arXiv:astro-ph/0505216v2 24 Jul 2005 butyt tee eln tteegso h galax- outer the the of of brightness (face-on edges surface region faint the disk the at principle, decline quite In change steeper ies. to a seemed to disks inner abruptly the sur- in in decline late-type, brightness exponential face mostly the that of found (1981a, spirals, Searle images edge-on & 2004b). photographic Kruit al. der using et van morphologi- 1981b), Pohlen and (1979) common see Kruit review, a der recent Van be a (for to feature thought cal are galaxies disk eyeg-nntr ae tdffiutt eemn fde- if determine truly to are difficult truncations it tected makes nature their However, edge-on very brightness. integration surface line-of-sight observed where the disks, increases edge-on in detect to eg,Bro hmsn19;Wie ta.2001; al. et a Weiner truncation as 1997; without traced the Thompson radii be & where Barton larger can to galaxies (e.g., profile exponential inclined brightness single there surface Wesson moderately 2002), & Grijs disk’s Kregel, de or Grijs, & de Kruit, Fig- face-on der 2001; late-type van Kregel, Pohlen (principally 3–5 disk. 2001; galaxies e.g., at inner edge-on truncations spirals; in the of truncation. lengths “classical” for prevalence a scale apparent such that about of the by than example Despite an steeper two shows 1 exponential, of ure de- an factor break the by truncation a that described the and best outside disks, brightness can is surface face-on truncations in in that cline identified showed clearly galaxies, be Sbc–Sc specific face-on disk. to the related in hard features be it dynamical makes might and also truncations morphological galaxies whether edge-on see in to (spirals, arms. etc.) features spiral rings, disk strong identifying bars, clearly e.g., of to, difficulty The due variations azimuthal to lcrncades [email protected] [email protected] pohlen@ia [email protected], [email protected], address: Electronic rpittpstuigL using typeset Letters Preprint ApJ by Accepted rnain nteselrpplto tteegsof edges the at population stellar the in Truncations olne l 20) sn eyde xoue fthree of exposures deep very using (2002), al. et Pohlen ue ih ssilpr fteds.Asbe fteepolsaei ga in are profiles these of subset headings: anti-truncations. intera A Subject disklike that the spiral suggesting disk. of visible spirals), the some one-armed include of or isophotes part (lopsided outer still isophotes sig the is not galaxies, light are bulge Sab–Sb surface-brightne outer which the the isophotes sharp of in outer rather extent crucially, disk; have suggestiv outer and, profiles profile the isophotes, the gala exponential either of disk S0–Sb — majority barred main zone of the the outer sample spheroidal than gradual larger fairly more rounder a a show of progressively te one-third 25% We are About least profile. anti-truncations. at disk of main in types the found than with are slope galaxies sc in S0–Sb they inner-disk shallower barred distinctly of (3–5 number are radius a profiles certain of profiles a brightness beyond radial sharply steepens profile h ik fsia aaisaecmol huh ob truncated: be to thought commonly are galaxies spiral of disks The 1. nttt eAto´sc eCnra,C i ´ce /,3 L´actea s/n, Via C/ Canarias, Astrof´ısica de de Instituto NITUCTO FDSSI AL-YEBRE GALAXIES BARRED EARLY-TYPE IN DISKS OF ANTI-TRUNCATION µ A INTRODUCTION T B E tl mltajv 6/22/04 v. emulateapj style X & aais tutr aais litcladlniua,c galaxies: — cD lenticular, and elliptical galaxies: — structure galaxies: 5 ae rnain easier truncations makes 25) radial ee ri,Jh .Bcmn n ihe Pohlen Michael and Beckman, E. John Erwin, Peter nntr,a opposed as nature, in c.es cetdb p Letters ApJ by Accepted ABSTRACT are ardS–bglxe rmteUpaaGlx Catalog Galaxy with Uppsala 1973) the (Nilson from galaxies S0–Sb barred 20) rey tcnit falnrhr ( northern al. all et of Erwin consists and (2005) it Erwin Briefly, in detail (2005). more in scribed in- recent of result the represent the may in teractions. so spirals) and still one-armed disk, strong but or outer show shallower lopsided galaxies these (e.g., with of asymmetries disk appear Some the they at profiles. of Instead, exponential dominating continuations cannot. bulge) halo be majority the or to the of envelope radii, outer extent larger round be outer can a the profiles from (perhaps these light (inner) of the as some of explained While projection outward profile. the exponential above light of cess profile brightness surface the becomes “anti-truncation”: of kind a least (at ex- an shows 1 Figure 2005); ample. Beckman & Pohlen, Erwin, 60 u h imtrlmtpoue isi ao flumi- of favor obviously against in is It bias biased galaxies. a high-surface-brightness produces and nous limit diameter galaxies. the 65 (see but of bars total without a galaxies had few we 2005), a galax- Erwin eliminating these After of is there classification ies.) since inconsistent however, Cluster, for did, Virgo evidence the (We no from galaxies galaxies. S0 1998), Cluster include Kenney are Virgo & tend excluded Cluster Koopman galaxies galaxies; we Sa Virgo Sc Virgo the field (e.g., resemble in to spirals spirals inconsistency field with for when compared evidence types is Hubble there here- in Because 1991, al. RC3). et Vaucouleurs after de from classifications and h vrl apeo hc hssuyi ae sde- is based is study this which on sample overall The subset significant a to attention draw we Letter this In h apei opeewti t eeto criteria, selection its within complete is sample The ◦ ,adheliocentric and ), 20L aua eeie Spain Tenerife, Laguna, La 8200 ∼ shallower stastost h hloe,outer shallower, the to transitions ss tosmyb epnil o tleast at for responsible be may ctions 2. 5)o ardS–bglxe hc exhibit which galaxies S0–Sb barred of 25%) rniinadotriohtswhich isophotes outer and transition APESLCINADDATA AND SELECTION SAMPLE l egh) eew rsn the present we Here lengths). ale h poiebhvo:terouter their behavior: opposite the icnl one hntemain the than rounder nificantly unbarred fads meddwti a within embedded disk a of e rasprt tla ao But halo. stellar separate a or rs hssget htthe that suggests This arms. aiswt smercouter asymmetric with laxies h ailsraebrightness surface radial the is hr r w distinct two are There xies. tlrerdi ota hr sa ex- an is there that so radii, large at D mtee“anti-truncations”; these rm 25 V ≥ aais u ic eincluded we since but galaxies; ≤ 2 . 00k s km 2000 0 ′ xsratio axis , spiral − 1 (measurements a/b > δ ≤ 2 . − ( 0 10 i ◦ . ), 2 Erwin et al.

is the opposite of a classical truncation — and happens at similar radii (3.2–6.0 inner scale lengths) — we dub them “anti-truncations”; by extending Freeman’s (1970) classification scheme we can also refer to them as “Type III” profiles. Because these are early-type galaxies (many of them S0), an obvious interpretation would be that the excess light at large radii is not from the disk, but rather from the luminous bulge or halo. This appears to be true for about one-third of the galaxies with anti-truncations (left-hand column of Figure 2) — all of them S0 — which have two characteristics in common. First, the inflection in the profiles is smooth and curved, which suggests that the outer light is from a separate component, added to the inner light from the disk. Second, the isophotes are elliptical in the inner region and become progressively rounder at larger radii (Figure 3a); this is clear evidence for an inclined disk embedded in a more spheroidal outer Fig. 1.— Truncated and untruncated surface-brightness profiles component. of barred galaxies: azimuthally averaged surface-brightness profiles But the majority of the galaxies (center and right- for NGC 4596 (SB0) and NGC 2273 (SBa), along with exponential fits (dashed lines). NGC 4596 shows a Type I exponential profile hand columns of Figure 2) have outer isophotes which extending to at least 5.7 scale lengths with no sign of a truncation; are not significantly rounder than the inner isophotes NGC 2273 shows a truncation at r ∼ 3.3 scale lengths. (Figure 3b). Morphologically at least, the excess outer light is still part of the disk. In several of the Sab and Sb galaxies (e.g., NGC 4319, 4699, 5740, 5806), sym- both strong and weakly barred galaxies (SB and SAB metric or asymmetric spiral arms are clearly visible in classifications from RC3), the result is representative of the region outside the transition, which again argues for the majority of local, early-type disk galaxies. the excess outer light being part of the disk. Finally, in We obtained azimuthally averaged, R-band surface most of these profiles the transition is quite sharp, so we brightness profiles of all the objects. Details of the obser- are probably not seeing the addition of light from two vations, calibrations, and reduction procedures are given overlapping components, but rather a single disk with in Erwin et al. (2005). Particular attention was paid to a fairly abrupt change in its density profile. The outer ensuring flat and accurately subtracted sky backgrounds, part of the profile is usually quite exponential; we show with the halos of bright stars and neighboring galax- exponential fits to the outer profiles in Figure 2. ies carefully masked out. The orientation of the outer Could we be seeing thick disks, which are exponen- disk (position angle and projected ellpiticity) was de- tial but with larger scale lengths than the brighter thin rived from free-ellipse fits to the images, or from kine- disks? The combined profile of a thin disk and a thick matic studies in the literature (see Erwin & Sparke 2003; disk should be the sum of two exponentials; observations Erwin 2005). The final profiles are from logarithmically of edge-on galaxies show that scale length ratios (thick spaced, concentric ellipses with constant position angle to thin) are almost always less than 2.0 (Pohlen et al. and ellipticity, matching that of the outer disk. 2004a). For some of our galaxies, particularly those with smooth transitions between the inner and outer regions, 3. ANTI-TRUNCATIONS IN SURFACE BRIGHTNESS the profile is indeed well fit by a double-exponential (e.g., PROFILES NGC 3489, 3941, and 4143). However, the resulting scale Figure 1 shows surface brightness profiles for two mod- length ratios (outer to inner, or “thick” to “thin”) range erately inclined galaxies which exhibit “conventional” be- from 2.1 to 6.3, with a mean of 3.8, which is clearly out- havior: NGC 2273 shows a classical truncation at 3.3 side the range of known thin/thick-disk systems. Fur- scale lengths of the inner (r . 95′′) disk, while NGC 4596 thermore, for the galaxies with the sharpest transitions shows a single exponential profile which extends to at (e.g., NGC 3982, 4371, 4691, and 5740), the best-fitting least > 6 scale lengths with no hint of a truncation. The double-exponential produces a distinct excess above the majority of galaxies in our sample have similar profiles, observed profile in the transition region. We conclude with untruncated profiles being far more common (see that the outer excesses we see are probably not thick Erwin et al. 2005 for the complete set of profiles). disks. At least 25% of the galaxies have profiles with qual- So what’s going on? One clue may be the presence of itatively different behavior: there is an excess of light lopsided/asymmetric spiral arms in the outer regions of at large radii, above the projected exponential disk pro- some of the “disky” galaxies. NGC 4319 has a dramatic, file. The transition happens at surface brightness levels one-armed outer spiral (responsible for the excess light − ′′ ′′ µR ∼ 22.6–25.6 mag arcsec 2 (mean = 24.1 ± 1.0; for between 50 and 100 in the profile; Figure 2) which may comparison, the truncations in Pohlen et al. 2002 are at be connected to a faint structure further out resembling a mean µR = 24.6). Figure 2 presents profiles for 12 of a tidal tail. NGC 5740 and NGC 5806 have weaker, lop- the 16 galaxies which unambiguously show this behav- sided spiral distortions in their outer isophotes. So one ior (there are other galaxies which may have excess light possibility is that interactions are responsible in some at large radii, but S/N and sky-subtraction uncertain- fashion for the outer excess light in the disks. ties make it more difficult to be certain). Because this In two galaxies (NGC 4045 and NGC 4612), there are Anti-Truncations in Early-Type Disks 3 outer rings coincident with the transition between inner inflow, which are necessarily smaller than the bars. and outer slopes; in three more (NGC 4371, NGC 4699, These surface-brightness profiles with outer excesses, and NGC 5806), the transition happens at ∼ 2–2.5 times which we have termed anti-truncations or Type III pro- the bar radius, which is the typical size of outer rings (e.g, files, are found in ∼ 25% of our sample; this should be Buta & Crocker 1993). This suggests that the transition regarded as a lower limit, since not all of our images might be happening at or near the bar’s outer Lindblad reach the same limiting surface brightness and we could resonance. be missing similar features at fainter surface-brightness We note that in some galaxies (e.g., NGC 4612 and levels. Even if we exclude those galaxies whose isophote NGC 4699), the transition to the outer profile happens shapes indicate we are probably seeing luminous halos or at a relatively small radius and high surface brightness outer bulges (left-hand column of Figure 2; 8% of the full level; since the outer disk is easily detected in these sample, or 20% of the S0 galaxies), at least 17% of the galaxies, it might be tempting to classify the inner ex- galaxies appear to have disks with anti-truncation behav- ponential slope as the galaxy’s “bulge” (the r . 40′′ ior. Disks with anti-truncations thus appear to be more region of NGC 4699 is in fact referred to this way in common than disks with classical truncations (. 12% of Sandage & Bedke 1994). But these inner regions are the sample; see Erwin et al. 2005), at least in barred, where the bars (and, in NGC 4699, numerous spiral S0–Sb disk galaxies. arms and dust lanes) are found and are thus not clas- sical (kinematically hot) bulges; smaller central excesses (e.g., r . 10′′ for NGC 4699) may be better candidates We would like to thank the anonymous referee for a for the bulge proper. The inner disk regions of these two careful reading, and especially for suggesting we take a galaxies, at least, can thus be viewed as “pseudobulges” closer look at the issue of thick disks. This research was (e.g., Kormendy & Kennicutt 2004), albeit much larger partly supported by grant AYA2004-08251-CO2-01 from than those pseudobulges thought to arise from bar-driven the Spanish Ministerio de Ciencia y Educaci´on.

REFERENCES Barton, I. J., & Thompson, L. A. 1997, AJ, 114, 655 Pohlen, M., Dettmar, R.-J., L¨utticke, R., & Aronica, G. 2002, Buta, R., & Crocker, D. A. 1993, AJ, 105,1344 A&A, 392, 807 de Grijs, R., Kregel, M., & Wesson, K. H. 2001, MNRAS, 324, 1074 Pohlen, M., Balcells, M., L¨utticke, R., & Dettmar, R.-J. 2004a, de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H. G., Buta, R. A&A, 422, 465 J., Paturel, G., Fouqu´e, P. 1991, Third Reference Catalogue of Pohlen, M., Beckman, J. E., H¨uttemeister, S. H., Knapen, J. H., Bright Galaxies. (New York: Springer-Verlag) (RC3) Erwin, P., & Dettmar, R.-J. 2004b. In Penetrating Bars through Erwin, P. 2005, MNRAS, submitted Masks of Cosmic Dust: The Hubble Tuning Fork Strikes a New Erwin, P., & Sparke, L. S. 2003, ApJS, 146, 299 Note, ed. D. L. Block, I. Puerari, K. C. Freeman, R. Groess, & Erwin, P., Pohlen, M., & Beckman, J. E. 2005, in prep E. K. Block (Dordrecht: Springer), 713 Freeman, K. C. 1970, ApJ, 170, 811 Sandage, A., & Bedke, J. 1994, The Carnegie Atlas of Galaxies Kregel, M., van der Kruit, P. C., & de Grijs, R. 2002, MNRAS, (Washington, D.C.: Carnegie Institution) 334, 646 van der Kruit, P. C. 1979, A&AS, 38, 15 Koopmann, R. A. & Kenney, J. D. P. 1998, ApJ, 497, L75 van der Kruit, P. C, & Searle, L. 1981a, A&A, 95, 105 Kormendy, J., & Kennicutt, R. C., Jr. 2004, ARA&A, 42, 603 van der Kruit, P. C, & Searle, L. 1981b, A&A, 95, 116 Nilson, P. 1973, Uppsala General Catalog of Galaxies, Uppsala Weiner, B. J., Williams, T. B., van Gorkom, J. H., & Sellwood, J. Astron. Obs. Annals, 5, 1 A. 2001, 546, 916 Pohlen, M., 2001, Ph.D. Thesis, Ruhr-University Bochum, Germany 4 Erwin et al.

Fig. 2.— Type III (“anti-truncation”) R-band surface-brightness profiles. Left column: Outer excess light is associated with halo/spheroid. Center and right columns: Outer excess light is probably part of the disk. Vertical dashed lines mark lower- and upper-limit measurements of the bar semi-major axis (deprojected); vertical dot-dashed lines indicate approximate size of outer rings, if any; and slanted dashed lines are exponential fits to regions of the profile delimited by the small squares. The arrows indicate R25 (one-half of D25). No photometric calibration was possible for NGC 4319 or NGC 4699. Anti-Truncations in Early-Type Disks 5

a b

Fig. 3.— Azimuthally averaged surface-brightness profiles (top) and isophotal ellipticity profiles (bottom) for two galaxies with Type III ′′ profiles. a. NGC 3412 shows a clear “disk + spheroid” profile: the outer excess light (r & 120 ) is associated with isophotes which become progressively rounder at larger radii, consistent with the idea of a (projected) elliptical disk embedded within a rounder spheroid, e.g., a ′′ halo or outer bulge. b. NGC 4371 has outer isophotes (r & 190 ) with approximately constant ellipticity, roughly the same as the inner ′′ isophotes, so the excess light at r & 190 is probably part of the disk.