A&A 382, 488-494 (2002)
DOI: 10.1051/0004-6361:20011647
E. M. Corsini1 - A. Pizzella2 - F. Bertola2
1 -
Osservatorio Astrofisico di Asiago, Dipartimento di Astronomia,
Università di Padova, via dell'Osservatorio 8, 36012 Asiago, Italy
2 -
Dipartimento di Astronomia, Università di Padova,
vicolo dell'Osservatorio 2, 35122 Padova, Italy
Received 2 September 2001 / Accepted 25 October 2001
Abstract
We present major and minor-axis kinematics of stars and
ionized gas as well as narrow and broad-band surface photometry of
the Sa spiral NGC 2855. In the nuclear regions of this unbarred and
apparently undisturbed spiral galaxy the gas is rotating
perpendicularly to the galaxy disk. We suggest that this
kinematically-decoupled component is the signature of an acquisition
process in the history of this galaxy.
Key words: galaxies: individual: NGC 2855 - galaxies: kinematics and dynamics - galaxies: spiral - galaxies: formation - galaxies: structure
Like fossils, multiple stellar and gaseous components with a misaligned or even opposite angular momentum with respect to the host galaxy conserve the memory of processes driving galaxy formation and evolution. In disk galaxies apart from polar rings they usually are hidden from morphological inspection and are serendipitously discovered by measuring detailed kinematics (see Bertola & Corsini 1998 for a review). This is the case for a variety of kinematically-decoupled components ranging from the counter-rotating disks of ionized gas observed in a large fraction of S0's (Bertola et al. 1992; Kuijken et al. 1996) as well as in one Sa, NGC 3626 (Ciri et al. 1995; Haynes et al. 2000), to the two counter-rotating stellar disks found in the S0 NGC 4550 (Rubin et al. 1992; Rix et al. 1992), in the early-type spirals NGC 3593 (Bertola et al. 1996), NGC 4138 (Jore et al. 1996; Haynes et al. 2000), and NGC 7217 (Merrifield & Kuijken 1994), to the orthogonally-rotating stellar cores discovered in two Sa spirals, namely NGC 4698 (Bertola et al. 1999) and NGC 4672 (Sarzi et al. 2000), and to the two counter-rotating gaseous components as in the S0 NGC 7332 (Fisher et al. 1994; Plana & Boulesteix 1996) and in the Sab NGC 4826 (Braun et al. 1992, 1994; Rubin 1994).
The recent findings by Bettoni et al. (2001) on the higher gas content of counter-rotators and polar ring galaxies with respect to normal galaxies gives further support to the idea that kinematically-decoupled components are the end result of one or more second events that occurred in the history of the host. Yet, even with this insight, the undisturbed appearance with no evident signatures of interaction characterizing most of galaxies with kinematic peculiarities (Bettoni et al. 2001) raises questions about the effective rate of second events and on their role in determining Hubble types. Accretion of external gas and minor mergers seem to be the favoured mechanisms to form decoupled components preserving the pre-existent disk (e.g. Thakar & Ryder 1996; Thakar et al. 1997); nevertheless major mergers can not be ignored since for a narrow range of initial conditions they still represent a viable alternative in building massive counter-rotating disks (Pfenniger 1999) and polar ring galaxies (Bekki 1998). Bulge growing, disk heating and spiral-pattern smoothing are among the observable consequences of second events in disk galaxies, as predicted by numerical experiments (e.g. Aguerri et al. 2001; Quinn et al. 1993; Thakar & Ryder 1998). They result in a change of the host morphology toward earlier spiral types. Recently the presence of a velocity gradient along the minor axis of the disk has been reported in the innermost region of two Sa spirals, namely NGC 4698 (Bertola et al. 1999) and NGC 4672 (Sarzi et al. 2000), characterized by an uncommon and remarkable orthogonal geometrical decoupling between bulge and disk. In NGC 4698 the minor-axis velocity gradient is observed in both stars and ionized gas, while in NGC 4672 only in the stellar component. Such a stellar rotation along the disk minor axis indicates the presence of a kinematically isolated core, which is rotating perpendicularly with respect to the disk component. The analysis of the HST images of the nucleus of NGC 4698 shows that in this case the isolated core is a nuclear stellar disk with a scalelength of a few tens of pc (Pizzella et al. 2001). According to Bertola & Corsini (2000) the presence of these orthogonally-rotating isolated cores indicates that the entire disk of the galaxy could be the end result of the acquisition of external material in polar orbits around a pre-existing oblate spheroid, which became the bulge of the present galaxy. In this scenario the isolated core has been formed by the gaseous material settled down in the symmetry plane of the oblate spheroid during the acquisition process.
In this paper we present a new case of kinematical orthogonal
decoupling in the bulge of early-type spiral galaxies which
characterizes the gaseous component only. In fact the ionized gas
observed in the innermost regions (
kpc) of the Sa
NGC 2855 is in orthogonal rotation with respect to the rest
of the galaxy.
The paper is organized as follows. We present our photometric and spectroscopic observations of NGC 2855 in Sect. 2 and we analyze its morphological and kinematical properties in Sects. 3 and 4, respectively. Our conclusions are discussed in Sect. 5.
The narrow and broad-band imaging of NGC 2855 were carried out at the Roque de los Muchachos Observatory in La Palma with the Telescopio Nazionale Galileo (TNG) on April 18, 1999 and at the European Southern Observatory (ESO) in La Silla with the 3.6-m ESO telescope on March 27, 2001. The instrumental setup and the observing log are collected in Table 1.
Parameter | April 18, 1999 | March 27, 2001 | |
Instrument | TNG+OIG | ESO 3.6-m+EFOSC2 | |
CCD | 2![]() |
#40 Loral/Lesser | |
Pixel size |
![]() ![]() |
![]() ![]() |
|
Scale |
![]() |
![]() |
|
Field of view |
![]() |
![]() |
|
Filter | R | #438 H![]() ![]() |
#642 R |
Exposure time |
![]() |
![]() |
![]() |
Seeing FWHM![]() |
![]() |
![]() |
![]() |
Using standard MIDAS routines the images were bias
subtracted, corrected for flatfield using sky flats taken at the
beginning and end of each observing night, and cleaned of cosmic
rays. The sky background level was removed by fitting a second order
polynomial to the regions free of sources in the images. The images
obtained with the same filter at the same telescope were shifted and
aligned using the common field stars and then they were averaged
(after checking the their PSF's were comparable). The map of the
H
+[N II] emission was obtained by subtracting the ESO R-band
image, suitably scaled, from the narrow-band image (Fig. 1).
To confirm the spiral classification of NGC 2855 we built an
unsharp masked image of the galaxy, by dividing its narrow-band frame
by itself after being convolved with a circular two-dimensional
Gaussian of
(Fig. 1).
This procedure enhanced the spiral pattern of the galaxy as well as
any surface-brightness fluctuation and non-circular structure
extending over a spatial region comparable with the size of the
smoothing Gaussian.
Absolute calibration was performed only for the R-band images.
![]() |
Figure 1:
Continuum-subtracted H![]() ![]() |
Open with DEXTER |
We analyzed the isophotal profiles of the galaxy by masking the field
stars and fitting ellipses to the isophotes using the
IRAF task
ELLIPSE. We first allowed the centers of the ellipses to vary, to test
whether the optical disk is disturbed. Within the errors of the fits,
we found no evidence of a varying center, therefore the ellipse fits
were repeated with the ellipse centers fixed. The resulting surface
brightness, ellipticity and position angle radial profiles obtained
from both the ESO and TNG R-band images are given in Fig. 2.
![]() |
Figure 2: R-band surface-brightness, ellipticity and position angle radial profiles of NGC 2855 measured on the TNG ( filled circles) and 3.6-m ( open circles) images. Errorbars smaller than symbols are not plotted. |
Open with DEXTER |
The spectroscopic observations of NGC 2855 were carried out at the ESO in La Silla with the 1.52-m ESO telescope on March 17-22, 1999 and New Technology Telescope (NTT) on May 22, 2001. The details of the instrumental setup and the log of these observing runs are given in Table 2.
At the beginning of each exposure the galaxy was centered on the slit using the guiding camera. During the March 1999 observing run, we took spectra of several giant stars whose spectral type ranged from G5III to K7III to be used as templates in measuring the stellar kinematics. Comparison spectra were taken before every object exposure. The reduction was carried out with MIDAS. All the spectra were bias subtracted, flatfield corrected, and wavelength calibrated. The contribution of the sky was determined from the outermost 10'' at the two edges of the resulting spectra, where the galaxy light was negligible, and then subtracted. The spectra taken along the same axis in the same run were co-added using the center of the stellar continuum as reference, and cosmic rays were cleaned up in the process.
Parameter | March 17-22, 1999 | May 22, 2001 | |
Instrument | ESO 1.52-m+B&C | NTT+EMMI | |
Grating/Grism | #33 1200
![]() |
#7 600
![]() |
|
CCD | #39 Loral/Lesser | #36 TEK 2048 | |
Pixel size |
![]() ![]() |
![]() ![]() |
|
Scale |
![]() |
![]() |
|
Dispersion |
![]() |
![]() |
|
Slit width |
![]() |
![]() |
|
Wavelength range | 4812-6824 Å | 5630-6968 Å | |
Instrumental
![]() |
52
![]() |
28
![]() |
|
Position angle | ![]() |
![]() |
![]() |
Exposure time | ![]() |
![]() |
30 min |
Seeing FWHM |
![]() ![]() |
![]() |
We measured the stellar kinematics from the galaxy absorption features
present in the wavelength range from 5064 Å to 5540 Å
and centered on the Mg line triplet (
Å).
We used the Fourier Correlation Quotient method (FCQ, Bender 1990)
following the prescriptions of Bender et al. (1994). The
spectra were rebinned along the spatial direction to obtain a nearly
constant signal-to-noise ratio larger than 20 per resolution element
(with a peak of 80 in the innermost regions of the spectra). The
galaxy continuum was removed row-by-row by fitting a fourth to sixth
order polynomial as in Bender et al. (1994). HR 2035 was
adopted as a kinematical template. This allowed us to derive, for each
spectrum, the line-of-sight velocity distribution (LOSVD) along the
slit and to measure its moments, namely the radial velocity v, the
velocity dispersion
and the values of the coefficients h3and h4. At each radius, they were derived by fitting the
LOSVD with a Gaussian plus third- and fourth-order Gauss-Hermite
polynomials H3 and H4, which describe the asymmetric and
symmetric deviations of the LOSVD from a pure Gaussian profile (van
der Marel & Franx 1993; Gerhard 1993).
We derived errors on the LOSVD moments from photon statistics and CCD
read-out noise, calibrating them by Monte Carlo simulations as done by
Bender et al. (1994). These errors do not take into account possible
systematic effects due to any template mismatch. The resulting
stellar kinematics was reported in Table 3 and plotted in Fig. 4.
The ionized-gas kinematics was measured by the simultaneous Gaussian
fit of the emission lines present in the spectra (namely [N II]
,
H
,
and [S II]
)
with an "ad hoc'' procedure written within the IDL
environment. The galaxy continuum was removed from the spectra as
done for the stellar kinematics. We fitted in each row of the
continuum-subtracted spectrum a Gaussian to each emission line,
assuming them to have the same line-of-sight velocity and velocity
dispersion (corrected for heliocentric velocity and instrumental FWHM,
respectively). An additional absorption Gaussian has been added in the
fit to take into account for presence of the H
absorption line and
the flux ratio of the [N II]
lines have been fixed to 1:3. Far from
the galaxy center (for
)
we averaged adjacent spectral
rows in order to increase the signal-to-noise ratio of the relevant
emission lines.
We calibrated the formal velocity errors obtained from the
least-squares fit to match those derived by measuring the rotation
curve of night-sky emission lines (cf. Corsini et al. 1999).
The resulting ionized-gas kinematics is given in Table 4 and shown in
Fig. 4.
![]() |
Figure 3: Deep image of the NGC 2855 field adapted from Malin & Hadley (1997) and processed to enhance the ring structure surrounding the galaxy. Our ESO R-band image is shown in the inset. The frame is 15' height. North is up and east is left. |
Open with DEXTER |
NGC 2855 is an early-type spiral galaxy classified as
Sa(r) by Sandage & Tammann (1981, RSA hereafter) and RS0/a(rs) by de Vaucouleurs et al. (1991, RC3 hereafter).
Its total B-band magnitude is
mag (RC3) which
corresponds to
MB=-19.11 mag for a distance D=22.3 Mpc. We
derive the distance from the heliocentric systemic velocity we
measured (
)
assuming H0 = 75
Mpc-1 and following the precepts of RSA. The measured systemic
velocity is in agreement within the errors with the values listed by
RSA, Tully (1988) and RC3. The inclination
was obtained from
the observed axial ratio (whose mean value between 15'' and 30''is q=0.79, Fig. 2) after correcting for the
typical intrinsic flattening of S0/a galaxies (
q0 = 0.18, Guthrie
1992).
![]() |
Figure 4: The stellar ( filled circles) and ionized-gas kinematics ( open symbols) measured along the major ( left panel) and minor axis ( right panel) of NGC 2855. All velocities are plotted as observed after subtracting the value of the systemic heliocentric velocity and without applying any inclination correction. Errorbars smaller than symbols are not plotted. In the lower panel open circles and open squares represent data obtained with the ESO 1.52-m telescope and NTT, respectively. |
Open with DEXTER |
Sandage (1961) in The Hubble Atlas of Galaxies considered the galaxy
as transition case from S03 to Sa(r) due to its circular central
absorption ring. Sandage & Bedke (1994) included NGC 2855 in the
section of Carnegie Atlas of Galaxies devoted to Sa galaxies with a
spiral pattern entirely defined by dust lanes (Panels 73 and 74). The
presence of tightly wound spiral arms in the disk is confirmed by
our continuum-subtracted H+[N II] emission image of NGC 2855 and by
the unsharp-masked version of its narrow-band image (Fig. 1).
The morphology NGC 2855 appears relatively undisturbed (Rudnick & Rix
1998) and the galaxy has no close companions (RC3) suggestive of
interaction phenomena. However enhanced IIIa-J plates with
deep exposures of the NGC 2855 field (Malin & Hadley 1997) reveal the
presence of a structure of low surface brightness (
28 mag arcsec-2), which surrounds the galaxy. We processed this image,
which was kindly put at our disposal by D. Malin, to enhance the
contrast of the faint structure (Fig. 3). We found that
this almost complete loop is elongated in a direction close to the
galaxy minor axis (
)
and it extends out
to about 6' and 8' from the galaxy center on its NE and SW side,
respectively. It resembles an offset ring with an inclination of about
.
At fainter levels the stellar disk is nearly face on,
confirming a trend already present in the ellipticity profile shown in
Fig. 2. Indeed we derive an inclination
at r=141'', corresponding to the outermost observed radius in our
ESO R-band image. This change of inclination is associated with a
variation of the isophotal position angle
(
between 53'' and 141'', and
at
from Fig. 3), and is indicative of a warped stellar disk.
The large extension and possible warp of the disk component could
explain the discrepancy between the available bulge-disk
decompositions of the surface brightness of NGC 2855. Indeed the disk
results are embedded in the bulge at all the observed radii according
to the V (Baggett et al. 1998) and K-band
photometric decompositions (Möllenhoff & Heidt 2001). On the other
hand, the disk dominates the galaxy light for
in the B and
J-band photometric decomposition performed by Boroson (1981) and
Möllenhoff & Heidt (2001), respectively.
The stellar kinematics does not show any peculiarity on both the
observed axes. It extends out to 28'' (corresponding to 3.0 kpc)
and 20'' from the center (corresponding to 2.8 kpc after
correcting for inclination) along the major and minor axis,
respectively. The velocity dispersion declines from a central value of
about 190 to 150
.
Along the major axis we observe a
maximum
at
.
Along the
minor axis there is no significant stellar velocity gradient and the
velocity dispersion ranges from a maximum of about 190 to 130
at
the farthest measured radius.
The major-axis kinematics of the ionized gas is measured out to 57''(6.2 kpc) from the nucleus. Gas rotates slower than stars in the
inner
(0.4 kpc). The opposite is true further out. Ionized
gas reaches its maximum
at
.
At larger radii the gas rotation velocity decreases
to about 100
on both sides of the nucleus. We interpret this
velocity trend as the result of a projection effect rather an
intrinsic phenomenon and therefore we conclude that the gaseous as
well as the stellar one tends to be more face-on at larger radii. The
gas velocity dispersion peaks at a central value of about 150
,
and outwards it decreases, remaining lower than 60
for
.
We derived the minor-axis kinematics of the gaseous component out to
11'' (1.5 kpc after deprojecting) and to 41'' (5.7 kpc after
deprojecting) from the NTT and ESO 1.52-m spectrum, respectively.
In the NTT spectrum the velocity curve shows a steep gradient in the
nucleus rising to a maximum observed rotation of about 75
at
.
At larger radii the gas velocity drops to zero and no
rotation is detected for
.
The presence of a sharp rotation
of the gas component along the galaxy minor axis is confirmed by the
kinematics obtained from the ESO 1.52-m spectrum, although these data
are characterized by a lower spectral and spatial resolution with
respect to the NTT ones. The gas velocity dispersion along the minor
axis shows the same behaviour observed along the major one.
The analysis of the velocity curves of stars and ionized gas along
both the major and minor axis of NGC 2855 shows a kinematical
decoupling between the gas in the innermost regions of the galaxy and
the remaining gas.
Indeed the innermost gas shows a central steep velocity gradient along
the minor axis (for
)
where stars do not have any
significant rotation. The gas rotates slightly slower than the stars along
the galaxy major axis (for
). On the contrary, the outer
gas rotates faster than the stars along the major axis and shows no
rotation along the galaxy minor axis.
We exclude that the gas velocity gradient we measure along the galaxy minor axis is due to non-circular motions in a barred potential. Indeed we do not observe any bar structure in low-inclined disk of the galaxy either on our R-band image or in the near-infrared images of Peletier et al. (1999) and Möllenhoff & Heidt (2001). In addition, the radial profiles of ellipticity and position angle do not show the signatures of the presence of a nuclear bar, as discussed by Wozniak et al. (1995).
Since the minor-axis gas rotation is measured inside the
bulge-dominated region and since the intrinsic shape of bulges is
generally triaxial (Bertola et al. 1991) we can consider
the kinematical decoupling observed in NGC 2855 as an effect due to
the bulge triaxiality.
If the gas component is in one of the equilibrium planes of the bulge (as
in the case of the Sa NGC 4845 studied by Bertola et al. 1989; Gerhard et al. 1989) the sharp rise and fall of the gas
rotation velocity observed for
along the minor axis might
be caused by strong non-circular motions. However this is unlikely,
since we expect a mild triaxiality for the bulge of NGC 2855 because
of the lack of a significant isophotal twisting for
(Fig. 2).
We therefore interpret the observed ionized-gas velocity field as
being due to the presence of two kinematically-decoupled gaseous
components, which are rotating around two roughly orthogonal
axes. They are the shortest and the longest axes of the triaxial
bulge. In this picture the innermost gas is moving onto the
equilibrium plane orthogonal to the equatorial one, where both the
galaxy disk and outer gas component are settled. If this is the case,
the kinematical decoupling is also consistent with an abrupt inner
warp of the ionized-gas component.
Kinematical decoupling between two components of a galaxy suggests the
occurrence of a second event, so it is easy to explain the existence
of the innermost orthogonally-rotating gas as being due to external
material acquired from a direction close to the bulge equilibrium
plane orthogonal to that of the galaxy disk. This scenario is
supported by the presence of the faint structure which surrounds the
galaxy and is elongated in a direction close to the galaxy minor
axis. Part of the gaseous material associated with this loop could be
sinked towards the galaxy nucleus sweeping away the gas of the disk
and giving rise to the decoupled component.
Although it involves a second event, the mechanism of formation of the
kinematically-decoupled component in NGC 2855 seems to be different
from that of NGC 4698 and NGC 4672, where the entire disk could be
formed in the acquisition process (Bertola & Corsini 2000). In fact
it has to be pointed out that in NGC 2855 the innermost gas has not
yet turned into stars, while it is partially and completely
transformed into stars in NGC 4698 and NGC 4672, respectively.
According to this result, although the low inclination of NGC 2855
does not preclude that its bulge could be an oblate spheroid
protruding out of the disk plane but we have to explain why star
formation did not occur in the inner orthogonally-rotating gaseous
component. Such a kinematically-decoupled gas may represent an early
stage in the formation of an orthogonally-rotating stellar core.
The case of NGC 2855 gives further support to the mounting evidence that in nearby galaxies even with an apparently undisturbed morphology we find the signatures of on-going or recent acquisition phenomena if we study them in detail. A better knowledge of second events in present-day galaxies is a key to understanding such processes in the early universe, when they were more frequent due to the higher galaxy density.
Acknowledgements
We acknowledge Ralf-Jürgen Dettmar for the ESO 3.6-m narrow-band imaging of NGC 2855. We are indebted to R. Bender and R. Saglia for providing us with the FCQ package we used for measuring the stellar kinematics. We are also grateful to D. Malin for the image of NGC 2855 we used in Fig. 3.