The Astrophysical Journal, 576:L25–L28, 2002 September 1 䉷 2002. The American Astronomical Society. All rights reserved. Printed in U.S.A. A NEW HIGH-REDSHIFT Lya EMITTER: POSSIBLE SUPERWIND GALAXY AT z p 5.691 Masaru Ajiki,2 Yoshiaki Taniguchi,2 Takashi Murayama,2 Tohru Nagao,2 Sylvain Veilleux,3 Yasuhiro Shioya,2 Shinobu S. Fujita,2 Yuko Kakazu,4 Yutaka Komiyama,5 Sadanori Okamura,6,7 David B. Sanders,4 Shinki Oyabu,8 Kimiaki Kawara,8 Youichi Ohyama,5 Masanori Iye,9 Nobunari Kashikawa,9 Michitoshi Yoshida,10 Toshiyuki Sasaki,5 George Kosugi,5 Kentaro Aoki,8 Tadafumi Takata,5 Yoshihiko Saito,9 Koji S. Kawabata,9 Kazuhiro Sekiguchi,5 Kiichi Okita,5 Yasuhiro Shimizu,10 Motoko Inata,8 Noboru Ebizuka,11 Tomohiko Ozawa,12 Yasushi Yadoumaru,12 Hiroko Taguchi,13 Hiroyasu Ando,5 Tetsuo Nishimura,5 Masahiko Hayashi,5 Ryusuke Ogasawara,5 and Shin-ichi Ichikawa9 Received 2002 May 29; accepted 2002 July 19; published 2002 August 7 ABSTRACT During the course of our deep optical imaging survey for Lya emitters at z ≈ 5.7 in the field around the z p 5.74 quasar SDSSp J104433.04⫺012502.2, we found a candidate strong emission line source. Follow-up optical spectroscopy shows that the emission-line profile of this object is asymmetric, showing excess red wing emission. These properties are consistent with an identification of Lya emission at a redshift of z p 5.687 Ⳳ 0.002. The observed broad line width (Dv FWHM ⯝ 340 km s⫺1) and excess red wing emission also suggest that this object hosts a galactic superwind. Subject headings: galaxies: formation — galaxies: individual (LAE J1044⫺0130) — galaxies: starburst line profile shows a sharp blue cutoff and broad red wing emission, both of which are often observed in star-forming systems with prominent wind outflows. These features are also expected from radiative transfer in an expanding envelope. Therefore, Dawson et al. (2002) suggested that the Lya profile of J123649.2⫹621539 is consistent with a superwind with a velocity of ∼300 km s⫺1. Galactic superwinds are now considered to be one of the key issues for understanding the interaction and evolution of both galaxies and intergalactic matter (e.g., Heckman 1999; Taniguchi & Shioya 2001). In order to improve our knowledge of galactic superwinds at high redshift, a large sample of superwind candidates at z 1 3 is needed. During the course of our new search for Lya emitters at z ≈ 5.7, we found a candidate superwind galaxy at z p 5.69. In this Letter we report its observed properties. We adopt a flat universe with Q matter p 0.3, QL p 0.7, and h p 0.7, where h p H0 /(100 km s⫺1 Mpc⫺1) throughout the Letter. 1. INTRODUCTION Recent progress in deep optical imaging with 8–10 m class telescopes has enabled new searches for star-forming galaxies beyond z p 5. In particular, imaging surveys using narrow passband filters have proved to be a particularly efficient way to find such galaxies (Hu & McMahon 1996; Cowie & Hu 1998; Steidel et al. 2000; Kudritzki et al. 2000; Hu et al. 2002; Y. Taniguchi et al. 2002, in preparation). Indeed, the most distant Lya emitter known to date is HCM 6A at z p 6.56 (Hu et al. 2002), and more than a dozen Lya emitters beyond z p 5 have been discovered (Dey et al. 1998; Spinrad et al. 1998; Weymann et al. 1998; Hu, Cowie, & McMahon 1998; Hu, McMahon, & Cowie 1999; Hu et al. 2002; Dawson et al. 2001, 2002; Ellis et al. 2001), most by using this technique. One interesting object is J123649.2⫹621539 at z p 5.190, which was found serendipitously in the Hubble Deep Field– North flanking fields (Dawson et al. 2002). Its Lya emission- 2. OPTICAL DEEP IMAGING 1 Based on data collected at the Subaru Telescope, which is operated by the National Astronomical Observatory of Japan, and at the W. M. Keck Observatory. 2 Astronomical Institute, Graduate School of Science, Tohoku University, Aramaki, Aoba, Sendai 980-8578, Japan; ajiki@astr.tohoku.ac.jp. 3 Department of Astronomy, University of Maryland, College Park, MD 20742-2421. 4 Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, Honolulu, HI 96822. 5 Subaru Telescope, National Astronomical Observatory, 650 North A’ohoku Place, Hilo, HI 96720. 6 Department of Astronomy, Graduate School of Science, University of Tokyo, Tokyo 113-0033, Japan. 7 Research Center for the Early Universe, School of Science, University of Tokyo, Tokyo 113-0033, Japan. 8 Institute of Astronomy, Graduate School of Science, University of Tokyo, 2-21-1 Osawa, Mitaka, Tokyo 181-0015, Japan. 9 National Astronomical Observatory of Japan, 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan. 10 Okayama Astrophysical Observatory, National Astronomical Observatory, Kamogata-cho, Asakuchi-gun, Okayama 719-0232, Japan. 11 Communications Research Laboratory, 4-2-1 Nukui-Kitamachi, Koganei, Tokyo 184-8795, Japan. 12 Misato Observatory, 180 Matsugamine, Misato-cho, Amakusa-gun, Wakayama 640-1366, Japan. 13 Department of Astronomy and Earth Sciences, Tokyo Gakugei University, 4-1-1 Nukui-Kitamachi, Koganei, Tokyo 184-8501, Japan. We have carried out a very deep optical imaging survey for faint Lya emitters in the field surrounding the quasar SDSSp J104433.04⫺012502.2 at redshift 5.7414 (Fan et al. 2000; Djorgovski et al. 2001; Goodrich et al. 2001), using the prime-focus wide-field camera Suprime-Cam (Miyazaki et al. 1998) on the 8.2 m Subaru Telescope (Kaifu 1998) on Mauna Kea. SuprimeCam consists of 10 CCD chips of 2k # 4k and provides a very wide field of view: 34  # 27  (0⬙. 2 pixel⫺1). In this survey, we used the narrow passband filter, NB816, centered on 8160 Å with a passband of Dl FWHM p 120 Å; the central wavelength corresponds to a redshift of 5.72 for Lya emission. We also used broad passband filters B, RC, IC, and z⬘. A summary of the imaging observations is given in Table 1. All observations were done under photometric conditions, and the seeing was between 0⬙. 7 and 1⬙. 3 during the entire run. Photometric and spectrophotometric standard stars used in the flux calibration 14 The discovery redshift was z p 5.8 (Fan et al. 2000). Since, however, the subsequent optical spectroscopic observations suggested a bit lower redshift of z p 5.73 (Djorgovski et al. 2001) and z p 5.745 (Goodrich et al. 2001), we adopt z p 5.74 in this Letter. L25 L26 POSSIBLE SUPERWIND GALAXY AT z p 5.69 Vol. 576 TABLE 1 Journal of Imaging Observations Band B ............ RC . . . . . . . . . . . IC . . . . . . . . . . . . NB816 . . . . . . z . . . . . . . . . . . . Observation Date (UT) 2002 2002 2002 2002 2002 Feb Feb Feb Feb Feb 17 15, 16 15, 16 15–17 15, 16 Tinta (s) mlim(AB)b FWHMstarc (arcsec) 1680 4800 3360 36000 5160 27.1 26.8 26.2 26.6 25.4 1.2 1.4 1.2 0.9 1.2 a Total integration time. Limiting magnitude (3 j) within a 2⬙ aperture. c FWHM of stellar objects in the final image. b are SA101 for the B, R C, and IC data and GD50, GD108 (Oke 1990), and PG 1034⫹001 (Massey et al. 1996) for the NB816 data. The z  data were calibrated by using the magnitude of SDSSp J104433.04⫺012502.2 (Fan et al. 2000); since any quasar is a potentially variable object, the photometric calibration of the z  data may be more unreliable than those of the other band data. To avoid delays in obtaining follow-up spectroscopy, we analyzed only two of the CCD chips, one of which included the quasar SDSSp J104433.04⫺012502.2. The individual CCD data were reduced and combined using IRAF and the mosaicCCD data reduction software developed by Yagi et al. (2002). The total size of the reduced subfield is 11⬘. 67 # 11⬘. 67, corresponding to a total solid angle of ≈136 arcmin2. The volume probed by the NB816 imaging has (comoving) transverse di⫺1 mensions of 27.56 h⫺1 0.7 Mpc # 27.56 h 0.7 Mpc, and the FWHM half-power points of the filter correspond to a comoving depth along the line of sight of 44.34 h⫺1 0.7 Mpc (z min ≈ 5.663 and z max ≈ 5.762; note that the transmission curve of our NB816 filter has a Gaussian-like shape). Therefore, a total volume of 3 33,700 h⫺3 0.7 Mpc is probed in our NB816 image. Source detection and photometry were performed using SExtractor version 2.2.1 (Bertin & Arnouts 1996). Our detection limit (a 3 j detection with a 2⬙ diameter aperture) for each band is summarized in Table 1. For source detection in the NB816 image, we used the criterion that a source must be an object with at least a 13 pixel connection at the 5 j level. (Note that given the pixel resolution of 0⬙. 2 pixel⫺1, an extended source observed in ∼1⬙ seeing must be observed as a source with more than a 13 pixel connection.) Adopting the criterion for the NB816 excess (IC ⫺ NB816 1 1.0 mag), we have found a strong emission line source located at a(J2000.0) p 10h44m37s, d(J2000.0) p ⫺01⬚30⬘34⬙ (hereafter LAE J1044⫺0130). In this Letter we report on the this source. The sky position of LAE J1044⫺0130 is shown in Figure 1. The optical thumbnail images of LAE J1044⫺0130 are given in Figure 2. As shown in this figure, LAE J1044⫺0130 is clearly seen in only the NB816 image; the observed equivalent width is EWobs 1 310 Å. The NB816 image reveals that LAE J1044⫺ 0130 is spatially extended. It is interesting to note that this object shows not a circular shape but an irregular shape. The angular diameter is 2⬙. 3 (above the 2 j noise level). The size of the pointspread function in the NB816 image is 0⬙. 90. Correcting for this spread, we obtain an angular diameter of 2⬙. 1 for the object. In the cosmology adopted here, this corresponds to a diameter of d ⯝ 12.4 h⫺1 0.7 kpc. 3. OPTICAL SPECTROSCOPY In order to investigate the nature of LAE J1044⫺0130, we obtained optical spectroscopy. First, we used the Subaru Faint Fig. 1.—Finding chart of LAE J1044⫺0130 in our NB816 image. Field size is 2 # 2. North is up and east is to the left. Object Camera and Spectrograph (FOCAS; Kashikawa et al. 2000) with the lowest spectral resolution grism, 150 lines mm⫺1, blazed at l p 6500 Å together with an order-cut filter SY47 on 2002 March 11 (UT). The wavelength coverage was ∼4900– 9400 Å. The use of an 0⬙. 8 wide slit gave a spectroscopic resolution of R ∼ 400 at 8000 Å. The total integration time was 1800 s. The spectrum is shown in the upper panel of Figure 3. We detected a single emission line at l ≈ 8130 Å. This observation was done under photometric conditions, and we used the spectroscopic standard star, Feige 34 (Massey et al. 1988), to Fig. 2.—Thumbnail images of LAE J1044⫺0130 (upper panels), also displayed as contours (middle panels). Angular size of the circle in each panel corresponds to 8⬙. Lower panel shows the spectral energy distribution (in magnitude units). From our optical spectroscopy, we find that a galaxy located at 2⬙. 4 northeast of LAE J1044⫺0130 shows two emission lines at l ∼ 6710–6720 Å. These lines can be identified as the [O ii] l3727 doublet redshifted to z p 0.802 Ⳳ 0.002. This foreground galaxy might enhance the image of LAE J1044⫺0130 owing to gravitational lensing. However, since there is no counter image of the LAE J1044⫺0130 in our deep NB816 image, the magnification factor might be less than 2. No. 1, 2002 AJIKI ET AL. Fig. 3.—Optical spectra of LAE J1044⫺0130 obtained with FOCAS on Subaru. Upper panel: Spectrum (R ∼ 400) obtained on 2002 March 11. Lower panel: Spectrum (R ∼ 1000) obtained on 2002 March 13. calibrate the spectrum. The emission-line flux was calculated to be (1.5 Ⳳ 0.3) # 10⫺17 ergs s⫺1 cm⫺2. A second spectrum was obtained with FOCAS on 2002 March 13 (UT) in nonphotometric conditions. This time, we used the 300 lines mm⫺1 grating blazed at 7500 Å together with the same order-cut filter SY47, giving a spectral resolution of R ⯝ 1000 with the same 0⬙. 8 wide slit. This setting gave the same wavelength coverage as the lower resolution observation. Again the total integration time was 1800 s. The spectrum is shown in the lower panel of Figure 3. The emission-line profile marginally shows the sharp cutoff at wavelengths shortward of the line peak. However, the spectral resolution is insufficient to confirm this feature. To clarify the shape of the line profile seen in the FOCAS observations, additional observations were obtained with the Keck/Echelle Spectrograph and Imager (ESI; Sheinis et al. 2000) in echellette mode using a 1⬙ slit, which provided a spectral resolution of R ⯝ 3400 at l p 8000 Å. Two 1800 s integrations were obtained in photometric conditions on 2002 March 15 (UT). The spectra were calibrated using the spectroscopic standard stars Feige 34 and HZ 44 (Massey et al. 1988). The combined spectrum is shown in the middle panel of Figure 4. Again only one emission line, at l ≈ 8130 Å, is found in the ESI optical spectrum, which covers the wavelength range 4000–9500 Å. The emission-line profile at 8130 Å is confirmed to be slightly asymmetric, showing a cutoff at wavelengths shortward of the line peak. Although a single emission line in the wavelength range 4000–9500 Å may alternatively be identified as [O ii] l3727, the fact that the observed profile appears to be skewed suggests that this line is in fact Lya emission (e.g., Stern et al. 2000; Dawson et al. 2002). Therefore, we conclude that the emission L27 Fig. 4.—Optical spectrogram (upper panel) and one-dimensional spectrum (middle panel) of LAE J1044⫺0130 obtained with ESI on Keck II (R ∼ 3400); note that 5 pixel binning was applied to the one-dimensional spectrum. Model profile fit is shown by the thick solid curve (see text). Sky (OH airglow) emission lines are shown in the lower panel. line at 8129.3 Ⳳ 3.0 Å must be Lya, which then yields a redshift of 5.687 Ⳳ 0.002. The rest-frame equivalent width of the putative Lya emission is estimated to be EW0 1 46 Å. The observed Lya flux is (1.49 Ⳳ 0.33) # 10⫺17 ergs cm⫺2 s⫺1. This appears consistent with our FOCAS observations made on 2002 March 11. 4. RESULTS AND DISCUSSION 4.1. Star Formation and Superwind Activities The observed Lya flux is f(Lya) p (1.49 Ⳳ 0.33) # 10⫺17 ergs cm⫺2 s⫺1 based on the ESI spectrum. Given the cosmology adopted in this Letter, we obtain an absolute Lya luminosity of ⫺1 L(Lya) ⯝ (5.3 Ⳳ 1.2) # 10 42 h⫺2 0.7 ergs s . This Lya luminosity is comparable to those of other z 1 5 galaxies: (1) 3.3 # 1042 ergs s⫺1 for HCM 6A at z p 6.56 (Hu et al. 2002); (2) 6.1 # 10 42 ergs s⫺1 for SSA22-HCM1 at z p 5.74 (Hu et al. 1999); (3) 3.4 # 10 42 ergs s⫺1 for HDF 4-473.0 at z p 5.60 (Weymann et al. 1998); and (4) 8.5 # 10 42 ergs s⫺1 for J123649.2⫹621539.5 at z p 5.19 (Dawson et al. 2002). Note that all the above luminosities are estimated by using the same cosmology as that used here. We note that approximately half of the intrinsic Lya emission from LAE J1044⫺0130 could be absorbed by intergalactic atomic hydrogen (e.g., Dawson et al. 2002). In order to reproduce the observed Lya emission-line profile, a two-component profile fit was made using the following assumptions: (1) the intrinsic Lya emission-line profile is Gaussian, and (2) the optical depth of the Lya absorption increases with decreasing wavelength L28 POSSIBLE SUPERWIND GALAXY AT z p 5.69 shortward of the rest-frame Lya peak. The resulting fit is shown in Figure 4 (thick curve) and corresponds to the following emission and absorption line parameters: (1) Lya emission: line center l c, em p 8030.70 Å, line flux fem ⯝ 2.42 # 10⫺17 ergs s⫺1 cm⫺2, and line width FWHM em ⯝ 650 km s⫺1; and (2) Lya absorption: line center l c, abs p 8122.73 Å, optical depth at the absorption center tabs ⯝ 9.85, and line width FWHM abs ⯝ 175 km s⫺1. This analysis suggests that the total Lya emission-line flux amounts to 1.73 # 10⫺17 ergs s⫺1 cm⫺2. This is larger by a factor of 1.16 than the observed flux, giving a total Lya luminosity of ⫺1 L(Lya) ∼ 6.1 # 10 42 h⫺2 0.7 ergs s . We then estimate the star formation rate (SFR) of LAE J1044⫺0130. Using the relation SFR p 9.1 # 10⫺43L(Lya) M, yr⫺1 (Kennicutt 1998; Brocklehurst 1971) and the total Lya luminosity, we obtain SFR p ⫺1 5.6 Ⳳ 1.1 h⫺2 0.7 M, yr . Note that SFR may be overestimated because part of the Lya emission may arise from shock-heated gas if the superwind interpretation is applicable to this object. Although the signal-to-noise ratio of our ESI spectrum is not high enough to analyze the profile shape in great detail, the presence of the excess red wing emission seems secure (Fig. 4). The FWHM of the Lya emission is measured to be 340 Ⳳ 110 km s⫺1 and the full width at zero intensity is estimated to be 890 Ⳳ 110 km s⫺1. These properties are similar to those of the Lya emitter at z p 5.190, J123649.2⫹621539, found by Dawson et al. (2002). Vol. 576 4.2. Comments on Possible Association with the Quasar SDSSp J104433.04⫺012502.2 and the Lyman Limit System at z p 5.72 The observed redshift of LAE J1044⫺0130, z p 5.687, is close both to the quasar redshift z p 5.74 (see footnote 14) and to that of a Lyman limit system (LLS) at z LLS p 5.72 in the quasar spectrum reported by Fan et al. (2000). The redshift difference between LAE J1044⫺0130 and the quasar corresponds to the velocity difference of Dv ≈ 2370 km s⫺1, and that between LAE J1044⫺0130 and the LLS corresponds to Dv ≈ 1476 km s⫺1. The angular distance between LAE J1044⫺0130 and the quasar is approximately 330⬙, giving a comoving separation of ∼13 h⫺1 0.7 Mpc. This separation seems too large to identify LAE J1044⫺0130 as a counterpart of the LLS. It seems also unlikely that LAE J1044⫺0130 is associated with the large-scale structure in which the quasar SDSSp J104433.04⫺012502.2 resides. We would like to thank both the Subaru and Keck Telescope staff for their invaluable help. We would also like to thank the referee for useful comments. This work was financially supported in part by the Ministry of Education, Culture, Sports, Science, and Technology (grants 10044052 and 10304013). REFERENCES Bertin, E., & Arnouts, S. 1996, A&AS, 117, 393 Brocklehurst, M. 1971, MNRAS, 153, 471 Cowie, L. L., & Hu, E. M. 1998, AJ, 115, 1319 Dawson, S., Spinrad, H., Stern, D., Dey, A., van Breugel, W., de Vries, W., & Reuland, M. 2002, ApJ, 570, 92 Dawson, S., Stern, D., Bunker, A. J., Spinrad, H., & Dey, A. 2001, AJ, 122, 598 Dey, A., Spinrad, H., Stern, D., Graham, J. R., & Chaffee, F. H. 1998, ApJ, 498, L93 Djorgovski, S. G., Castro, S., Stern, D., & Mahabal, A. A. 2001, ApJ, 560, L5 Ellis, R., Santos, M. R., Kneib, J.-P., & Kuijken, K. 2001, ApJ, 560, L119 Fan, X., et al. 2000, AJ, 120, 1167 Goodrich, R. W., et al. 2001, ApJ, 561, L23 Heckman, T. M. 1999, in AIP Conf. Proc. 470, After the Dark Ages: When Galaxies Were Young (the Universe at 2 ! z ! 5), ed. S. S. Holt & E. P. Smith (Woodbury: AIP), 322 Hu, E. M., Cowie, L. L., & McMahon, R. G. 1998, ApJ, 502, L99 Hu, E. M., Cowie, L. L., McMahon, R. G., Capak, R., Iwamuro, F., Kneib, J.-P., Maihara, T., & Motohara, K. 2002, ApJ, 568, L75 Hu, E. M., & McMahon, R. G. 1996, Nature, 382, 231 Hu, E. M., McMahon, R. G., & Cowie, L. L. 1999, ApJ, 522, L9 Kaifu, N. 1998, Proc. SPIE, 3352, 14 Kashikawa, N., et al. 2000, Proc. SPIE, 4008, 104 Kennicutt, R. C., Jr. 1998, ARA&A, 36, 189 Kudritzki, R.-P., et al. 2000, ApJ, 536, 19 Massey, P., Strobel, K., Barnes, J. V., & Anderson, E. 1988, ApJ, 328, 315 Miyazaki, S., Sekiguchi, M., Imi, K., Okada, N., Nakata, F., & Komiyama, Y. 1998, Proc. SPIE, 3355, 363 Oke, J. B. 1990, AJ, 99, 1621 Sheinis, A. I., Miller, J. S., Bolte, M., & Sutin, B. 2000, Proc. SPIE, 4008, 522 Spinrad, H., Stern, D., Bunker, A., Dey, A., Lanzetta, K., Yahil, A., Pascarelle, S., & Fernández-Soto, A. 1998, AJ, 116, 2617 Steidel, C. C., Adelberger, K. L., Shapley, A. E., Pettini, M., Dickinson, M., & Giavalisco, M. 2000, ApJ, 532, 170 Stern, D., Bunker, A., Spinrad, H., & Dey, A. 2000, ApJ, 537, 73 Taniguchi, Y., & Shioya, Y. 2001, ApJ, 547, 146 Weymann, E. J., Stern, D., Bunker, A., Spinrad, H., Chaffee, F. H., Thompson, R. I., & Storrie-Lombardi, L. 1998, ApJ, 505, L95 Yagi, M., Kashikawa, N., Sekiguchi, M., Doi, M., Yasuda, N., Shimasaku, K., & Okamura, S. 2002, AJ, 123, 66