Starburst galaxies in filaments

Starburst galaxies in cluster-feeding filaments unveiled by Spitzer


We report the first direct detection with Spitzer of galaxy filaments. Using Spitzer and ancillary optical data, we have discovered two filamentary structures in the outskirts of the cluster Abell 1763. Both filaments point toward Abell 1770 which lies at the same redshift as Abell 1763 (), at a projected distance of 13 Mpc. The X-ray cluster emission is elongated along the same direction.

Most of the far-infrared emission is powered by star formation. According to the optical spectra, only one of the cluster members is classified as an active galactic nucleus. Star formation is clearly enhanced in galaxies along the filaments: the fraction of starburst galaxies in the filaments is more than twice than that in other cluster regions. We speculate that these filaments are feeding the cluster Abell 1763 by the infall of galaxies and galaxy groups. Evidence for one of these groups is provided by the analysis of galaxy kinematics in the central cluster region.

Subject headings:
galaxies: clusters: individual (Abell 1763) – galaxies: kinematics and dynamics – infrared: galaxies

1. Introduction

Thanks to large redshift surveys, it is presently well established that baryonic matter is arranged in the Universe in a network of voids and filamentary structures (e.g. de Lapparent, Geller & Huchra, 1986; Doroshkevich et al., 2001; Pimbblet, Drinkwater & Hawkrigg, 2004). Galaxy clusters are found at the intersections between filaments. Hierarchical models of galaxy formation (e.g. Benson, 2005) and observations (e.g. Zabludoff & Franx, 1993; Ceccarelli et al., 2005; Cortese et al., 2006) show that clusters constantly accrete small galaxy groups through filaments. These accreted galaxies undergo an accelerated evolution when they meet the denser cluster environment, experiencing a short starburst phase followed by the quenching of their star formation activity as their gas reservoirs are consumed and/or stripped off (e.g. Bekki, 1999; Balogh et al., 2000; Fujita, 2004). These events are traced by differences in morphology and star formation rate of galaxies inside and outside clusters (e.g. Dressler, 1980; Balogh et al., 1998; Goto et al., 2003; Kodama et al., 2004).

Much of what we have learned so far about the evolution of galaxies in clusters comes from detailed studies of the inner cluster regions, where the contrast with respect to the field is high, so that most spectroscopically targeted galaxies have a high probability of being cluster members. We are therefore limited to study the later stages of the environmental processes that drive galaxy evolution in clusters. We can follow the entire evolution of these galaxies, only observing the external regions where infalling field galaxies first encounter the cluster potential and the intracluster gas. Unfortunately, studies of this kind are extremely rare since they require wide-field images coupled with intensive spectroscopic follow-up observations (see, e.g., Porter & Raychaudhury, 2007).

Moreover, in order to better characterize galaxy properties through their spectral energy distributions (SED), multi-wavelength observations are needed. Until recently, most studies of the external regions of galaxy clusters have been based on optical/near-infrared data (see, e.g., Treu et al., 2003; Gerken et al., 2004; Rines et al., 2005). On the other hand, infrared studies of clusters have been limited to the inner regions (see the review of Metcalfe, Fadda & Biviano, 2005). With the advent of Spitzer (Werner et al., 2004), the outskirts of galaxy clusters have now started to be observed (Bai et al., 2006, 2007; Geach et al., 2006; Marcillac et al., 2007).

In this letter we report our discovery with Spitzer of two galaxy filaments in the outer region of the cluster Abell 1763 (at a redshift ). The two filaments are rich in actively star-forming galaxies, which are probably exhausting their gas supply while entering the virialized cluster environment. Throughout this paper we use km s Mpc, , .

2. Spitzer and Ancillary Observations

Abell 1763 was surveyed with the InfraRed Array Camera (IRAC Fazio et al., 2004) and the Multiband Imaging Photometer for Spitzer (MIPS Rieke et al., 2004) as part of a program to observe three medium-redshift clusters up to large distances from the center study the star formation activity in infalling galaxies. A field of 40’55’ was observed with MIPS at 24, 70, and 160 m. An overlapping field of 39’39’ was observed at 3.6, 4.5, 5.8, and 8.0 m with IRAC. The entire field is covered by the Sloan Digital Sky Survey (SDSS) in the and bands. To detect fainter objects in preparation for the spectroscopic follow-up observations, we obtained an image of the 24 m field with the Large Format Camera on the Palomar 200in telescope, reaching two magnitudes deeper than the SDSS image. We obtained 382 spectra with the Hydra spectrograph on the WIYN telescope at Kitt Peak for most of the sources brighter than 0.3 mJy at 24 m, corresponding to an average signal-to-noise ratio of 5, with optical counterparts brighter than . To obtain a better description of the velocity distribution of the cluster, we obtained additional 198 spectra mainly in the central region with the DOLORES multi-object spectrograph on the TNG telescope in La Palma, Spain. In this case, we targeted fainter galaxies with including faint infrared sources and early-type galaxies. Finally, to obtain estimates of the stellar mass of the galaxies as well as a better description of their SEDs, we obtained and images of most of the 24 m field with the Wide InfraRed Camera on the Palomar 200in telescope. These images are typically 3 mags deeper than the 2MASS (Skrutskie et al., 2006) images in this field. The data and their analysis will be presented in forthcoming papers.

3. Spatial and redshift distributions

The number of redshifts from our WIYN and TNG observations, as well as that from SDSS, are reported in Table 1. Cluster members were identified using the procedure developed by den Hartog & Katgert (1996).

Figure 1.— Spatial distribution of the 181 spectroscopically selected members of the cluster Abell 1763. Target selection was done exclusively with 24m sources in the region outside the 1.5 Mpc radius central region. In the inner region, early-type galaxies were targeted also to better describe the velocity distribution of the cluster. The dotted rectangle corresponds to the 24 m MIPS field. Starburst galaxies and AGN are marked with filled dots and triangles, respectively. Circles of 1.5 Mpc radius indicate the centers of Abell 1763 and Abell 1770, connected by a dashed line. The shaded regions have been traced by eye to highlight two filamentary structures. In these regions there is a clear excess of starburst galaxies.

The spatial distribution of the selected cluster members (Figure 1) presents a remarkable asymmetry outside ( Mpc), the radius at which the mean interior overdensity is 500 times the critical density. We computed using the relationship by Carlberg et al. (1997) between and velocity dispersion and a Navarro, Frenk & White (2005) profile with concentration to obtain from . The asymmetry is caused by the presence of two galaxy filaments oriented at and counterclockwise from the North direction, hereafter referred to as the Northern (N) and Southern (S) filaments, respectively. The S (and perhaps also the N) filament is directed toward the neighboring cluster Abell 1770, located at approximately the same redshift of Abell 1763 (), but Mpc away. The N and S filaments have different mean velocities: and  km/s, respectively, relative to the cluster mean velocity, suggesting that the two filaments are entering the cluster from opposite directions (see Figure 2).

Within , the filaments are no longer distinguishable from the inner cluster. However, the galaxy distribution in the inner region and the X-ray emitting intra-cluster gas are elongated along roughly the same direction of the two filaments. The X-ray image also shows an emission tail south-west of the center which leads Lima-Neto & Durret (2007) to suggest that Abell 1763 is undergoing accretion events. Moreover, the X-ray temperature map conserves traces of a previous merger from the south-east (Prokhorov & Durret, 2007). An ongoing accretion scenario is reinforced by the radio observations of the brightest cluster galaxy (BCG hereafter) which is a wide-angle tail source with bent radio emission (Lima-Neto & Durret, 2007). The bending is probably caused by the pressure of the intra-cluster gas as the BCG moves relative to it, heading toward Abell 1770. Although the direction of motion is opposite to what would be naively expected if the filaments were feeding the cluster, it can be explained as the result of a non-dissipative collision with an infalling group reaching the cluster from the filaments which displaced the BCG from its original location (the galaxy is now 0.1 Mpc off the X-ray peak emission).

Additional supporting evidence for an ongoing merger event in the cluster center comes from the analysis of the velocity distribution of the cluster member galaxies. In the inner region, the velocity distribution deviates from gaussianity (see Figure 2). The value of the kurtosis, , is significantly less than that expected for a Gaussian distribution (% confidence level). Such a platykurtic distribution can be interpreted as the superposition of two Gaussians with different means (see Figure 2). A similar decomposition is obtained also through a non-parametric adaptive-kernel technique (Fadda, Slezak & Bijaoui, 1998). The two systems differ in mean velocity by  km/s, and have velocity dispersions and  km/s. These correspond to X-ray temperatures and  keV, respectively, using the relationship (e.g. Girardi et al., 1996). These values are close to the observational estimates by Lima-Neto & Durret (2007),  keV for the main cluster and  keV for the detected south-west emission tail. It is therefore tempting to identify the two systems detected in X-ray with the two systems detected in velocity space.

Survey N N N Flux limits
Total Cluster Cluster
& 24m
WIYN 360 80 80 r’20.5, f0.3 mJy
TNG 157 101 34 r’21.5
SDSS 96 18 11 r’20.8
Total 581 181 106
Table 1Redshifts measured

4. Galaxy properties

We have fitted the SEDs of the 181 spectroscopically identified cluster members using a set of templates built with GRASIL (Silva et al., 1998) which includes elliptical, spiral, starburst, and post-starburst SEDs (Biviano et al., 2004). Using an altenative set of templates (Polletta et al., 2007) we obtained similar results. SEDs of active galactic nuclei (AGN) were not considered since only one cluster member, which lies on the S filament (see Figure 1), is an AGN according to its optical spectrum. Moreover, the sources with IRAC fluxes do not occupy the color diagram region where AGN are expected (Lacy et al., 2007) and none of the cluster members corresponds to point sources in the XMM image of the field.

Using the best-fit templates we have then determined the galaxy stellar masses () and star-formation rates (SFR). are determined from the rest-frame -band luminosities by adopting and 0.5, for early- and late-type templates, respectively (Lin, Mohr & Stanford, 2003). SFR are determined from total infrared luminosities using the relationship of Kennicutt (1998).

We define the ratio , where is the timescale over which the galaxy is assumed to form stars at the currently observed SFR. We take  Myr as typical durations of starburst episodes (Barton et al., 2000). High values of identify galaxies with elevated SFRs given their stellar mass. Hereafter, galaxies with are called starburst galaxies. Given that a galaxy gas content is at most 25% of its total baryonic mass (Young & Scoville, 1991), starburst galaxies will consume their entire gas content in the current starburst episode, if it lasts indeed Myr.

Starburst galaxies preferentially inhabit the two filaments (see Figure 1). This is confirmed by the following statistical analyses, which are performed on the subsample of cluster members with 24 m emission, in order to ensure completeness. A bidimensional Kolmogorov-Smirnov test (Fasano & Franceschini, 1987) indicates that the spatial distribution of starburst galaxies is indeed significantly different from that of other cluster members, (% confidence level). Moreover, the fraction of starburst galaxies is , , and in the filaments, the central region, and the outer region excluding the filaments, respectively. In the same regions, the average is , , and , respectively. The distributions of the values are nearly identical in the two filaments.

5. Discussion and conclusions

Using infrared Spitzer photometric data, complemented with optical/near-infrared photometric and spectroscopic data, we have discovered two galaxy filaments joining into the cluster Abell 1763. The 24 m selection has made the filaments stand out very clearly against the background and the dense cluster region. This is due to the enhanced star-forming activity among the filament galaxies, as most of the 24 m emission is of stellar origin (see also Geach et al., 2006; Marcillac et al., 2007).

Figure 2.— Rest-frame velocity distributions for different cluster regions. Upper panel: galaxies within 1.5 Mpc of the X-ray cluster center. The solid line represents a two-Gaussian fit to the velocity distribution, and the dashed lines represent the two separate components of the fit, with velocity dispersions of km s (left) and km s (right). The vertical arrow indicates the BCG velocity. Middle and lower panels: galaxies outside 1.5 Mpc. The shaded histogram shows the velocity distribution of galaxies outside the filaments. The grey line represents a fit to this velocity distribution with two Gaussians with velocity dispersions of km s (left) and km s (right). The solid line histograms represent the velocity distributions of galaxies in the N-filament (middle) and S-filament (lower). The solid lines represent robust Gaussian fits to the filament velocity distributions, with velocity dispersions of km s and km s (N-filament and S-filament, respectively). Errors on the velocity dispersions are 1-, based on 1000 bootstrap resamplings.

Both filaments extend in the direction toward the neighboring cluster Abell 1770, which is located Mpc away. The Abell 1763 intra-cluster gas is elongated in the same direction, as indicated by its X-ray morphology (Lima-Neto & Durret, 2007). The cluster is probably being fed by the two filaments along which galaxies and galaxy groups accrete onto the main cluster. Evidence for ongoing accretion is provided by the detection of substructures both in the X-ray emitting intra-cluster gas (Lima-Neto & Durret, 2007) and in the velocity distribution of cluster galaxies (see Figure 2). Since the spatial distribution of the galaxies identified as part of the low-velocity subcluster is not compact, the subcluster-cluster collision has probably occurred already, leading to the disruption of the colliding subcluster. This scenario could explain the displacement of the BCG from the cluster center, and its relative motion with respect to the intra-cluster gas (Lima-Neto & Durret, 2007).

Galaxies belonging to the two filaments have higher SFR, on average, than other cluster members, both those in the inner region and those outside but not in filaments. We therefore conclude that filaments are environments favorable to the onset of a galaxy starburst activity. Processes related to the dense intra-cluster gas are not effective in filaments, leaving gravitational interactions among galaxies as the most promising candidate for the stimulation of starbursts. In fact, the relatively high density of galaxies in filaments compared to the general field, and their relatively low velocity dispersion ( km/s) compared to the cluster enhances the tidal effect of galaxy encounters (Mamon, 1996), and hence the probability of an induced star-forming activity (Barnes & Hernquist, 1996).

Our results and conclusions find support in recent Spitzer investigations of other clusters. Marcillac et al. (2007) have found that the mid-infrared selected galaxies in a distant cluster () are associated with infalling galaxies. Bai et al. (2007) have suggested that the cluster environment is able to stimulate the star-formation activity in infalling field galaxies before they enter the cluster central regions where gas is stripped and star-formation suppressed. Our results are also consistent with those of Porter & Raychaudhury (2007), who have used optical data to discover an enhanced star-forming activity among galaxies associated with filaments in the nearby Pisces-Cetus supercluster. They also claim that the SFR in the filament galaxies peaks at 3–4 Mpc from the cluster center which is consistent with our findings. However, we could better assess this result by extending our coverage further to the East.

Since our results are based on the study of a single cluster, it would be hazardous to generalize them to the entire cluster population, as different clusters may follow different evolutionary paths. With the completion of our observational program, another two clusters at will be analysed to confirm that galaxy evolution is indeed speeding up in cluster-feeding filaments.

This work is based in part on observations made with Spitzer, a space telescope operated by the Jet Propulsion Laboratory, California Institute of Technology under a contract with NASA. Support for this work was provided by NASA through an award issued by JPL/Caltech. Funding for the SDSS and SDSS-II has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, the US Department of Energy, NASA, the Japanese Monbukagakusho, the Max Planck Society, and the Higher Education Funding Council for England. The SDSS is managed by the Astrophysical Research Consortium for the Participating Institutions (see list at Facilities: Spitzer (MIPS), Spitzer (IRAC), Palomar 200in (LFC), Palomar 200in (WIRC), WIYN (Hydra), TNG (Dolores).


  1. affiliation: Visiting Astronomer, Kitt Peak National Observatory, National Optical Astronomy Observatories, which is operated by the Association of Universities for Research in Astronomy, Inc. (AURA) under cooperative agreement with the National Science Foundation.
  2. affiliation: Visiting Astronomer, Italian Telescopio Nazionale Galileo (TNG) operated on the island of La Palma by the Fundación Galileo Galilei of the INAF (Istituto Nazionale di Astrofisica) at the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofisica de Canarias
  3. slugcomment: To appear in ApJ Letters


  1. Bai, L. et al. 2006, ApJ, 639, 827
  2. Bai, L. et al. 2007, ApJ, 664, 181
  3. Balogh, M. L. et al. 1998, ApJ, 504, L75
  4. Balogh, M. L., Navarro, J. F., & Morris, S. L. 2000, ApJ, 540, 113
  5. Barnes, J. E. & Hernquist, L. 1996, ApJ, 471, 115
  6. Barton, E. J., Geller, M. J., & Kenyon, S. J. 2000, ApJ, 530, 660
  7. Bekki, K. 1999, ApJ, 510, L15
  8. Benson, A. J. 2005, MNRAS, 358, 551
  9. Biviano, A., et al. 2004, A&A, 425, 33
  10. Carlberg, R. G., et al. 1997, ApJ, 485, L13
  11. Ceccarelli, M. L., et al. 2005, ApJ, 622, 853
  12. Cortese, L., et al. 2006, A&A, 453, 847
  13. de Lapparent, V., Geller, M. J., & Huchra, J. P. 1986, ApJ, 302, L1
  14. den Hartog, R. & Katgert, P. 1996, MNRAS, 279, 349
  15. Doroshkevich, A. G., et al. 2001, MNRAS, 322, 369
  16. Dressler, A. 1980, ApJ, 236, 351
  17. Fadda, D., Slezak, E., & Bijaoui, A. 1998, A&AS, 127, 335
  18. Fasano, G. & Franceschini, A. 1987, MNRAS, 225, 155
  19. Fazio, G. G., et al. 2004, ApJS, 154, 10
  20. Fujita, Y. 2004, PASJ, 56, 29
  21. Geach, J. E., et al. 2006, ApJ, 649, 661
  22. Gerken, B., et al. 2004, A&A, 421, 59
  23. Girardi, M., et al. 1996, ApJ, 457, 61
  24. Goto, T., Yamauchi, C., Fujita, Y., et al. 2003, MNRAS, 346, 601
  25. Kennicutt, Jr., R. C. 1998, ARA&A, 36, 189
  26. Kodama, T., et al. 2004, MNRAS, 354, 1103
  27. Lacy, M. et al. 2007, AJ, 133, 186
  28. Lima-Neto, G. & Durret, F. 2007, A&A, submitted
  29. Lin, Y.-T., Mohr, J. J., & Stanford, S. A. 2003, ApJ, 591, 749
  30. Mamon, G. 1996, in Third Paris Cosmology Colloquium, H. J. de Vega & N. S´anchez eds., p. 95, arXiv:astro-ph/9511101
  31. Marcillac, D., et al. M. 2007, ApJ, 654, 825
  32. Metcalfe, L., Fadda, D., & Biviano, A. 2005, Space Sci. Rev., 119, 425
  33. Navarro, J. F., Frenk, C. S., & White, S. D. M. 1997, ApJ, 490, 493
  34. Pimbblet, K. A., Drinkwater, M. J., & Hawkrigg, M. C. 2004, MNRAS, 354, L61
  35. Polletta, M., et al. 2007, ApJ, 663, 81
  36. Porter, S. C. & Raychaudhury, S. 2007, MNRAS, 375, 1409
  37. Prokhorov, D. A.& Durret, F. 2007, A&A, in press (preprint: arXiv:0708.2166)
  38. Rieke, G. H., et al. 2004, ApJS, 154, 25
  39. Rines, K., et al. 2005, AJ, 130, 1482
  40. Silva, L., et al. 1998, ApJ, 509, 103
  41. Skrutskie, M. F., et al. 2006, AJ, 131, 1163
  42. Treu, T., et al. 2003, ApJ, 591, 53
  43. Werner, M. W., et al. 2004, ApJS, 154, 1
  44. Young, J. S. & Scoville, N. Z. 1991, ARA&A, 29, 581
  45. Zabludoff, A. I. & Franx, M. 1993, AJ, 106, 1314
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minumum 40 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description