Recent Results of the MPE Infrared/Submillimeter Group
The SINS survey: spatially-resolved dynamics and star formation
Observations and theoretical simulations have now established a
framework for galaxy formation and evolution in the young Universe.
Galaxies formed as baryonic gas cooled at the center of collapsing
dark matter halos, and subsequently grew through mergers and
collisions leading to the hierarchical build-up of galaxy mass.
It remains unclear, however, when and how disks and spheroids
— the primary components of today's galaxies —
were formed, over what timescales, and which processes were
driving the early evolution of galaxies.
The major limitation is our incomplete knowledge of the relevant
mechanisms that control the phase, angular momentum, cooling,
and dynamics of the baryonic matter.
The SINS survey now sheds new light on these issues, from detailed
information on the dynamics, morphologies, and physical properties
of massive star-forming galaxies at early stages of evolution.
The survey focusses on the crucial epochs at lookback times of
8–11 billion years ago (cosmological redshifts z ~ 1–3),
when a major fraction of the mass in stars seen in present-day
galaxies is believed to have been put in place.
Using the near-infrared integral field spectrometer SINFONI at the ESO
Very Large Telescope, we targeted primarily the Hα line emission
arising mainly from photoionized nebulae around young massive stars.
The Hα morphology and kinematics from our SINFONI observations
trace the distribution of active star-forming sites and the gas motions
on typical scales of 4 &ndash 5 kpc. Our full SINS sample includes
over 60 detected star-forming galaxies of different classes at
z ~ 1.5 – 3. For about 15 of them, we also carried
out adaptive-optics assisted observations, revealing the structure
and kinematics of those galaxies on scales as small as 1 &ndash 2 kpc.
Further information about the SINS survey can be found on the
SINS web pages.
Evidence for a significant number of massive rotating disks at z ~ 2
One of the key questions that we can now address with such data is whether
the dynamics of distant star-forming galaxies are dominated by regular ordered
motions as in rotating disks, or rather by strongly disturbed/irregular motions
expected for violent collisions between galaxies. This distinction has direct
implications on the mechanisms through which galaxies assemble their mass:
smooth but rapid accretion through cold infall along filaments or rapid series
of minor mergers (mass ratio > 3:1 between the progenitors), or alternatively
violent dissipative major mergers (mass ratio < 3:1), respectively.
Qualitatively, the Hα kinematics of a majority of the SINS galaxies are
consistent with disk rotation, albeit in clumpy, turbulent systems.
In order to strengthen these conclusions, we have developed a set of
quantitative kinematic criteria, based on templates from observations of
nearby galaxies and from simulations, and the technique of "kinemetry".
Applying these criteria to our highest-quality data, we find that ~ 3/4
of the resolved systems (with half-light radius larger than 4 kpc) display
no dynamical evidence of having had a recent major merger (see Figure 1).
This provides evidence that there exists a significant population of rapidly
star-forming systems in regularly-rotating, unperturbed configurations.
The implication is that the high star formation rates
(~ 100 Msun/yr) in these young but rapidly
evolving galaxies are not driven by major mergers, and that they must have
formed via smoother accretion processes, such as gas inflow along cold
filamentary streams, or rapid series of minor mergers.
First comparison of the dynamical properties of different
galaxy classes at z ~ 1.4 – 3.4
Having established the predominance of rotating disk-like systems among
the SINS survey sample, an immediate question is whether these high redshift
disks follow a relation between velocity and size as do spiral galaxies in
the present-day universe. Using our SINFONI data of 32 galaxies, probing the
bulk of z ~ 1.5 – 3 star-forming galaxies at the high-mass end,
we found that the sub-samples of rest-frame UV- and optically-selected
(16 galaxies each) are dynamically similar, and follow a similar velocity-size
relation as nearby spiral galaxies (see Figure 2).
We further combined these results with data of 13 bright submillimeter-selected
galaxies (SMGs) at similar redshifts, observed as part of a long-term program
involving several members of our team to map the molecular gas through CO line
emission with the IRAM/Plateau de Bure millimeter interferometer. SMGs represent
the most luminous, dusty, and intensely star-forming systems at these cosmological
epochs. These SMGs occupy a very different region of the velocity-size diagram,
with significantly larger velocity widths and much smaller sizes. Their properties
imply higher central matter densities by nearly an order of magnitude and lower
lower angular momenta compared to the galaxies from the SINS sample.
Together with the spatially-resolved CO line data obtained for several
of them showing strongly perturbed kinematics on scales of ~ 1 – 2 kpc,
these results suggest that major mergers are more frequent among the bright SMG
population compared to more "normal" star-forming galaxies at high redshift.
Diagnostic diagram used to distinguish between regularly-rotating
"disk-like" galaxies and systems undergoing major mergers (right).
Our templates for each group, shown in blue and red respectively, have
been analyzed as if they were observed at redshift z ~ 2 with
the VLT/SINFONI set-up used for the SINS survey observations.
The total kinematic asymmetry is defined as the sum in quadrature
of the measured asymmetry in the velocity and velocity dispersion
fields, and the probability distribution functions of this parameter
for non-merging and merging systems (inset) is used to identify the
boundary between unperturbed and merging systems (black line).
Performing this analysis on SINS galaxies (black points), we find
that the majority (8/11) of our best resolved systems are disk-like.
Visual analysis of the velocity fields of SINS galaxies (shown at left)
confirms the efficacy of our classification.
The centers of stellar continuum emission in each system is indicated
with a black cross, and the maxima and minima of the Hα velocities
are indicated (in km/s).
First comparison of the dynamical properties of galaxy samples
at z ~ 1.4 - 3.4 : the velocity-size diagram of
rest-frame UV- and optically-selected star-forming galaxies from
the SINS survey (blue and red points, respectively), and bright
submillimeter-selected galaxies (SMGs) observed as part of a program
to map the CO molecular gas line emission with millimeter interferometry
The location of the SINS samples, with a majority of rotating disk-like
systems, follows a velocity-size relation with large overlap with that
for nearby spiral galaxies (grey crosses, from Courteau 1997).
In contrast, the SMG sample is characterized by much larger velocity
widths and smaller sizes, implying much higher matter densities and
lower angular momenta consistent with a majority of them undergoing
dissipative major mergers between gas-rich galaxies.
Publications from this work:
- Kinemetry of SINS high redshift star-forming galaxies:
Distinguishing rotating disks from major mergers.
Shapiro, K. L., et al. 2008, ApJ, in press (arXiv:0802.0879)
- Dynamical properties of z~2 star-forming galaxies and
a universal star formation relation.
Bouché, N., et al. 2007, ApJ, 671, 303
- The rapid formation of a large rotating disk galaxy
three billion years after the Big Bang.
Genzel, R., et al. 2006, Nature, 442, 786
- SINFONI integral field spectroscopy of z~2 UV-selected galaxies:
rotation curves and dynamical evolution.
Förster Schreiber, N. M., et al. 2006, ApJ, 645, 1062
Author: N.M. Förster Schreiber, Date: 07/07/2008
Back to the top
Star formation in the hosts of high-z QSOs
Intense star formation in the hosts of high-z QSOs is expected from many scenarios
of AGN-galaxy co-evolution but very hard to quantify by observations, because the
powerful AGN outshines the host at most wavelengths.
Following our use of Spitzer Space Telescope spectroscopy to disentangle AGN and
host emission in nearby QSOs and reconstruct the infrared SED of the AGN proper,
we have now extended this technique to high redshift by searching for the mid-infrared
'PAH' emission features in twelve z~2 mm-bright type 1 QSOs, selected from unlensed and
lensed QSO samples. On top of the AGN continuum, we detect PAH emission from luminous
star formation in nine objects individually as well as in the composite spectrum for
the full sample. This provides strong evidence for intense star formation in the hosts
of these mm-bright QSOs, sometimes exceeding 1000 solar masses per year and dominating
their rest frame far-infrared emission. The PAH-based limit on star formation rates is
lower for luminous z~2 QSOs that are not preselected for their mm emission. Partly
dependent on systematic changes of the AGN dust covering factor and the dust spectral
energy distribution of the AGN proper, the spectral energy distributions of such mm-faint
high-z QSOs may be AGN dominated out to rest frame far-infrared wavelengths.
The average Spitzer mid-infrared spectrum of mm-bright z~2 QSOs shows clear PAH emission.
PAH and rest-frame far-infrared emission correlate and extend to higher luminosities
the similar correlation for local PG QSOs. These results indicate intense star
formation in the hosts of mm-bright high-z QSOs, dominating their rest frame far-infrared
Publications from this work:
Quasar and ULIRG evolution study (QUEST): I. The origin of the far infrared continuum of QSOs, M. Schweitzer, D. Lutz, E. Sturm, A. Contursi, L.J. Tacconi, M.D. Lehnert, K. Dasyra, R. Genzel, S. Veilleux, D. Rupke, D.-C. Kim, A.J. Baker, H. Netzer, A. Sternberg, J. Mazzarella, S. Lord ApJ 649, 79 (2006)
Spitzer Quasar and ULIRG Evolution Study (QUEST). II. The Spectral Energy Distributions of Palomar-Green Quasars, H. Netzer, D. Lutz, M. Schweitzer, A. Contursi, E. Sturm, L.J. Tacconi, S. Veilleux, D.-C.Kim, D. rupke, A.J. Baker, K. Dasyra, J. Mazzarella. S. Lord ApJ 666, 806 (2007)
Dust covering factor, silicate emission, and star formation in luminous QSOs, R. Maiolino, O. Shemmer, M. Imanishi, H. Netzer, E. Oliva, D. Lutz, E. Sturm, A&A, 468, 979 (2007)
PAH Emission and Star Formation in the Host of the z~2.56 Cloverleaf QSO, D. Lutz, E. Sturm, L.J. Tacconi, E. Valiante, M. Schweitzer, H. Netzer, R. Maiolino, P. Andreani, O. Shemmer, S. Veilleux, ApJ, 661, L25 (2007)
Star formation in the hosts of high-z QSOs: Evidence from Spitzer PAH detections, D. Lutz, E. Sturm, L.J. Tacconi, E. Valiante, M. Schweitzer, H. Netzer, R. Maiolino, P. Andreani, O. Shemmer, S. Veilleux, ApJ, in press (arXiv0805.2669) (2008)
Author: Dieter Lutz, Date: 11/06/2008
Back to the top
Extended Silicate Dust Emission in QSOs
Unified models for active galaxies postulate that all active galactic
nuclei (AGN) harbour a central, accreting massive black hole surrounded
by dusty, obscuring material in form of a (clumpy) torus that absorbs or
shadows emission from the nuclear region. The tori are predicted to
exhibit prominent silicate dust features at 9.8 and 18 micron in either
absorption or emission, depending on whether an AGN is viewed with the
torus edge-on (Type 2) or face-on (Type 1) (cp. Fig. 1). However,
silicate emission in Type 1 AGN has not been observed until recently.
The Spitzer Space Telescope (Spitzer), with its good mid-infrared
(mid-IR) wavelength coverage and sensitivity, has drastically changed
our view of this problem. Prominent silicate emission features have been
detected now in the mid-IR spectra of luminous quasars and also in less
luminous AGN. This allows a fresh look at the properties of AGN dust and
opens new routes to test AGN unification theories. However, the
apparently cool (~200 K) dust is inconsistent with theoretical
expectations of much hotter torus walls. Furthermore, not all Type 1
objects are silicate emission sources. Alternatively, the silicate
emission may originate in dust not directly associated with a torus,
e.g. in a dusty Narrow Line Region (NLR).
In order to test this alternative explanation we have performed mid-IR
spectroscopy of 23 QSOs with Spitzer-IRS. These spectra, and especially
the silicate emission features at 10 and 18 micron, can be fitted using
dusty narrow line region (NLR) models and a combination of black bodies
(and small contributions from starburst templates). The bolometric
luminosities of the QSOs allow us to derive the radial distances and
covering factors for the silicate-emitting dust. The inferred radii are
100-200 times larger than the dust sublimation radius, much larger than
the expected dimensions of the inner torus. Our QSO mid-IR spectra are
consistent with the bulk of the silicate dust emission arising from the
dust in the innermost parts of the NLR.
Illustration of a dusty AGN torus,BLR and NLR, and the role of
the line of sight (Padovani & Urry).
Fit results (2 examples): NLR model (blue, solid curve); M82 (blue,
dashed curve); blackbodies (green, dotted curves); total model (red
curve); observed spectrum (black curve)
NLR cloud distance (R_dust) versus bolometric QSO luminosity,
compared to the dust sublimation distance R_sub (solid line)
of a putative dusty torus. On average, R_dust is 170 times
larger than R_sub.
Publications from this work:
Extended Silicate Dust Emission in Palomar-Green QSOs,
M. Schweitzer, B. Groves, H. Netzer, D. Lutz, E. Sturm, A. Contursi,
R. Genzel, L.J. Tacconi, S. Veilleux, D.-C: Kim, D. Rupke, A.J. Baker,
2008, ApJ 679, 101
Silicate emissions in active galaxies - From LINERs to QSOs,
E. Sturm, M. Schweitzer, D. Lutz, A. Contursi, R. Genzel, M. D. Lehnert,
L. J. Tacconi, S. Veilleux, D. S. Rupke, D.-C. Kim, A. Sternberg,
D. Maoz, S. Lord, J. Mazzarella, D. B. Sanders
2005, ApJ Letter, 629, L21
Author: Eckhard Sturm, Date: 11/06/2008
Back to the top
Mid-infrared spectra of submillimeter galaxies: Luminous high-z starbursts
is becoming increasingly clear that
characterizing the population of high redshift submillimeter galaxies
an important step towards an understanding of the formation of massive
galaxies. One of the methods best suited to study these dust-rich and
galaxies is mid-infrared spectroscopy, using methods which our group
developed and already extensively applied to infrared galaxies in the
universe. We have completed the first such study using the Infrared
Spectrograph (IRS) on board the Spitzer Space Telescope for a sample of
majority of spectra are
well fitted by a starburst template or by the superposition of PAH
features and a weak mid-infrared continuum, the latter a tracer of
galactic nuclei (AGNs; including Compton-thick ones). We obtain
spectroscopic redshifts for all nine sources detected with IRS,
previously unknown ones. The median value of the redshift distribution
is z ~2.8 if we
assume that the four IRS nondetections are
at high redshift – lower redshift would require rest-frame
mid-IR obscuration even
larger than is seen in local ULIRGs. This suggests a modest extension
higher redshift of the optical-based SMG redshift distribution. The
rest frame mid- to far-infrared spectral energy
distributions are consistent with these submillimeter galaxies being
versions of local ultraluminous infrared galaxies. The mid-infrared
support the scenario that submillimeter galaxies are sites of extreme
formation, rather than dominated by X-ray obscured AGNs, and represent
phase in the formation of massive galaxies.
The average Spitzer-IRS spectrum of 9 submillimeter galaxies is well
fitted by the superposition of a starburst template (M82) plus a weak
Publications from this work:
- Mid-Infrared Spectroscopy of Two Luminous Submillimeter
Galaxies at z ~ 2.8: Lutz, D.; Valiante, E.; Sturm, E.; Genzel, R.;
Tacconi, L. J.; Lehnert, M. D.; Sternberg, A.; Baker, A. J., 2005, ApJ,
- A Mid-Infrared Spectroscopic Study of Submillimeter
Galaxies: Luminous Starburst at High Redshift: Valiante, E., Lutz, D.,
Sturm, E., Genzel, R., Tacconi, L.J., Lehnert, M. D., Baker, A.J. 2007,
ApJ, 660, 1060
Author: Dieter Lutz, Date: 09/06/2008
Back to the top
Nuclear streaming in NGC 1068
We are pursuing a programme to map the distribution and kinematics of
stars and gas in nearby active galactic nuclei, using adaptive optics
to reach high spatial resolution.
This time, we report the first direct observations of neutral,
molecular gas streaming in the nucleus of NGC1068 on scales of
<30 pc using SINFONI near-infrared integral field spectroscopy.
At a resolution of 0.075", the flux map of molecular hydrogen emission
around the nucleus reveals two prominent linear structures leading to
the AGN from the north and south. The kinematics of the gas in these
features are dominated by non-circular motions and indicate that
material is streaming towards the nucleus on highly elliptical or
parabolic trajectories whose orientations are consistent with that of
the disk plane of the galaxy. This is interpreted as evidence for
fuelling of gas to the central region from scales of 30pc to scales of only a few parsecs. One of the infalling clouds lies
directly in front of the central engine. This is understood as a
streamer tidally disrupted that form the optically thick outerpart of
an amorphous clumpy molecular/dusty structure which contributes to the
Flux map of the H_2 1-0S(1) emission in the central 0.4"x0.4" of
NGC1068 with a resolution of 0.075" (~5 pc). The peak of the
non-stellar continuum is represented by a crossed circle. The open
triangles show the projected trajectory of the northern concentration
of gas (the tongue). The half-crosses show the past trajectory of the
gas which is currently located in front of the AGN (the core).
Fig 2:Velocity map of the molecular gas extracted
SINFONI datacube in the central 0.4"x0.4" of NGC1068. A crossed circle
indicates the peak of the continuum emission. The open triangles show
the projected trajectory of the northern concentration of gas (the
tongue). The half-crosses show the past trajectory of the gas which
is currently located in front of the AGN (the core).
Publications from this work:
- Molecular gas streamers feeding and obscuring the active nucleus
Mueller Sanchez F., Davies R., Genzel R., Tacconi L., Eisenhauer F., Hicks E., Friedrich S., Sternberg A., 2008
A Close Look at Star Formation around Active Galactic Nuclei,
Davies R., Mueller Sanchez F., Genzel R., Tacconi L., Hicks E., Friedrich S., Sternberg A., 2007, ApJ, 671, 1388
- The Role of Molecular Gas in Obscuring AGN,
Hicks E., Davies R., Malkan M., Genzel R., Taaconi L., Mueller Sanchez F., Friedrich S., Sternberg A., 2008
Starburst and Torus Evolution in AGN,
Vollmer B., Beckert T., Davies R., 2008
Author: Francisco Müller-Sanchez, Date: 09/06/2008
Back to the top
Zooming into Nearby Active Galactic Nuclei: NGC3227
We are pursuing a programme to map the distribution and
kinematics of stars and gas in nearby active galactic nuclei, using adaptive
optics to reach high spatial resolution.
The main aims of the project are
(i) determine the extent and history of star formation and its relation to the
(ii) measure the properties of the molecular gas, and understand its relation
to the obscuring torus;
(iii) derive black hole masses from spatially resolved stellar kinematics.
As part of this study, we have performed a detailed study of the
Seyfert 1 galaxy NGC3227 at a spectral resolution of 70km/s and a spatial
resolution of 0.085", equivalent to only 7 pc.
Line and continuum maps of the central 75 pc of NGC3227, from SINFONI
Top left: the 2.1μm continuum is dominated by non-stellar light
associated with the AGN, which is unresolved.
Top centre: the stellar continuum is easily resolved.
Top right: the molecular gas exhibits a wealth of detail in its morphology.
Bottom left: much of the ionised gas originates from the recent star
Bottom right: the fairly weak coronal lines excited by the AGN are only seen
close around it.
The nuclear star forming region around the Seyfert nucleus is
spatially resolved, on scales of a few parsecs to a few tens of
The most recent episode of intense star formation began ~40 Myr
ago but has now ceased.
Within 30 pc of the AGN this starburst still accounts for 20-60% of
the galaxy's bolometric luminosity.
Despite showing evidence for moderate rotation, the stars' kinetic
energy is dominated by random motions indicating that they lie in a
Black Hole Mass Measurement
Schwarzschild modelling of the stellar kinematics lead to a black hole
mass in the range 7x106 to 2x107 solar masses.
The upper end is consistent with (although still less than) previous
estimates made using other techniques.
The large range arises through a degeneracy in whether mass is
attributed to the black hole or the stars and gas, which can be
resolved with better kinematic line profiles.
The gas in the central 80 pc exhibits several critical
properties that are expected of a molecular obscuring torus: its
spatial extent is a few tens of
parsecs, it is geometrically thick, and it has a high column density.
This argues that the gas we have observed is the torus.
Moreover, based on the similarity of their spatial extents and their
kinematics, it is likely that the gas and stars are physically mixed.
Thus the torus also supports episodes of active star formation.
It seems unlikely that the current level of AGN or star forming
activity can inject sufficient energy into the ISM to maintain the
vertical thickness of the torus.
However, this was possible when the star formation rate was at its
We speculate that the torus may heat and subsequently cool, changing
its vertical profile, as the star
formation and AGN go through active and quiescent phases.
Publications related to this work:
- The Star-Forming Torus and Stellar Dynamical Black Hole Mass in
the Seyfert 1 Nucleus of NGC3227:
Davies R., Thomas J., Genzel R., Mueller Sanchez F., Tacconi L.,
Sternberg A., Eisenhauer F., Abuter R., Saglia R., Bender R., 2006
ApJ, 646, 754
SINFONI adaptive optics integral field spectroscopy of the Circinus
Mueller Sanchez F., Davies R., Eisenhauer F., Tacconi L., Genzel R., 2006,
A&A, 454, 481
Author: Ric Davies, Date: 04/12/2006
Back to the top
Evolution of colour and polarisation parameters of emission from SgrA*
The supermassive black hole in the Galactic Centre, SgrA*, was detected in the
near infrared not before 2002 when 8-m class telescopes with AO camera systems
became available. Since then, NIR emission has been observed photometrically,
spectroscopically, and polarimetrically several times. This emission appears
in form of outbursts of radiation (flares), typically lasting for 1 to 3 hours.
In June 2005 we observed a long (more than 2.5 h), bright (K~15) flare in K
band with the AO assisted integral field spectrometer SINFONI at the VLT.
The emission showed variations on time scales of about 20 minutes. Even more
exciting, we found that the spectral colour index of the emission and the
brightness of the source were correlated: the brighter the flare, the bluer
its colour. This result is illustrated in the figure below.
SINFONI observations of the SgrA* flare in June 2005. Left: Narrow band images
extracted from SINFONI data cubes, separated into dim and bright flare phases.
In the dim phases (bottom row), SgrA* appears brighter in the longer wavelength
regime, whereas the difference is smaller in bright phases (top row). Right:
Correlation between colour index β (defined via νLν ∝ νβ)
and flux for SgrA* (dots and triangles with error bars) and S7, a
comparison star (blue diamonds). The brighter the flare is, the bluer is its
Using the NIR camera NAOS/CONICA in polarimetry mode, we were also able to
observe polarised emission from a flare in May 2006. Again variability on time
scales of ~15 minutes could be observed in several parameters like flux,
polarised flux, and polarisation fraction. Additionally the polarisation angle
showed a spectacular swing of about 70° within 15 minutes at the end of the
Observations of a polarised flare in May 2006. Top left: Sum image
of the polarimetric channels 0 and 90 degrees obtained during the flare peak.
Bottom left: Difference image of these two channels. The residual flux
at the position of SgrA* nicely illustrates the strong polarisation
(up to ~40%) of the light. Right: Evolution of source flux
(top) and polarisation angle (bottom)
with time. The strong polarisation and the swing in polarisation angle are
These recent observations allow a deeper understanding of the emission from
SgrA*. They support the dynamical emission model of a plasma hot spot orbiting
the black hole at (or close to) the innermost stable circular orbit. In this
picture a plasma bubble arises from the accretion disk due to infall of matter
or magnetic reconnection. This bubble orbits the black hole with a period
about 15-20 min, gets sheared, cools down and vanishes after few cycles.
But the observed plarimetric properties also tell us that a "pure" plasma spot
model is probably too simple; there might also be a jet component.
Publications from this work:
- Eckart A., et al. 2006b, A&A, 455, 1
- Eisenhauer F., et al. 2005, ApJ, 628, 246
- Genzel R., et al. 2003, Nature, 425, 934
- Gillessen S., et al. 2006, ApJ 640, 163
- Paumard T., et al. 2006, in prep.
- Trippe S., et al. 2006, MNRAS submitted
Author: Sascha Trippe, Date: 25/10/2006
Back to the top
High-Resolution Millimeter Imaging of Submillimeter Galaxies
Submillimeter galaxies at redshifts 1-4 are the location of extremely intense star formation - observing these systems we are witnessing
the formation of massive galaxies. Over the past years, we have
conducted a series of high resolution studies of submm galaxies at mm and near-infrared wavelengths, in order to determine their dynamics, sizes, masses, and other properties like metallicities. Determining when and
how such massive galaxies formed in the buildup of structure in the universe is an important step towards understanding galaxy evolution.
As most recent step, we have obtained ~0.6" resolution IRAM PdBI interferometry
of eight submillimeter galaxies at z~2-3.4, where we detect continuum at 1 mm
and/or CO lines at 3 and 1 mm. The CO 3-2/4-3 line profiles in five of the
sources are double-peaked, indicative of orbital motion either in a single
rotating disk or of a merger of two galaxies. The millimeter line and continuum
emission is compact; we marginally resolve the sources or obtain tight upper
limits to their intrinsic sizes in all cases. The median FWHM diameter for
these sources is <=0.5" (4 kpc). The compactness of the sources does not
support a scenario in which the far-IR/submillimeter emission comes from a
cold (T<30 K), very extended dust distribution. These measurements clearly
show that the submillimeter galaxies (SMGs) we have observed resemble
scaled-up and more gas-rich versions of the local universe, ultraluminous
galaxy population. Their central densities and potential well depths are much
greater than those in other z~2-3 galaxy samples studied so far. They are
comparable to those of elliptical galaxies or massive bulges. The SMG
properties fulfill the criteria of ``maximal'' starbursts, in which most
of the available initial gas reservoir of 1010-1011 Msolar is converted to
stars on a timescale ~3-10tdyn~a few times 108 yr.
Composite of the millimeter sizes (FWHM) of 8 SMGs. In each case, we show
an optical or near-IR image, with the SMG position and its astrometric
uncertainty (of optical/IR vs. millimeter/radio frames) shown as a thin cross.
A circle/ellipse denotes our best estimate of the FWHM intrinsic mm diameter
of the source, as estimated from our line and/or continuum data. Dotted lines
denote upper limits. A thick bar (or cross, for lensed sources with a
preferential lensing direction) marks a length of 10 kpc in the source frame.
Publications from this work:
- High-Resolution Millimeter Imaging of Submillimeter Galaxies,
Tacconi, L. J., Neri, R., Chapman, S. C., Genzel, R., Smail, I.,
Ivison, R. J., Bertoldi, F., Blain, A., Cox, P., Greve, T., Omont, A.,
2006, ApJ, 640, 228
- An interferometric CO survey of luminous submillimetre galaxies,
Greve, T. R., Bertoldi, F., Smail, Ian, Neri, R., Chapman, S. C.,
Blain, A. W., Ivison, R. J., Genzel, R., Omont, A., Cox, P.,
Tacconi, L., Kneib, J.-P., 2005, MNRAS, 359, 1165
- SPIFFI Observations of the Starburst SMM J14011+0252:Already Old, Fat, and Rich by z=2.565, Tecza, M., Baker, A. J., Davies, R. I.,
Genzel, R., Lehnert, M. D., Eisenhauer, F., Lutz, D., Nesvadba, N.,
Seitz, S., Tacconi, L. J., Thatte, N. A., Abuter, R., Bender, R., 2004,
ApJ 605, L109
- Interferometric observations of powerful CO emission from three
submillimeter galaxies at z=2.39, 2.51 and 3.35, Neri, R., Genzel, R.,
Ivison, R.J., Bertoldi, F.,Blain, A.W., Chapman, S.C., Cox, P., Greve, T.R.,
Omont, A., Frayer, D.T., 2003, ApJ, 597, L113
- Spatially resolved millimeter interferometry of SMM J02399-0136:
A very massive galaxy at z=2.8, Genzel, R., Baker, A.J., Tacconi, L.J.,
Lutz, D., Cox, P., Guilloteau, S., Omont, A. 2003, ApJ, 584, 633
Author: Dieter Lutz, Date: 12/09/2006
Back to the top
Messages from the Abyss
During observations of the Galactic Centre with the new Very Large
Telescope adaptive optics (AO) imager NACO on 9 May 2003, we
observed a powerful flaring by a factor of 5 in the H-band (1.65µm) emission
towards SgrA* (Fig. 1). The flare lasted for 30 min. Its rise and
decay can be well fitted by an exponential of timescale 5 min. In a second
observing run in June 2003, we observed two more flares on two consecutive
days, this time in the Ks band (2.16µm). The K-band flares rose to a factor of 3
above the quiescent level, and each lasted for 85 min. Their characteristic
rise/decay times were 2 to 5 min. Both flares exhibited significant and similar
temporal substructure (Fig. 2). The 16 June flare showed five major peaks
spaced by 13 to 17 min, resembling a 15-40%, quasi-periodic modulation of the
overall flare profile. The 15 June flare had three major peaks separated by 14
and 17 min, followed by a weaker peak 28 min later.
Detection of variable near-infrared emission from SgrA*. Raw H-band (1.65-µm)
AO images (40 mas full-width at half-maximum, FWHM) of the central 1" of the Milky
Way, obtained with the NACO AO imager9,10 on UT4 (Yepun) of the ESO VLT, before and
during the H-band flare on 9 May 2003. The image scale is linear. The integration time for
each image was 60 s, from six 10-s individual exposures. The time (in minutes from the
beginning of the set at 6 h 59 min 24 s (UT) on 9 May 2003) is shown in the box in the
upper right of each image. The images were sky-subtracted, flat-fielded and corrected for
bad pixels. North is up and east to the left, scales are for an assumed distance of 7.94 kpc
(25,880 light years). The unique infrared wavefront sensor was used to close the loop of
the AO system on the bright supergiant IRS7, 5.5" north of SgrA*. The fraction of the
power in the diffraction-limited core (Strehl ratio) is about 50% (visible seeing 0.45"
FWHM). The position of the 15-yr-orbit star S2 is marked by a cross, and the
astrometric location of the black hole is marked by a circle.
The power spectrum analysis for both flares (Fig. 2) exhibits a
significant peak with a time period of 16.8 ± 2 min, thus confirming the
reality of the structure seen directly in the light curves. The comparison star
S1 clearly does not show such a quasi-periodic structure. However, the question
arises whether this periodicity is truly a fundamental and significant property
of all SgrA* flares, or whether it is
caused by fluctuations in a ‘red noise’ power spectrum. More data will tell,
but the fact that two events separated by about 93 periods show similar
substructure is very suggestive. Because of the lack of continuous time
coverage, we cannot make a statement on the substructure of the May H-band
flare. Finally we found a fourth flare in re-analysing earlier archival L0-band (3.76µm) NACO
data taken on 30 August 2002. This flare rose to 70% above the quiescent
emission (Fig. 2), and had a decay time of 10 min. The infrared flares all
originated from within a few milliarcseconds, or a few hundred Schwarzschild
radii, of the black-hole position (Table 1). That position was determined from
the focus of the best-fitting Kepler orbit of the star S2.
Light curves of the SgrA* infrared flares and quiescent emission in 2002-03.
Flux densities were extracted from the Lucy deconvolved and beam restored images
with two aperture sizes. Error bars (±1sigma) indicate the combined statistical and
systematic uncertainties. SgrA* data are shown as filled blue circles (connected with a
solid curve). For comparison, the light curves of the nearby star S1 are shown as light red
crosses (0.2" southwest of SgrA*, left panel of Fig. 1). S1 has flux densities comparable to
the SgrA* flare state. In all cases the times are relative to the UT time listed above
each graph. Arrows in d and e mark the substructure peaks discussed in the text.
a, Light curve of the 30 August L0 -band flare. As S2 and SgrA* cannot be spatially
separated in this epoch, we have subtracted 8.8 mJy to account for the contribution of S2.
b, Light curve of the 9 May H-band flare. The gap in the data between t - t0 = 23 and
37 min was due to sky observations during that time. c, Power spectra of the flares in
panels d and e, and S1, normalized by their high frequency noise. d, Ks-band light curve
on 15 June 2003. Between t - t0 = 37 and 46 min, the AO system was not operational.
Because of our choice of the dichroic for the AO system, the signal-to-noise ratio of this
flare is not as good as in the 16 June flare. For better presentation, the flux densities of S1
were multiplied by a factor of 2. e, Ks-band light curve on 16 June 2003. The time
structure of this flare may have the tendency to chirp (periods decrease with time).
Author: Thomas Ott, Date: 10/29/2003
Back to the top
© Infrared and Submillimeter Astronomy Group at MPE
09/06/2008, editor of this page: Thomas Ott