Kinematics of planetary nebulae in the outskirts of galaxies, from slitless FOCAS radial velocities

1
Kinematics of planetary nebulae in the outskirts
of galaxies, from slitless FOCAS radial velocities

• Roberto H. Mendez
• Institute for Astronomy,
• University of Hawaii, Honolulu

2

• Collaborators
• Ana M. Teodorescu (IfA, Univ. of Hawaii)
• Rolf-Peter Kudritzki (IfA, Univ. of Hawaii)
• Lent C. Johnson (University of Wyoming)

3
List of topics

• Motivations kinematics of elliptical galaxies
• Basic PN discovery technique
• Slitless velocities with Subaru FOCAS
• Calibration mask and the local PN NGC 7293
• New PNs discovered in NGC 4697
• On the Keplerian decline of line-of-sight
  velocity dispersion (losvd)
• Future activities

4
(No Transcript)
5
Motivation

• Kinematics of elliptical galaxies are difficult
  to study they lack cold H gas, and
  absorption-line studies of the integrated stellar
  spectrum are restricted to the central regions,
  where the surface brightness is sufficiently
  high.
• At the same time, any information about the
  dynamical behavior of the outer regions of
  ellipticals is valuable to test and even to guide
  theoretical ideas about galaxy formation. For
  example, ellipticals are expected to be
  surrounded by dark matter halos, and to show high
  rotation in the outskirts.
• Planetary nebulae (PNs) are excellent kinematic
  probes for dark matter and angular momentum
  distribution, particularly at large angular
  distances from the galaxys center, where they
  are easier to discover.

6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
The Subaru FOCAS version

• Dispersing element echelle grism, used in 4th
  order. We get 0.5 Angstroms/pixel.
• Calibration of displacement using a rigid mask
  with approximately 1000 holes.
• Width of images if 5 pixels, 140 km/s.
• Error in velocities if 0.4 pixel, 10 km/s.

11
NGC 4697 FORS (ESO VLT, Chile) and FOCAS field
s
12
Calibration Fun
Thorium-Argon Lamp through an Engineering
Mask 970 points ?100 pixel spacing
13
5039.230 Å
5028.655 Å
5017.254 Å
5009.350 Å
5002.097 Å
14
Radial velocity calculations

• Identify four closest calibration points that
  form a square around PN
• Calculate the redshifted wavelength of the O
  III emission feature at each of the four points
• Obtain the final wavelength by using a bilinear
  interpolation calculation combining the four
  values according to the spatial relation of each
  of the 4 calibration points to the PN
• Final wavelength and heliocentric correction are
  used to calculate the radial velocity

15
NGC 7293, a local PN
16
(No Transcript)
17
Quality control using NGC 7293

• 1.Illuminate the calibration mask using NGC 7293
  and take an image
• 2.Insert the grism and take another image
• 3.Illuminate the mask using the Th-Ar lamp and
  take another image through the grism.
• 1 and 2 give displacements that can be
  transformed into radial velocities. We call these
  slitless velocities.
• 2 and 3 give a velocity for each hole in the
  mask,
• using the holes as slits. We call these
  classical velocities.

18
Slitless vs. classical velocities
19
FORS vs. FOCAS

• Earlier data obtained with FORS VLT (ESO, Cerro
  Paranal, Chile) yielded velocities for 535 PNs in
  the central regions (Mendez et al. 2001 ApJ 563,
  135)
• New FOCAS data (218 PNs) in two fields partly
  overlapping the FORS data but extending outwards
  along the major axis of NGC 4697

20
FORS vs FOCAS velocities

• We compare 162 PNs in NGC 4697, measured
  with both FORS and FOCAS. The FORS errors are 35
  km/s. Comparing with random number simulations,
  we confirm that the FOCAS errors are 10 km/s.

21
Radial vels of PNs in NGC 4697

• The 218 slitless PN velocities measured with
  FOCAS, as a function of projected distance along
  the major axis. The effect of rotation is
  visible, and the line-of-sight velocity
  dispersion decreases markedly outside. The solid
  lines indicate the escape velocity due to visible
  matter.

22
Rotation in NGC 4697

• Plus signs absorption-line data
• Diamonds PN data
• As expected, the PNs rotate in the same sense
  as the stellar population. We detect no
  significant rotation beyond 5 Re.

23
NGC 4697 is almost edge-on
24
Line-of-sight velocity dispersion in NGC 4697

• Plus signs absorption-line data
• Squares PN data. The
• outer 4 points are based on
• the FOCAS velocities.
• The solid line is a Hernquist
• model with a constant M/L
• ratio and a total mass of
• 1.5x1011 Msun.
• We find no evidence of dark
• matter within 5 Re.

25
Other cases

• Romanowsky et al. (2003 Science, 301, 1696)
  using the PN.S at La Palma (Herschel telescope)
  have reported more examples of elliptical
  galaxies showing no evidence of dark matter
  halos
• NGC 3379 (very convincing)
• NGC 4494 and NGC 821 (not so convincing, more
  data are needed)
• They originally claimed definite proof of
  absence of dark matter, but all these cases,
  including NGC 4697, can be interpreted as a
  consequence of radial anisotropy (predominance of
  radial orbits implies that far from the center we
  see less dispersion than for an isotropic
  system). More advanced modeling is in progress
  De Lorenzi et al. 2007, MNRAS 376, 71 and 2008,
  MNRAS in press.

26
Not all ellipticals misbehave

• There are some intermediate-mass ellipticals
  that do show evidence of a dark matter halo
• NGC 5128 (Hui et al. 1995 ApJ 449, 592)
  confirmed by Peng, Ford and Freeman 2004, ApJ
  602, 685. This one shows also significat halo
  rotation.
• NGC 1344 (Teodorescu et al. 2005, ApJ 635,
  290).
• So now we have to answer another question
  how frequent is the Keplerian behavior?

27
At the present time

• As we collect more information, a variety in the
  amount and/or distribution of dark matter seems
  to be emerging. So far, the FOCAS data on PNs in
  NGC 4697 provide the best evidence of a Keplerian
  decline in the kinematics of an elliptical
  galaxy. We also confirm the same behavior in NGC
  821 (Teodorescu et al., in preparation).
• Massive giant ellipticals Gerhard et al. (2001
  AJ 121, 1936) have reported a spread in the
  luminous-to-dark matter ratio.
• Are we seeing the same kind of behavior in
  ordinary ellipticals?

28
Alternatively

• If we insist that dark matter has to be
  present everywhere, then
• in some cases these galaxies have such
  peculiar kinematics that they manage to conceal
  their dark matter halos we can try to build
  consistent dynamical models and learn about their
  properties.

29
Next FOCAS project

• We would like next to study PNs in the Virgo
  cluster elliptical M 60. This galaxy is more
  massive, and it shows hot X-ray emitting gas,
  giving direct evidence of the existence of a dark
  matter halo. Will the PNs show a correspondingly
  high losvd in the outskirts? Our first observing
  run in 2007 was affected by bad seeing, so we
  detected only about 40 PNs. We will try again
  next May.