated the mass of gas in the central ∼ 1 kpc (in diameter) using the flux of the beam (∼ 110 Jy beam⁻¹ km s⁻¹) at the galactic center, along with the Galactic CO-to-H₂ conversion factor (X_CO) of 2 × 10²⁰ cm⁻² (K km s⁻¹)⁻¹ (Bolatto et al. 2013), and a correction factor of 1.36 for He and heavy elements. Here we assumed that X_CO is constant and similar to that of the Milky Way disk across the whole of NGC 5430 when the gas properties are averaged over several 100 parsec to kiloparsec scale (e.g., Donovan Meyer et al. 2013; Bolatto et al. 2013). The derived total mass is ∼(8.3 ± 1.5) × 10⁸ M_⊙.² The value is considerably higher than the average value of galaxies with the same morphological type, ∼ 3.3 × 10⁸ M_⊙ (Sakamoto et al. 1999).³

Kuno et al. (2007) show that the central concentration of molecular gas is proportional to the bar strength because a strong bar exerts a large gravitational torque and would be most effective in driving gas inflows (see their figure 49). The central concentration of molecular gas (f_con) is defined as

\begin{equation} f_{\mathrm{con}}=\frac{M_{\mathrm{H_{2}}}(R_{\mathrm{K20}}/8)}{M_{\mathrm{H_{2}}}(R_{\mathrm{K20}}/2)}, \end{equation}

(1)

where R_K20 is the radius of 20 mag arcsec⁻² in the 2MASS K_S band, and |$M_{\mathrm{H_{2}}}\ (R_{\mathrm{K20}}/8)$| and |$M_{\mathrm{H_{2}}}\ (R_{\mathrm{K20}}/2)$| represent the gas masses enclosed within the radius of R_K20/8 and R_K20/2, respectively. By using equation (1), the f_con of NGC 5430 is estimated to be ∼ 0.34. The value of f_con is indeed higher that the average value (0.20) of barred galaxies in Kuno et al. (2007).⁴ Therefore, NGC 5430 likely has a stronger bar than the visually isolated galaxies in Kuno et al. (2007) in terms of f_con.

The strong bar is also implied by the high deprojected bar ellipticity. There is increasing evidence that bar morphology is determined primarily by the bar strength. Various studies, including both observations and simulations, have reported the existence of correlation between the bar strength and the bar ellipticity (Athanassoula 1992; Martinet & Friedli 1997; Abraham & Merrifield 2000; Block et al. 2001, 2004; Comerón et al. 2010; Kim et al. 2012). The Spitzer project of S⁴G (Sheth et al. 2010) has decomposed the galactic structure of NGC 5430 by using the GALFIT package and the Spitzer IRAC 3.6 μm image, by assuming a three-component bulge–disk–bar model (Salo et al. 2015). The derived bar ellipticity is ∼ 0.6. The bar ellipticity of NGC 5430 is again on the higher side of the observed and simulated bar ellipticity, which ranges from ∼ 0.2 to 0.8.

It is certainly possible that the formation of the strong bar in NGC 5430 is induced by past galaxy interactions. Simulations have shown that even a small perturbation, such as galaxy flyby and minor merger, can induce a bar (e.g., Noguchi 1987; Berentzen et al. 2004; Lang et al. 2014; Łokas et al. 2014).

3.2 Velocity fields

3.2.1 Velocity fields of CO and a comparison with Hα

Figures 5a and 5b show the velocity fields of ¹²CO and ¹³CO, respectively. With two plus signs mark the galactic center (upper plus) and the W–R region (lower plus). The receding and approaching sides are located in the south and north, respectively. The velocity fields of ¹²CO and ¹³CO are consistent with each other on the global scale.

Fig. 5.

Velocity fields of NGC 5430. The color scale of the velocity fields is shown within a range of 2700 to 3100 km s⁻¹. Contours of velocity fields are plotted in levels of 2720, 2770, 2820, 2870, 2920, 2970, 3020, and 3070 km s⁻¹ in all panels. The galactic center and the W–R region are marked with a plus sign in all panels as well. (a) Velocity field of ¹²CO(1–0) (color scale and contours). (b) Velocity field of ¹³CO(1–0) (color scale and contours). (c) Comparison of velocity fields of Hα (color scale and black contours) and ¹²CO (white contours). (d) Smoothed velocity field of Hα with velocity and spatial resolutions of ¹²CO field. (e) Residual velocity of the smoothed velocity field of Hα (panel d) minus the velocity field of ¹²CO (panel a). (Color online)

The detailed velocity distribution in the ¹³CO field is affected by the low sensitivity. For this reason, the kinematic discussion is focused on the ¹²CO field. The central region reveals parallel velocity contours in ¹²CO. However, velocity fields of disk galaxies usually exhibit either the “spider” diagram of circular motion or the “S-shape” twisted velocity gradient of noncircular motion. Our CO emission may be smeared out owing to the large beam size and low velocity resolution (12 km s⁻¹) at the galactic center, where the velocity field changes rapidly within a single beam. Therefore, neither the spider nor the S-shape pattern is seen. It is marginally observed from the ¹²CO map that the velocity gradient is almost normal to the CO offset ridges (namely, the contours are parallel to the ridges), implying a large velocity jump across the bar, which is predicted by theoretical studies (e.g., Athanassoula 1992). For ¹²CO, this velocity jump is ∼ 100 km s⁻¹ per 1 kpc.

Figure 5c compares the velocity field of ¹²CO (white contours) to the Hα from the GHASP survey (color scale and black contours). Spatial and velocity resolutions of the Hα field are ∼ 4^″ (820 pc) and 5 km s⁻¹, respectively. The CO field reveals the global variation in velocity that is observed in Hα. Because of the higher resolution, the velocity field of Hα more clearly reveals velocity structures at the galactic center. The galactic center reveals twisted S-shape velocity contours, which is evidence of noncircular motion of the bar (Epinat et al. 2008). This pattern is barely seen if the velocity field of Hα is smoothed out with the velocity and spatial resolutions of the CO maps (figure 5d). Figure 5e shows the velocity residual (color scale) of the smoothed Hα field with the CO field subtracted from the Hα field, superimposed on the emission boundary of the ¹²CO integrated intensity map. The majority of the residuals are within ±10 km s⁻¹, while some are as much as ±50 km s⁻¹. The regions with large residuals are located at around the edge of the CO emissions or are associated with weak CO emission. Thus, these large residuals are not significant for judging the difference between molecular and ionized gases.

Velocity dispersions of molecular and ionized gases appear to be more different. Figure 6a shows the velocity dispersion of the molecular gas, while panels (b) and (c) show the velocity dispersion of the ionized gas on a color scale. Velocity dispersion of the molecular gas reaches a peak of ∼ 90 km s⁻¹ at the galactic center, with an average of ∼ 65 km s⁻¹ within the central 1 kpc. In the bar region, large velocity dispersion appears at the leading side of the bar. The eastern bar exhibits a slightly larger dispersion than the western bar, ∼ 15–40 km s⁻¹ compared with ∼ 12–30 km s⁻¹. The asymmetric spiral shows a low velocity dispersion of 5–12 km s⁻¹. We found that this distribution of velocity dispersion is consistent with that observed in low-resolution ¹²CO and can be explained by a secular origin. The high velocity dispersion (> 50 km s⁻¹) at the galactic center results from the accumulation of molecular gas in the collision region where x₁ (bar) and x₂ (circumnuclear region) orbits intersect, and from the spatial superposition of circular and noncircular components (e.g., Hüttemeister et al. 2000; Schinnerer et al. 2000; Hsieh et al. 2011). The intermediate velocity dispersion (a few tens km s⁻¹) of the bar, peaking at the leading side on both sides of the bar, is due to the crowded orbits (e.g., Athanassoula 1992; Hüttemeister et al. 2000). The low velocity dispersion (< 15 km s⁻¹) in the asymmetric spiral arm is consistent with that of the unresolved giant molecular clouds (associations) in the galactic disks (e.g., Tosaki et al. 2007; Donovan Meyer et al. 2013; Faesi et al. 2014).

Fig. 6.

Velocity dispersions of NGC 5430. The galactic center and the W–R region are marked with a black plus sign in all panels. (a) Velocity dispersion (second moment) of ¹²CO(1–0). (b) Velocity dispersion of Hα (color scale) overlaid with the boundary of ¹²CO integrated intensity map (red contour). (c) Velocity dispersion of Hα (color scale) overlaid with the Hα velocity dispersion of 38 km s⁻¹ (red contour). Two white lines mark the locations of the intersection between x₁ and x₂ orbits (contact points). (Color online)

In contrast to the CO, velocity dispersion of the ionized gas exhibits stronger local variation because of the higher resolution. Figures 6b and 6c show the velocity dispersion of Hα using color scale. The CO boundary is overlaid with a red contour in figure 6b. Two regions with high velocity dispersion are observed in the central region, located at opposite sides of the galactic center. These regions correspond to the intersection of the x₁ and x₂ orbits, the “contact points” (indicated by two white lines in figure 6c). The velocity dispersion of Hα peaks at the northern intersection at 72 km s⁻¹, while the maximal velocity dispersion at the southern intersection is 64 km s⁻¹. The intersections are connected by two high-dispersion dust lanes of the bar. The maximal velocity dispersion at the dust lanes reaches ∼ 65 km s⁻¹. Beyond the dust lanes, the W–R region exhibits larger velocity dispersion than the opposite region in the western bar. The peak velocity dispersion in the W–R region is 61 km s⁻¹ with an average of 45 km s⁻¹ within the surrounding 1 kpc region. The large velocity dispersion of Hα compared to the CO is perhaps owing to the bias toward the diffuse ionized gas disturbed by supernova explosions and expanding nebulae around the massive populations, such as the W–R stars (Melnick et al. 1999; Moiseev & Lozinskaya 2012; Kam et al. 2015).

3.2.2 Position–velocity diagrams and rotation curve

Figure 7 compares the position–velocity (PV) diagrams of Hα (color scales, gray contours) and ¹²CO (black contours). The PV diagrams are obtained along the major axis of the velocity fields at 181|$_{.}^{\circ}$|9. The PV diagram of the ionized gas has the same resolution as that of the molecular gas has. Both PV diagrams are composed of a circumnuclear disk, with a steep increase in velocity within ±20^″ (∼ 4 kpc), and a strong emission at ∼−30^″ (∼ 6 kpc) and 2800 km s⁻¹ tracing the asymmetric spiral arm.

Fig. 7.

Position–velocity (PV) diagram of NGC 5430. The PV diagram is cut along the major axis of NGC 5430 at a position angle of 181|$_{.}^{\circ}$|9. The PV diagram of ¹²CO(1–0) is indicated with black contours. The PV diagram of Hα is shown on a color scale and by gray contours. The PV diagram of Hα has the same spatial and velocity resolutions as that of ¹²CO(1–0) has. The blue curve denotes the rotation curve traced by Hα created by Epinat et al. (2008). In order to compare the two sides of the galaxies, the reversed rotation curve with pink dots is overlaid. (Color online)

The blue curve in figure 7 represents the rotation curve of the NGC 5430 measured in the ionized gas by Epinat et al. (2008). In order to make a comparison between the receding and approaching sides, the reversed rotation curve is overlaid with pink dots. The rotation curve reaches a plateau within a few arcsec on both sides, fluctuating at about 3100 km s⁻¹ (+140 km s⁻¹ relative to the V_sys of 2960 km s⁻¹) and 2800 km s⁻¹ (−160 km s⁻¹). The two sides of the rotation curve are symmetric up to ∼±20^″ (∼ 4 kpc). This radius includes the central region and the majority of the bar. A small bump is observed on the receding side at the radius just beyond the W–R region, ∼ 30^″ (∼ 6 kpc), where the velocity at the receding side is ∼ 10 to 20 km s⁻¹ larger than that at the approaching side. The kinematic asymmetry around the W–R region might be owing to the inhomogeneous mass distribution caused by the extra gas contributed by the satellite collisions and the subsequent active star formation (note that the rotation curve is derived from Hα and therefore it is sensitive to the star formation history).

In terms of the rotation curve, NGC 5430 has almost relaxed and shows no signs of interaction in the inner disk. Overall, we found that the rotation curve beyond the central region (>|±1 kpc|) is flat, fluctuating within ∼ 60 km s⁻¹ on both sides. Moreover, the difference of asymmetry between the two sides is small (< 50 km s⁻¹). Simulations reveal that systems with ongoing minor galaxy interaction or encounters have nonflat rotation curves. Velocity increases with galactocentric radius by at least 100 km s⁻¹ in the galactic disk. Moreover, the rotation curves are asymmetric, differing by > 100 km s⁻¹ between the two sides of the disk (Kronberger et al. 2006). Therefore, the observed rotation curve of NGC 5430 implies that the system has almost relaxed, exhibiting no obvious signs of kinematic distortion.

The outer disk shows a larger degree of asymmetry than the inner disk. Beyond the asymmetric spiral arm (±40^″ or ∼ 8 kpc), the two sides of the rotation curve exhibit a larger deviation of ∼ 30 km s⁻¹. It is possible that the outer disk is not completely stabilized from the past interaction with HOLM 569 B. Moreover, because NGC 5430 is a member of a galaxy group (Geller & Huchra 1983), the perturbed outer disk can also result from ongoing interaction with other group members.

The kinematics of NGC 5430 help to constrain the timescale on which the two galaxies encounter. Both simulations and observations suggest that global kinematic disturbances of minor mergers fade within ∼ 1–2 Gyr after the first encounter of two galaxies, and the rotation curves exhibit no more severe distortions (e.g., Rubin et al. 1999; Dale et al. 2001; Kronberger et al. 2006). Thus, NGC 5430 likely captured HOLM 569 B about a few Gyr ago. However, the merger system is not yet fully settled, as three observational results suggest to us: by the large velocity dispersion of the ionized gas, inhomogeneous mass distribution at the collision site (the W–R region), and probably the asymmetric rotation curve at the outer disk as well.

4 Physical properties of molecular gas and star formation activity

4.1 Physical properties of molecular gas

4.1.1 Line ratios

Because of the different excitation conditions associated with each molecular line, the line ratio constrains the physical properties of the molecular gas. In addition to the ¹²CO(1–0)-¹³CO(1–0) ratio (R_1–0) a ¹²CO(2–1)-¹²CO(1–0) ratio (R_12CO) and ¹²CO(2–1)-¹³CO(2–1) ratio (R_2–1) are quoted by using the measurements reported in the literature. Table 1 shows the line intensities and ratios of the galactic center and the W–R region of NGC 5430. All values have been measured on a scale of 22^″ (in diameter), centering on the galactic center and the W–R region. The scale was chosen by the lowest resolution among all measurements, which is the IRAM 30 m single dish observation in ¹³CO(2–1).

Table 1.

Line intensities and ratios of the central (second row) and W–R regions (third row).*

12CO(1–0)	13CO(1–0)	12CO(2–1)	13CO(2–1)	HCN(1–0)	|$R_{\mathrm{1-0}} = \frac{^{12}\mathrm{CO\, (1-0)}}{\mathrm{^{13}CO\, (1-0)}}$|	|$R_{\mathrm{12CO}} = \frac{^{12}\mathrm{CO\, (2-1)}}{\mathrm{^{12}CO\, (1-0)}}$|	|$R_{\mathrm{2-1}} = \frac{^{12}\mathrm{CO\, (2-1)}}{\mathrm{^{13}CO\, (2-1)}}$|	Rdense=|$\frac{\mathrm{HCN\, (1-0)}}{\mathrm{^{12}CO\, (1-0)}}$|
[K km s−1]	[K km s−1]	[K km s−1]	[K km s−1]	[K km s−1]
117.4 ± 2.1	12.7 ± 0.1	81.6 ± 0.5	6.5	2.1 ± 0.3	10.3 ± 0.1	0.70 ± 0.01	12.6	0.018 ± 0.001
11.4 ± 0.3	1.9 ± 0.1	6.1 ± 0.5	…	< 0.8	6.0 ± 0.1	0.54 ± 0.01	…	< 0.070

*Measurements of ¹²CO and ¹³CO(1–0) are from this work with the CARMA telescope. ¹²CO(2–1) and HCN(1–0) are obtained from the IRAM 30 m telescope by Contini et al. (1997). ¹³CO(2–1) is measured using the IRAM 30 m telescope as well by Kruegel, Chini, and Steppe (1990). The listed numbers have a resolution of 22^″ in diameter. The original resolution of HCN(1–0) is 27^″. The total flux (in jansky) within the central 27^″ has been converted into the brightness of a 22^″ beam based on the assumption that the spatial size of dense gas cannot be larger than that of CO emission.

The line ratios of the galactic center of NGC 5430 are similar to those of the Galactic star-forming regions, starburst nucleus, and unresolved starburst galaxies, while the values for the W–R region are consistent with the galactic disk clouds. R_1–0 values of the galactic center and the W–R region of NGC 5430 are ∼ 10 and ∼ 6, respectively. A high R_1–0 (10–20) is observed in the Galactic star-forming regions, starburst nucleus, and unresolved starburst galaxies, while a low R_1–0 (∼ 6) occurs in the galactic disk, relatively quiescent molecular clouds (e.g., Aalto et al. 1995; Paglione et al. 2001; Simon et al. 2001; Tan et al. 2011; Papadopoulos et al. 2012). The high R_1–0 is explained by the change in opacity of the molecular gas owing to the high temperature, velocity dispersion, and column density of active star-forming regions and galaxies. R_12CO also increases with increasing star formation activity. The variation stems from the fact that the upper energy boundary (∼ 16.6 K) and the effective (including radiative trapping) critical density (∼ 230 cm⁻³) of ¹²CO(2–1) are slightly higher than the typical temperature (∼ 10 K) and density (∼ 100 cm⁻³) of molecular clouds. Therefore, R_12CO increases from ∼ 0.5 in quiescent molecular clouds to > 0.7 in dense and/or warm star-forming clouds (e.g., Chiar et al. 1994; Oka et al. 1998; Sawada et al. 2001; Koda et al. 2012). The line ratio in the W–R region (0.54) is similar to those of the quiescent clouds, while a higher ratio (0.7), close to that of star-forming regions and galaxies, is found for the central region. Finally, R_2–1 appears to increase from < 10 to 10–60 when moving from quiescent to star-forming regions and galaxies (e.g., Aalto et al. 1995; Sawada et al. 2001; Papadopoulos et al. 2012; Israel et al. 2015). The trend is similar to that of R_1–0. R_2–1 is available at the central region of NGC 5430. The measured value of 12.6 was in the range for star-forming regions and galaxies.

4.1.2 Large velocity gradient calculations

Gas temperature and volume density can be quantified by comparing the observed line ratios with those obtained by using one-zone large velocity gradient (LVG) models based on local photon trapping and the escape probability method (Goldreich & Kwan 1974; Scoville & Solomon 1974). In the LVG calculations, molecular excitation conditions depend on the local kinetic temperature (T_k), line opacity (column density of CO per unit velocity, N_CO/dv), and volume density (⁠|$n_{\mathrm{H_{2}}}$|⁠), as low-J CO is excited by the collisions with H₂. The CO–H₂ collisional cross-sections reported by Yang et al. (2010) were used in the present work.

The column density of CO per unit velocity was assumed by using our observations, and it is log (N_CO/dv) [cm⁻² (K km s⁻¹)⁻¹] = 16.6–17.2. Thus, the calculation was simplified to consider the variables of T_k and |$n_{\mathrm{H_{2}}}$|⁠. Assuming the width of the velocity range for giant molecular cloud associations to be 40–100 km s⁻¹, and assuming a surface density of ∼ 1000 M_⊙ pc⁻² and a CO-to-H₂ abundance of 8 × 10⁻⁵, the resultant log (N_CO/dv) is in the above range. We note that the following results of our LVG calculations were based on the above assumptions, and therefore should be treated with caution.

Figure 8 shows the results of our LVG calculations. Panels (a) and (b) show the results for the central and W–R regions, respectively. The results are shown as T_k (y-axis) versus |$n_{\mathrm{H_{2}}}$| (x-axis). The observed line ratios are shown as contours. The dashed, solid, and dot-dashed contours denote R_1–0, R_12CO, and R_2–1, respectively. R_2–1 is available for the central region only. For each line ratio, two contours are plotted, corresponding to log (N_CO/dv) = 16.6 in thin contour and 17.2 in thick contour. Solutions were found in the areas enclosed by all contours; these areas are marked in gray.

Fig. 8.

Result of LVG calculations in (a) the central region and (b) the W–R region. The dashed, solid, and dot-dashed contours represent the ratios of ¹²CO(1–0)/¹³CO(1–0) (R_1–0), ¹²CO(2–1)/¹²CO(1–0) (R_12CO), and ¹²CO(2–1)/¹³CO(2–1) (R_2–1), respectively. The ratio of R_2–1 is available for the central region only. For each ratio, two contours are drawn, representing log (N_CO/dv) = 16.6 (thin contour) and 17.2 (thick contour) cm⁻² (K km s⁻¹)⁻¹. Solutions of LVG calculations are highlighted by gray shadows.

The LVG calculations suggest that the bulk molecular gas is not heated by the compact starburst in the W–R region, whereas the bulk molecular gas is generally warm in the central region. Bulk molecular gas associated with the W–R region has a low temperature of ∼ 10 K, similar to that of the galactic disk molecular gas. Starburst is ongoing in the W–R region (subsection 4.2); thus, the temperature is expected to be high, presumably at or above ∼ 100 K (e.g., Weiß et al. 2001; Matsushita et al. 2010). The low calculated temperature implies that the off-center starburst is spatially small so that the bulk molecular gas seen in low-resolution observations (average over the scale of kpc) remains cold. The T_k in the central region is 10–40 K, warmer than the galactic disk molecular gas (∼ 10 K). Although the star-forming activity in the central region is not as drastic as that in the W–R region, the distribution of star-forming regions is more extended. Thus, the molecular gas is generally warm in the central region, yielding a high average T_k over the length of kpc.

The LVG calculations show that the density of bulk molecular gas traced by the CO lines is 200–1600 cm⁻³ in both regions. The gas density is consistent with the effective critical density of the CO lines.

4.2 Star formation activity

4.2.1 Star formation rate

Star formation history can be quantified by using star formation rates (SFRs) tracing different timescales. In this work, we used the SFRs estimated from the infrared [SFR(IR)], radio continuum [figure 9a, SFR(RC)], and Hα [figures 9b and 9c, SFR(Hα)] to constrain the star formation history. The three wavelengths respectively trace the average SFRs during the last ∼ 100 Myr, several tens of Myr, and several Myr, respectively (Kennicutt 1998a; Murphy et al. 2011; Calzetti 2013). All SFRs in this work were estimated based on the Kroupa initial mass function (IMF).

Fig. 9.

Star formation tracers of NGC 5430. Multiwavelength intensity maps are shown on color scale and in contours, where contours are plotted with levels of 10%, 15%, 40%, and 70% of the maximum intensity. Resolution of the observations is shown in the right-hand lower corner of each panel. (a) Radio continuum in 20 cm (1.4 GHz) observed by VLA (Condon et al. 1998). (b) Hα image from the GHASP survey (Epinat et al. 2008). (c) H ii regions (circles) identified by Brière, Cantin, and Spekkens (2012) overlaid on panel (b). The sizes of the circles denote the physical size of the H ii regions. (d) H ii regions (circles) of Brière, Cantin, and Spekkens (2012) overlaid on ¹²CO image. The sizes and colors of the circles indicate the physical size and star formation rate density (Σ_SFR) of H ii regions, respectively. The color denotes the following range: magenta, Σ_SFR > 0.05 M_⊙ yr⁻¹ kpc⁻²; yellow, 0.005 < Σ_SFR < 0.05 M_⊙ yr⁻¹ kpc⁻²; and cyan, Σ_SFR < 0.005 M_⊙ yr⁻¹ kpc⁻². (Color online)

Infrared observations suggest that the average SFR during the last ∼ 100 Myr is significantly higher than that of the Milky Way. Infrared emission originates from the dust heated by stars younger than 10⁸ yr; therefore, SFR(IR) traces a relatively long history of star formation. The luminosity-to-SFR relation, calibrated by Calzetti (2013), is

\begin{equation} \frac{\textit{SFR}}{[M_{\odot }\:\textrm {yr}^{-1}]}=2.84\times 10^{-44}\frac{L_{\mathrm{IR}}}{[\mathrm{erg\,s^{-1}}]}, \end{equation}

(2)

where L_IR is the infrared bolometric luminosity in the range of 5 to 1000 μm. The L_IR of NGC 5430 is 10^10.88 L_⊙ as obtained from IRAS, resulting in an SFR(IR) of ∼ 8 M_⊙ yr⁻¹, ∼ 8 times higher than that of the Milky Way.

The value of the SFR(RC) suggests that an instantaneous starburst occurred ∼ 10 Myr ago. The radio continuum at 1.4 GHz is mainly the nonthermal synchrotron radiation (∼ 90%) produced by relativistic electrons (S). These electrons are accelerated by the supernovae of massive stars. Therefore, the radio continuum traces the star formation history on the timescale corresponding to the lifetime of the lowest massive stars (8 M_⊙) that could become supernovae, which is ∼ 40 Myr. Condon, Cotton, and Broderick (2002) and Heesen et al. (2014) suggested a derivation of SFR based on the radio continuum:

\begin{equation} \frac{\textit{SFR}}{[M_{\odot }\:\mathrm{yr}^{-1}]}=0.75\times 10^{-21}\frac{L_{\mathrm{1.4\, GHz}}}{[\mathrm{W\, Hz^{-1}}]}, \end{equation}

(3)

where L_1.4 GHz is the luminosity at 1.4 GHz. A small fraction of the radio continuum originates from free–free emission (S_th). To subtract the thermal emission from the total flux, we use the following relation:

\begin{equation} \frac{S}{S_{\mathrm{th}}}=1+10\left( \frac{\nu }{\mathrm{GHz}} \right)^{0.1-\alpha }, \end{equation}

(4)

where α is the nonthermal spectral index and ν is the observed frequency. We adopt a typical value of α ≈ 0.8 (Condon et al. 2002). The total flux at 1.4 GHz measured by the VLA was 65.9 mJy (Condon et al. 1998). After subtracting S_th from S_1.4GHz, L_1.4GHz becomes 1.3 × 10²² W Hz⁻¹. The corresponding SFR is ∼ 10 M_⊙ yr⁻¹, one order of magnitude higher than that of the Milky Way. Among the global SFR(RC), ∼ 1 M_⊙ yr⁻¹ is contributed by the W–R region. The instantaneous starburst may be a result of the satellite collision. Simulations predict that the highest SFR occurs during the final merging phase of two galaxies and during the post-merging phase (Cox et al. 2008; Lotz et al. 2010).

The derived SFR(RC) is a lower limit owing to the mass loss of massive stars in the W–R region. Keel (1982) argued that the mass loss of massive stars in the W–R region is too extreme, so that only ∼ 4% of the massive stars could become supernovae and contribute to the radio continuum. If true, the SFR(RC) is considerably underestimated. If we naively scale the observed SFR(RC) by the fraction of stars that could not become supernovae, the real global SFR during the past ∼ 10 Myr will be ∼ 10 [the observed global SFR(RC)] + 1 [the observed SFR(RC) in the W–R region]/0.04 = 35 M_⊙ yr⁻¹, and ∼ 25 M_⊙ yr⁻¹ will be contributed by the W–R region.

In contrast to the past, the recent SFR (a few Myr) indicated by the total luminosity of the ionized gas is not particularly dramatic, comparable to that of the Milky Way. The Hα from the ionized gas traces the most recent star formation owing to the short lifetime of massive stars. Figure 9b shows the Hα image obtained from the Observatoire Haute-Provence 1.92 m-diameter telescope as part of the kinematical 3D Gassendi Hα survey of spirals survey (GHASP). The Hα luminosity (L_Hα) suggested by the survey data is 2.4 × 10⁴¹ erg s⁻¹ (Epinat et al. 2008). SFR(Hα) is derived from the luminosity–SFR calibration of Calzetti (2013):

\begin{equation} \frac{\textit{SFR}}{[{M_{\odot }\: \mathrm{yr}^{-1}}]}=5.5\times 10^{-42}\frac{L_\mathrm{H\alpha }}{[\mathrm{erg\, s^{-1}}]}. \end{equation}

(5)

The derived global SFR(Hα) is 1.3 M_⊙ yr⁻¹. This value suggests that the recent global SFR is comparable to that of the Milky Way.

The Galactic-level SFR is confirmed by adding up the local extinction-corrected SFR of individual H ii regions identified by other independent observations. Figure 9c shows the H ii regions identified by Brière, Cantin, and Spekkens (2012) by using the imaging Fourier transform spectrograph SpIOMM on the 1.6 m-diameter Ritchey–Chretien telescopes. The total SFR in the H ii regions is 2.4 M_⊙ yr⁻¹, including ∼ 0.6 M_⊙ yr⁻¹ from the central region and 1 M_⊙ yr⁻¹ from the W–R region. The values are higher than the SFRs suggested by GHASP observations, probably owing to the uncertainty of the indirect flux calibration in the GHASP survey (see Epinat et al. 2008 for the details). In spite of the difference, both global SFR(Hα)'s are significantly lower than the global SFR(IR) and SFR(RC). Moreover, the SFR in the W–R region decreases by more than 10 times between the actual SFR(RC) and SFR(Hα).

NGC 5430 is experiencing the end of the recent W–R starburst. Evolutionary stellar synthesis studies suggest that the age of the W–R region in NGC 5430 is ≥5.3 Myr (Fernandes et al. 2004). It is at the relatively late stage because W–R starbursts typically fade out within 10 Myr (Schaerer & Vacca 1998). Our derived SFRs also suggest that the star formation activity peaked around ∼ 35 M_⊙ yr⁻¹, and is rapidly decreasing to the Milky Way value (1–2 M_⊙ yr⁻¹). Moreover, theoretical models suggest that the ionized flux generated by massive stars decreases by 1–2 orders at the end of a W–R starburst (Schaerer & Vacca 1998). The difference between the SFRs of actual SFR(RC) and SFR(Hα) in the W–R region is consistent with this theoretical prediction.

Finally, if the W–R episode occurred immediately after the satellite collision, the collision time should be between ∼ 5.3 and 10 Myr ago because the W–R period (the time during which the ionized flux decreases to the pre starburst level or the duration of the appearance of the W–R stars) theoretically does not exceed 10 Myr (Schaerer & Vacca 1998).

4.2.2 Relation between the current star formation activity and molecular gas

In this sub-subsection, we compare star formation activity and molecular gas. Since the Hα flux is not spatially calibrated on the image of the GHASP survey (figure 9b) due to the strategy of the survey, we used the calibrated flux of H ii regions from Brière, Cantin, and Spekkens (2012) and the ¹²CO flux of the beam at the same positions for the analyses in this sub-subsection. The typical size of H ii regions is 1–2.5 kpc², comparable to our beam area of ∼ 1.5 kpc², but we suggest that the results still have to be treated with care.

We found that the ¹²CO distribution of NGC 5430 is correlated with star formation rate surface density. Figure 9d compares the H ii regions of Brière, Cantin, and Spekkens (2012) (circles) and our ¹²CO image (color scale). The physical areas of H ii regions are indicated by the circle sizes. To fairly compare the star formation activity, we normalized the SFRs of H ii regions by their areas, to obtain the quantity that is known as the star formation rate surface density (Σ_SFR), measured in units of M_⊙ yr⁻¹ kpc⁻². The Σ_SFR of the H ii regions is indicated with colored circles. Cyan, yellow, and magenta denote Σ_SFR below 0.005 M_⊙ yr⁻¹ kpc⁻², in the range of 0.005 to 0.05 M_⊙ yr⁻¹ kpc⁻², and above 0.05 M_⊙ yr⁻¹ kpc⁻², respectively. The ¹²CO emissions are mostly associated with moderate and high Σ_SFR (yellow and magenta circles). The missing ¹²CO in the low Σ_SFR regions can be explained by insufficient sensitivity, by assuming the ratio of Σ_SFR to ¹²CO flux to be the same in the moderate- and low-Σ_SFR H ii regions (in other words, the same star formation efficiency).

Current global star formation efficiency (SFE) of NGC 5430 is consistent with those of other disk galaxies. The current SFE is quantified as SFR(Hα)|$/M_{\mathrm{H_{2}}}$|⁠, where |$M_{\mathrm{H_{2}}}$| is the molecular gas traced by the ¹²CO. The total SFR(Hα) in the ¹²CO region is ∼ 2.2 M_⊙ yr⁻¹ obtained by using the measurements of Brière, Cantin, and Spekkens (2012). The total |$M_{\mathrm{H_{2}}}$| in our map is ∼ 4.7 × 10⁹ M_⊙, for the total flux of 5.7 × 10⁴ Jy beam⁻¹ km s⁻¹ and the CO-to-H₂ conversion factor of 2 × 10²⁰ cm⁻² (K km s⁻¹)⁻¹. The global SFE of the galaxy is therefore ∼ 4.6 × 10⁻¹⁰ yr⁻¹. The value is consistent with those of the nearby disk galaxies (Kennicutt 1998b; Bigiel et al. 2008; Huang & Kauffmann 2015).

The local SFE exhibits a large regional variation. The SFE at the galactic center is ∼ 9 × 10⁻¹⁰ yr⁻¹. The SFE is calculated by using the total SFR(Hα) of the two central H ii regions from Brière, Cantin, and Spekkens (2012) and the total ¹²CO flux of the beams associated with them. We found that the SFE is slightly lower compared with other circumnuclear regions, for which the SFEs are typically of the order of 10⁻⁹ yr⁻¹ (e.g., Kennicutt 1998b).

The SFE of the W–R region is 10 times higher than the global SFE and the SFE at the galactic center, and is ∼ 4.5 × 10⁻⁹ yr⁻¹, as obtained by using the SFR(Hα) of the giant H ii region at the W–R region and the ¹²CO flux of the beam associated with it. The large beam of CO observations may include molecular gas irrelevant to the starburst. Therefore, the derived SFE is a lower limit.

4.2.3 The externally triggered starburst at the bar end

Watanabe et al. (2011) suggested that the enhancement of star formation efficiency at the bar ends is due to the effective cloud–cloud collisions triggering the star formation. This questions whether HOLM 569 B is indispensably required to explain the W–R starburst at the eastern bar end. W–R stars do appear at the bar end of the Milky Way and are more concentrated toward the spiral arms, suggesting a possible promotion of massive star formation via galactic structures (e.g., Davies et al. 2012). The western bar end is lacking unusual star formation. Because the W–R starburst is transient, it is possible that it took place and terminated at the opposite side end. In other words, the two bar sides do not evolve simultaneously (Keel 1982). This is certainly a possible alternative.

Nonetheless, the W–R region exhibits several star formation features that do not favor an internal origin. The surface density of the W–R stars in the W–R region of NGC 5430 is more than 100 times higher than that of the Milky Way (Massey 2003; Mauerhan et al. 2011; Davies et al. 2012; de la Fuente et al. 2013). Moreover, the H ii region associated with the W–R region deviates from the global H ii regions in terms of the luminosity function, which is similar to Milky Way (Keel 1982). Such a distortion of the H ii region luminosity function is not observed in internally triggered extreme starburst environments such as circumnuclear starbursts (e.g., Kennicutt et al. 1989; Feinstein 1997; Alonso-Herrero & Knapen 2001). Finally, the star-forming region in the W–R region is particularly young (∼ 5 Myr) compared with the rest of the bar area, which is 10–12 Myr old. All these features suggest that the W–R region has likely been formed by a collision against HOLM 569 B.

4.2.4 Dense gas content and future star formation activity

Dense gas content (> 10⁴ cm⁻³) of molecular gas can be estimated in terms of dense gas fraction. Table 1 lists the intensity of dense gas HCN(1–0) and the ratio of HCN/¹²CO(1–0) (R_dense) in the central and W–R regions. R_dense increases from 0.01–0.06 in nonstarburst nuclei to 0.06–0.2 in starburst nuclei (Matsushita et al. 2010). The R_dense in the central region of NGC 5430 is within the range of non-starburst nuclei, 0.018. The value for the W–R region remains undetermined, < 0.070. R_dense is often used to infer the fraction of dense gas (f_dense) because the critical density of HCN(1–0) is as high as 10^4–5 cm⁻³. Together with the adopted X_CO and the HCN-to-H₂ conversion factor (X_HCN) of 1.3 × 10²¹ cm⁻² (K km s⁻¹)⁻¹ (Solomon et al. 1992), R_dense = 0.018 corresponds to f_dense as low as ∼ 12%. In the W–R region, f_dense is nearly four times higher than that in the central region (46%), by using the upper limit of 0.070 on R_dense.

The dense gas fraction in the central region of NGC 5430 is lower than that in the circumnuclear regions of other strong-barred galaxies (∼ 50%–100%, e.g., Kohno et al. 1999; Pan et al. 2013), implying that even though the strong bar is efficiently transporting molecular gas to the center, the majority of the gas does not become dense gas. The low density gas fraction explains the above-mentioned low SFE of the central region, compared with other circumnuclear regions. According to the relation of HCN luminosity to SFR that was suggested by Gao and Solomon (2004), the future SFR formed by this dense gas will be ∼ 0.4 M_⊙ yr⁻¹, comparable to the current SFR in the central region indicated by Hα.

The future SFR, arising from the dense gas in the W–R region, will be at most 0.15 M_⊙ yr⁻¹, nearly 10 times lower than the current SFR in the W–R region, which is ∼ 1 M_⊙ yr⁻¹. This is consistent with the previous discussion suggesting that the star formation activity in the W–R region is decaying and will continue to decrease in the future.

5 Initial mass ratio

We roughly estimate the initial mass ratio of NGC 5430 and its companion based on the observed morphology, velocity, and star formation properties. Major mergers can destroy galaxy disks and produce remnants with properties similar to those of observed giant ellipticals (but some studies have reported on the possibility of formation of a disk galaxy in a major merger (e.g., Springel & Hernquist 2005). We rule out this possibility since the candidate of companion has been suggested to be embedded in NGC 5430 (Keel 1982). Intermediate-mass mergers (4:1–7:1) tend to produce both ellipticals and hybrid systems, which are disk-like galaxies supported by velocity dispersion rather than rotation (Bournaud et al. 2004, 2005). The value of rotation velocity to velocity dispersion of these hybrid systems (V_rot/σ) is small, 0 < V_rot/σ ≤ 2 (Jog & Chitre 2002; Naab & Burkert 2003; Bournaud et al. 2004, 2005). The maximum Hα velocity dispersion of the galaxy is ∼ 70 km s⁻¹, and the deprojected Hα rotational velocity in the galactic disk is ∼ 300 km s⁻¹ (corrected with an inclination of 32° from Epinat et al. 2008), resulting in a V_rot/σ of ∼ 4. Hence, NGC 5430 is strongly rotationally supported as a regular disk galaxy. Therefore the initial mass ratio between NGC 5430 and its satellite is likely larger than 7:1. Simulations show that, for a primary galaxy with a stellar mass of NGC 5430 value,⁵ enhanced star formation only occurs in pairs with a mass ratio < 20:1 (Cox et al. 2008; Lotz et al. 2010). Therefore the initial mass ratio of NGC 5430 is likely between 7:1–20:1, and probably around 10:1 to generate a star formation timescale (inverse of SFE) of < 2 Gyr at the initial coalescence phase (Cox et al. 2008).

6 Summary

In this paper, we constrain the minor-merging and star formation history of the three-spiral Wolf–Rayet (W–R) galaxy NGC 5430 and its satellite HOLM 569 B embedding in the eastern bar of NGC 5430.

Molecular gas tracers ¹²CO(1–0) and ¹³CO(1–0) were observed by CARMA with a resolution of ∼ 1.3 kpc (section 2). The observed area is ∼ 16 kpc.

The main results are as follows (section 3):

• ¹²CO(1–0) is confidently detected in the central region, entire bar, and an asymmetric spiral arm (the third spiral) of NGC 5430. Solid detections of ¹³CO(1–0) are shown at the central region and the eastern bar with W–R region.

• The molecular bar shows the same offset ridges as seen in typical barred galaxies. The offset ridges of the two sides of the bar are symmetric around the galactic central region, but the eastern bar is stronger in both CO lines.

• NGC 5430 has a strong bar suggested by the high central concentration of molecular gas and the high ellipticity of the bar.

• The velocity field of ¹²CO(1–0) matches the velocity field and the position–velocity diagram of Hα, which was observed with higher spatial and velocity resolutions.

• The rotation curve of Hα shows no obvious distortion by galaxy merger. However, inhomogeneous mass distribution is suggested by a small bump of the rotation curve around the W–R region, implying that the merger system has nearly, but not fully, stabilized. The outer rotation curve is more asymmetric between the two sides of the galaxy either due to the past interaction with HOLM 569 B and/or to ongoing interaction with members in the galaxy group.

• Since the minor merger systems kinematically stabilize after 1–2 Gyr after the first encounter, NGC 5430 likely captures HOLM 569 B on this timescale.

Physical properties of molecular gas at the central region and the W–R region are inferred by the Large Velocity Gradient calculations (LVG) and the line ratios of ¹²CO(1–0)/¹³CO(1–0) (R₁₋₀), ¹²CO(2–1)/¹²CO(1–0) (R_12CO), and ¹²CO(2–1)/¹³CO(2–1) (R₂₋₁) (section 4). The R_12CO and R₂₋₁ ratios are cited from literatures. The ratio of R₂₋₁ is only available in the central region. The main results are as follows:

• The line ratios of the central region are R₁₋₀ ≈ 10, R_12CO ≈ 0.7, and R₂₋₁ ≈ 12.6. The ratios are in the range of starburst regions, but close to boundaries between the starburst region and the relatively quiescent molecular regions (e.g., the Galactic disk molecular gas). The ratios in the W–R region are consistent with the galactic disk molecular gas, with R₁₋₀ (∼ 6) and R_12CO (0.54).

• The temperature of the central region derived from the LVG calculations is 10–40 K, warmer than that of the W–R region, 10 K. These results imply that the W–R region is small and compact; thus, the bulk molecular gas measured by low-resolution observations includes a significant amount of low-temperature molecular gas irrelevant to the W–R starburst. In contrast, in spite of a lower degree of star formation, star formation is more spatially extended in the central region, leading to a higher derived temperature.

• Gas density is 200–1600 cm⁻³ in both regions, consistent with the critical density of CO lines.

Star formation history is constrained using star formation rates (SFRs) that trace different timescales, including infrared [SFR(IR)] for the average of the past hundred Myr, radio continuum [SFR(RC)] for the past twenty or thirty Myr, and Hα [SFR(Hα)] for the past two or three Myr (section 4). The main results are as follows:

• The observed global SFR(IR), SFR(RC), and SFR(Hα) are ∼ 8, 10, and 2.4 M_⊙ yr⁻¹, respectively. The observed SFR(RC) might be a lower limit due to the mass loss of massive stars. The realSFR(RC) is probably as high as ∼ 35 M_⊙ yr⁻¹. These SFRs suggest that there is an instantaneous starburst about twenty or thirty Myr ago. Together with the previous stellar synthesis study of the W–R region and the theoretical model of W–R starburst, we suggest that the starburst started about 5.3–10 Myr ago, likely triggered by the collision with the satellite.

• We found that the current star formation efficiency of the central region of NGC 5430 is low, compared with other galaxies. Moreover, the current dense gas fraction of the central region inferred by the ratio of HCN/¹²CO(1–0) is also low, implying a low SFR in the future as well.

• The current SFE of the W–R region is 10 times higher than that of the central region. However, SFR of the W–R region is rapidly decreasing by more than 10 times from the collision time to the present time. The future SFR implied by the amount of dense gas suggests that the SFR of the W–R region will continue to decrease.

We discussed a possible initial mass ratio of NGC 5430 and its satellite to be 7:1–20:1 based on the observed morphology, velocity, and star formation properties (section 5).

Finally, we note that the theoretical simulation is still necessary for constraining the merger history of the NGC 5430 system. Each system may have unique satellite orbits that would significantly affect their current morphology and star formation activity. Models of NGC 5430 and direct comparison of the multiwavelength observations will yield important constraints on this system. Future high resolution H i observations would be useful for revealing large-scale dynamic structures beyond the molecular and stellar disk and provide a knowledge of possible interaction orbits. Wide-field CO observations are also indispensable for revealing the gas distribution and the gas supply to star formation in detail, as well as the possible influence of galaxy collisions toward the central region, e.g., the asymmetric SFE at the contact point of the circumnuclear ring.

The authors thank the anonymous referee for the careful reading of the manuscript and comments. HAP and MU thank the instructors and the local organizing committee of the CARMA summer school 2013 for help with observations and the organization of the summer school. Authors thank Jin Koda for providing the code for LVG calculations. Support for CARMA construction was derived from the states of California, Illinois, and Maryland, the James S. McDonnell Foundation, the Gordon and Betty Moore Foundation, the Kenneth T. and Eileen L. Norris Foundation, the University of Chicago, the Associates of the California Institute of Technology, and the National Science Foundation. Ongoing CARMA development and operations are supported by the National Science Foundation under a cooperative agreement, and by the CARMA partner universities. This research has made use of the Fabry Perot database, operated at CeSAM/LAM, Marseille, France. This research has made use of the NASA/IPAC Extragalactic Database which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

1

In the Holmberg galaxy catalog of galaxy groups, NGC 5430 is HOLM 569.

2

The uncertainty is dominated by the uncertainty of flux calibration of CARMA observations.

3

Corrected for the adopted X_CO in this work.

4

Galaxies in Kuno et al. (2007) have distances of < 25 Mpc and an observing resolution of 15^″. NGC 5430 is farther than those galaxies by ∼ 2 times, but observed with a resolution ∼ 2 times finer. Thus these two studies sample the same galactocentric ranges in equation (1).

5

The stellar mass of NGC 5430 is ∼ 5.3 × 10¹⁰ M_⊙, from the Spitzer S⁴G project (Sheth et al. 2010).