Extended Tidal Structure in Two Lyα-Emitting Starburst Galaxies

Evan D. Skillman (University of Minnesota), John M. Cannon (MPIA, Heidelberg), Daniel Kunth (Institut d’Astrophysique, Paris) ,

Claus Leitherer

(STScI), Miguel Mas-Hesse (Centro de Astrobiologia CSIC-INTA, Spain), Göran Östlin (Stockholm Observatory, Sweden), Artashes Petrosian (Byurakan

Astrophysical Observatory and Isaac Newton Institute of Chile, Aremian Branch, Armenia)

The Propagation of Ly

a

Photons

These two systems were chosen for imaging because they exhibit prominent Ly α emission (Leitherer et al. 1995, Giavalisco et al. 1996, Kunth et al. 1998, Leitherer et al. 2002). Hydrogen Ly a is one of the most important diagnostic emission lines in astrophysics. It is predicted to be luminous in star-forming galaxies (Charlot & Fall 1993), and could potentially be used as an indicator of star formation activity in distant galaxies where the line is shifted into the visual or near-infrared region.

However, this application is not straightforward, since the appearance of Ly α emission from starburst galaxies does not correlate well with the strength of the burst, the metallicity of the ionized gas or the dust content as measured by extinction. Rather, the geometry of the neutral gas and the presence of outflows appear to be important factors in determining the appearance of Ly a emission.

.

It is well known that due to high resonant scattering by neutral hydrogen in the ISM, Ly a photons can be attenuated by even small amounts of dust. It is then expected that only young, relatively dust-free galaxies should be prodigious sites of Ly a emission. Low-metallicity starburst galaxies in the local universe may be considered nearby analogs to such objects which are expected in greater numbers at higher redshifts.

Thus it was surprising that HST observations of the most metal-poor galaxy known, I Zw 18, showed only damped Ly a absorption and no emission (Kunth et al. 1994). In stark contrast, the more metal-rich starburst galaxy Haro 2 showed prominent Ly a emission (Lequeux et al. 1995). Further observations have revealed the importance of ISM structure in determining the escape fraction of Ly a photons (Giavalisco et al. 1996, Kunth et al.

1998).

The studies of Kunth et al. (1998), Tenorio-Tagle (1999), and Mas .

Hesse (2003) have elucidated the characteristics which appear to govern the propagation of Ly a photons in star-forming galaxies. If static, homogeneous neutral gas with column densities  10 18 cm -2 shields the ionized gas, no emission will be detected. The resonant scattering of the Ly a photons will lead to increased probability of destruction by any dust which is present. On the other hand, there may be diffuse Ly a emission which is detectable on sightlines not coincident with the sources of UV photons. Similarly, if the areal coverage of the neutral gas is not uniform but clumpy, some Ly a emission may be detectable on favorable sightlines.

Finally, if the velocity structure of the neutral gas is not static but rather outflowing from the ionizing regions (outflow velocities photons to the red of 1216 Å can escape and Ly significant. This explains the strong Ly a  a 200 km sec -1 ), Ly a emission may be emission detected in some starburst galaxies with complete spatial coverage by neutral gas which is also comparatively rich in both metals and dust.

.

Astrophysical Journal, 2004, 608, 768 Abstract

We present new VLA C-configuration H I imaging of the Ly

a-

emitting starburst galaxies Tol 1924-416 and IRAS 08339+6517. The effective resolution probes neutral gas structures larger than 4.7

kpc in Tol 1924-416, and larger than 8.1 kpc in IRAS 08339+6517.

Both systems are revealed to be tidally interacting: Tol 1924-416 with ESO 338-IG04B (6.6

΄ = 72 kpc minimum separation), and IRAS 08339+6517 with 2MASX J08380769+6508579 (2.4

΄ = 56 kpc minimum separation).

The H I emission is extended in these systems, with tidal tails and debris between the target galaxies and their companions.

Since Ly α emission has been detected from both of these primary systems, these observations suggest that the geometry of the ISM is one of the factors affecting the escape fraction of Ly α emission from starburst environments. This could be a very important factor to take into account when calculating the escape fraction of ionizing radiation from starbursts in the early universe.

Furthermore, these observations argue for the importance of interactions in triggering massive star formation events.

…………………………………………………………………….

Figure 1.

(a) DSS image of Tol 1924-416, overlaid with contours of the H I zeroth-moment image. Contours correspond to column densities of (7.3, 29, 51, 73) x 10 20 cm -2 .

Each galaxy is labeled; beam size is shown at bottom left. (b) Intensity-weighted velocity field of Tol 1924-416.

figure it is apparent that H I From this is being removed from one or both systems. Also, there remains a component of solid-body rotation within the optical extent of both galaxies. Beam size is labeled at lower left.

.

Starburst Triggering Mechanisms

. In some systems, regardless of total mass, gravitational interactions can initiate powerful starburst episodes. In Taylor (1997), it was found that H II galaxies (i.e., systems undergoing significant, concentrated star formation events) have more H I -rich companions detected at small separations than a similar sample of low surface brightness galaxies. This suggests that at least some bursts of star formation in low-mass systems are tidally triggered. Examples abound of higher-mass systems where interactions have initiated prolific star formation events (e.g., the ``Antennae'' galaxies) and have even produced new galactic systems in their own right (“tidal dwarf galaxies”). With these data we demonstrate the extended tidal structure of two starburst systems, providing strong evidence for triggered star formation episodes in these galaxies.

.

Figure 3.

DSS image of Tol 1924-416, overlaid with contours of H I the 3, 4.5, 6, 7.5 and 9 s emission at levels; this corresponds to column densities of (2.3, 3.4, 4.6, 5.7, 6.9) x 10 20 cm -2 , respectively. Each galaxy is labeled in the upper left plane; beam size is shown at bottom left of each frame, and heliocentric velocities are labeled in the upper right.

.

Distance (Mpc) Tol 1924-416

37.5

ESO 338-IG04B

37.5

IRAS 08339+6517

80

Figure 2.

(a) DSS image of IRAS 08339 + 6517, overlaid with contours of the H I zeroth-moment image. Contours correspond to column densities of (5.5, 15, 24, 33, 42, 51) x 10 20 cm -2 . Each galaxy is labeled; beam size is shown at bottom right. (b) Intensity-weighted velocity field of IRAS 08339 + 6517. From this figure it is apparent that H I is being removed from one or both systems.

The companion galaxy appears to retain a component of solid-body rotation in neutral gas; clear signs of rotation are less prominent in IRAS 08339 + 6517, however, suggesting that this interaction has completely disrupted the neutral gas in this system. Beam size is labeled at lower right.

.

Conclusions

VLA H I imaging of the starburst galaxies Tol 1924-416 and IRAS 08339 + 6517 has been presented. These two systems are remarkably similar in H I content, mass, and current evolutionary state. In each, we find extended neutral gas between the target and nearby neighbors, suggesting that interactions have played an important role in triggering the massive starbursts in the primary galaxies. The close proximity of the companions suggests that the interactions were recent, and the similar velocities of both primary and secondary galaxies argues that these systems may end up gravitationally bound.

.

…………………………………… Since both primary systems are intense Ly a emitters, these data support the interpretation that the ISM kinematics are an important mechanism that controls the escape of Ly a photons from starburst regions.

This has immediate implications for the use of the strength of Ly a emission in determining star formation rates, since the results will be dependent on the geometry of the ISM and not on properties inherent to the starburst being considered. Further H I observations of Ly a -emitting galaxies (and, conversely, of starburst systems with no apparent Ly a emission) are certainly warranted to further explore the role of the ISM in regulating the escape of Ly a photons from starburst environments.

….

2MASX J08380769+6508579

80

M

V

V

SYS

(km sec

-1

) HI Mass (M



) System Mass (M



) Tidal Fraction (%)

-19.57

-- (1.4±0.2) x 10

9

2830 (9.3±1.2) x 10

8

(4.2±0.5) x 10

9

~ 40 -20.35

-- 5750 (1.1±0.2) x 10

9

(5.6±0.7) x 10

9

~ 70 (7.0±0.9) x 10

8 Charlot & Fall 1993,

ApJ

, 415, 580 Giavalisco et al. 1996,

ApJ

, 466, 831 Heckman et al. 2001,

ApJ

, 558, 56 Kunth et al. 1994,

A&A

, 282, 709

Figure 4.

DSS image of IRAS 08339 + 6517, overlaid with contours of H I emission at the 3, 4.5, 6, 7.5 and 9 s levels; this corresponds to column densities of (6.9, 10.3, 13.8, 17.2, 20.7) x 10 20 cm -2 , respectively. Each galaxy is labeled in the upper left plane; beam size is shown at bottom right of each frame, and heliocentric velocities are labeled in the upper left.

.

Relevance for Early Universe Studies

. Currently it is thought that QSOs alone are not capable of reionizing the early universe (Madau et al. 1999). Starbursts then become a likely candidate for providing the necessary ionizing photons. However, studies of low-redshift starbursts indicate that only a small fraction of the ionizing radiation can escape from starbursts (Heckman et al. 2001). The present observation that Ly α emission escapes more easily from interacting systems implies that Lyman continuum photons may also escape more easily. Since the frequency of interactions is thought to be much higher in the early universe, this may imply that starbursts are more efficient at contributing to the reionization of the universe than inferred from observations of low redshift s tarburs t s ys tems .

.

Acknowledgements

The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. Support for this work was provided by NASA through grant number GO-9470 from the Space Telescope Science Institute, which is operated by AURA, Inc., under NASA contract NAS5-26555. J.M.C. was supported by NASA Graduate Student Researchers Program (GSRP) Fellowship NGT 5-50346. E.D.S. acknowledges partial support from NASA LTSARP grant NAG5-9221 and the University of Minnesota.

..

.

References

Kunth et al. 1998,

A&A

, 334, 11 Leitherer et al. 1995,

ApJL

, 454, L19 Leitherer et al. 2002,

ApJS

, 140, 303 Lequeux et al. 1995,

A&A

, 301, 18 Madau et al. 1999,

ApJ

, 514, 648 Mas-Hesse et al. 2003,

ApJ

, 598, 858 Taylor 1997,

ApJ

, 480, 524 Tenorio-Tagle et al. 1999,

MNRAS

, 309, 332