1

1. Introduction

NAOS+CONICA (hereafter NACO)saw first light on November 25, 2001, atVLT UT4 (YEPUN). NACO partially com-pensates the effects of atmospheric tur-bulence (seeing) and provides diffrac-tion-limited resolution for observingwavelengths from 1 to 5 µm, resulting ina gain in spatial resolution by a factor of5 to 15 (diffraction limit of an 8-m-classtelescope in K-band corresponds to 60mas). This article gives an overview ofthe main characteristics and sciencedrivers of NACO and briefly summaris-es the first results obtained during com-missioning. Prospective users of NACOare kindly asked to cite [7] and [13] asa general reference to NACO in theirscientific papers.

2. History

Adaptive Optics (AO) has a long his-tory at ESO, stretching back to the earlyoperations of COME-ON, COME-ON+,

No. 107 – March 2002

NAOS+CONICA at YEPUN:First VLT Adaptive Optics System Sees First LightW. BRANDNER, G. ROUSSET, R. LENZEN, N. HUBIN, F. LACOMBE, R. HOFMANN, A. MOORWOOD, A.-M. LAGRANGE, E. GENDRON, M. HARTUNG, P. PUGET, N. AGEORGES, P. BIEREICHEL, H. BOUY, J. CHARTON, G. DUMONT, T. FUSCO, Y. JUNG, M. LEHNERT, J.-L. LIZON, G. MONNET, D. MOUILLET, C. MOUTOU, D. RABAUD, C. RÖHRLE, S. SKOLE, J. SPYROMILIO, C. STORZ, L. TACCONI-GARMAN, G. ZINS

Figure 1: NAOS (light blue) and CONICA (red) attached to the Nasmyth B adapter ofYepun/UT4.

Dear Readers,With this issue, I am leaving the Editorship of THE MESSENGER, a task which I valued and enjoyed. TheEditorship now passes to my ESO colleague Peter Shaver. MARIE-HÉLÈNE DEMOULIN

2

tween NAOS and CONICA is alsotaken into account. Last but not least,NAOS can handle VLT M2 beamchopping in order to provide CONICAwith the best observing conditionsin the thermal infrared.

The NAOS Real Time Computer(RTC) constitutes a highly optimisedsymbiosis of hardware and softwareelements ([12]), and is capable of con-trolling the Adaptive Optics at fre-quencies up to 444 Hz. It continuouslymonitors the turbulence and the op-timisation of the correction, taking intoaccount the detected brightness of thereference source. It has been designedto easily facilitate the upgrade for aLaser Guide Star (see [2] for moredetails on the VLT Laser Guide Starfacility).

Initial testing indicates that NAOSis delivering full correction (on-axisStrehl ratio of up to 50% in K-band foraverage atmospheric conditions onParanal) for a reference source with V= 12 mag. Still higher Strehl ratiosmight be achievable after various in-strument parameters have been finetuned. The adaptive optics loop hasbeen closed on point sources as faintas V = 17.5 mag (vis. WFS) and K = 12mag (IR WFS).

3.2 CONICA

CONICA ([7]) operates with the f/15beam, which is passed through NAOSfrom the Nasmyth B focus of UT4.CONICA is equipped with a 1024 × 1024pixel ALADDIN2 InSb detector (pixelsize 27 × 27 µm). The ALADDIN2 de-tector is sensitive to radiation between0.9 and 5 µm, and is operated at a tem-perature of 27K. Instrumental back-ground is below 1 e–/s. CONICA uses

3. Instrumental Characteristicsand Science Drivers

3.1 NAOS

NAOS ([13]) is a state-of-the-art AOsystem optimised for operation at theVLT telescopes, and designed to pro-vide a Strehl ratio (SR) of ≥ 70% in theK band when wavefront sensing on abright natural reference source underaverage atmospheric conditions onParanal (yet not compromising its com-pensation capabilities on very faintsources – V ≥ 16 mag and K ≥ 12 mag).Active optical elements of NAOS in-clude a tip-tilt plane mirror and a de-formable mirror (DM) with 185 actua-tors. NAOS is equipped with twoShack-Hartmann type wavefront sen-sors (WFS) for wavefront sensing in theoptical (450 to 950 nm) and in thenear-infrared (0.8 to 2.5 µm).

Both the visual and infrared wave-front sensor can be operated either in14 × 14 array configuration with 144 ac-tive subapertures (in the pupil) or in 7 ×7 array configuration with 36 activesubapertures. A range in fields of viewper subaperture, temporal samplingfrequencies, and pixel scales (binning)of the WFS are available in order toprovide optimal performance over avery wide range of observing conditionsand reference source characteristics.Dichroics split the light between thewavefront sensing channel (NAOS IRor vis WFS) and the science channel(CONICA). Reference sources forwavefront sensing can be selected atangular separations of up to 60″ fromthe on-axis science target. Guiding onreference sources with a relative mo-tion is of course possible, and the dif-ferential atmospheric refraction be-

and finally ADONIS at the ESO 3.6-mtelescope on La Silla (see [1], [6]).Based on this experience, an early de-cision was made to equip the VLTcoudé foci with AO systems ([9]).

CONICA (the COudé Near-InfraredCAmera), a 1 to 5 µm infrared camerafor diffraction-limited imaging and spec-troscopy, was proposed by a consor-tium of the Max-Planck-Institut für As-tronomie, Max-Planck-Institut für Extra-terrestrische Physik, and the Osser-vatorio Astronomico di Torino in re-sponse to an ESO Call for Proposals. Itwas selected in 1991 as one of thefirst-generation VLT instruments. Sub-sequent re-organisations in the VLTschedule led to the (temporary) can-cellation of the AO equipped coudéfoci and the loss of the OsservatorioAstronomico di Torino to the CONICAproject. In 1994 studies for imple-menting AO at the VLT Nasmyth fociwere initiated ([5]) and, in parallel, aneffort started to re-design CONICAand transform it into an instrumentfor a Nasmyth focus. In 1997, a con-sortium consisting of Office Nationald’Etudes et Recherches Aérospatiales(ONERA), Observatoire de Paris andLaboratoire d’Astrophysique de l’Ob-servatoire de Grenoble was awarded acontract to develop and build theNasmyth AO System (NAOS) for theVLT.

After a period of integrated testingat the CNRS facilities in Bellevue, nearParis, NACO was shipped to Para-nal in October 2001 and achieved itsfirst light at UT4 at the end of November2001.

Figure 1 shows NACO mounted atNasmyth B of UT4. More detailed infor-mation can be found athttp://www.eso.org/instruments/naco

Figure 2: The triple system T Tauri. Shown on the left is an open-loop image (obtained through a narrow-band filter at 2.16 µm with an expo-sure time of 0.4 sec). Only the 0.7″ binary T Tauri North and South is resolved, while the close binary T Tauri South itself remains unresolved.The image in the middle was obtained with the same instrumental set-up, but with the AO loop closed. The 0.1″ binary T Tauri South is nownicely resolved. The image to the right finally shows a 2D long-slit K-band spectrum of both components of T Tauri South.

3

page http://www.eso.org/instruments/naco)

The operation of NAOS and its inter-nal configuration will be largely trans-parent to the user. This is achieved bythe NAOS Preparation Software (PS,[8]). The PS determines the optimalNAOS configuration given externalconstraints such as brightness of thereference source used for wavefrontsensing, angular separation betweenreference source and science target, oratmospheric conditions. The result ofthis optimisation is saved in an AO-con-figuration file (*.aocfg), which in turn isread by the Phase 2 PreparationSoftware (P2PP). In addition, it pro-vides the user with information on theexpected performance taking into ac-count observing conditions. Finally, itdelivers to the CONICA ETC the pa-rameters characterising the imagepeak intensity and the width after AOcorrection.

Estimates for required exposuretimes can be obtained from the CONI-CA exposure time calculator (ETC).The CONICA ETC is closely related toother ETCs for VLT instruments andshould thus feel familiar to most users.As a rule of thumb, the overall through-put of NACO is about half that ofISAAC, and required exposure timescan be scaled accordingly. Because ofthe smaller “footprint” of the diffraction-limited Point Spread Function (PSF) ofNACO compared to the seeing limitedPSF of ISAAC, point source sensitivityin the sky-background limited casetends to be higher for NACO than forISAAC.

Similar to other VLT instruments,NACO observations are supported by adedicated set of acquisition, observingand calibration templates. Using P2PP,these templates can be grouped in ob-serving blocks in order to define observ-ing sequences. For each observing block,the AO-configuration file created by thePS has to be specified only once in thecorresponding acquisition template.

4.2 Observing with NACOat the VLT

At the time of writing, NACO is still inits commissioning phase, and not all in-strument modes have been validated,yet. Initially it is expected to offer imag-ing and polarimetry from 1 to 4 µm aswell as long-slit spectroscopy (R = 500to 2000) and coronography, though thelatter two only with a restricted set ofslits and grims, and coronographicmasks, respectively. The cryogenicFabry-Perot and observations in theM-band (still requiring more testing ofthe chopping mode) will very likely onlybe offered at a later stage.

A typical observing sequence con-sists of presetting the telescope to thescience target, acquisition of a guidestar by the telescope’s active optics,

group identified both Galactic and ex-tragalactic science drivers:

Galactic astronomy

• Outflows and disks of young stellarobjects

• Search for low-mass/substellar com-panions of nearby stars

• Structure of young embedded objects• Ionisation fronts in HII regions• Galactic centre• Close companions and circumstellar

material around T Tauri stars• Structure of Red Giant envelopes

Extragalactic astronomy

• Quasars and host galaxies • Emission line imaging of super-lumi-

nous IRAS galaxies • Search for Black Holes in centres of

galaxies • Resolved images of radio jets and hot

spots • Obscured quasars • The cosmic distance scale

In addition, NACO will facilitate highspatial resolution studies of Solar sys-tem objects and astronomical targets inthe solar neighbourhood.

4. Observing with NACO

4.1 Preparation of Observations

All AO observations depend on theavailability of a reference source suit-able for wavefront sensing. Thus acrucial part in the preparation of obser-vations with NACO is to check for sucha reference source in the proximity tothe science target. The minimal re-quirement for partial correction byNACO will be a point source with V <17.5 mag or K < 12 mag within 60″ ofthe science target. Due to the presenceof angular anisoplanatism (see [14]),the best AO correction is achieved inthe on-axis case (science target identi-cal with reference source). Online ver-sions of the Guide Star Catalog 2 andthe 2MASS point source catalog aidthe prospective user of NACO in thesearch for suitable reference sources(follow links on the main NACO web

the ESO-developed IRACE readoutelectronics and acquisition software([10]). The read-out noise is 40e– fordouble correlated read-out (ReadResetRead), and can be reduced to ≤ 10e–

for longer exposures by using Fowlersampling (FowlerNSamp). Sustainabledata rates can be as high as 2 MByte/s(i.e., one full frame saved to disk everytwo seconds), and exposure timescan be as short as 0.2 sec for fullframe read-out (or even shorter for win-dowed read-out) and for short se-quences of up to 8 exposures or byusing the NDIT > 1 option (framesaveraged by IRACE).

CONICA houses seven cameras,which provide Nyquist sampling for thecorresponding bands between 1 and 5µm (see Table 1). Four of the camerasare optimised for observations be-tween 1 and 2.5 µm with pixel scales of13.25 mas/pixel (C50S camera), 27.03mas/ pixel (C25S camera), 54.1 mas/pixel (C12S camera) and 107 mas/pixel (C06S), and three of the camerasare optimised for observations between2 and 5 µm (C25L, C12L, C06L withcorresponding pixel scales). The fieldof view for individual cameras rangesfrom 13″ × 13″ (C50S) to a circularaperture of 73″ diameter (C06S andC06L).

A large variety of optical analysingelements can be rotated into theCONICA beam. For broad- and nar-row-band imaging one can selectamong 34 different filters. Polarimetricobservations are supported both by twoWollaston prisms (split approx. 3.4″ inK-band for dual imaging observations)and four wiregrids (“wide field”). Forspectroscopy one can choose betweenfour grisms (R = 500 to 2000, with var-ious orders covering the range from 1to 5 µm) and a cryogenic Fabry-Perotwith a resolution of around 1000.Coronography is supported by aLyot-type coronograph with mask diam-eters ranging from 0.7″ to 1.4″.

3.3 Science Drivers

The science drivers for NACO wereearly on defined by the CONICAInstrument Science Team chaired byMarie-Hélène Ulrich. The working

Camera f-ratio Scale

MagnificationFOV Spectral

[mas/pixel] [arcsec] range

C50S 52.7 13.25 3.50 13 × 13 1–2.5 µmC25S 25.8 27.03 1.72 27 × 27 1–2.5 µmC12S 12.8 54.3 0.85 54 × 54 1–2.5 µmC06S 6.36 109.3 0.425 73 Ø 1–2.5 µm

C25L 25.7 27.12 1.71 27 × 27 2.5–5.0 µmC12L 12.7 54.7 0.85 54 × 54 2.5–5.0 µmC06L 6.33 109.9 0.42 73 Ø 2.5–5.0 µm

Table 1: List of available Cameras with f-ratios, scales, magnification, field of view, wave-length range.

4

K-band spectrum of the two compo-nents of T Tauri South.

Io, the innermost of the four Galileanmoons of Jupiter and the most vol-canically active place in the SolarSystem was imaged with NACO onDecember 5, 2001. At the time of ob-servations, Io’s diameter of 3660 kmcorresponded to an angular diameterof 1.2″. Despite this relatively smalldiameter, NACO was able to resolvea wealth of details. Many of the vol-canic peaks clearly stand out (Fig.3). With VLT instruments like NACO itwill be possible to continue the “vol-cano watch” on Io from the ground nowthat the Galileo spaceprobe has ex-pired.

Observations of the Solar System’s“Lord of the Rings” were even morechallenging than the observations of Io.While CONICA’s and hence Yepun’sfield of view had to be steadied onSaturn, NAOS had to track the smallSaturn moon Tethys, which was usedas a reference source for wavefrontsensing, while the active optics ofYepun had to track a star used for de-termining active optics corrections andautoguiding. The initial experimentworked very well, and validated theability of NACO to do off-axis wavefrontsensing on “moving targets”, whilesteadying the science beam on a targetwith different sidereal motion. Observa-tions like these can be used to monitorweather patterns on Saturn, study thefine structure of the rings, and track themovement of smaller moons.

In the morning of February 4, 2002,in the final hour of the 2nd commis-sioning run, Yepun was pointed to-

be made available through the eclipselibrary ([4]).

5. Initial Results and ScienceProspects

While commissioning of NACO is stillongoing, it has been delivering impres-sive astronomical results virtually fromits first hour of operation. Initial resultshave been publicised in two press re-leases aimed at providing the ESOcommunity and the general public witha first impression on the unique capa-bilities of NACO at ESO VLT:

http://www.eso.org/outreach/press-rel/pr-2001/pr-25-01.html

http: //www. eso.org/outreach/press-rel/pr-2002/phot-04-02.html

Science demonstration targets ob-served include solar-system objectssuch as Io and Saturn, a nearbyfree-floating binary brown dwarf, near-by star-forming regions such as theBecklin-Neugebauer object with the as-sociated Kleinmann-Low nebula, TTauri stars, evolved stars such asFrosty Leo or Eta Car, crowded stellarfields in starburst clusters, and extra-galactic objects.

A series of 2.16 µm narrow-band im-ages of the prototypical pre-main se-quence triple star T Tauri nicely demon-strates the effect of adaptive optics.Shown on the left-hand side of Figure 2is an open-loop image. The 0.4-sec ex-posure freezes the speckle pattern,which in a longer exposure will averageout to a seeing-limited image with a res-olution of 0.5″. The middle image hasbeen obtained with the AO correctionturned on. The right-hand side shows a

then the acquisition of the referencesource with NAOS, check of exposuretimes, and finally the start and execu-tion of the observing sequence.Because of the need to acquire a refer-ence source (and to frequently openand close the loop – for instance in thecourse of dithering or autojitter se-quences) instrumental time overheadsfor NACO tend to be higher than, e.g.,for observations with ISAAC. All thesesteps are, however, handled by thetemplates and the software controllingNACO, and are hence transparent tothe user ([15]).

Calibration data for the most com-monly used modes will be obtained ona regular basis by ESO. For more spe-cialised observing modes the individualusers are requested to include the cali-bration needs in the time request.

4.3 Archiving and pipeline

As soon as a NACO data file is savedon the instrument workstation, it is au-tomatically transferred to the archivecomputer on Paranal. All NACO datawill be incorporated in the VLT archive,and be searchable through the data-base. After the end of the proprie-tary period, data will become avail-able to astronomers from ESO mem-ber countries by access to the VLTarchive.

A data pipeline is being developed inorder to facilitate quick-look analysis ofthe data and assessment of the dataquality. Pipeline-reduced data will alsobe distributed together with the rawdata files through the VLT archive([11]). Dedicated software tools will

Figure 3: left: Io, the volcanic moon of Jupiter, as imaged with NACO on December 5, 2001, through a near-infrared, narrow-band filter at2.166 µm. Despite the small angular diameter of Io of only about 1.2″, many features are visible at this excellent angular resolution of 60 masas provided by NACO. Shown on the right is a composite of the same exposure with another obtained at a longer wavelength (L’-filter at3.8 µm), with a latitude-longitude grid superposed. Many of the volcanic peaks clearly stand out in the longer-wavelength image.

5

wards the free-floating binary browndwarf 2MASS1426316+155701 ([3]).Despite atmospheric seeing as deter-mined by the Astronomical Site Moni-tor at a wavelength of 500 nm varyingbetween 0.8″ and 1.2″, and an averageairmass of 1.4, NAOS was able to

close the loop on the R = 15.6 mag(B–R = 1.0 mag) binary brown dwarf.Figure 5 shows the closed-loop K-bandimage (60 sec exposure time), anda 5-min K-band spectrum nicely sepa-rating both components of the 0.15″binary. For a distance of 18 pc to

2MASS1426316+155701, 0.15″ corre-spond to a projected separation of 3AU. Observations like these can beused to monitor the orbital motion ofnearby binary brown dwarfs.

Already in the second night of thefirst commissioning run, NACO closed

Figure 4: This image shows the giant planet Saturn, as observed with the VLT NAOS-CONICA Adaptive Optics instrument on December 8,2001. The distance between Earth and Saturn was 1209 million km. It is a composite of exposures in two near-infrared wavebands (H andK) and displays well the intricate, banded structure of the planetary atmosphere and the rings. Note also the dark spot at the south pole atthe bottom of the image.

Figure 5: The faint (R = 15.6 mag) free floating binary (0.15″) brown dwarf 2MASS1426316+155701 nicely resolved by NACO under non-op-timal observing conditions. Left: 60-sec K-band image. Right: difference of two 300-sec K-band spectra of the brown dwarf binary. The twobinary components with K = 12.2 mag and K = 12.8 mag are clearly detected.

6

• From ESO: P. Amico, P. Balles-ter, J. Beckers, J. Brynnel, C. Cava-dore, L. Close, F. Comerón, J.-G. Cuby,C. Cumani, N. Devillard, R. Dorn, G.Finger, M. García, G. Gillet, G. Huster,J. Knudstrup, M. Meyer, M. Peron, A.Prieto, L.P. Rasmussen, D. Silva, J.Stegmaier, M.-H. Ulrich, O. van derLühe, G. Wiedemann.

References

[1] Beuzit, J.-L., Brandl, B. et al. 1994, TheMessenger 75, 33.

[2] Bonaccini D., Hackenberg, W. et al.2001, The Messenger 105, 9.

[3] Close, L., Potter, D. et al. 2002, ApJ566, 1095.

[4] Devillard, N. 1999, ADASS 8, 333.[5] Hubin, N., Bertrand, T., Petitjean, P.,

Delabre, B. 1994a, SPIE 2201, 34.[6] Hubin, N., Beuzit, J.-L., Gendron, E.,

Demailly, L. 1994b, ICO-16 SatelliteConference on Active and AdaptiveOptics, ed. F. Merkle, p. 71.

[7] Lenzen, R., Hofmann, R. et al. 1998,SPIE 3354, 606.

[8] Marteau, S., Gendron, E. et al. 2000ADASS 9, 365.

[9] Merkle, F., Hubin, N., Gehring, G.,Rigaut, F. 1991, The Messenger 65, 13.

[10] Meyer, M., Finger, G., Mehrgan, H.,Nicolini, G., Stegmaler, J. 1998, SPIE3354, 134.

[11] Quinn, P.J., Albrecht, M.A. et al. 1998,SPIE 3349, 2.

[12] Rabaud et al. 2000 SPIE 4007, 659.[13] Rousset, G., Lacombe, F. et al. 2000,

SPIE 4007, 72.[14] Tessier, E., Bouvier, J., Beuzit, J.-L.,

Brandner, W. 1994, The Messenger 78,35.

[15] Zins, G., Lacombe, F. et al. 2000, SPIE4009, 395.

CONICA proposal was selected morethan a decade ago. Adaptive Optics willenable the VLTI to reach still faintermagnitudes, opening a wide range ofastronomical objects for research.More details on VLT AO systems anddevelopment will be presented in futurearticles in The Messenger.

These projects ensure that the VLTwill stay competitive throughout thisdecade, and give European astronomya leading edge in studies of the uni-verse.

Acknowledgements:Contributors to the CONICA andNAOS Projects

The NACO project would not havebeen possible without the dedication ofmany more people than included in thelist of authors of the present article oncommissioning activities. An (undoubt-edly incomplete) list includes:

• From ONERA: J.M. Conan, L.Rousset-Rouviere, L. Te Toullec.

• From LAO Grenoble: J.-L. Beuzit,G. Chauvin, P. Feautrier, P. Kern, P.-Y.Magnard, P. Rabou, E. Stadler.

• From Observatoire de Paris: R.Arsenault, D. Bauduin, G. Calvet, C.Collin, J. Deleglise, P. Gigan, B. Lefort,M. Lopez, C. Marlot, S. Marteau, G.Michet, J. Montri, G. Nicol, S. Pau, V.D.Phan, B. Talureau, D. Rouan, O. Ron-deaux, S. Servan, B. Taureau, P. VanDuoc.

• From MPIA Heidelberg: H. Becker,A. Böhm, P. Granier, W. Laun, K. Meix-ner, N. Muench, A. Tuscher, K. Wagner.

the loop on the blue supergiant Sher25, which is located at a distance of 7kpc in the NGC 3603 starburst region.A comparison to an HST/WFPC2 im-age of the same region highlights that –provided a reference source suitablefor wavefront sensing is available –NACO at Yepun can achieve in theK-band the same spatial resolution asHST/WFPC2 in the I-band (Figure 6).

6. Outlook: NACO and Beyond

A 3rd (and final) commissioning runof NACO is planned for the periodMarch 22 to April 4, 2002. After this,two runs for instrument “Paranalisa-tion”, and subsequently dry runs (in-cluding science verification) areplanned before the new VLT instrumentis deemed ready for science opera-tions. At the time of writing, it is plannedto offer NACO, the 5th VLT instrumentto become operational, from Period 70on to the general ESO community.More details on the proposal processcan be found in the Call for Pro-posals.

In the meantime, work on a panoplyof other advanced AO systems for theVLT and VLTI is in progress. The LGSfacility at UT4 is expected to come on-line in 2003. Both NACO and SINFONI(an integral field NIR spectrograph fedby a curvature sensing AO system) willgreatly benefit from the increased skycoverage facilitated by a laser guidestar. With the implementation ofMACAO-VLTI the coudé foci of all thefour VLT Unit Telescopes at last willalso be equipped with AO as was ini-tially envisioned when the original

Figure 6: Left: this image of NGC 3603 was obtained with the WFPC2 camera on the Hubble Space Telescope (HST) in the I-band (800 nm).It is a 400-sec exposure and shows the same sky region as in the 300-sec NACO image shown to the right. The image nicely demonstratesthat Yepun with NACO is capable of achieving in the K-band the same spatial resolution as HST/WFPC2 at three times shorter observingwavelength of 800 nm (I-band). The HST image was extracted from archival data. HST is operated by NASA and ESA.

This article is based on the Executive Summary of the European proposal for ALMA, which was preparedwith input from many persons, including J. Baars, R. Bachiller, A. Benz, A. Blain, D. Bockelée-Morvan, R.Booth, C. Carilli, F. Combes, J. Conway, P. Cox, J. Crovisier, A. Dutrey, E. Falgarone, T. Forveille, M. Guélin,S. Guilloteau, C. Henkel, R. Kurz, E. Lellouch, J. Lequeux, J-F. Lestrade, R. Lucas, K. Menten, R. Pallavicini,J. Richer, P. Shaver, L. Tacconi, E. van Dishoeck, M. Walmsley, T. Wiklind.

7

ALMA will be comprised of 64 12-me-tre diameter antennas of very high pre-cision, with baselines extending up to12 km. Figure 1 shows an artist’s im-pression of a portion of the array in acompact configuration. The array of an-tennas will be reconfigurable, givingALMA a zoom-lens capability. The high-est resolution images will come fromthe most extended configuration, andlower resolution images of high surfacebrightness sensitivity will be providedby a compact configuration in which allantennas are placed close to each oth-er. The instrument thus combines theimaging clarity of detail provided by alarge interferometric array together withthe brightness sensitivity of a large sin-gle dish. The large number of antennasprovides over 2000 independent inter-ferometer baselines, making possibleexcellent imaging quality with “snap-shot” observations of very high fidelity.The receivers will cover the atmospher-ic windows at wavelengths from 0.3 to10 millimetres. ALMA will be located onthe high-altitude (5000 metre) Llano de

frequency domain for redshifts largerthan 3.

Progress in mm receiver perfor-mance, which now approaches quan-tum limits, in fast digital electronics, in-cluding photonics techniques, and inantenna design coincide to make pos-sible an order of magnitude jump inproject scale for millimetre arrays. Thetime is ripe for ALMA to be a major stepfor astronomy, making it possible to studythe origins of galaxies, stars, and plan-ets. With a capability of seeing star-forming galaxies across the Universeand star-forming regions across the Gal-axy, it will open new horizons in science.

ALMA will be a millimetre/submillime-tre counterpart of the VLT and HST,with similar angular resolution and sen-sitivity but unhindered by dust opacity.It will be the largest ground-based as-tronomy project after the VLT/VLTI,and, together with the Next GenerationSpace Telescope (NGST), one of thetwo major new facilities for world as-tronomy coming into operation at theend of this decade.

1. Introduction

The Atacama Large Millimetre Array(ALMA) is one of the highest priorityprojects in astronomy today, combiningthe aspirations of scientists in Europe,the U.S. and elsewhere around theglobe. The project has now reached acritical milestone: this is the year whenthe project officially enters the con-struction phase. It has already been ap-proved for construction funding by theU.S., and the European decision to ap-prove the construction phase is expect-ed later this year.

ALMA will image the Universe at mil-limetre and submillimetre wavelengthswith unprecedented sensitivity and an-gular resolution. This frequency rangeis unique in providing access to the low-est transitions from most simple mole-cules, and in its sensitivity to the ther-mal emission from cold gas and dust inthe clouds where stars are forming. Theexpansion of the universe also bringsthe mid-infrared peak of emission fromactive star-forming galaxies into this

Figure 1. An artist’s impression of several of ALMA’ s antennas in a compact configuration.

The Atacama Large Millimetre ArrayR. KURZ1, S. GUILLOTEAU2, P. SHAVER1

1European Southern Observatory, Garching, Germany2Institut de Radio Astronomie Millimétrique, Grenoble, France

8

sees the activities of an ALMA Execu-tive Committee (AEC) and severaltechnical project teams, with advicefrom international Scientific and Man-agement Advisory Committees.

The design and development Phase1 of the project is being completed atpresent, and the construction phase(Phase 2) is scheduled to begin, pend-ing final decisions, with completionforeseen in 2011. The total estimatedcost for Phase 2 of the bilateral projectis 552 million (year 2000 US$), to beshared equally between Europe andNorth America.

Japan has also been working to-wards a project of this kind, the LargeMillimetre and Submillimetre Array(LMSA). The Japanese astronomicalcommunity has decided that it would bebest to fully merge this project with thebilateral project. A Japanese designand development phase, which isclosely co-ordinated with the Europeanand North American efforts, will becompleted in 2003. Japanese participa-tion will provide significant scientific en-hancements to the bilateral project, andthis possibility has been extensivelystudied. If it is realised, ALMA will be-come a truly global project, the firstever in ground-based astronomy.

3. Science with ALMA

3.1 Introduction

The scientific case for ALMA is over-whelming. The main science driversare the origins of galaxies, stars andplanets: the epoch of first galaxy for-mation and the evolution of galaxiesat later stages including the dust-obscured star-forming galaxies thatthe VLT and HST cannot see, and all

that of a 50-50 partnership betweenEurope and the U.S., with joint overalldirection. Three aspects were studiedin detail – scientific, technical and man-agement – and a feasibility study waspublished in April 1998.

The framework for the formal Euro-pean collaboration in Phase 1 of thisproject (the 3-year design and develop-ment phase) was established in De-cember 1998. A European Co-ordina-tion Committee (ECC) was created todirect the European effort, with partici-pation and funding from the EuropeanSouthern Observatory (ESO), the Cen-tre National de la Recherche Scienti-fique (CNRS), the Max-Planck-Gesell-schaft (MPG), the Netherlands Foun-dation for Research in Astronomy(NFRA) and Nederlandse Onderzoek-school Voor Astronomie (NOVA), andthe United Kingdom Particle Physicsand Astronomy Research Council(PPARC). In early 2000 the SwedishResearch Council (VR), and the Insti-tuto Geográfico Nacional (IGN) andMinisterio de Ciencia y Tecnología(MCYT) of Spain were added to theEuropean collaboration.

In June 1999 a formal agreement be-tween Europe and the U.S. regardingcollaboration on Phase 1 of a bilateralproject, now called the Atacama LargeMillimetre Array (ALMA), was signed.The U.S. side of the partnership is ledby the National Radio Astronomy Ob-servatory (NRAO), operated by Asso-ciated Universities, Inc. (AUI) under aco-operative agreement with the Na-tional Science Foundation (NSF). Re-cently, Canada has formally joined theU.S. in the North American collabora-tion. The overall direction for the proj-ect is provided by an ALMA Co-ordi-nation Committee (ACC), which over-

Chajnantor, east of the village of SanPedro de Atacama in northern Chile.This is an exceptional site for milli-metre astronomy, possibly unique inthe world.

2. Project Development

To make such an ambitious projectpossible, ALMA has become a joint en-deavour involving many nations andscientific institutions, and it is likely thatit will become the first global project inground-based astronomy – an essen-tial development in view of the ever-in-creasing complexity and cost of front-line astronomical facilities.

For years Europe has had a major in-volvement in millimetre astronomy, withseveral of the leading facilities in theworld. Many institutions throughoutEurope have active research pro-grammes in this field, and several ofthem have developed technical expert-ise in this area together with Europeanindustry, so it is natural that Europeshould be a primary participant inALMA.

The idea of a large European south-ern millimetre array (“Large SouthernArray”, LSA) has been discussed since1991, and in 1995 an LSA project col-laboration was formally established toexplore the possibility in a two-yearstudy which included site surveys inChile and critical technical studies. A re-port published in April 1997 concludedthis first phase.

An important step was taken in June1997. A similar project had also beenunder study in the U.S. since 1984, the“Millimetre Array” (MMA), and anagreement was made to explore thepossibility of merging the two projectsinto one. The basic principle was

Figure 2: Left: Sensitivity of ALMA, compared with some of the world’s other major astronomical facilities, for typical integration times of sev-eral hours. Right: Angular resolution of ALMA, compared with other major telescopes. The top of the band shown for ALMA corresponds tothe compact 150-m configuration, and the bottom corresponds to the large array with 12 km baselines. For the VLT, the solid line correspondsto the seeing-limited case, and the dashed line to the diffraction-limited case with adaptive optics.

9

tant Universe: the discovery of COemission in a z = 2.3 ultraluminous in-frared galaxy, the discovery of the far-infrared background radiation, and thediscovery of a large population of star-forming galaxies that probably domi-nate the luminosity of the Universe athigh redshift. The most remarkable dis-covery is that large amounts of dustand molecules were present already atz = 4.7. This redshift corresponds to alook-back time of 92% of the age of theUniverse and shows that enrichment ofthe interstellar medium occurred at veryearly epochs.

It is now clear that the mm/submmwavebands are exceptionally well suit-ed for the study of the distant Universe.Whereas the broadband flux from dis-tant galaxies is diminished in the UVand optical due both to the redshift andobscuration by internal dust, the samedust produces a large peak in the rest-frame far-infrared, which, when red-shifted, greatly enhances the millimetreand submillimetre emission from theseobjects. Thus, ALMA may provide oneof the best ways to find the first galax-ies that formed after the “dark ages”.

Current studies are limited to the verybrightest objects. ALMA will make itpossible to detect objects one hundredtimes fainter, and will make a decisivecontribution to one of the key questionsin current astronomy: the origin of theinfrared background and the star-for-mation history of the Universe. The“ladder” of molecular transitions essen-

phases of star and planet formationhidden away in dusty cocoons and pro-toplanetary disks. But ALMA will go farbeyond these main science drivers – itwill have a major impact on virtually allareas of astronomy. It will be a truemm/submm counterpart of the othermajor facilities for world astronomy, asillustrated by its relative performance inFigure 2.

3.2 Galaxies and Cosmology

Three dramatic events over the lastdecade have spectacularly opened upthe mm/submm wavebands to the dis-

Figure 3: A detailed view of the typical spectral energy distribution of a star-forming galaxy atredshifts of 0.1 to 10. In the millimetre and submillimetre bands the observed flux is almostindependent of redshift, and ALMA will be sensitive to such objects out to redshifts well be-yond 10 (Guiderdoni et al., 2001, in “The Birth of Galaxies”, eds. B. Guiderdoni et al., TheGioi Publ., p. 95).

wavelength λ (in µm)

Figure 4. The submillimetre and opticalwavebands provide complementary views ofthe Universe. This figure shows submillime-tre contours superimposed on an optical im-age of the galaxy cluster A1835 (Ivison et al.,2000, MNRAS 315, 209). It is clear that thesubmillimetre sources are very faint opticalobjects, while the red cluster galaxies arenot prominent submillimetre sources. Themm/submm wavebands are particularly sen-sitive to the most distant galaxies; thesource 14011+ 0253 is at a redshift of 2.55,as determined directly from CO line obser-vations. ALMA will provide submillimetre im-ages with much finer resolution than the op-tical image

10

galaxies can be resolved with linearresolutions of a few parsec. ALMA willbe able to map both the gas and thedust that obscure the nuclei. The pres-ence of central black holes can be stud-ied kinematically in a large number ofgalaxies. The centre of our own Galaxycan be observed free of obscuration, inparticular the gas dynamics of the 1-pccircumnuclear disk around the galacticcentre source Sgr A*.

ALMA will make observations of nor-mal galaxies at z = 1–2 with the samedetail as is presently possible in nearbygalaxies. The main dynamical featuresof nearby spirals will be observed withenough resolution and sensitivity toconstrain theoretical scenarios ofgalaxy evolution. The mass spectrumof molecular clouds in galaxies of dif-ferent types will be determined. De-tailed studies of nearby mergers and IRluminous galaxies will be important toserve as templates for objects foundat high redshifts. In the MagellanicClouds, star-formation processes canbe compared with those in our Galaxy.This would be highly interesting, be-cause star formation is closely relatedto the ambient radiation field, dust con-tent, and metallicity.

3.3 The Formation of Starsand Planets

A major astronomical goal of the 21stcentury is an understanding of howstars and planets form. Studying star-and planetary-system formation re-quires very high angular resolution, be-cause protoplanetary disks are small(100–500 AU) and the nearest star for-mation regions are ~ 100 pc away.ALMA will provide a linear resolution asfine as 1 AU at these distances.

Because it can observe at wave-lengths free of extinction, ALMA will bethe premier instrument for studying howgas and dust evolve from a collapsingcloud core into a circumstellar disk thatcan form planets. The array will be ableto directly observe astrophysical phe-nomena that have until now only beenconjectured in theoretical models of theearly stages of star formation. ALMAwill yield new unique information on thegravitational contraction of protostellarcloud cores, with accurate kinematicsand mass distributions inside the coresand their envelopes. It will give newclues on the role of the magnetic field inthe cloud cores, the circumstellar en-velopes, and the accretion disk. Ob-servations of high-excitation submmlines of various molecules will allow usto study the physics and chemistry ofthe shocks in the ubiquitous outflow jetsthat carry away the original angular mo-mentum.

For the later stages, when the newly-formed stars are surrounded by proto-planetary disks, imaging the gas anddust on scales of several AU will be

goals of ALMA, and one of the mainreasons for a very large collecting area.

Many gravitational lenses will befound by ALMA, possibly more numer-ous and at higher redshifts than in theoptical or radio wavebands because ofthe very steep source count. Gravita-tional arcs will be mapped in molecularlines.

Molecular absorption lines will be ob-served in the spectra of many quasars.This is a new field with great potential.Already, in the few sources brightenough, over 30 transitions from 18 dif-ferent molecules have been observedin absorption systems up to z = 0.9.This opens up the study of detailedchemistry at cosmological distances,and makes possible direct measure-ment of the cosmic background tem-perature at high redshifts. Thousandsof sources will be accessible to ALMA.

Active galactic nuclei can be studiedin depth at millimetre wavelengths be-cause of the low synchrotron and dustopacity and the unprecedented angularresolution of millimetre VLBI. The opti-cally-obscured molecular tori and thecircumnuclear starbursts of nearby

tially guarantees that a redshifted spec-tral line will appear in one of the ob-serving bands. ALMA will be thus ableto obtain the redshifts of distant galax-ies, and study their detailed morpholo-gy and kinematics. It will be able to de-tect not only molecular lines from theseobjects, but potentially also the atomicfine-structure lines of carbon, oxygen,and nitrogen, which, at high-z, are red-shifted into the submm bands.

At present, even the strongest sub-millimetre sources are very difficult toidentify. Most of them are not asso-ciated with previously known bright ob-jects, yet this population probably dom-inates the luminosity of the distantUniverse. With ALMA’s high angularresolution, precision and sensitivity, itwill be possible to accurately locatesuch sources in minutes, and mea-sure their redshifts through the detec-tion of CO lines in less than an hour.ALMA may also detect a population ofmore distant, optically obscured, ob-jects that would escape detection atother wavelengths.

The study of the early epochs ofgalaxy formation is one of the main

Figure 5. The detectability of the powerful [C II] line by ALMA. The coloured lines representthe integrated line flux densities one would observe for the 158 mm C II fine structure linefrom the sample of ultraluminous infrared galaxies observed by Luhman et al. (1998, ApJ 504,L11) using the ISO satellite if those galaxies were at the redshifts indicated by the abscissa.The thin line indicates the typical 5-sigma noise level of ALMA in two hours of integration, fora velocity resolution of 300 km s–1, and assuming the precipitable water vapour content of theatmosphere is 0.8 mm.

11

outflow speeds and high densities, sothat matter easily condenses into dustgrains. ALMA will image the distributionof matter in the outflows at distances ofa few stellar radii, to solve the long-standing problem of dust formation andstudy the interaction between stellarpulsations and wind acceleration. It willbe possible to study such objectsacross the Galaxy, and even in theMagellanic Clouds

Supernovae and gamma-ray burstswill both be important targets for ALMA.In both cases mm and submm obser-vations provide unique and importantinformation. Radio supernovae first ap-pear at these wavelengths, where theflux is relatively unaffected by free-freeand synchrotron self absorption. SN1987A in the LMC will be a prime targetfor ALMA. In the case of gamma-raybursts, mm/submm observations pro-vide unique information on the peak ofthe burst and important constraints onthe physical parameters. ALMA will al-low detection of all GRBs detectable inthe optical.

3.5 The Solar System

Because of its high angular resolu-tion, its fast imaging capabilities, and itswide instantaneous bandwidth, ALMAwill represent a major step forward inthe study of comets, asteroids andplanets. ALMA’s highest angular resolu-tion at a distance of 1 AU correspondsto a linear resolution of less than tenkilometres.

Observations of comets with ALMAwill greatly increase our understandingof their nature and origin, and comple-ment the planned space probes that will

receiver design. If successful, ALMAcould address two of the most interest-ing solar physics problems: the accel-eration of the highest energy electronsin flares, and the thermal response ofthe low chromosphere to waves andshocks from the interior.

ALMA will yield fundamental knowl-edge for our understanding of the dy-namics and chemistry of the envelopesof evolved, oxygen-rich and carbon-richstars, where important scientific goalsare the understanding of dust formationand the enrichment of the interstellarmedium with heavy elements. Thewinds of these red giant stars rapidlyremove the outer layers, terminatingfurther evolution. The winds have low

the only way to study the earlieststages of planet formation. Current mmarrays have revealed large (hundredsof AU) rotating disks around single TTauri stars, but the angular resolutionnecessary to resolve the inner regions,where planets are expected to form,will only be provided by ALMA. ALMAwill be able to reveal, within proto-planetary disks, the gaps that aretidally cleared by Jovian sized planetsat distances of a few AU from theiryoung, central stars. Multi-wavelengthstudies of such objects will be power-ful tools for analysing the dust and gasproperties on the scale of the SolarSystem. Maps of dust and optically thinmolecular lines with 0.1″ to 0.05″beams will provide crucial data on thechemistry, the reservoirs of the bio-genic elements, and the timescales onwhich planets form. ALMA will providethe masses of the pre-main-sequencestars through the measurement of theKeplerian motion in the protoplanetarydisks.

The high sensitivity of the array willallow us to make unbiased surveys ofpre-main-sequence stars to obtain thestatistics of disk properties and fre-quency of protoplanetary systems indifferent star-forming regions. Com-parison between isolated star formationand dense clusters will become possi-ble. ALMA will also address the high-mass star-formation problem. Dense,hot cores are known to exist aroundmassive (proto)-stars, but the existenceof circumstellar disks analogous tothose found around low mass star re-mains uncertain.

The study of star and planet forma-tion is another major goal of ALMA, andone of the drivers for the highest angu-lar resolution.

3.4 Stars and their Evolution

With its coverage of the millimetreand submillimetre ranges, ALMA willgreatly expand the field of stellar as-tronomy. It will detect tens of thousandsof stars over the entire H-R diagram. Itwill cover the full life cycle of stars.ALMA may provide the only way to ob-serve some massive stars that spendtheir entire brief lives hidden by dust intheir parent molecular clouds. It willprovide unique information on thewinds of hot stars, novae, the photos-pheres of giants and supergiants, andnon-thermal processes in flare stars,Be stars, and dust formation in super-novae and in the outflows of planetarynebulae. It will resolve the photos-pheres and chromospheres of giantand supergiant stars within a few hun-dred parsec.

ALMA is designed to be able to ob-serve the nearest star, our sun. This isa challenging problem because of thethermal constraints on the antenna and

Figure 6. ALMA will reveal the details of planet formation. The figure shows a hydrodynamicmodel of a protostellar accretion disk in which a giant protoplanet is forming (Bryden et al.,1999, ApJ 514, 344). The newly-formed protostar resides (invisibly) at the centre of the ac-cretion disk and a Jupiter-mass protoplanet orbits around it at the Jupiter–sun distance. A gapin the disk is cleared out by the protoplanet. Using ALMA it will be easy to image such gapsin protostellar disks in nearby star-forming regions. In this plot surface density is coded as“height”.

Figure 7. Distributions of the SiS and CN 3-mm emission in the CSE IRC+10216. Thecolours represent the line intensities inte-grated in a narrow velocity interval centredon the systemic velocity of the star (Guélinet al., 1996, in “Science with Large Milli-meter Arrays”, ed. P. Shaver; Springer Publ.,p. 276).

12

both. There were thus several trade-offs to be considered. Small antennashave higher precision and give betterwide-field imaging. The use of large an-tennas maximises the collecting area,and reduces the number (and there-fore cost) of receivers and the demandson the correlator. In view of thesupreme importance of high sensi-tivity, the largest possible antenna sizewas chosen. The combination of point-ing and surface accuracy required forefficient operation at submillimetrewavelengths (respectively 0.6″ and 25µm rms or better) is difficult to achievefor antenna diameters greater thanabout 12 metres, so this determinedthe antenna size. An array of 6412-metre diameter antennas providesa total collecting area of over 7000square metres, satisfying the sensitivityrequirements given above. This largenumber of antennas also provides ex-cellent high-resolution imaging capabil-ity, which will be very important for thescience objectives of ALMA. The 2016independent baselines remove theneed to use Earth rotation to provideaperture synthesis, and allows highresolution “snapshot” images of high fi-delity.

Thus, the angular resolution and sen-sitivity requirements are satisfied byALMA, an array of 64 12-metre diame-ter high-precision antennas extendingover a region up to 12 km across. Thereceivers should provide completewavelength coverage of the atmos-pheric bands over the range 0.3 to 10mm; the dewars will be built to accom-modate all ten receivers needed, andwill be populated initially with thefour of highest priority. Other require-ments include wide instantaneousbandwidth (16 GHz per antenna), aflexible correlator system allowingspectral resolution as high as 5 kHz(0.01 km s–1 at 100 GHz), and a com-plete polarisation capability providingall Stokes parameters. The site mustobviously be large, flat, and very high(to minimise the atmospheric attenua-tion at these wavelengths), and the5000-metre high Llano de Chajnantorin the Atacama desert region of north-ern Chile is ideal. Even on such an ex-ceptional site, atmospheric pathlengthfluctuations are critical for imaging per-formance, and ALMA is designed withwater vapour radiometers to compen-sate for these variations, as well asa fast switching capability to “freeze”the atmosphere.

More information on the ALMA proj-ect can be found in earlier issues of TheMessenger (March 1996, March 1998;December 1998; and June 1999), andon the following websites:

http://www.eso.org/projects/alma/http://www.alma.nrao.eduhttp://www.nro.nao.ac.jp/~lmsa/

index.html.

the stratosphere of Titan with highspectral resolution will provide thevertical and latitudinal distributions ofthese constituents, giving better con-straints on the photochemistry thatoccurs in Titan’s atmosphere and itsresponse to seasonal effects. ALMA willhave sufficient sensitivity to detect andmap CO, and perhaps other species,such as HCN, in the tenuous atmos-pheres of Pluto and Triton. This will pro-vide clues on the nature of the interac-tion between their icy surfaces and theiratmospheres.

The scientific reach of ALMA thus ex-tends from the most distant objects inthe Universe to details of the nearestobjects in our solar system. It will beone of the major astronomical facilitiesof the 21st century.

4. Scientific and TechnicalRequirements

High angular resolution is of greatimportance both for observations of thedistant Universe and for detailed stud-ies of the processes of star and planetformation nearby in our own Galaxy. Itis clear from HST observations thatan angular resolution of at least 0.1″ isneeded for high-redshift studies. Simi-larly, an angular resolution of 10 milli-arcsec or better is required to resolvethe gaps in protoplanetary disks cre-ated by forming planets, and suchresolution should be achieved at leastat the shorter (submillimetre) wave-lengths accessible to ALMA. Both re-quirements imply baselines of 10 km orgreater.

Such high angular resolution cannotbe exploited without adequate sensi-tivity. The noise in brightness tempera-ture increases as the square of thebaseline. However, millimetre astrono-my is the domain of cold matter, so thebrightness temperatures to be ob-served are low. In the case of spectrallines, bandwidths are limited by the linewidths, so increasing detector band-width does not help. Furthermore, mod-ern receiver performance is approach-ing quantum limits and/or the atmos-pheric noise limits. Therefore, the onlyway to increase the sensitivity is to in-crease the collecting area of the array.An angular resolution of < 0.1″ can onlybe achieved for thermal lines with acollecting area approaching 104 squaremetres. The other main driver for veryhigh sensitivity is the detection of themost distant galaxies in the Universe. Ifgalaxies formed by successive mergersof sub-galactic objects, the highest pos-sible sensitivity will be needed to detectthe first luminous objects.

The large size and collecting areacan only be achieved with a large arrayof antennas. The collecting area of anarray can be enhanced by increasingthe number of antennas, their size, or

be able to sample only a few comets.Over 20 molecular species have so farbeen discovered in comets. ALMA willmake it possible to search for lessabundant molecules, radicals, and newions, and to investigate isotopic ratiosin several species. Such studies willprovide key information on the origin ofcomets and the formation of the SolarSystem. It will be possible to detectmolecules in distant comets and studythe evolution of their outgassing as theyapproach the Sun. The fast high-reso-lution imaging capability of ALMA willallow us to study structures in the innercoma of comets, and maps of the dis-tribution of rotational temperatures ofdifferent molecular species will help usstudy the thermodynamics, excitationprocesses, and physical conditions inthese objects.

Asteroids and cometary nuclei ofsmall sizes, and even distant objectssuch as Centaurs and trans-Neptunianobjects, will be detectable in the mmand sub-mm continuum. Together withobservations in other wavebands,ALMA will allow us to probe the tem-perature of these objects at variousdepths and to measure their albedoand size. Imaging thermal emissionfrom the planetary satellites, the Pluto/Charon system and the largest aster-oids will provide clues to their thermalproperties and the degree of hetero-geneity of their surface.

ALMA will be able to map planetaryatmospheres on short timescales.Maps of CO and HDO in Mars andVenus will give data on wind, tempera-ture, CO and water distribution, and at-mospheric dynamics on spatial scalescomparable to regional weather scales.The analysis of meteorological and cli-matic variations in the atmosphere ofMars will be a valuable complement tofuture space missions. Searching formolecular trace species likely to bepresent in these planets, such as sul-fur-bearing compounds in Venus andorganic species in Mars, will becomepossible. Wide bandwidth capabilitieswill allow us to probe the deep atmos-phere of Venus. Mapping HCN and CO,and searching for other nitriles onNeptune, will provide information onwhether the origin of such molecules isinternal or external. It will be possible todetect and map tropospheric speciessuch as PH3 in the Giant Planets.During very dry conditions at the high-altitude site proposed for ALMA, themapping of H2O and HDO on the fourgiant planets will provide clues on theorigin of water.

ALMA will also observe the atmos-pheres of Pluto and the satellites of thegiant planets. ALMA will be able todetect SO2 and SO in the plumesof the volcanoes on Jupiter’s moon Ioand may discover other trace con-stituents. Mapping the millimetre linesof CO, HCN, HC3N, and CH3CN in

13

News from the 2p2 TeamL. GERMANY

Service Mode

The percentage of time allocated toservice mode observations at the 2.2-mis increasing yet again in Period 69. Wewould just like to take this opportunity toremind the PIs of upcoming ServiceMode programmes: please read the web-page describing our Standard Calibra-tions before submitting your Phase 2package. It is different to the StandardCalibrations performed at the VLT. Wenow have 2xU filters, 2xB filters and 2xIfilters but only one of each is part of theStandard Calibrations. Please check:http://www.ls.eso.org/lasilla/Telescopes/2p2T/E2p2M/WFI/CalPlan/StdCal/to learn which of the filters is coveredin the standard calibrations.

We also have a “Service Mode Tips”page at: http://www.ls.eso.org/lasilla/Telescopes/2p2T/E2p2M/WFI/SMTips.html with suggestions on how best to organ-ise your OBs and observing pro-grammes in order to maximise thechances of them being executed.

New Filters

We have recently mounted a newB filter in the ESO/MPG Wide FieldImager. This filter has superseded theB/99 filter as part of the StandardCalibrations for service mode pro-grammes. Service Mode PIs note: if youwish to use the B/99 filter for yourprogramme, you can no longer requestStandard Calibrations in your READMEfile.

The zeropoints for this filter and forthe new U360 filter can be found at:http://www.ls.eso.org/lasilla/Telescopes/2p2T/E2p2M/WFI/zero-points/ and the transmission curves at:http://www.ls.eso.org/lasilla/Telescopes/2p2T/E2p2M/WFI/filters/

NGAS

To deal with the vast quantities ofdata that will be produced by ALMA andthe new instruments to arrive at theVLT, the NGAS (Next GenerationArchive System) project was initiatedby ESO’s Data Management Division.In this system, magnetic disks are usedto store and transport the data betweenChile and Garching, rather than thelower capacity CDs, DLTs and DVDsthat are currently used.

Since the ESO/MPG Wide Field Im-ager already produces large amountsof data in a short period of time, the 2.2-m was chosen as the test bench for theimplementation of the NGAS. We havebeen using this system since July 2001with the final commissioning and ac-ceptance of NGAS at the 2.2-m per-formed in December 2001. In the sec-ond half of 2001, the 2p2 team wassending both magnetic disks and DLTsto Garching for archiving; however, asof January 2002, only the magneticdisks of the NGAS now make thejourney.

To learn more about the NGAS sys-tem please visit: http://arcdev.eso.org/NGAST/

ESO 1.52-m Telescope

Fans of the Boller and Chivens spec-trograph may be relieved (or perhapssorry) to see the old HP1000 interfacego and the new gui interface come on-line. The new system is more user-friend-ly and as of October 2001 has seen theend of an era as the big pizza-wheels(magnetic tapes) are no longer thread-ed – all data are now directly written toDAT tape. Never fear though – theHP1000 is still on hand, sleeping in thebackground, and the pizzas are readyto spin again if the new technology fails.

October 2001 also saw the end of theold REOSC control unit (which con-trolled dome movement and the wind-screens) in the dome. For those of uswho don’t speak French, we can nowread and understand the positions ofthe switches on the new control panel.

Danish 1.54-m Telescope

Service MissionIn November 2001, and in collabora-

tion with a team from CopenhagenUniversity Observatory, a completemaintenance and upgrade of DFOSCand the systems at the Danish 1.54-mtelescope was undertaken.

The full report can be found at: http://www.ls.eso.org/lasilla/Telescopes/2p2T/ D1p5M/RepsFinal/#dfosc

We would like to thank the team fromCopenhagen University Observatorywho came out to La Silla to work on thiswith us: Anton Norup Soerensen, Mor-ten Liborius Jensen and Jens Klougart.

A view of the Residencia at sunset,

as seen from the Telescope Platform at the top of Paranal. A narrow path connects the top withthe Residencia and invites to a nice 45-min downhill walk! Photo: Massimo Tarenghi.

14

2. Place Chilean astrophysics in an in-ternationally highly competitive posi-tion in several major areas of currentinterest.

3. Educate and train a new generationof young astronomers to take full ad-vantage of the unique opportunitiesavailable in Chile to carry out world-class research, and help create ca-reer opportunities for young as-tronomers in Chilean institutions.

4. Make the Chilean public highlyaware of work and discoveries bynational astrophysicists and use as-trophysical research as a model topromote national pride.The Centre will be hosted by the

Universidad de Chile at its AstronomyDepartment in Cerro Calán, and willhave as associated organizations theAstronomy Department of the PontificiaUniversidad Católica and the PhysicsDepartment of the Universidad de Con-cepción. In the next section I describethe present status of Astronomy inthese institutions, as well as other re-cent centres.

II. Existing Institutes

Universidad de Chile

Astronomy is one of the scientific dis-ciplines with the longest tradition at theUniversidad de Chile. It began in 1852,when the Chilean Government estab-lished the National Observatory in as-sociation with the Universidad de Chile.During the early 60’s the build up ofmodern international observatories(Cerro Tololo, Cerro La Silla, and CerroLas Campanas) prompted the Univer-sidad de Chile to create the AstronomyDepartment at the Observatorio Nacio-nal, starting up the first programme ofLicenciatura en Astronomía in 1966and a decade later the current Magisterin Astronomy programme. Many gradu-ates of these programmes went toobtain their Ph.D. degrees in USA,Canada, and Europe, returning in theearly 80’s to the Department, making adeep impact in astrophysical researchand teaching. Two of these early grad-

leaders in the area. The Centre mem-bers will tackle, through the use of thecurrently available mega-telescopes,various problems of fundamental scien-tific importance, covering the origins ofa broad range of objects. They will in-vestigate origins from the largestscales, by studying aspects of galaxyformation and evolution, to the smallestscales, by studying aspects of the col-lapse of an individual star, facing can-dent problems such as: What did theuniverse look like 10 billion years ago?How do galaxies form? How do low-mass and high-mass stars form?Where do planets, complex moleculesand the beginnings of life came from?Significant progress toward the mainscientific goals of the Centre will onlybe achieved if individuals studying dif-ferent aspects of origins in astrophysicsare brought together for the fruitful andmuch needed interchange of differentviews and ideas.

The Centre’s mission is to pave theway, through the research of theirmembers and the education of the newgenerations of astronomers, for identi-fying and setting up the basis of thenew problems to be tackled with thenew generation of instruments to be setin Chile, in particular with ALMA. Thisworld’s largest array will produce someof the most exciting science in astrono-my over the following 10–20 years. Itwill be the best instrument for detectingproto-galaxies at very high redshifts inthe early Universe, as well as both pro-tostars and proto-planetary disks innearby star-forming clouds. With a res-olution higher than the Hubble tele-scope, and a sensitivity 10 times largerthan anything else currently in exis-tence, it will be the largest and mostpowerful millimetre array in the world.

To accomplish its mission, the FON-DAP excellence centre has set for itselfthe following strategic objectives in sci-ence, education, and outreach:1. Transform the nature of Chilean as-

trophysics from individual researchefforts into a coherent, co-ordinatedand collaborative endeavour in sci-ence and education.

Chile is becoming the astronomicalcapital of the world. The Atacamadesert encompasses the best locationson Earth to build astronomical observa-tories. Paranal, Las Campanas, Tololo,Pachón, and La Silla are well-knownplaces that contain top-quality opticalastronomical facilities open to Chileanastronomers. In the last few years anew generation of astronomical facili-ties, the mega-telescopes, have beenconstructed in those sites: VLT,Magellan, and Gemini. In addition,there will be near San Pedro deAtacama at a height of 5000 m, themost powerful radio synthesis tele-scope of the world: the Atacama LargeMillimetre Array (ALMA). This tele-scope will open up a new, unexplored,window of the electromagnetic spec-trum for astrophysical studies. It will op-erate in an spectral range where cloudsof cold gas which are the placental ma-terial of mostly every object we know inthe universe, have their characteristicspectral signatures.

The privileged access of Chilean as-tronomers to this unparalleled suite ofinstruments provides them with aunique opportunity to address some ofthe most fundamental problems inmodern astrophysics. In a time span ofa few years ALMA will permit them to in-vestigate the origins of celestial ob-jects. To face these new challenges inthe best possible way it has becomenecessary that the nation provides theastronomers with the proper environ-ment and tools to undertake the re-search which should put them at thefrontiers of astrophysics.

I. Recent Developments

The Chilean CONICYT (ConsejoNacional de Ciencia y Tecnología) re-cently approved a Centre of Excellencein Astronomy, within the FONDAP(Fondo Nacional de Desarrollo deAreas Prioritarias) programme. ThisCentre constitutes a new approachamong astronomers to generate theconditions to boost Chilean astro-physics, and place it among the world

The Messenger continues its publication of reports from the Chilean astronomical community with thePanorama of Chilean Astronomy by Professor Leonardo Bronfman, President of the Sociedad Chilena deAstronomía, followed by an article by A. Reisenegger et al. on the on-going investigation of the Shapley super-cluster.

CHILEAN ASTRONOMY

A Panorama of Chilean AstronomyL. BRONFMAN, Universidad de Chile

President, Sociedad Chilena de Astronomía

15

and physics. There are 8 graduate stu-dents and of the order of 100 under-graduate students. The DAA carries outresearch in both observational and the-oretical astrophysics. Most of the re-search is based on observations atthe telescopes of the internationalobservatories operating in NorthernChile. A significant fraction of the De-partment’s research projects involvescollaborations with European, NorthAmerican, Australian and other LatinAmerican astronomers, with fundingfrom several international agencies(NASA, NSF, etc).

The main areas of research are:Planetary Astronomy (search for extra-solar planets); Stellar Astronomy (su-pernovae, Be stars, Cepheid and RRLyrae variables; distances, ages, andmetallicities of clusters in the Galaxyand the Magellanic Clouds); Extraga-lactic Astronomy (dynamics of groupsand clusters of galaxies, evolution andstructure of Dumbell galaxies, structureand dynamics of superclusters, faintand low-surface brightness galaxies,gravitational lenses in clusters of galax-ies, quasars, large-scale structure andcosmology); and Theoretical Astro-physics (stellar evolution, compact ob-jects, extragalactic astrophysics andcosmology, hot gas and metals in clus-ters of galaxies, formation of structurein the Universe). DAA members havehad important participation in the dis-covery of the accelerated expansion ofthe Universe; the disclosure of UltraCompact Galaxies in clusters; the dis-covery and understanding of MA-CHOS; the detection of the overall col-lapse of the Shapley Superclusters; thestrong clustering of galaxies in pairsand in small groups.

With the support of FundaciónAndes, who also funds several fellow-ships, the DAA is involved in a partner-ship programme with the Department ofAstrophysical Sciences of PrincetonUniversity (USA). This programme in-cludes a joint postdoctoral prize fellow-ship in observational astronomy, facul-ty and graduate student exchange, jointorganisation of international meetingsin astrophysical areas of common in-terest, and the construction of an ob-servatory in the outskirts of Santiagofor student training. The PUC is one offour international affiliate members ofAURA (Association of Universities forResearch in Astronomy), and has builta number of partnerships with severalacademic institutions in Chile andabroad.

Universidad de Concepción

The astronomy group was created in1995, as part of the Physics Depart-ment. Its current staff includes DouglasGeisler, Wolfgang Gieren, Ronald Men-nickent, and recently Tom Richtler. Thereare 4 postdocs funded by the Comité

maximum and their rate of decline, andis the basis for studies of the expansionof the Universe; (iii) the first derivationof the mass and distribution of H2 in theGalaxy using a complete CO survey(Bronfman et al. 1988, ApJ 324, 248);(iv) the first complete CO survey of theLMC (Cohen et al. 1988, ApJ 331, 95);(v) recent searches for infall motions to-ward young stellar objects (Mardoneset al. 1997, ApJ 489, 719); and (vi) athorough review on the environmentand formation of massive stars (Garayand Lizano 1999, PASP 111, 1049).

The Astronomy Department is active-ly engaged in collaboration with a num-ber of international institutions, present-ly holding agreements with the Asso-ciation of Universities for Research inAstronomy (AURA), the CarnegieSouthern Observatory (CARSO), theUniversity of Florida, the NationalRadio Astronomy Observatory (NRAO),CalTech, the National AstronomicalObservatory of Japan (NAOJ), and theInstituto Astrofísico de Canarias (IAC).A 45-cm telescope donated by the gov-ernment of Japan, with a CCD camera1k × 1k, will arrive at Cerro Calán in2002, to be used for public outreachand student training. The Departmentoperates the Estación AstronómicaCerro El Roble, with a 90-cm Maksutovtelescope.

Pontificia Universidad Católicade Chile

The Departamento de Astronomía yAstrofísica (DAA) is part of the PhysicsFaculty of the Pontificia UniversidadCatólica de Chile (PUC). Astronomy atthe PUC started in 1929 when theUniversity received, as a donation, a36-inch Mills telescope located then atthe Observatorio Manuel Foster on theCerro San Cristóbal in Santiago. TheInstitute for Physics and Astrophysicswas founded to teach in these areas;the Institute later gave rise to the pres-ent Faculty of Physics. The Departmentof Astronomy and Astrophysics wascreated in 1996, and operates at theCampus San Joaquín in Santiago. ThePUC is promoting the growth of astro-physics by the creation of faculty andpostdoctoral positions, as well as un-dergraduate and graduate programmesof study: a Certificado Académico (mi-nor), a Licenciatura (B.Sc.), a M.Sc.,and a Ph.D in Exact Sciences.

The University, recognizing the majorcomparative advantages, has givenhigh priority to astronomy. Since 1995the number of members at the DAA hasincreased steadily. Currently, workingat the DAA there are 8 faculties: L. In-fante (chair), F. Barrientos, M. Catelan,A. Clocchiatti, G. Galaz, D. Minniti, H.Quintana and A. Reisenegger, as wellas 8 postdoctoral fellows. All membersat the DAA are heavily involved in re-search and teaching, both astronomy

uates, M.T. Ruiz and J. Maza, have ob-tained the National Science Prize. Thepresent staff of the Astronomy Depart-ment includes 14 faculty members:M.T. Ruiz (Chair), H. Alvarez, L. Bronf-man, L. Campusano, S. Casassus, E.Costa, G. Garay, P. Lira, S. López, J.May, J. Maza, D. Mardones, F. Noël,and M. Rubio, as well as 5 postdoctor-al fellows. The Astronomy Departmentis part of the Facultad de CienciasFísicas y Matemáticas, which attracts afreshman class of 500 from the top 2%of all graduating high school students,a most important asset. It is located atCerro Calán, Santiago, in a large build-ing with adequate teaching and re-search facilities, as well as the oldestand most complete astronomical libraryin the country.

The Universidad de Chile created in1999, with support from FundaciónAndes, the first Ph.D. programme in as-tronomy in the country. The programmehas currently 8 students enrolled, andis expected to significantly increase thenumber of young scientists starting acareer in Astronomy. To strengthen theareas of astrophysics not currentlycovered by the members of the De-partment, during the first few years ofthe programme part of the teaching isdone in association with Yale Uni-versity. It is expected that the pro-gramme will timely change its centre ofgravity to the Universidad de Chile. Akey for such change will be the incor-poration to the staff of theoretical astro-physicists in the next two years,through a joint programme of thePhysics and Astronomy Departments,with support of the Comité Mixto ESO-Chile. The Ph.D. programme was re-cently awarded an important grant fromthe Ministry of Education, through aMECESUP project, to hire new facultyand fund student fellowships over thenext 3 years.

Research is carried on in a number ofareas, including Quasars and AGNs(evolution of galactic nuclear regionsand the formation of massive blackholes); Large-Scale Structure and Cos-mology; Starburst Galaxies (dwarf sys-tems, metal-poor galaxies) ; High-Massand Low-Mass Star Formation (galacticdistribution, hot molecular cores, bipo-lar outflows, gas kinematics at highdensities, embedded pre-main se-quence stars); Supernovae; FaintStellar Objects (faint white dwarfs,brown dwarfs, extrasolar planets, bary-onic dark matter); ISM in the Galaxyand the Magellanic Clouds; RedGiants; Planetary Nebulae; and SolarAstrometry. Some research highlightsare: (i) the discovery of the first freefloating brown dwarf (Ruiz et al. 1997,AJ 491, 107); (ii) the Calán-Tololo su-pernovae survey (Hamuy et al. 1995,AJ 109, 1), which provided the first cal-ibration of the relationship between theabsolute magnitude of type Ia SNe at

16

the understanding of the others. Eachof the areas will be led by a PrincipalInvestigator (P.I.) which will be respon-sible for the proper advance to achievethe goals expected in his/her area of re-search. In what follows we shortly sum-marise the research areas, putting em-phasis in the connectivity betweenthem.

1. Birth and Evolution ofStructures in the Universe(P.I.: L. Infante)

One of the greatest challenges in ex-tragalactic astrophysics is to under-stand the formation and evolution ofgalaxies. Understanding the physicsunderlying the processes by whichthese structures formed and evolved isthe main thrust of modern cosmology.To reconstruct the star-formation histo-ry of the Universe beginning with theearly epochs of galaxy formation andreaching the present is one of the keyquestions in astrophysics. The Centrewill carry out studies in the followingtopics:

• Primeval galaxies. – How, at theearliest epochs, the seeds observed inthe cosmic microwave background ra-diation lead to the formation of the firstgalaxies (and stars), and how these pri-mordial low-mass galaxies merge toproduce the more massive systems wesee today.

• Galaxy clusters as tracers of Large-Scale Structure. – Galaxies cluster onall scales by gravitational forces andthis clustering evolves with time. Asgalaxies cluster on smaller scales, theyinteract more and more, leading to theirgrowth via a merging process. How,during early times, clustering process-es assemble larger galaxies by merg-ers of smaller or not fully developedones, leading to the present-day largegalaxies; and how gas with a mass anorder of magnitude larger than those ingalaxies, is trapped in the growinggravitational potential wells.

• Starburst galaxies. – The currentgeneration of mega-telescopes andunique instrumentation will make possi-ble the identification of hundreds of ex-treme star-forming galaxies at redshifts~ 3 and higher, thereby initiating de-tailed studies of star-forming galaxieswhen the Universe was still young.These studies will permit to charac-terise and trace the evolution of star-burst galaxies on a firm statistical basisand, in particular, to determine the peakepoch of star formation activity, helpingto constrain structure formation models.

• Globular Clusters and GalaxyFormation. – Being the living probes ofthe earliest epoch when most galaxiesformed, globular clusters (GCs) are anideal tool for the study of galaxy forma-tion. Centre members plan to obtain re-liable age, chemical abundance, andkinematic information for globular clus-

students unable to move away from theregion. Research is mostly oriented toExtragalactic Astronomy (intermediatered-shift clusters, X-Ray clusters).

III. The Next Decade

The FONDAP Centre of Excellenceis expected to play a major role sup-porting the formation of human re-sources in astrophysics at different uni-versities in Chile. The Ph.D., M.Sc.,and/or B.Sc. programmes in astronomyat the Universidad de Chile, PontificiaUniversidad Católica, and Universidadde Concepción will benefit stronglyfrom the efforts of the astronomers as-sociated with the Centre. They will offergraduate courses jointly to students ofall institutions, supervise students fromany institution, and visiting professorswill be encouraged to have stays atmore than one site. The Centre will alsosupport and be involved in summerschools to attract the young, science-oriented minds, to astrophysics. TheCentre is expected to become, in a ten-year frame, a world-wide recognised in-stitution that provides graduate studenttraining at the highest level of excel-lence and performs frontier research inastrophysics.

The Centre will encourage membersto perform and be involved in the exe-cution of large surveys. Chilean as-tronomers are in a unique position to doprominent surveys, maximising tele-scope time and fostering cross collabo-rations. This can be achieved by or-ganising the Chilean community, anddedicating a fraction of the available(human and technical) resources.These surveys, along with the soliddoctoral programmes at ChileanUniversities, would perhaps becomethe long-lasting legacy of the newCentre. Its present members are G.Garay (Director), M.T. Ruiz (Sub-director), L. Bronfman, D. Geisler, W.Gieren, L. Infante, D. Mardones, J.Maza, D. Minniti, H. Quintana, and M.Rubio.

The first goal of the Centre is tobroaden the research base in each as-tronomy site within the country, giving aparticular emphasis to the developmentof research areas related to the study ofthe origins of celestial objects. The sec-ond, and equally important, goal of theCentre is to strengthen the teaching oftheoretical astrophysics in all astrono-my Ph.D. programmes in the country sothat they will be fully conducted byChilean universities.

There will be five main areas of as-trophysics to be cultivated at theCentre. All of them have a distinct inte-grator: The study of origins of celestialobjects ; and a common window for fu-ture progress: Sub-millimetre wave-length observations with ALMA. Theadvance in knowledge in any of theseareas will have an immediate impact in

Mixto ESO-Chile, the Alexander VonHumboldt Foundation, and a NASAproject. Research is funded mostly byFONDECYT, as well as by FundaciónAndes and several international agen-cies.

Current fields of research include:Star clusters (galactic and extragalac-tic); Stellar populations (in the localgroup and other nearby groups of gal-axies); Distance scale calibration (Ce-pheids, RR Lyraes, red clump stars,blue supergiants, PNLF, GCLF, SN Ia);Physics of stellar standard candles;Cataclysmic variables and accretiondisks; Dwarf novae and long-termvariables; Close binary evolution andbrown dwarfs in binary systems; Oldgiants in the galactic halo; Galacticstructure. A full account of the ongoingwork can be found in Gieren et al. 2002(The Messenger 106, 15)

The Universidad de Concepción re-cently created a Doctoral programme inphysics, including astronomy. Presentlythere is one student enrolled in astron-omy and several more are expected toenter the programme in 2002. In 2003a Licenciatura en Física con Menciónen Astronomía will begin.

Universidad Católica del Nortein Antofagasta

The Instituto de Astronomía at theUniversidad Católica del Norte, of re-cent formation, operates the Observa-torio Cerro Armazones, with two tele-scopes of 84 cm and 41 cm. In 2002,with the support of the Physics De-partment also from the Facultad deCiencias, an undergraduate pro-gramme, Licenciatura en Física conmención en Astronomía, will start. TheInstitute operates, with the help of ESO,an important outreach centre, theCentro de Divulgación de la Astrono-mía, serving the northern Chile com-munity. There is presently one full-timestaff member, L. Barrera, and anotherone will be appointed this year with thehelp of the Comité Mixto ESO-Chile.The research carried on at the Instituteincludes Ground Support for SpaceMissions; Quasar Monitoring; Trans-Neptunian Objects; Craters and Me-teorites in the II Region of Chile.

Universidad de La Serena

The Universidad de La Serena hascreated a new astronomy group at thePhysics Department, with two staffmembers, A. Ramírez and H. Cuevas.Another one will be hired with the helpof the Comité Mixto ESO-Chile. ThePhysics Department, traditionally de-voted to teaching, is presently stronglysupporting the development of re-search in astronomy. The group hasfurther impact over the IV Region ofChile by providing the possibility ofstarting a career in astronomy to local

17

comprehensive view of the infallprocesses leading to either the forma-tion of an isolated low-mass star, asseen in dark globules, or to the forma-tion of a cluster of massive stars, asseen in dense molecular cores. Dolow- and high-mass stars form in aself-similar way? or do they follow adifferent formation path, possibly dueto differences in the initial conditionsof the parental gas? In the latter case,which is the determinant physical pa-rameter of the ambient gas that es-tablishes the different modes of col-lapse? With ALMA they will study thekinematic of the infalling gas close tothe forming star, the structure and kine-matics of protostellar disks, and theirrole in the formation of binary stars andplanets.

• Outflows. – One of the major astro-nomical results of the last two de-cades has been the discovery thatstar formation is accompanied by en-ergetic, collimated mass outflow. Theoutflow mechanism is unknown and re-quires study on scales as close to thestar as possible. High-velocity flowsfrom recently formed stars will be ob-served with unprecedented sensitivity,helping to elucidate the mechanismthat drives the outflow and its influencein limiting the growth of a star. They willalso investigate how the outflow phe-nomena that appear in the earlieststages of star formation affect the phys-ical and chemical properties of the en-vironment.

• Star formation in the MagellanicClouds. – The Large and SmallMagellanic Clouds (LMC, SMC), beingthe two nearest external galaxies, pro-vide the best opportunity for a detailedstudy of the stellar and interstellar mat-ter component of any external galaxy.Understanding the process of formationof stars, their interaction with the sur-rounding gas and dust, the chemicalenrichment, as well as the older popu-lation of stars in these galaxies, is thelink between the knowledge in ourGalaxy and the rest of the Universe.

5. Brown Dwarfs and PlanetarySystem Studies (P.I.: M.T. Ruiz)

• Brown Dwarfs. – Sub-stellar ob-jects, like brown dwarfs, are likely toprovide unique information on the frag-mentation processes that accompanythe formation of a single star or a clus-ter of stars. Although the first free float-ing brown dwarf was discovered onlyfive years ago, current surveys areshowing that there are plenty of theseobjects, implying that we might be im-mersed in a sea of small dark bodies.How many and how massive are they?The answer to these questions willhave an impact on the theory of stel-lar formation and might have some dy-namical consequences for the galaxyas a whole. Using the new optical/IR

tinguish between the mechanismsleading to “radio-loud” and “radio-quiet”quasars.

3. The Extragalactic DistanceScale (P.I.: W. Gieren)

• Cepheid variables as standard can-dles. – Cepheid variables are thoughtto be one of the most reliable standardcandles, and hence one of the mostpowerful distance indicators. There are,however, uncertainties, in particular theeffect of metallicity on the Cepheid fun-damental physical parameters and ontheir pulsation properties. To determinethe effect of metallicity on Cepheidproperties, Centre members will carryout a programme to discover Cepheidvariables in nearby galaxies with steepmetallicity gradients in their disk, andthen calibrate the metallicity effect onthe period-luminosity relationship. Thisprogramme is expected to lead to thefirst truly accurate empirical determina-tion of the effect of metallicity onCepheid-based galaxy distances.

• The distance to the LMC. – Al-though the knowledge of the distanceto the Large Magellanic Cloud (LMC) isa key step in the determination of thedistance ladder, its value is still a mat-ter of debate. Centre members will em-ploy the surface brightness techniquecalibrated on Cepheid variables, and atinfrared wavelengths to minimise prob-lems with interstellar absorption andmetallicity to obtain the distance with aprecision of 3%. Such a precision willmean a big leap forward in the calibra-tion of the distance scale, and theHubble constant.

4. Star Formation (P.I.: G. Garay)

A comprehensive theory of both high-and low-mass star formation is an es-sential requirement if we are to under-stand galaxy formation and evolutionand the formation of Sun-like stars andplanets. However, our knowledge ofhow stars form is still rudimentary.Several questions of a basic nature re-main unanswered: (a) What motionsoccur before and during the gravitation-al collapse of a molecular cloud core toform stars? (b) How do stars acquiretheir main-sequence mass, or howdoes the collapse and accretion stop?and (c) What is the role of disk-likestructures in the formation of singlestars, binary stars, and planets? (d)How are high-velocity bipolar jets driv-en away from the central protostar? Toaddress these questions, Centre mem-bers have identified 850 individual re-gions of massive star formation in theMilky Way, the most complete samplepresently available.

• Infall. – Members of the Centre willstudy, at sub-millimetre wavelengths,the long-sought evidence of gravitation-al infall onto very young stars, to get a

ter systems, which will provide criticalclues to reveal the formation and chem-ical evolution history of galaxies, andwill help to tightly constrain galaxy for-mation models.

• Nearby galaxies. – The nearestgalaxies are key in the understandingof galaxy evolution, since they are theonly galaxies that can be studied con-sistently on a star-by-star basis, provid-ing direct information on the distributionof ages and metallicities. Their studieswill help us to constrain the integrated-light population synthesis models usedin the interpretation of distant galaxiesnot resolved into stars.

• The Milky Way. – Our own galaxy,the Milky Way, is a unique place tostudy in great detail the formation andevolution of galaxies. We can deter-mine with exceeding resolution the in-terplay between its stellar, molecular,and neutral components, which shouldbe representative of similar galaxies.Centre members are involved in thedetermination of the large-scale distri-bution of massive star formation in ourown Galaxy, aiming to derive the bestrendition of the spiral structure of theMilky Way, using the most adequatetracers, which will become a key stonefor future comparison with the distribu-tion in external galaxies at the scale ofspiral arms.

2. Quasars and Active GalacticNuclei (P.I.: J. Maza)

Being the most luminous objects ofthe Universe, quasars and active galac-tic nuclei are splendid, but puzzling,probes of the earlier stages of cosmicevolution. Members of the Centre willcarry out research in three main topicsof this fascinating area of astrophysics.They will seek to understand how theluminosity function for quasars evolveswith time (or redshift), which is of para-mount importance in order to under-stand how massive black holes can de-velop in galactic centres and how theybecome bright quasars. They will alsoinvestigate the mechanism of the enor-mous energy production in compact ob-jects, including quasars, nuclei of giantelliptical galaxies, BL Lac objects, andcores of radio galaxies. In addition, theywill undertake observations to deter-mine the metal abundance of thefaintest population of starburst galax-ies, a project of utmost importancesince it will permit to obtain the most ac-curate value of the primordial Heliumabundance, a basic cosmological pa-rameter. Once ALMA comes on line, theresearchers in this field will get accu-rate measurements of the sub-millime-tre emission from quasars and activegalactic nuclei to clarify the relationshipbetween these enormously energeticobjects and to learn about the mecha-nism of their energy production. Thesemeasurements would be crucial to dis-

18

such systems with evolutionary modelsof our own solar system.

The present article could not havebeen written without the contribution ofthe FONDAP Centre of ExcellenceDirector, Guido Garay, and of its P.I.Members. I thankfully acknowledge thecontribution from M.T. Ruiz, Director ofthe Astronomy Department at Universi-dad de Chile; L. Infante, Chairman ofthe Pontificia Universidad Católica deChile Department of Astronomy andAstrophysics; W. Gieren, Head of theAstronomy Group at the Universidad deConcepción Physics Department; L.Barrera, from the Instituto de Astrono-mía at Universidad Católica de Anto-fagasta; and A. Ramírez, from Universi-dad de La Serena.

modern techniques such as radial ve-locities, planetary occultations (transits)and micro-lensing. Once ALMA is avail-able we will be able to undertake mo-lecular line observations of the atmos-pheres of planets and other bodieswhich will give new knowledge of plan-etary “weather”, the structure of atmos-pheric wind and the variations in chem-ical constituents. Studies of proto-plan-etary disks will be carried out using therecently available IR facilities. ALMA,with its sensitivity and resolving power,will be the ideal instrument to providedefinite answers regarding the forma-tion and evolution of proto-planetarydisks. Their images will have enoughdetail to allow astronomers to seechemical variations in proto-planetarysystems and to permit them to compare

mega-telescopes and their IR instru-mentation it will be possible to in-vestigate the physical characteristicsof these objects, particularly those inorbit around nearby stars which will al-low us to obtain their masses. ALMAwill be a perfect instrument for the fol-low-up studies of brown dwarfs found inthese studies.

• Extrasolar planets and proto-plane-tary disks. – One of the great appealsof astronomy is undoubtedly its poten-tial to help us understand the origin ofour planet. The Centre will foster thedevelopment of the area of planetaryscience, currently non-existent in thecountry, starting from available humanresources. This would be accomplishedby joining and developing searches forextrasolar planetary systems using

Dynamics and Mass of the Shapley Supercluster,the Largest Bound Structure in the Local UniverseA. REISENEGGER1, H. QUINTANA1, D. PROUST 2, and E. SLEZAK3

areisene@astro.puc.cl

1Departamento de Astronomía y Astrofísica, Facultad de Física, Pontificia Universidad Católica de Chile, Chile2DAEC – Observatoire de Meudon, France3Observatoire de la Côte d’Azur, France

Introduction

The Shapley Supercluster is thelargest bound structure identified in thelocal Universe (z < 0.1). In this article,we discuss the role of superclusters aspresent-day “turning points” in thegrowth of structure in the Universe. Wereview observations of the ShapleySupercluster and their interpretation,particularly with regard to its dynamicsand the determination of its mass,much of which has been done by ourgroup, centred at Pontificia UniversidadCatólica de Chile. Finally, we describeour recent application of a sphericalcollapse model to the supercluster, anddiscuss possibilities of future progress.

1. Cosmological StructureFormation and Superclusters

Observations of the cosmic micro-wave background radiation show mat-ter in the early Universe to be very uni-formly distributed, with large-scale den-sity perturbations as small as 1 part in105 (Smoot et al. 1992). This is in strongcontrast with the present-day Universeand its highly overdense condensa-tions, such as galaxies and clusters ofgalaxies. A natural and widely acceptedexplanation for the growth of the densi-ty perturbations is that initially slightlyoverdense regions attract the surround-

ing matter more strongly than under-dense regions. Therefore, the expan-sion of overdense regions is sloweddown with respect to underdense re-gions, and the density contrast grows.Eventually, regions of large enoughoverdensity can stop their expansionaltogether and start recontracting. Theircollapse is then followed by a processof relaxation or “virialisation”, afterwhich the resulting object is in an ap-proximate equilibrium state, in which itsstructure is only occasionally perturbedby merging with other objects. This stateis well described by the virial theoremof classical mechanics, which statesthat in such an equilibrium state thegravitational potential energy of the ob-ject is proportional to the total kineticenergy of the smaller objects randomlymoving inside it (stars in a galaxy,galaxies in a cluster of galaxies). Thistheorem allows to infer the mass of theobject (which determines the gravita-tional potential) from the measured ve-locity dispersion and size of the col-lapsed structure.

Inflationary models for the earlyUniverse predict a well-defined relationbetween the amplitude of the “initial”density fluctuations on different spatialscales. The prediction, corroborated byseveral sets of observations, impliesthat the fluctuations are largest on thesmallest scales. Since fluctuations on

aIl significant scales grow at the samerate, those on the smallest scales willfirst reach turnaround and virialisation.Thus, the chronological order of forma-tion of objects proceeds from small tolarge, i.e., from globular clusters1 togalaxies, groups of galaxies, and final-ly to galaxy clusters, the largest viri-alised known structures at present. Thenext larger objects, namely groupingsof clusters of galaxies, or superclusters,should presently be undergoing gravi-tational collapse, whereas even largerstructures should still be expanding andonly slightly denser than average.

On the largest scales, the Universe isstill undergoing a nearly uniform expan-sion. For a very distant galaxy, its red-shift factor 1 + z, defined as the ratio ofthe observed wavelength of a given linein the spectrum to the wavelength ofthe same line at its emission, is a goodapproximation to the factor by whichthe Universe expanded in all spatial di-rections while the radiation was travel-ling through it. From a Newtonian pointof view, applicable to regions muchsmaller than the Hubble length2, this is

1Structures much smaller than globular clusterscannot collapse spontaneously, since their gravita-tional attraction is not strong enough to overcomethe pressure of the intergalactic gas. Therefore,stars are formed only within collapsing or alreadycollapsed larger structures.

19

ies in Centaurus that appears to be oneof the most populous yet discovered,[...], oval in form with dimensionsroughly 2.8° by 0.8°”, and centred at theposition of the very rich cluster Shapley8 (Shapley 1933). It was later rediscov-ered, identified with an X-ray source,called SC 1326-311 (Lugger et al.1978), and is now more commonlyknown as A3558 (Abell et al. 1989).

This structure gained attention whena dipole anisotropy in the cosmic micro-wave background radiation (CMBR)was detected and interpreted as due tothe Earth’s motion with respect to thehomogeneous background frame de-fined by the CMBR (Smoot & Lubin1979 and references therein). Correct-ing for small-scale, local motions, thevelocity of the Local Group of galaxies3

with respect to the CMBR is 600km/s (e.g., Peebles 1993), approxi-mately in the direction of the structurefound by Shapley, but long forgotten.Initially, it was thought that this motionwas produced by the gravitational at-traction of the Hydra-Centaurus Super-

unrelated to their distance from us. Inthis limit, structure along the line ofsight is difficult to discern, but the dis-persion of redshifts allows to determinethe object’s mass, as discussed above.

In the intermediate regime, of interesthere, the situation is much more murky.Deviations from the uniform expansionare large, some parts of a structure maybe expanding while others are contract-ing at the same time, but no virial equi-librium has been reached. More de-tailed modelling is generally required todisentangle the three-dimensional mor-phology and the internal dynamics of thestructure. If this can be done, one of itsby-products are the masses of thesestructures, which can be used to con-strain the spectrum of initial fluctuations,an important ingredient of all models ofcosmological structure formation.

2. The Shapley Supercluster

According to recent catalogues of ag-glomerations of clusters of galaxies(Zucca et al. 1993; Einasto et al. 1997),the so-called Shapley Supercluster isby far the largest such structure in thelocal Universe, out to z ~ 0.1. Its coreregion was first pointed out by Shapley(1930), who noticed a “cloud of galax-

equivalent to write the present-day re-cession velocity of the galaxy as vr = cz= H0d, where c is the speed of light, z isthe same as in the previous definition,H0 is the Hubble parameter (see foot-note 2), and d is the (Euclidean) dis-tance between us and the object, there-fore allowing us to (approximately) inferthe distance from the measurement ofz. Knowledge of an object’s sky co-or-dinates α and δ and its redshift z allowits approximate positioning in the three-dimensional space, and large cata-logues of objects with this informationcan be used to trace the three-dimen-sional large-scale structure.

In virialised structures, on the otherhand, there is no expansion, and red-shift differences between constituentobjects are due to the local kinematicDoppler effect resulting from relativemotions within the structure, which are

Figure 1: Two projections of the distribution of galaxies with measured redshifts in the region of the Shapley Supercluster. The radial co-or-dinate, the recession velocity determined from the redshift, cz, measured in km/s, is an imperfect surrogate (see text) for the galaxy’s dis-tance to us, which would be at the vertex. The angle in panel (a) is right ascension α in hours (1 h = 15°), in panel (b) it is declination δ in de-grees. Both angles are expanded relative to their true size.

2The Hubble length is c/H0 5000 Mpc, whereH0 is the “Hubble parameter”, i.e., the present ex-pansion rate of the Universe, and 1 Mpc (mega-parsec = 3.086 × 1019 km = 3.261 million light-years, roughly the distance between our Galaxyand Andromeda or the size of a cluster of galaxies.The Hubble length is roughly the distance tra-versed by a light ray during the age of theUniverse, much larger than the size of any of thestructures being discussed here.

3This group contains our own Galaxy (the “MilkyWay”), M 31 (“Andromeda”), and 20 or so smallergalaxies, such as M32, M33, and the MagellanicClouds.

(a) (b)

20

at declination δ –30°, telescopes inChile are ideally suited for its study. Infact, Bardelli and collaborators havemade extensive use of the ESO 3.6-mfor a detailed study of the very densecondensations around the clusters A3558 and A 3528, whereas Quintanaand collaborators mostly took advan-tage of the large field (1.5° × 1.5°) of the2.5-m telescope at nearby Las Campa-nas, in order to cover a much widerarea, ~ 8° × 10°. Their papers discussquantitative and qualitative propertiesof individual clusters and the morpholo-gy of the SSC (under the assumptionthat redshifts indicate distances, at leaston the global scale of the supercluster).The most detailed discussion so far ofthe latter was given by Quintana et al.(2000), who find the SSC to consist ofseveral subcondensations with filamen-tary structures emerging and connect-ing them. Two projections of the “three-dimensional galaxy distribution” in rightascension α, declination δ, and reces-sion velocity cz are given in Figure 1.

Different methods have been used toput bounds on the mass and density ofparts of the SSC6:

(1) adding up the masses of indi-vidual clusters determined through thevirial theorem (e.g., Quintana et al. 1995,1997) or from X-ray observations (Ray-chaudhury et al. 1991; Ettori et al. 1997),which of course gives a lower bound onthe total mass, as intercluster matter isnot taken into account;

(2) estimates of the overdensity onthe basis of galaxy counts on the sky,e.g., in 2 dimensions, having to makeeducated guesses as to the SSC’sdepth and the fraction of the observedgalaxies belonging to the SSC as op-posed to the foreground or background(Raychaudhury 1989);

(3) estimates of the overdensity fromcounts in 3-dimensional redshift space,assuming that the redshift is a good dis-tance indicator (Bardelli et al. 1994,2000), i.e., that the SSC is still in the ini-tial expansion phase; and

(4) virial estimates applied to thewhole or a large part of the SSC (Mel-nick & Moles 1987; Quintana et al.1995; Ettori et al. 1997), which requires,at least conceptually, that the SSC hasalready ended its collapse phase andrelaxed to an equilibrium state.

We emphasise that methods 3 and 4place the supercluster at opposite endsof its dynamical evolution, which areboth not correct according to the dis-cussion in Section 2. For the sake ofargument, consider a homogeneousstructure exactly at turnaround. At thatinstant, there are no internal motions,and all galaxies within the structureare therefore at the same redshift.Analysing this structure with method 3,

was Raychaudhury (1989) who pointedout that the concentrations found by allthese authors were the same, andcoined the name “Shapley Super-cluster”. He confirmed a strong concen-tration of galaxies in that direction,based on a much larger catalogue thanthat used by Shapley, estimated its to-tal luminosity, L, and found that anenormous mass-to-light ratio,5 M/L 6000hM/L, is required to producethe observed Local Group motion.

We will not attempt to review all thesubsequent work carried out on theShapley Supercluster, but only highlighta few contributions which are importantto the present purpose. Spectroscopicobservations, with the main purpose ofobtaining redshifts, in order to deter-mine masses and trace the 3-dimen-sional structure of the SSC, were carriedout by several groups, but most inten-sively and systematically by two of them:Quintana and collaborators (Quintanaet al. 1995, 1997, 2000; Drinkwater et al.1999) and Bardelli and collaborators(Bardelli et al. 1994, 1998, 2000, 2001;early work reported in The Messenger:Bardelli et al. 1993). Since the SSC is

cluster, at cz 3000 km/s. However,Dressler et al. (1987) found a coherentstreaming velocity beyond this struc-ture, out to cz 6000 km/s, implyingthat, if there was one single or domi-nant “attractor”, this had to be furtheraway. Melnick and Moles (1987), usingredshift data gathered in “SC 1326-311”by Melnick and Quintana (1981) andpreliminary data taken over a more ex-tended region by Quintana and Melnick(later published in Quintana et al. 1995),showed that there is a much larger con-centration at cz 14,000 km/s, whichthey called the “Centaurus Superclus-ter”, giving a first mass estimate for itscentral region, based on the virial theo-rem, as 2.5 × 1015 h–1 M within 1° or~ 2.5 h–1 Mpc,4 and finding that this isfar from the mass needed, by itself, toproduce the required acceleration ofthe Local Group over the age of theUniverse. Scaramella et al. (1989)pointed out that there is a large con-centration of clusters in the direction ofthe Local Group motion, which theycalled “α-region”, without reference toeither Shapley or Melnick & Moles. It

Figure 2: Schematic representation of the spherical collapse model. A discussion is given inthe text.

4Here, h = H0/(100 km/s/Mpc) 0.5 – 0.8 is thestandard parametrisation of our ignorance regard-ing the exact value of the Hubble parameter, andM, is the mass of the Sun, 2 × 1030 kg.

5L = 4 × 1026 watts is the luminosity of the Sun.The typical mass-to-light ratio of a cluster of galax-ies is ~ 300hM/L (Peebles 1993).

6We will not go here into the still unresolved andperhaps somewhat academic question of whatwould be meant by “the whole SSC”. See Quintanaet al. (2000) for a recent discussion.

21

panding, to the moment at which thestructure emitted the light currently be-ing observed), the speed of each shellcan be written as

Equations (1) can also be combined toyield

the mass enclosed within the shell ofcurrent radius r. Therefore, having anestimate for t1 (depending on the cos-mological model), a measurement ofthe radial velocity r• for each shell would

any given shell’s radius, r(t), to be writ-ten in the familiar parametric form,

(e.g., Peebles 1993, chapter 20), whereA and B are constants for any givenshell, determined by the enclosed massM and the shell’s total energy per unitmass, G is Newton’s gravitational con-stant, and η labels the “phase” of theshell’s evolution (initial “explosion” at η= 0, maximum radius or “turnaround” atη = π, collapse at η = 2π). As we are ob-serving many shells at one given cos-mic time t1 (measured from the BigBang, at which all shells started ex-

one would argue that there is a finitenumber of objects within a vanishingvolume, and therefore an infinite densi-ty would be inferred. With method 4, thevanishing kinetic energy of the structureis taken to reveal a vanishing potentialenergy, and therefore a vanishing massdensity. In practice, this state of zeromotion is of course never realised,mostly because superclusters are neverhomogeneous, but always contain sub-structure (most prominently clusters ofgalaxies) whose internal velocity dis-persions produce a spread in redshiftspace, in this way making both esti-mates give finite results. However, it isnot fully clear whether these finite re-sults are close to the true values to bedetermined7. So far, details in the defini-tion of the volume to be considered ap-pear to be more important than thechoice of method. The average densitywithin a large radius, ~ 10h–1 Mpc,around the SSC’s centre generallycomes out to be a few times the cos-mological critical density ρc = 3H0

2/(8πG), not far from the density expect-ed at turnaround (~ 5ρc). This makesthe discussion above particularly rele-vant, as more and more data on thisand other superclusters are being ac-cumulated.

3. Spherical Collapse Models

In order to avoid considering only theextreme limiting cases of pure Hubbleexpansion or a time-independent equi-librium state, Reisenegger et al. (2000)considered a simplified, spherical mod-el that allows to calculate the full, non-linear dynamical evolution from the ini-tial (Big Bang) expansion through turn-around until the final collapse. Thismodel, pioneered by Regos & Geller(1989) and applied by them to the in-falling galaxies in the outskirts of indi-vidual clusters, treats the matter ascomposed of concentric sphericalshells, each of which first expands andthen contracts under the gravitationalpull of the enclosed matter (composedof the smaller shells). If the structurehas an outwardly decreasing densityprofile, the innermost shells will col-lapse most rapidly, and no crossing ofshells will occur at least until the firstshells have collapsed, so the mass Menclosed in each shell is a constant intime.8 This allows the time evolution of

Figure 3: Distribution of galaxies and clusters of galaxies in the Shapley Supercluster, withreference to the central cluster A 3558. The abscissa represents the projected angular dis-tance θ of each object to the centre of A 3558, related to the projected physical distance byr⊥ = 2.5 θ [degrees] h–1 Mpc, where our distance to A 3558 is taken to be cz/H0 = 143 h–1

Mpc. The vertical axis is the line-of-sight velocity, cz. Dots are individual galaxies, circles rep-resent the centres of clusters and groups of galaxies. Note the dense region extending hori-zontally from the location of A 3558 (circle at θ = 0, cz 14,300 km s–1), which is interpret-ed as the collapsing structure. (Figure reproduced from Reisenegger et al. 2000.)

7Simulations by Small et al. (1998) of super-clusters near turnaround within popular cosmolog-ical models show at least the virial mass estimateto be surprisingly accurate. It remains to be deter-mined how sensitive this result is to the dynamicalstate (or density) of the supercluster and to theamount of structure on smaller scales, andwhether the galaxy overdensity method is similarlyaccurate.

8Since the density in a (proto-)cluster or super-cluster is always much larger than the presentlypopular value of the “cosmological constant” or“exotic energy” density, the effect of the latter onthe dynamics can be safely ignored, except in itscontribution to the age of the Universe, i.e., thetime available for the collapse to occur.

(1)(2)

(3)

22

radii, i.e., that all galaxies within thedense region are gravitationally boundto the structure. One has to assume fur-ther that the radial density profile liesbetween ρ ∝ r–3 and ρ ∝ r–2, as in theoutskirts of simulated clusters of galax-ies (e.g., Navarro, Frenk, & White1997). Nevertheless, tests of this mod-el in numerical simulations shows it towork quite well in the infall regionsaround clusters (Diaferio & Geller 1997;Diaferio 1999).

Figure 4 shows the enclosed massas a function of radius, M (r), as deter-mined by the two methods discussed,together with a third determination,namely the cumulative mass of theclusters enclosed in the given projectedradius. The cluster mass estimates,M500, taken from Ettori et al. (1997) forthe most important clusters, are mass-es within a radius enclosing an averagedensity 500 times the critical density ρc.This is substantially higher than thestandard “virialisation density” of ~ 200ρc, and therefore gives a conservativelower limit to the total virialised mass,which may be increased by a factor~ (500/200)1/2 1.58 for a more realis-tic estimate.

Given the simplifications and uncer-tainties involved, there seems to be fairagreement among the different massdeterminations, and it seems safe tosay that the mass enclosed by radius r= 8h–1 Mpc lies between 2 × 1015h–1 M

and 1.3 × 1016h–1 M, corresponding toa density range ρ/ρc ~ 3–20. It is inter-esting, nevertheless, that Diaferio &Geller’s method gives results that differso little from the lower limit to the viri-alised mass in clusters. Therefore, ifthis method is applicable to the SSC,either there is very little mass outsidethe clusters of galaxies, or the clustermass estimates are systematicallyhigh.

For comparison, the mass requiredat the distance of the SSC to producethe observed motion of the Local Groupwith respect to the cosmic microwavebackground is Mdipole 2.8 × 1017h–1

M, where a standard value for thecosmological density parameter, Ωm ~0.3, has been assumed.9 The masswithin 8h–1 Mpc can therefore produceat most ~ 5% of the observed LocalGroup motion, which makes it unlikelythat even the whole SSC would domi-nate its gravitational acceleration. Onthe other hand, consistent models ofthe density and velocity distribution onlarge scales (where density fluctuationsare small) in the local Universe can nowbe built (e.g., Branchini et al. 1999). Inthese, the SSC figures prominently, al-though the Local Group motion origi-nates from a combination of several “at-tractors”.

formally divergent, and it should also bevery high in the enclosed region, mak-ing the identification of the caustics rel-atively unambiguous. Figure 3 shows r⊥and cz for the galaxies in the SSC’s(projected) central area. The dense re-gion enclosed by the caustics clearlystands out.

The caustics amplitude, A(r⊥), is adecreasing function of r⊥, related to theradial velocity, r• (r), by a transformationakin to the Legendre transform familiarfrom thermodynamics and classicalmechanics (Reisenegger et al. 2000,Appendix). This transformation can ingeneral not be directly inverted. GivenA(r⊥), one can in the general case onlyobtain an upper bound on r• (r), andtherefore an upper bound on the en-closed mass M (r).

Furthermore, as Diaferio & Geller(1997) pointed out, even if thelarge-scale shape of the structure isroughly spherical, it is expected to havesubstructure with random velocities.These add to the purely radial infall ve-locities, washing out the caustics andexpanding the redshift range coveredby galaxies in the collapsing structure,which further increases the estimatedmass. To cure this problem, they pro-pose the alternative relation:

There is no rigorous derivation for thisresult, although it can be justifiedheuristically by assuming that A(r⊥) re-flects the escape velocity at different

allow to solve for η and, thus, M for thesame shell. Since redshift measure-ments yield velocity components alongthe line of sight, the determination of r•

is possible in an indirect and somewhatlimited way, as follows.

Along a given line of sight throughthe collapsing structure, passing at adistance r⊥ from its centre, there can begalaxies at different distances from theobserver, whom we take to be at thebottom of Figure 2. A galaxy at thesame distance as the centre of thestructure falls towards the latter per-pendicularly to the line of sight, andtherefore has the same redshift as thecentre. Galaxies farther away from theobserver have a velocity component to-wards the observer, so their redshift issomewhat lower than that of the struc-ture’s centre. The opposite happenswith galaxies closer to the observer (notshown), qualitatively inverting the red-shift-distance relation familiar fromHubble’s law. The redshift differencebetween a galaxy and the structure’scentre is small both when the galaxy isclose to the centre (because the veloc-ity is perpendicular to the line of sight),and when it is far away (because thegravitational pull is weak), so it has totake its extreme (positive or negative)value at an intermediate distance fromthe centre (rm in Fig. 2), thus definingtwo symmetric “caustics” in redshiftspace, of amplitude vmax = A(r⊥).

Note that, generally, there will begalaxies also outside the caustics, cor-responding to the expanding Universefar away from the collapsing structure.However, the density at the caustics is

Figure 4: Enclosed mass as a function of radius around A 3558 given by different methods.The solid line is the sum of masses of clusters and groups within the given radius, taking theirprojected distance as the true distance to A 3558. The dashed and dot-dashed lines are up-per bounds to the total mass based on the pure spherical infall model, for a Universe domi-nated by a cosmological constant and for a Universe with critical matter density, respective-ly. The solid-dotted line is the estimate from Diaferio & Geller’s (1997) escape-velocity mod-el. (Figure reproduced from Reisenegger et al. 2000.)

(4)

9Ωm is the ratio of the average mass density inthe Universe to the critical density. The requiredMdipole is proportional to Ωm

0.4.

23

Drinkwater, M J., Proust, D., Parker, Q.A.,Quintana, H., & Slezak, E. 1999, PASA,16, 113.

Einasto, M., Tago, E., Jaaniste, J., Einasto,J., & Andernach, H. 1997, A&AS, 123,119.

Ettori, S., Fabian, A.C., & White, D.A. 1997,MNRAS, 289, 787.

Goldberg, D.M. 2001, ApJ, 552, 413.Goldberg, D.M., & Spergel, D.N. 2000, ApJ,

544, 21.Lugger, P.M., et al. 1978, ApJ, 221, 745.Melnick, J., & Quintana, H. 1981, A&AS, 44,

87.Melnick, J., & Moles, M. 1987, RMxAA, 14, 72.Navarro, J.F., Frenk, C.S., & White, S.D.M.

1997, ApJ, 490, 493.Peebles, P.J.E. 1993, Principles of Physical

Cosmology (Princeton: Princeton Univer-sity Press).

Quintana, H., Ramirez, A., Melnick, J., Ray-chaudhury, S. & Slezak, E. 1995, AJ, 110,463.

Quintana, H., Melnick, J., Proust, D., &Infante, L. 1997, A&AS, 125, 247.

Quintana, H., Carrasco, E.R., & Reisen-egger, A. 2000, AJ, 120, 511.

Raychaudhury, S. 1989, Nature, 342, 251.Raychaudhury, S., Fablan, A.C., Edge, A.C.,

Jones, C., & Forman, W. 1991, MNRAS,248, 101.

Regos, E., & Geller, M.J. 1989, AJ, 98, 755.Reisenegger, A., Quintana, H., Carrasco,

E.R., & Maze, J. 2000, AJ, 120, 523.Scaramella, R., Baiesi-Pillastrini, G., Chin-

carini, G., & Vettolani, G. 1989, Nature,338, 562.

Shapley, H. 1930, Harvard Coll. Obs. Bull.874, 9.

Shapley, H. 1933, Proc. Nat. Acad. Sci., 19,591.

Small, T.A., Ma, C.-P., Sargent, W.W.L., &Hamilton, D. 1998, ApJ, 492, 45.

Smoot, G.F., & Lubin, P.M. 1979, ApJ, 234,L83.

Smoot, G.F., et al. 1992, ApJ, 396, L1.Zucca, E., Zamorani, G., Scaramella, R. &

Vettolani, G. 1993, ApJ, 407, 470.

preparation). A more homogeneouscatalogue might also allow to attempt athree-dimensional, non-spherical mod-el of the SSC, perhaps along the linesof recent work on recovering the initialdensity fluctuations from the present-day redshift-space density distributionin a mildly nonlinear density field (Gold-berg & Spergel 2000; Goldberg 2001).Numerical simulations of superclusterscan also be run in order to test and cal-ibrate dynamical models to be appliedto the SSC.

Our recent research on this topic andthe writing of this article were financial-ly supported by the co-operative pro-gramme ECOS/CONICYT C00U04, bya 1998 Presidential Chair in Science,and by DIPUC project 2001/06PF.

References

Abell, G.O., Corwin, H.G. & Olowin, R.P.1989, ApJS, 70, 1.

Bardelli, S., Scaramella, R., Vettolani, G.,Zamorani, G., Zucca, E., Collins, C.A., &MacGillivray, H.T. 1993, The Messenger,71, 34.

Bardelli, S., Zucca, E., Vettolani, G., Zamo-rani, G., Scaramella, R., Collins, C.A., &MacGillivray, H.T. 1994, MNRAS, 267, 665.

Bardelli, S., Zucca, E., Zamorani, G., Vetto-lani, G., & Scaramella, R. 1998, MNRAS,296, 599.

Bardelli, S., Zucca, E., Zamorani, G., Moscar-dini, L., & Scaramella, R. 2000, MNRAS,312, 540.

Bardelli, S., Zucca, E., & Baldi, A. 2001,MNRAS, 320, 387.

Branchini, E., et al. 1999, MNRAS, 308, 1.Diaferio, A. & Geller, M.J. 1997, ApJ, 481,

633.Diaferio, A. 1999, MNRAS, 309, 610.Dressler, A., et al. 1987, ApJ, 313, L37.

4. Conclusions and FurtherWork

The SSC is undoubtedly a remark-able structure in a very interesting dy-namical state, which deserves furtherstudy. A first attempt at a truly dynami-cal model of the supercluster has beenmade, but much further progress ispossible, at least in principle. A largeamount of information is available,namely the sky co-ordinates and red-shifts of ~ 6000 galaxies (e.g., Bardelliet al. 2000; Quintana et al. 2000), animportant fraction of which is still un-published, and which is still being ex-panded, in order to fill in gaps andcover an even wider area (Quintana,Proust et al., in preparation), in order toassess whether a “boundary” of thesupercluster can somewhere be dis-cerned. Only a very limited part of theavailable information, namely the posi-tion of the caustics on the (r⊥, z) plane,has been used in the present sphericalcollapse models. In principle, the gal-axy density at each point on this planecan be used to refine the model, eitherconstraining it more strongly by assum-ing that the galaxies trace the mass, ordetermining both the mass density andthe galaxy density from the full two-dimensional information (Reiseneggeret al., in preparation). However, this willrequire a uniform redshift catalogue,not available at present, since the avail-able data are a collection of observa-tions taken by different astronomers fordifferent purposes. In order to assessthe incompleteness of the present red-shift catalogue, a complete and accu-rate photometric catalogue of the re-gion is being prepared (Slezak et al., in

Swimming pool of the Residencia

The swimming pool at the lowest floor of the Residenciawas introduced into the project as a part of the humidi-fication system. However, italso serves as an importantpsychological element thathelps to overcome the harshliving conditions, especially forthe permanent staff. Photo: Massimo Tarenghi.

24

The wavelength coverage from thenear-UV to the near-IR meant that bothstellar and circumstellar phenomenacould be studied simultaneously. With aCCD as detector, even a 50-cm tele-scope would suffice to obtain within 30minutes spectra of 5th-magnitude starswith a signal-to-noise ratio of 100 and aspectral resolving power of 20,000.ESO’s Observing Programmes Com-mittee could be convinced to recom-mend the allocation of the ESO 50-cmtelescope on La Silla for several runs ofup to 3 months each. The time was tobe shared among several projects, ofwhich Be stars formed only one.

This has resulted in an until then un-precedented database of almost 2000high-quality echelle spectra of threedozen Be stars (some 500 additionalones were obtained from northern-hemi-sphere observatories). Since meanwhilethe ESO 50-cm telescope was de-com-missioned in 1997 and after some tem-porary revival was terminally moth-balled in 1999, a summary in The Mes-senger of the many exciting results thatcould be extracted may be appropriate.Before embarking on the story, we em-phasise that the picture of Be starssketched below is probably not repre-sentative of late-type Be stars, whichare much more inert. Moreover, it is notcomplete because, for the sake ofbrevity, only a minimum of non-HEROSresults is mentioned.

The Velocity Law in the Disks

A few broad-lined B stars were notedsome years ago to show a weak centralreversal, dubbed central quasi-emission(CQE), in some spectral lines (Fig. 1).From the comprehensive HEROS data-base, two points quickly became ap-parent: (i) All these stars are actuallyso-called shell stars, i.e. Be stars whosecircumstellar disk (‘shell’) is intersectedby the line of sight, thereby imprintingnarrow and often very deep absorptionlines onto the stellar spectrum. (ii)CQE’s only occur in spectral lines thathave also some component due to ab-sorption in the disk. (The latter is cool-er than the central star so that the twospectra do not have all spectral lines incommon.)

not fundamentally differ from the one ofother early-type stars. But variousprocesses are more pronounced andinteract with one another more strong-ly. This renders Be stars attractive lab-oratories for the exploration of early-type stars in general, which make thelargest stellar contribution to the chem-ical and dynamical evolution of late-type galaxies.

The comparison with a laboratory isappropriate because variability on time-scales of hours to decades is the com-mon theme of virtually all investigationsthat are based on repeated observa-tions of Be stars. Obviously, the role ofactively controlled experiments is to betaken over by observational monitoring.With many modern observing facilitiesthis is getting ever more difficult.Therefore, when the PI of the FLASH/HEROS spectrographs (e.g., Wolf, B.,Mandel, H., Stahl, O., et al. 1993, TheMessenger, No. 74, p. 19), Prof. Bern-hard Wolf from Heidelberg, offered theopportunity of a long-term collabora-tion, we immediately jumped on it.

Introduction

Be stars were discovered as early as1866 by the famous Jesuit astronomerAngelo Secchi. Not only was γCas thefirst known Be star but the first star everto be seen displaying emission lines.About half-way between then and to-day, the work by Otto Struve and othersled to the picture that Be stars differfrom supergiant B-type stars, whichalso feature emission lines, in that theyare much less evolved, rotate extreme-ly rapidly (up to 450 km/s at the equa-tor), and their emission lines arise froma circumstellar disk.

Roughly 1–2% of the 10,000 visuallybrightest stars belong to this class, andonly few other types of stars have beenobserved so intensively and over a sim-ilar period of time. Therefore, every ad-dition to the arsenal of observing toolsand facilities resulted in an at least pro-portional expansion of the empiricalknowledge base about Be stars. Thismade it increasingly clear that the phys-ics of the atmospheres of Be stars does

REPORTS FROM OBSERVERS

To Be or Not to Be and a 50-cm Post-Mortem EulogyD. BAADE1, T. RIVINIUS1 and S. ŠTEFL2

1European Southern Observatory, Garching b. München, Germany2Astronomical Institute, Academy of Sciences, Ond řejov, Czech Republic

Figure 1: Central quasi-emission components (CQEs) in four different stars. The CQEs them-selves are a purely circumstellar, geometric-kinematic phenomenon. The FeII line is formedonly in the disk; the other lines also have some photospheric contribution, which is broaderdue the rapid stellar rotation. The velocity has been reduced to the heliocentric scale with theCQEs marking the respective systemic velocities.

25

Some Be stars erupt relatively mildlyevery few weeks. Others prefer onemajor outburst per decade, and thereare all kinds of mixed cases. In starswith infrequent outbursts one can ob-serve the incipient demolition of thedisk at its inner edge. With time, a cav-ity eats into the disk, which can be re-plenished by one or more later out-bursts (Fig. 3).

The HEROS spectra have not yet en-abled us to elucidate the physicalprocess(es) underlying these mass-loss events. Otto Struve still conjec-tured that the Be stars are rotationallyunstable. Later, quantitative analyseshave rendered this picture untenable.Not considering a small number of starsspun up by mass transfer from a com-panion, the HEROS database did notlead to the identification of stars withmore than three quarters of the criticalequatorial velocity. It only confirmedthat rapid rotation is necessary but notsufficient for a B star to become a Bestar.

Periodic Short-Term Variability

The first report about periodic lineprofile variability in any Be star ap-peared in The Messenger (1979, No.19, p. 4). Many others have been pub-lished since (cf. Fig. 4). For a while, theproximity of the observed periods of 0.5to 2 days to calculated rotation periodsmade co-rotating surface structures ap-pear as a possible alternative to theoriginal explanation as non-radial pul-sation.

common such phases of enhancedmass ejection are. More interesting is,however, that these events seem to ba-sically follow one generalised temporalprofile, which may be scaled up ordown in strength and duration. This pat-tern includes that a significant part ofthe matter is ejected with super-criticalvelocities and eventually does manageto merge with the disk (Fig. 2).

This provided a strong hint that theexplanation of CQEs might already becontained in studies of the formation ofshell absorption lines. Indeed, calcula-tions by R. Hanuschik (1995, A&A, 295,423) have shown that apparent centralreversals arise from disks providedthere is no significant radial motion.Because only quasi-Keplerian rotationcan plausibly achieve such a well-tunedequilibrium, the velocity law at largecould be deduced. Other studies hadarrived at the same result before, butthe evidence was always rather indi-rect. This confirmation was importantbecause the problem with a Kepleriandisk is that its specific angular momen-tum is larger than at the stellar equatoreven if the star were rotationally unsta-ble. Models for the disk formation needto account for this.

The Disk Life Cycles

Be stars are too old for their disks tobe fossils of the formation of the centralstars. The stellar radiation pressureand intrinsic dynamical instabilitieshave cleared the circumstellar space along time ago. So the disks must be‘decretion’ disks, i.e. consist of matterlost from the central star. (In binaries,the disk matter may also originate froma Roche lobe overflowing companion.β Lyrae is the most prominent ex-ample. However, the existence of sin-gle Be stars and of Be stars inhigh-mass X-ray binaries with neutronstar secondaries mandate other expla-nations.)

It was already known before that Bestars undergo outbursts. The HEROSspectra could further illustrate how

Figure 2: The evolutionof the Si II 6347 emissionline of µ Cen over an out-burst cycle. The left andthe right peak respec-tively form in the ap-proaching and recessingparts of a rotating disk.The blue vertical linesmark the projected rota-tion velocity of the star(dashed) and theKeplerian velocity at thestellar surface (dotted),respectively. The initialoccurrence of line emis-sion at super-Keplerianvelocities suggests thatsome of the matter wasmoving fast enough toescape the star. Thesubsequent reduction inseparation of the twopeaks shows that theejecta were collected atlarge distances from thestar, where the rotationvelocities are lower, andmerge with the disk.

Figure 3: Two emissionsystems can be distin-guished in theseBalmer emissionprofiles of FV CMa.They indicate thesimultaneous existenceof two structuralentities, which must bespatially separatedfrom each other: Thepair with the higher ve-locity separationrotates faster and istherefore closer to thestar. Maybe the diskbears some resem-blance with Saturn’srings although the geo-metrical thickness ofthe disk is a consider-able fraction of the stel-lar diameter. However,the inner ring expandsand eventually reachesthe outer disk.

26

the different response of spectral tran-sitions to the atmospheric conditionsprevailing between the hot poles andthe equatorial regions with their rota-

added to the large rotational velocity,the observed line profiles are crude 1-Dmaps of the stellar surface. The partialdegeneracy of these maps is lifted by

Since the velocity fields of most non-radial pulsation modes deviate fromspherical symmetry by their own char-acteristic pattern, which furthermore is

Figure 4: Observed (left) and modelled (right) line profile variability of He I 4471 and Mg II 4481 in ωCMa. The stellar and atmospheric mod-el parameters are identical for both lines. And yet, the line-specific appearance of the observed Ipv is reproduced properly, like the forbiddencomponent contribution on the blue side of He I 4471 and the relatively sharp absorption core of Mg II 4481. The helium line forms preferen-tially at higher stellar latitudes, where the temperature is higher. The dependency of also the pulsational velocity field on latitude, then, resultsin different variability. This sets tight constraints on the possible models.

Figure 5: Selected time-averaged observed and modelled line profiles of µ Cen. Apart from crude initial guesses of the primary stellar pa-rameters, the only input to the modelling procedure were the residuals from the mean profile of HeI 4713. And yet, also the mean profile isnearly perfectly matched by the model. The fit is nearly as good also for the other He I lines (unless contaminated by emission). Si III 4553,too, is still reasonably well reproduced while the applied LTE model is known to produce too strong CII 4267 absorption. These results en-courage the idea that quantitative, absolute parameters can be derived from stellar pulsation modelling (‘astero-oscillometry’).

27

The disk of 28 CMa is seen almostface-on. Stars with edge-on disks areknown to fade during an outburst. Thebrightening of 28 CMa would there-fore suggest, as has been argued byothers before, that after an outburstthe disk is in some area optically thickeven in the continuum. This woulddivert light into directions above andbelow the disk. However, to explain anamplitude of 40% in this way is a realchallenge.

Confusing SMC

Both pulsations and mass loss de-pend sensitively on metallicity. It wouldbe extremely useful to check whetherthere are any systematic differencesbetween Be stars in the Galaxy and theSMC. In some fields and clusters of thelatter, one-half and more of all B-typestars show emission lines. Moreover, atleast Be stars in NGC 330 are known toexhibit photometric short-term variabili-ty of the same kind as in the solar neigh-bourhood. By sacrificing a little bit in S/Nand temporal resolution, the impressivethroughput of VLT+UVES should permitthe observation of stars that are 10,000times fainter than the ones we hadmonitored with HEROS and the ESO50-cm telescope, and thereby to reachNGC 330 spectroscopically.

We submitted a corresponding ob-serving proposal, which apparently pre-sented the picture emerging from thefirst sections of this article so affirma-tively that the OPC concluded that wewould not find any difference betweenGalaxy and SMC. This would not be in-teresting enough to confirm observa-tionally with as scarce a resource asthe VLT. Fortunately, the move ofFORS2 from Kueyen to Melipal openeda special observing window with UVES

which previous series of observationscould be continued.

Combining Past and Present

Comprehensive though the informa-tion contents of HEROS spectra is, it al-ways represents only a small fractionof the complete facts. Accompanyingother observations are therefore par-ticularly valuable. Sometimes they canbe secured even retroactively.

A simple joke claims: ‘The world’slargest telescope is when all peoplelook into the same direction’. In fact thecollective light-gathering power of alleyes of mankind exceeds even the oneof the VLT and will be topped only bysome ELT. By implication, the smallesttelescope is the eyes of a single per-son. However, what has been largely for-gotten since the days of a F.W.A. Arge-lander is that a well-trained pair of hu-man eyes is a surprisingly good meas-uring instrument. So, we have recentlyteamed up with Argentine amateur as-tronomer Sebastian Otero (Liga Ibero-americana de Astronomia, BuenosAires), who has taught himself thealmost extinct art of visual photometryand, in a good night, reaches an accu-racy of a few hundreds of a magnitude.

He has recently alerted us to a newoutburst of the bright Be star 28 CMa.We had observed the previous majoroutburst with HEROS in 1996. But,then, there was no way to obtain ac-companying photometry. Building up onour conclusion that outbursts of Bestars are very similar, we are nowstitching the two only partly overlappingseries of data together by occasionally‘borrowing’ a new FEROS spectrum asa safeguard against the much longertimescale of 28 CMa than of the Bestars observed by us before.

tionally reduced effective temperatureand gravity. HEROS was the first in-strument to produce dense series ofelectronic high-resolution spectra cov-ering the full optical wavelength range.Accordingly, this mapping potentialcould be fully exploited, especiallysince contamination by the disk is morereadily recognisable when numerouslines can be compared. In this way, thepulsational nature of the line profilevariability could be unambiguously es-tablished.

A separate Messenger article willdescribe the modelling techniques inmore detail. However, it is worthwhilementioning already here that by mod-elling the line profile variability, i.e.merely the residuals from the meanprofiles, global stellar parameters canbe derived. These, in turn, permit amodel spectrum to be calculated, thatmatches the observed spectrum stun-ningly well (Fig. 5). Since nonradialpulsations of various kinds are verywide-spread among early-type stars,this might lay the foundation to anentirely new stellar analysis tech-nique, which could be called astero-oscillometry.

Pulsation and Outbursts

The HEROS observations of Be starsinitially focused on the 3rd-magnitudestar µ Centauri, which seemed to un-dergo the (then still much more enig-matic) outbursts particularly often. Infact, this choice was highly fortunate asthe later analysis of the Hipparcosall-sky database has not furnished anymore active Be star than µ Cen. Thereal thrill, however, of this star unfoldedwith the discovery that its outbursts,which had been assumed to be sto-chastic, not only repeated periodicallybut with two periods, namely 29 and 52days, which correspond to beat periodsbetween the nonradial pulsation modeswith the largest amplitudes.

Would this be the century-longsought key to the understanding of theBe phenomenon? At most partly: Late-type Be stars are not known to be peri-odically variable or to undergo out-bursts and so need to be explained dif-ferently. Most of the repetition times ofHipparcos outbursts of Be stars are ofthe order of 100–200 days, sometimesmore. To establish such timescales asperiods requires observations with atime baseline of at least twice thatlength, whereas the ESO 50-cm wasde-commissioned already in 1997. So,we were fortunate to be able to arrangefor the relocation of HEROS from LaSilla to other places and eventually, in2000, to the 2-m telescope at theOndřejov Observatory of the CzechRepublic. However, in Ondřejov theweather is not much better than inGarching; and there are only very fewbright, active equatorial Be stars, for

Figure 6: The ratio of the peak height of the Hα emission to the adjacent continuum (top) andthe visual magnitude (bottom) of 28 CMa in series of observations covering 7 consecutive an-nual observing seasons.

28

complexities added, and the expansionand analysis of the HEROS databasecontinue.

Acknowledgements: We are indebt-ed to too numerous people at theLandessternwarte Heidelberg and theEuropean Southern Observatory (LaSilla and Garching) to give credit to allof them individually. We have not for-gotten the manifold technical, scientific,and administrative support.

course, the possibility of this result be-ing due to small number statistics.However, since we believe that thispossibility is probably very small, wesubmitted a letter to some major astro-nomical journal. Only to be told by thefirst referee that our data were not in-teresting. This problem has meanwhilebeen fixed. But the puzzle of the Bephenomenon persists with some oldquestions answered and some new

on Kueyen in August 2001, and theDirector General’s Discretionary TimeCommittee recommended some recon-naissance observations.

The results are shown in Figure 7:Not only is there a difference betweenGalactic and the two SMC Be stars weobserved but the latter are not variableat all at a level that would have beeneasily detected in Galactic observa-tions of the same quality! There is, of

Figure 7: Mean line profiles and their temporal RMS variance of Galactic (left; observed from La Silla) and SMC (right; observed with the VLTand UVES) early-type, pole-on Be stars compared. The red overplot demonstrates the expected RMS variance if the SMC stars underwentthe same variability as their Galactic counterparts. The differences in the strength of the Mg II 4481 line are due to the metallicity of the SMC,which is about 80% lower than in the Galaxy.

Spectroscopy of Quasar Host Galaxies at the VLT: Stellar Populations and Dynamics Down to theCentral KiloparsecF. COURBIN1, 2, G. LETAWE1, P. MAGAIN1, L. WISOTZKI3, P. JABLONKA4, D. ALLOIN5,K. JAHNKE6, B. KUHLBRODT6, G. MEYLAN7, D. MINNITI2

1Institut d’Astrophysique de Liège, Belgium; 2Universidad Católica de Chile, Chile; 3Potsdam University, Germany; 4GEPI, Observatoire de Paris, France; 5ESO-Vitacura, Chile; 6Hamburger Sternwarte, Germany; 7STScI, Baltimore, USA

1. Scientific Context

Discovered more than 40 years ago,and in spite of tremendous theoreticaland observational efforts, quasars andtheir host galaxies remain puzzling ob-jects. It seems now established that atleast all large galaxies harbour massiveblack holes (e.g., Magorrian et al.1998), and that quasar-like activity maybe a common but transient phenome-non in galaxy evolution. However, verylittle of the physical processes at work

during such episodes of nuclear activi-ty is actually understood. The followingquestions list some of the important is-sues still to be solved: What are thetime-scales involved in the fuelling (orrefuelling) of massive black holes?Does the material they burn come frommergers? Do all galaxies contain ablack hole or only the most massiveones? What is the exact relation be-tween galaxies and quasars? Is thefeedback from a luminous ActiveGalactic Nuclei (AGN) onto its host

galaxy important? Is it in the form ofhuge quantities of ionising radiation ordoes it manifest itself directly as me-chanical outflows such as jets?

Quasar host galaxies have beenstudied almost exclusively by imaging.Examples of such work are numerousand use a broad variety of telescopes,instruments and post-processing tech-niques. The Hubble Space Telescope(HST) data of Bahcall et al. (1997) haveshown that quasars occur in all types ofgalaxies. Stockton et al. (1998) used

29

quasar and for the only bright PSF staravailable in this field of view. Each ob-ject was observed with the three high-resolution grisms of FORS1 and thestandard collimator (0.2 arcsec pixel),giving a mean spectral resolution of700. The exposure time was 1200 sec-onds in each of the three grisms, ingrey time.

3. Spectra Decomposition andStellar Population

The method used for decontaminat-ing the host’s spectrum from the quasarlight is an adaptation of the MCS imagedeconvolution algorithm (Magain et al.1998) to spectroscopy (Courbin et al.2000a). It uses the spatial informationcontained in the spectrum of one orseveral PSF stars in order to spatiallydeconvolve the science spectrum. Thisresults in higher spatial resolutionacross the slit and in a spectrum whichis decomposed into several channels.One of these channels contains thespectrum of extended objects, whilethere is one individual channel for eachpoint source in the slit. In the applica-tion we already showed for the gravita-tional lens HE 1104-1805 (Courbin etal. 2000b and Lidman et al. 2001),there were two channels for each of thequasar images and the channel for theextended sources was used for the

rate decomposition of the data into theindividual spectra of the quasar and ofits host galaxy. We have chosen to car-ry out our study for optically bright ra-dio-quiet quasars selected from theHamburg/ESO survey (Wisotzki et al.2000). Our sample involves quasarswith MB < –24 and z < 0.33, for most ofwhich we also have sharp H-band im-aging obtained at the NTT with SOFI.The spectroscopic work in progress isbased on the VLT/FORS1 observationsof 18 of these objects, with ANTU atParanal observatory. The 5 observingnights allocated in the context of theprogrammes 65.P-0361(A) and 66.B-0139(A) were all clear and with seeingoften below 0.6 arcsec.

Decomposition of the quasar andhost spectra requires knowledge of theinstrumental Point Spread Function(PSF). We have therefore chosen theMOS mode of FORS1 in order to ob-serve simultaneously the quasars andPSF stars in the field of view. One 19arcsec long slit was placed on thequasar, several other slits were placedon PSF stars (note that there was al-ways at least one PSF available), andthe rest was used to observe galaxiesin the immediate environment of thequasar. We show in Figure 1 part of theFORS1 field of view around HE1503+0228, where are indicated thepositions of two slits used for the

adaptive optics near-IR data to map theplumes and jets around the quasar PG1700+518. Recent HST optical studieshave established that high-luminosityquasars generally reside in big ellipti-cals, irrespective of radio properties(McLure et al. 1999, McLeod & Rieke1995a, Disney et al. 1995). There alsoseems to be a trend that more luminousQSOs are hosted by more massivegalaxies (McLeod & Rieke 1995b).

A more detailed and quantitative un-derstanding of the physical conditionsin hosts galaxies can only be obtainedfrom spectroscopic observations. Thedata available in this field are veryscarce, and although extensive quasarhost spectroscopy was conducted al-ready in the early 1980s (e.g., Borosonet al. 1985), they have never reallybeen followed up with improved instru-mentation and analysis techniques, ex-cept for a few isolated objects (amongthem 3C 48 being the probablybest-studied case in this field – e.g.,Chatzichristou et al. 1999, Canalizo &Stockton 2000. See also Crawford &Vanderriest 2000 for similar work withother objects).

Nearly all spectroscopic observa-tions up to now were designed as long-slit, “off-nuclear” spectroscopy. Off-nu-clear observations attempt to minimisecontamination by the bright quasar, byplacing the slit of the spectrograph afew arcsec away from it. While thissurely contributes to minimise contami-nation by the quasar, it also minimisesthe signal from the host galaxy itself!In addition, only the outer parts of thehost are probed, hence giving a biasedidea of the overall stellar content, andloosing the dynamical information.

Recent developments in instrumen-tation and the advent of sophisticatedpost-processing techniques have madeit possible to tackle in a new way theproblem of light contamination by thecentral AGN. We describe in the pres-ent article the first results of a compre-hensive VLT spectroscopic campaignwhich, at variance with the traditionalapproach, is designed as “on-axis”spectroscopy combined with spectradeconvolution. The case of the low-red-shift quasar HE 1503+0228, at z =0.135 is taken as an example andplaced in a broader context which willinvolve optical and near-IR two-dimen-sional spectroscopy with instrumentssuch as GIRAFFE, SINFONI andFALCON.

2. VLT SpectroscopicObservations

Our goal is to study the stellar andgas contents of quasar host galaxies aswell as their dynamics, and to comparetheir properties with “normal” galaxieswithout quasar activity. This not only re-quires deep spectra with moderatelyhigh spectral resolution, but also accu-

Figure 1: Part of the VLT/FORS1 pointing image. The seeing is 0.62″ on this 30-sec R-bandexposure. The 1″ slits used to obtain the spectra of the target and of the PSF star are indi-cated.

30

nent) is used to separate the spectra ofthese two sources. A smoothing in thespatial direction is applied for this pur-pose to the extended component.However, such a method is not optimalfor deriving velocity curves, as spatialsmoothing (which is not preciselyknown as it varies with the local S/N)may modify the spectral position of thelines at a given spatial position, by av-eraging spatial components of differentradial velocities. We have therefore de-signed another method, for extractingnarrow emission lines in two dimen-sions, from the rest of the spectrum.The method decomposes the data in:(1) the narrow emission lines of thehost, and (2) the low-frequency signalcomposed of the spectrum of thequasar plus the continuum of the hostgalaxy. The result of the process isshown in Figure 3, where the rotation ofthe galaxy is conspicuous.

varies between10 and 25 per Åfrom the bluestparts to the red-dest. This allowsthe analysis ofthe stellar popu-lation. Measuringthe emission andabsorption linesin the spectrumof HE 1503+0228,and comparingthem with that ofnormal spiral andelliptical galaxies(Kennicutt, 1992a, b, Tragger et al.1998), we find that the spectrum of theHE 1503+0228 host is that of a normalearly-type spiral galaxy. Its interstellarmedium is ionised by star formationand only a tiny residual excitation canbe attributed to the central AGN.

4. Dynamics of the Host

Many of the quasar hosts we ob-served (about 30–40%) display promi-nent narrow emission lines that some-times extend over several arcseconds.These lines, and the fact that we ob-serve with the slit centred on the qua-sar, allow us to derive the rotation curveof the host, from the outer parts of thegalaxy, down to the central kiloparsec.

In the deconvolution process de-scribed above, the difference in spatialproperties between AGN (point source)and the host galaxy (extended compo-

lensing galaxy. For quasar hosts, thereis one channel for the quasar and onefor the host galaxy.

Figure 2 displays the integrated spec-trum of HE 1503+0228. Each panelshows the deconvolved/decomposeddata for the three FORS1 grisms. Thequality of the decomposition can bejudged from the many narrow emis-sions seen in the host and hidden in thequasar spectrum. In addition, there isno trace of residuals of the quasarbroad emission lines in the spectrum ofthe much fainter host. Courbin et al.(1999) already showed from simula-tions that such a decomposition waspossible, but there were no suitableMOS data at that time to test themethod on real spectra.

Luckily, almost all regions of astro-physical interest in the spectrum of HE1503+0228 are observed in two differ-ent grisms. This allows for doublecheck of the deconvolution results andto build a composite spectrum from4000 Å to 9500 Å, with higher signal-to-noise ratio in the regions where thegrisms overlap. Note in Figure 2 the ex-cellent agreement between the decon-volution done for each grism, in the re-gions of overlap. The data are plottedhere as they come from the deconvolu-tion process, without any adjustment toensure a good match between the dif-ferent grisms. Accurate spectrophotom-etry can actually be done on the de-convolved data.

The host galaxy spectrum we pre-sent here has a signal-to-noise that

Figure 2: One-dimensional deconvolved spectrum of HE 1503+0228, at z = 0.13552, and its host galaxy. Each panel correspondsto one FORS1 grism and displays the individual flux-calibratedspectrum of the quasar and its host galaxy. Note that the spectrumof the host shows no trace of the AGN broad emission lines. Theselines are seen narrow in the spectrum of the host. Note also the verygood agreement in the deconvolution of the 3 grisms in the over-lapping regions. The spatial resolution on the deconvolved 2D spec-tra is 0.4 arcsec.

31

local adaptive optics system (i.e.,around each IFU) the instrument willbe used to observe simultaneouslyquasars, PSF stars (for accurate post-processing techniques) and galaxies inthe immediate vicinity of the quasar.The FALCON concept (Hammer et al.2002) not only allows to map the veloc-ity field of the quasar host, but also theone of each component of the groupsor clusters of galaxies involved in thefuelling process of the central AGN.

The study of quasar host galaxies istherefore one more of the numerousareas that will highly benefit from highangular resolution now made possiblein a systematic way on large tele-scopes, either by using AO or by apply-ing post-processing techniques, or acombination of both.

Acknowledgements

We would like to thank the ESO stafffor their very efficient help at the tele-scope during our two VLT runs, and in

mination will become possible by allow-ing for accurate determination of theinclination of the galaxy. At low redshift,GIRAFFE observations are simple. A 1-hour shot will already reveal the velocityfield of the gas. Adding two hours to thiswill allow to follow the important Cal-cium absorption lines, hence the stellarvelocity field.

From our FORS observations, wecan safely predict that GIRAFFE willefficiently be used up to redshift 0.5,for an average seeing observation atParanal. SINFONI is similar to GI-RAFFE, but its adaptive optics systemand its near-IR detector will allow to fol-low quasar hosts at much higher red-shifts (up to z = 2), especially sincelaser guide stars will allow to accessthe entire sky or close to this, with theAO instruments.

On a longer term, FALCON, anear-IR integral field spectrograph us-ing multi-conjugate adaptive optics,shall take place at the VLT. With itsmany Integral Field Units (IFU) and a

The extracted emission lines areused to compute the rotation curve ofthe galaxy in HE 1503+0228 as shownin Figure 4. The first data point (greendots) is at only 0.5 kpc away from theAGN (assuming H0 = 65 km s–1 Mpc–1,Λ = 0, and ΩM = 0.3). In order to inferthe mass of the galaxy, and of its cen-tral regions, we use a three-componentmodel including a central mass con-centration, a thin disk, and a spherical-ly symmetric halo of dark matter. Fittingthis model to the data gives the blueline in Figure 4, which has to be cor-rected for the inclination of the galaxyand other geometrical effects (redcurve). This inclination is estimatedfrom our SOFI and FORS images. Wefind i ~ 46 ± 9° . This is certainly the ma-jor source of uncertainty on the mass ofthe galaxy. A preliminary estimate of themass we can infer for the host of HE1503+0228 is (1.3+0.4

–0.1) 1010 M within 1kpc. The total mass of the galaxy, by in-tegrating our mass models over the 10kpc is (1.4+0.5

–0.2) 1011 M, or about themass of a “normal” spiral galaxy.

5. Towards 2D Spectroscopyand High Spatial Resolution:GIRAFFE, SINFONI and FALCON

The study of quasar host galaxies isa fantastic application for high spatialresolution capabilities. We show from avery simple and short FORS/MOS ob-servation that a distant quasar host gal-axy can already be probed in spec-troscopy down to the central kiloparsec,making it possible to infer in a quanti-tative way the mass of the galaxy andto compare its stellar population withother non-AGN galaxies.

Our observation was taken underaverage seeing at Paranal, about0.6–0.7″ depending on the grism. Again of a factor 10, as is obtained withadaptive optics, will allow for similarstudies easily up to redshift 1 or slight-ly more. This means that we will be ableto follow the evolution with time of theinteractions between AGNs and theirhosts. It also means that the nearestAGNs will be resolved into great detail,maybe down to the central parsec,where the dust torus and the outerparts of AGN accretion disks becomevisible (e.g., Marco & Alloin, 2000,Alloin et al. 2001).

Modern observatories, and in partic-ular ESO, offer a broad range of instru-ments not only capable of high-resolu-tion imaging at NGST-like resolution,but also capable of producing two-dimensional spectra at such a spatialresolution. GIRAFFE is a step towardshigh-resolution 2D spectroscopy, andwill already allow us to map the wholevelocity field (gas and stars) of low-red-shift quasar-host galaxies. Stellar pop-ulation gradients might be seen acrossthe objects, and accurate mass deter-

Figure 3: Example of emission lines extraction for the Hα and N II lines in the I grism. Theoriginal data are shown on the left. The middle panel shows the emission lines of the hostgalaxy alone, whereas the right part shows the sum of the quasar spectrum and of the con-tinuum of the host galaxy.

Figure 4: Rotationcurve of thehost galaxy ofHE 1503+0228.The dots are thedata points obtainedusing simultaneouslythe Hα and NII linesin the I-band grism.The blue curve isthe fit of our threecomponent galaxymodel, and the redcurve shows the fitafter correction forthe inclination of thegalaxy and of anumber of slit andseeing effects (seeCourbin et al. 2002for more details).

32

Lidman C., Courbin F., Kneib J.-P., et al.2000, A&A 364, L62.

McLeod K.K., Rieke G.H., 1995a, ApJ 454,L77.

McLeod K.K., Rieke G.H., 1995b, ApJ 441, 96.McLeod K.K., Rieke, G.H., Storrie-Lombardi,

L.J., 1999, ApJ 511, L67.McLure R. J., Kukula M. J., Dunlop J. S., et

al. 1999, MNRAS 308, 377.Magain P., Courbin F., Sohy S., 1998, ApJ

494, 452.Magorrian J., Tremaine S., Richstone D., et

al. 1998, AJ 115, 2285.Marco, O., Alloin, D. 2000, A&A 353, 465.Percival W. J., Miller L., McLure R. J.,

Dunlop J. S., 2001, MNRAS 322,843.Stockton A., Canalizo G., Close L., 1998,

ApJ 500, L121.Trager S.C., Worthey G., Faber S.M., et al.

1998, ApJS 116, 1.Wisotzki L., Christlieb N., Bade N., et al.,

2000, A&A 358,77.

Canalizo G., Stockton A., 2000, ApJ 528,201.

Chatzichristou E.T., Vanderriest C., Jaffe W.,1999, A&A 343, 407.

Courbin, F., Letawe, G., Magain, P., et al.2002, in preparation.

Courbin F., Magain P., Kirkove M., Sohy S.,2000a, ApJ 539, 1136.

Courbin F., Lidman C., Burud I., et al. 2000,The Messenger 101, 17.

Courbin F., Magain P., Sohy, S. et al. 1999,The Messenger 97, 26.

Crawford, C. S., Vanderriest, C., 2000, MN-RAS 315, 433.

Disney M.J., Boyce P. J., Blades J. C., 1995,Nature 376, 150.

Hammer, F., Sayède F., Gendron, E., et al.2002, astro-ph/0109289, in press inScientific Drivers for ESO FutureVLT/VLTI lnstrumentation.

Kennicutt R.C., 1992a, ApJS 79, 255.Kennicutt R.C., 1992b, ApJ, 388, 310.

particular Claire Moutou, MeriemeChadid, Emmanuel Jehin, HermanBoehnhardt and Ricardo Scarpa. Gé-raldine Letawe is a teaching assistantsupported by the University of Liège,Belgium. Frédéric Courbin acknowl-edges financial support through Chi-lean grant FONDECYT/3990024 andfrom the European Southern Observa-tory. Marie Curie grant MCFI-2001-00242 is also gratefully acknowledged.

References

Alloin, D., Galliano, E., Cuby, J.G., 2001,A&A 369, L33.

Bahcall J.N., Kirhakos S., Saxe D.H.,Schneider D.P., 1997, ApJ 479, 642.

Boroson T.A., Persson S.E., Oke J.B., 1985,ApJ 293, 120.

The Crab Pulsar and its EnvironmentJ. SOLLERMAN (ESO, Stockholm Observatory) and V. FLYCKT (ESO, Luleå University of Technology)

1. Introduction

The Crab Nebula is a supernovaremnant. The supernova exploded in1054 AD, and was monitored by con-

temporary Chinese astronomers (seee.g., Sollerman, Kozma, & Lundqvist2001). Today the nebula offers a spec-tacular view, with a tangled web of line-emitting filaments confining an amor-

phous part ghostly shining in synchro-tron light. The beauty of the Crabmakes it repeatedly appear in PR pic-tures, even from a southern observato-ry like the VLT (Fig. 1).

At the heart of the nebula resides theenergetic 33-ms Crab pulsar. This mV ~16 object actually powers the whole vis-ible nebula. The Crab nebula and itspulsar are among the most studied ob-jects in the sky. This astrophysical lab-oratory still holds many secrets abouthow supernovae explode and abouthow pulsars radiate and energise theirsurrounding nebulae.

A main theme for pulsar researchhas been to understand the emissionmechanism for the non-thermal pulsarradiation. This is still to be accom-plished. No comprehensive model ex-ists that can explain all the observedfeatures of the radiation. Observation-ally, only recently was a broad rangeUV-optical spectrum of the pulsar pub-lished (Sollerman et al. 2000). We havenow extended the study into the infra-red (IR).

But even if most of the research onthe Crab pulsar has concerned theradiation mechanism, almost all of thespin-down energy actually comes outin the particle wind. This is the powersource of the Crab nebula. The stun-ning image of the pulsar environmentobtained with CHANDRA (Fig. 2) cap-tures a glimpse of the energeticprocesses at work. Direct evidence ofthe pulsar activity has long been seenin the system of moving synchrotronwisps close to the pulsar itself.

Figure 1: The VLT (UT2 + FORS2) view of the Crab nebula. A composite of images in B(blue),R and S II(red), taken in November 1999 as part of commissioning. ESO PR photo 40f.

33

In Figure 4 we plot our measure-ments as de-reddened fluxes togetherwith the optical-UV data from Soller-man et al. (2000). We note that our IRfluxes deviate significantly from themost recent results published by Eiken-berry et al. (1997). Our measurementsgive a fainter pulsar by some 0.3 mag-nitudes.

The pulsar magnitudes of Eikenberryet al. agree with those in the 2MASSpoint source catalogue. As our relative-ly isolated standards in the field showgood agreement with 2MASS, we donot think the difference in pulsar mag-nitudes is due to an offset in thezero-point. Instead, our measurementsof the Crab pulsar agree well with pre-vious time-resolved photometry of theCrab (Penny 1982; Ransom et al.1994). Such measurements generallyuse a large aperture and simply as-sume any non-varying contribution tobe due to the background. By integrat-ing under the pulsar light curve, theymeasure only the pulsating contributionof the flux. Our ISAAC photometry hasexcellent signal and image quality. Inthe complex region around the pulsar,this significantly improves the back-ground subtraction. In particular, PSFsubtraction excludes contributions from

mate that our magnitude measure-ments of the pulsar are correct to about0.05 magnitudes.

Figure 2: TheCHANDRA X-rayview of the pulsar.The spectaculartorus and a long jetis clearly seen inthis space-based45-minute exposurefrom August 1999.The field of view is2.5 arcminutes. Withsome imagination,the same structurescan be seen in theVLT 5-minuteB-band image(Fig. 1). PhotoNASA/CXC/SAO.

Figure 3: The central part of the Crab nebula in the infrared, Js, H and Ks. Observations withISAAC on 13 October 2000. The pulsar is the lower right (South Preceding) of the two brightobjects near the centre of the field.

Figure 4: Spectral energy distribution of the Crab pulsar. The optical and UV data are fromSollerman et al. (2000). The diamonds show the IR flux as published by Eikenberry et al.(1997). The triangles are our new ISAAC measurements. Also shown (squares) are the flux-es of the knot, here multiplied by a factor ten. The optical data for the knot is from HST. Allobserved fluxes were de-reddened using R = 3.1 and E(B – V) = 0.52 (Sollerman et al. 2000).

1http://oposite.stsci.edu/pubinfo/pr/1996/22.html

The detailed study of the wisps wasstarted by Scargle (1969) using obser-vations obtained before the authors ofthis article were born. Hester et al.(1995) used the HST to study the wispsat higher resolution, and presentedtheir observations as the spectacular‘The Crab Movie’1, where the constantactivity in the region around the pulsaris highlighted.

The most stunning discovery in theseHST images was the knot sitting just0.6 arcseconds from the pulsar. At 2kpc this amounts to a projected dis-tance of only 1000 AU. Hester et al. in-terpreted this feature as a shock in thepulsar polar wind. Our IR observationsalso allowed us to have a look at thesemanifestations of the magnetic rela-tivistic wind from the pulsar.

2. IR Photometry, Reductionsand Results

IR imaging in the short wavelength(SW) mode of ISAAC was obtainedin service mode on the VLT on Octo-ber 13, 2000. The exposures weretaken in Js, H and Ks with a totalexposure time of 156 seconds perband. The main goal of these short ex-posures was to properly calibrate ourIR spectroscopy. However, the imagequality provided by Paranal also al-lowed a detailed view of the centralregion of the Crab nebula. The near-IR images are displayed next to eachother in Figure 3.

Photometry was obtained of the Crabpulsar and some of the stars in the fieldusing PSF-fitting (DAOPHOT). We esti-

34

knot appears clearly in both frames,and amounts to a few per cent of thepulsar light at these wavelengths.

Data on the Crab pulsar are alsoavailable in the HST archive. Most ofthe observations are from Jeff Hester’scomprehensive monitoring programmeof the inner parts of the nebula, whichhas shown just how active the Crabnebula really is. These frames allow adetailed study of both the spectral andtemporal properties of the knot. InAugust 1995 the region was observedin three filters (F300W, F574M,F814W). We measured the de-red-dened spectral index for the knot to beαv ~ –0.8, which agrees well with our IRdata (Fig. 4).

Furthermore, a wealth of data in theF574M filter allows a study of the tem-

clearly red. In the Ks-band the flux fromthe knot amounts to about 8% of theflux of the pulsar. The stationary wispappears to have a flatter spectrum inthis regime.

3. Optical Data from VLT andHST

To extend the wavelength region overwhich to derive the spectral character-istics of the inner Crab components,several optical images are available inthe ESO and HST data archives.

The VLT PR image (Fig. 1) was tak-en with FORS2 on 10 November 1999,just two weeks after first light. On thefive-minute B-band and the one-minuteR-band exposures, we could PSF-sub-tract the pulsar to reveal the knot. The

the wisps and the nearby knot. We be-lieve that the difference with 2MASS issimply a matter of resolution, and thatwe are now able to subtract virtually allbackground from the pulsar.

The 3-colour image of the centralparts of the Crab (Fig. 5) is colour cod-ed with Js = blue, H = green and Ks =red. In an attempt to keep some physi-cal information in this image, the indi-vidual frames were scaled to make anobject with a flat de-reddened FV spec-trum appear white.

This image shows the well-knownfeatures of the inner Crab. The wispsare clearly visible in higher detail thanpreviously obtained in the IR. Some fil-aments are also seen, most strongly inthe Js band. This is most likely due tothe [FeII] 1.26µm emission line. The Ksband is instead dominated by amor-phous synchrotron emission (Fig. 3).With suitable cuts the pulsar image ap-pears slightly elongated. This is due tothe presence of the knot first identifiedby Hester et al. (1995) on a HST image.To reveal this structure in our imageswe constructed and subtracted a PSFfrom the stellar images. After subtrac-tion, the knot is clearly revealed in allthree bands. A colour image made outof the PSF subtracted frames is shownin Figure 6. The image directly givesthe impression that the knot is redderthan for example the wisps.

Quantifying this we have estimatedthe magnitudes of the knot as well as ofthe nearby wisp 1. The knot was meas-ured within an aperture of 0.9 arcsec-onds and the wisp was simply meas-ured with an aperture of 1.2 arcsec-onds. The spectral energy distributionof the knot is shown in Figure 4. It is

Figure 5: The Crab nebula in the infrared. This is a colour compo-site of the frames shown in Figure 3. North is up and East to theleft.

Figure 6: The centre of the Crab nebula in the infrared. After the cen-tral stars have been removed using PSF subtraction, the knot closeto the pulsar position is revealed. The FOV shown in this ISAAC im-age is 55.6 arcseconds, as provided by the NAOS/CONICA C12Scamera.

Figure 7: HST F547M images of the very centre of the nebula. The leftmost frame is obtainedon March 1994, and the right frame on October 2000, simultaneous with our IR imaging. Notethat the wisp structures are very dynamic, but that the knot seen south-east of the central pul-sar has persisted over more than 6 years. The FOV is 20 arcseconds, North is up and Eastto the left.

35

the pulsar environment with NAOS/CONICA. With a resolution supersed-ing HST we will be able to monitor thestructures close to the pulsar, with 2pixels corresponding to merely 50 AU.This would provide an unique opportu-nity to study the structure and dynam-ics of the inner pulsar wind and its in-teraction with the surroundings.

References

Eikenberry, S.S., Fazio, G.G., Ransom,S.M., Middleditch, J., Kristian, J., &Pennypacker, C.R. 1997, ApJ, 477, 465.

Golden, A., Shearer, A., & Beskin, G.M.2000, ApJ, 535, 373.

Hester, J.J, Scowen, P.A., Sankrit, R., et al.1995, ApJ, 448, 240.

Lou, Y.-Q. 1998, MNRAS, 294, 443.Middleditch, J., Pennypacker, C., & Burns,

M.S. 1983, ApJ, 273, 261.Oke., J.B. 1969, ApJ, 158, 90.Penny, A.J. 1982, MNRAS, 198, 773.Ransom, S.M., Fazio, G.G., Eikenberry, S.

S., Middleditch, J., Kristian, J., Hays, K., &Pennypacker, C.R. 1994, ApJ, 431, 43.

Scargle, J.D. 1969, ApJ, 156, 401.Shapakidze, D. & Machabeli, M. 1999,

ptep.proc, 371.Sollerman, J., Lundqvist, P., Lindier, D., et al.

2000, ApJ, 537, 861.Sollerman, J., Kozma, C., & Lundqvist, P.

2001, A&A, 366, 197.

for a plasma mechanism. None ofthese scenarios predict a very redspectral distribution.

Another area where caution may berequired is in the recent claims of weakand red off-pulse emission from the Crabpulsar in the visible (Golden, Shearer &Beskin 1999). It is clear that the knotclose to the pulsar has to be seriouslyconsidered in these kinds of studies.

5. Future Plans

The Crab pulsar and its environmentcontinue to be the prime astrophysicallaboratory for the study of the pulsaremission mechanism and the spin-downpowering of pulsar nebulae. Althoughmuch observational effort has been putinto this object, a modern re-investiga-tion is likely to clean up the many con-tradictory measurements. Optical imag-ing in good seeing would require only afew minutes with the VLT, and would di-rectly determine the knot-subtractedspectral energy distribution. ISAAC inthe LW range can in less than 3 hoursclarify if the knot is significantly con-tributing to the emission at these fre-quencies, and would establish if the IRdrop of the pulsar is real.

Most interesting is the possibilityto monitor the very central parts of

poral behaviour (Fig. 7). First of all wenote that the knot is indeed present inall frames. It thus appears quasi-sta-tionary for more than six years, al-though the position appears to vary atthe 0.1 arcsecond level. The de-red-dened flux of the knot (Fig. 4) is meas-ured to be 9 × 10–28 ergs s–1 cm–2 Hz–1

but variations of the flux by at least 50%are observed.

4. Discussion and Implications

For the pulsar itself we have addednew information in the IR. Together withthe optical-UV data in Sollerman et al.(2000) this significantly revises the ob-servational basis for the pulsar emis-sion mechanism. In fact, most of thetheoretical efforts have been based onthe old optical data from Oke (1969)and the IR continuation of Middleditch,Pennypacker & Burns (1983). Our newresults call for a fresh look on the emis-sion mechanism scenarios for youngpulsars.

For the knot, we have shown that thestructure is indeed quasi-stationary, andthat the emission has a red spectrum.Few models are available for the knot.Lou (1998) presented a formation sce-nario in terms of MHD theory, whileShapakidze & Machabeli (1999) argue

SOFI Discovers a Dust Enshrouded SupernovaL. VANZI, ESO, ChileF. MANNUCCI, CNR, ItalyR. MAIOLINO and M. DELLA VALLE, Observatory of Arcetri, Italy

1. An IR Search for SN

Luminous Infrared Galaxies (LIRGs)are characterised by luminosities largerthan 1011 L mostly emitted in the far-IRspectral range. Since the discovery of ahandful of these objects by Rieke &Low (1972), and afterward the exten-sive list produced by IRAS, the sourceof energy of the LIRGs phenomenonhas been matter of debate. The radia-tion observed is mostly thermal emis-sion by dust heated by some primarysource. Dust extinction in LIRGs canexceed several magnitudes in the opti-cal and this makes any optical studymore difficult than in normal galaxies.Recent IR observations, mostly by ISO,allowed to shed new light on the prob-lem and to identify the starburst (SB)activity as the main source of energy formost LIRGs (Genzel et al. 1998). How-ever, the presence of obscured AGNs,which are elusive in the optical, hasalso been identified in a few cases rais-ing again the issue on the relative con-tribution of active nuclei to the bolomet-ric luminosity. Yet, if most of the lumi-

Figure 1: SN2001db is detected in the Ks image of NGC 3256 observed on January 9, 2001with SOFI (right) when compared with an archival image (left).

36

Dec.(J2000) = –43 54′ 21″, 5″.7 to theWest and 5″.7 to the South of the Ksnucleus. The Ks magnitude of the SNin the image of January is 16.03. InFigure 1 we show the archival SOFIimage of NGC3256 (left) and the firstimage obtained in 2001 (right) whereSN2001db has been detected.

In Figure 2 the infrared light-curve ofSN2001db is compared with those ofSN1980k and SN1999em taken as rep-resentative of type IIL and type IIP, re-spectively (data from Dwek 1983,Barbon et al. 1982, Hamuy et al. 2001).The light-curve has been extinguishedby an AK = 0.55 (see next section). Theinfrared observations can be roughly fitboth with the light-curve of SN1980koffset by 65 days, and with the light-curve of the SN1999em offset by 90days and 1.0 mag. The average K-bandlight-curve for type II SNe obtained byMattila & Meikle (2001) is nearly iden-tical to the SN1980k light-curve inFigure 2, but offset by 55 days. Weestimate that the V-band magnitude ofthe SN was most likely fainter than 20at its maximum and, therefore, it wouldhave been missed by most of the opti-cal SN search programmes.

After this discovery we immediatelyapplied for Director Discretionary Timeto spectroscopically confirm our findingat the VLT. An optical spectrum was ob-tained on May 16, 2001, with FORS1at the ESO VLT-UT1 using grismGRIS_300V and 1″ slit yielding a spec-tral resolution of 500. The total integra-tion time was 15 minutes. Most of the

Eclipse package and then comparedwith each other and with archival im-ages.

The LIRG NGC 3256 was observedfor the first time on January 9, 2001,and subsequently on February 8,March 18 and April 1. SN2001db wasdetected on the first image, after com-parison with the following ones, at co-ordinates R.A.(J2000) = 10h27m50s.4,

nosity is powered by star formation, theSN rate should consequently be higherthan in normal galaxies. Thereforemeasuring the SN rate in LIRGs is anindirect way to measure the star-forma-tion rate.

Optical searches for SNe in standardstarburst galaxies however failed in de-tecting the expected enhanced SN rate.This is most likely due to the largeamount of extinction affecting thesegalaxies. A way around the problem isto observe in the near-IR where thedust optical depth is a factor of 10 low-er than in the visible. The first attemptsin this direction, however, have notbeen very promising (van Buren &Norman 1989, van Buren et al. 1994).LIRGs are an optimal sample for thiskind of observations; in fact their hugeluminosity requires star-formation ratesup to a few 100 M yr–1 and, conse-quently, a supernova rate up to a fewSNe per year.

We started a monitoring campaign inthe K (2.1µm) near-IR band of a sam-ple of 35 LIRGs aimed at detecting ob-scured SNe. The survey started in late1999 (not continuously) and carried onwith the ESO NTT, the TNG (GalileoNational Telescope) and the Kuiper/Steward Telescopes.

2. First Success with SOFI –SN2001db

A southern sample of galaxies wasobserved with SOFI at the NTT duringperiod 66. A total of 6 nights, about onemonth apart from each other, were allo-cated. Images were obtained in the Ksfilter using a total integration time onsource of about 30 minutes per object.The data were reduced using the ESO

Figure 2: Ks light-curve of SN2001db (blue points) compared with template curves from theliterature.

Figure 3: Optical spectrum of SN2001db observed with FORS1.

37

changed significantly (e.g. Danziger etal. 1991). If this is true the observed ra-tio between the broad components is1.73 which implies an extinction of AV =6.7 mag. for the case B theoretical ra-tio. There are, however, indications thata ratio higher than case B is more ap-propriate in SNe (Fassia et al. 2000, Xuet al. 1992) this would give AV = 5.3mag. In the case of variation of theemission line flux between the two ob-servations, conservative estimates givea lower limit on the extinction of AV > 4.6mag. We assume AV 5.6 mag, withan uncertainty of about 1 mag.

SNe discovered in the optical aremuch less extinguished. In the compila-tions given by Schmidt et al. (1994) andby Mattila & Meikle (2001) the opticalextinction is generally lower than AV <1.5 mag, as shown in Figure 5 wherethe distribution of extinction for optical-ly discovered SNe is reported. If thesecompilations are representative, thenthe extinction inferred for the IRSN2001db is probably the highestamong the SNe so far discovered.

4. The IR SN Rate

So far, our programme has detected4 SNe: two of them were detected alsoin the optical (SN1999gd and SN2000gb); another SN (1999gw) wasdiscovered at the TNG, but we couldnot obtain a spectroscopic identifica-tion; the fourth one is SN2001db dis-cussed in this paper. The number of SNevents found can be compared with theexpected detection rate in the LIRGs ofour sample. Assuming the conversionfrom blue luminosity to SN rate (alltypes) given in Cappellaro et al. (1999),i.e. SNr 10–12 (LB/L) yr–1, we wouldhave expected to detect ~ 0.5 SNe.Since we have detected 4 SNe, weroughly infer a SN rate about an orderof magnitude higher than estimated bythe conversion inferred by optical sur-veys. This is quite different from thenumber of SNe expected from thefar-IR luminosity. If most of the far-IR lu-minosity is due to star formation and we

results on the SN sub-type (IIL vs. IIP)can be driven from the spectroscopicdata.

3. Extinction

Extinction can be measured by usingdifferent methods. The ratio betweenthe narrow components of Hα and Hβis 12.0 which, for a Galactic extinctioncurve, gives an equivalent screen ex-tinction AV = 4.2 mag, or higher if as-suming a mixed case. The Na I inter-stellar absorption doublet at ~ 5890Åhas an equivalent width of 5.87Å thataccording to the relations of Barbon etal. (1990) implies an equivalent screenextinction AV ≥ 3.0–4.9 mag. To use theratio of the broadcomponents ofPaβ and Hα to di-rectly constrainthe reddening ofthe SN requiresthe assumptionthat during the 25days elapsed be-tween the infra-red and the opti-cal spectrum theline flux has not

emission lines detected are due to HIIregions or SN remnants (see next para-graph) while the continuum is due tothe background emission of the galaxy,however, the spectrum shows a clearbroad component of Hα which is a clearsignature of a type II SN. Hα has anasymmetric profile with a peak blue-shifted by about 2000 km/s with re-spect to the parent galaxy. The FWHMis ~ 5000 km/s. Broad emissions at7324Å and 8542Å–8662Å due to Ca IIrespectively are detected. In Figure 3we show the optical spectrum sub-tracted of the underlying background.At the same redshift of the peak of thebroad Hα we detect a FeII 7155Å line.Finally the spectrum shows indicationsfor two broad emission features at5893Å (NaI) and 6300Å ([OI]).

A near-IR spectrum was obtained inthe J band with ISAAC at the ESOVLT-UT1 on April 21, 2001 with a slit of1 arcsec and the low resolution gratingset at the 4th order, yielding a spectralresolution of 500. The total integrationtime was 45 minutes. The spectrum isshown in Figure 4. The broad compo-nent of Paγ is more prominent with re-spect to Hα and the profiles are almostidentical. There are indications of abroad component of Paγ and He I1.0083µm as well. It is worth noting thatthe strong [Fe II] emission relative toPap (narrow) indicates that the underly-ing emission is not simply due to HII re-gions, but must be contributed signifi-cantly by SN remnants. No conclusive

Figure 4: Near-Infrared spectrum of SN2001db observed with ISAAC.

Figure 5: Distributionof the extinction foroptically discoveredSNe.

38

Fassia, A., Meikle, W. P. S., Vacca, W. D., etal., 2000, MNRAS, 318, 1093.

Genzel, R., Lutz, D., Sturm, E., et al., 1998,ApJ, 498, 579.

Hamuy, M., Pinto, P. A., Maza, J., et al.,2001, ApJ, 558, 615.

Mattila & Meikle 2001.Rieke G.H. & Low F.J. 1972, ApJ 176, L95.Soifer, B.T., Neugebauer, G., Matthews, K.,

et al., 2001, ApJ, 559, 201.van Buren, D., & Norman, C.A., 1989, ApJ,

336, L67.van Buren, D., Jarrett, T., Terebey, S.,

Beichman, C., Shure, M., Kaminski, C.,1994, IAUC, 5960, 2.

Xu, Y., McCray, R., Oliva, E., & Randich, S.,1992, A&A, 386, 181.

limited angular resolution would haveprevented us to disentangle them fromthe peaked nuclear surface brightnessof the host galaxies.

References

Barbon, R., Benetti, S., Rosino, L., Cappel-laro, E., & Turatto, M., 1990, A&A, 237,79.

Cappellaro, E., Evans, R., & Turatto, M.,1999, A&A, 351, 459.

Danziger, I.J., Lucy, L.B., Bouchet, P., &Gouiffes, C., 1991, in Supernovae, S.E.Woosley ed. (Springer-Verlag), p. 69.

Dwek, E. 1983, ApJ, 274, 175.

adopt the conversion from LFIR to theSN rate given in Mattila & Meikle (2001),i.e. SNr 2.7 × 10–12 (LFlR/L) yr–1, wefind that our survey has missed about80% of the expected SNe. This can beexplained if most of the SNe are so em-bedded in dust that they are signifi-cantly obscured even in the near-IR or,alternatively, obscured AGNs may con-tribute substantially (~ 80%) to thefar-IR luminosity of these galaxies.Finally, there is growing evidence thatmost of the starburst activity is locatedin the nuclear region (Soifer et al.2001). If most SNe occur in the nucle-us (i.e. within the central 2″), then our

A Deep Look at an Active GalaxyThe image below shows the pe-

culiar edge-on spiral galaxy NGC 3628.It is situated in the constellation Leo

and forms a famous triplet of galax-ies together with M65 and M66, alsoknown as the Leo Triplet. Its dis-

tance is 35 million light-years/11 Mpc. NGC 3628 is interesting in several

respects: although classified as a spiral

39

er, air and water management, theParanal Residencia has already be-come a symbol of innovative architec-ture in its own right. Constructed withrobust, but inexpensive materials, it isan impressively elegant and utilitariancounterpart to the VLT high-tech facili-ties poised some two hundred metresabove, on the top of the mountain.

Ever since the construction of theESO VLT at Paranal began in 1991,staff and visitors have resided incramped containers in the “BaseCamp”. This is one of driest and mostinhospitable areas in the Chilean

Located in the middle of the AtacamaDesert, the Residencia incorporates asmall garden and a swimming pool, al-lowing the inhabitants to retreat fromtime to time from the harsh outside en-vironment.

Returning from long shifts at the VLTand other installations on the mountain,here they can breath moist air and re-ceive invigorating sensory impressions.With great originality of the design, ithas been possible to create an interiorwith a feeling of open space – this is atrue “home in the desert”.

Moreover, with strict ecological pow-

Summary

The Paranal Residencia at the ESOVLT Observatory is now ready and thestaff and visitors have moved into theirnew home.

This major architectural project hasthe form of a unique subterranean con-struction with a facade opening towardsthe Pacific Ocean, far below at a dis-tance of about 12 km. Natural daylightis brought into the building through a35-m wide glass-covered dome, a rec-tangular courtyard roof and variousskylight hatches.

galaxy of type Sbc (like our own MilkyWay galaxy) its massive dust showsdisturbances, possibly as a conse-quence of the fairly close proximity ofthe two other members of the Triplet.Furthermore, there seems to be alot of star formation going on, as onecan see in the upper right corner of theimage, where numerous star-formingregions with young massive blue starsare visible. The box-like bulge of thisgalaxy (visible at the top of the image)is also remarkable and could indicatethe presence of a central bar. A numberof globular clusters can be seen asfuzzy reddish spots in the halo of thegalaxy.

The field around NGC 3628 is rich invery faint galaxies, many of which canbe seen in this image as slightly diffuseobjects. Only a few foreground starsbelonging to our own Milky Way arevisible, sharp and point-like. However,

the conspicuously blue star-like objectjust SSW of the diffuse patch that ex-tends more or less along the minor axisof the main galaxy, is not a star but anX-ray emitting quasar at a redshift z =0.995.

Of special interest in this picture is anelongated low-surface-brightness fea-ture that seems to emerge along theminor axis of the large active galaxy. Itappears to be part of a chain of objectsthat coincides very exactly with an X-ray filament associated with ejection ofX-ray material from the centre of thegalaxy, as shown by observations withthe ROSAT and, very recently, theChandra satellites.

The high image quality of FORS2 onVLT/Kueyen enables resolution of thevarious objects along this optical fea-ture. A spectroscopic investigation iscurrently under way in order to studytheir possible physical relationship to

events in this conspicuously disturbed,nearby galaxy.

Technical information: The colour im-age was composed from five individualexposures through Bessel B, V, R and Ibroadband filters. Exposure times were120 + 600 sec in B, 300 sec in V, 600sec in R and 600 sec in I. They weretaken with FORS2 during the com-missioning period in February 2000 andretrieved from the ESO Science Ar-chive. The seeing on the differentframes was between 0.64 and 0.8 arc-sec. The size of the field is 6.8 × 6.8arcmin; north is up, east to the left. Pre-processing was done with the FORSpipeline in Garching. Observationswere carried out by G. Rupprecht, datareduction by F. Patat (both ESO/Garching), image composition by R.Hook and R. Fosbury (both ST-ECF),astronomical background provided byH. Arp (MPA). G. RUPPRECHT

Coming Home at ParanalUnique “Residencia” Opens at the VLT Observatory

(Taken from the ESO Press Release of 7 February 2002)

This photo shows the Residencia, looking towards west. The linear construction used to fill the natural depression of the ground in this areais evident. The 35-m central dome protrudes from the “filled-in” valley. Photo: Massimo Tarenghi.

40

Simple, but elegant furnishing andspecially manufactured carpeting com-plement a strong design of perspec-tives. The Republic of Chile, the hoststate for the ESO Paranal Observatory,is present with its emblematic painterRoberto Matta.

Additional space is also provided fora regional art and activity display.

The staff moved out of the containersand into their new home in mid-January2002. Today, the Paranal Residenciahas already become a symbol of inno-vative architecture in its own right, animpressively elegant and utilitariancounterpart to the VLT high-tech facili-ties poised some two hundred metresabove, on the top of the mountain.

covered dome, a rectangular courtyardroof and various skylight hatches. Thegreat originality of this design has madeit possible to create an interior with afeeling of open space, despite the un-derground location.

Facilities at the Residencia

To the visitor who arrives at theParanal Residencia from the harsh nat-ural environment, the welcoming feel-ing under the dome is unexpected andinstantly pleasant. This is a true “oasis”within coloured concrete walls and theair is agreeably warm and moist. Thereis a strong sense of calm and serenityand, above all, a feeling of cominghome. At night, the lighting below theroofing closure fabric is spectacularand the impression on the mind is over-whelming.

The various facilities are integratedover four floors below ground level.They include small, but nice and simplebedrooms, offices, meeting points, arestaurant, a library, a reception area, acinema and other recreational areas.The natural focal point is located nextto the reception at the entrance. Thedining room articulates the building atthe –2 level and view points throughthe facade form bridges between thesurrounding Paranal desert and the in-terior.

Atacama Desert, and eleven years is along time to wait. However, there wasnever any doubt that the construction ofthe telescope itself must have absolutepriority.

Nevertheless, with the major techni-cal installations in place, the time hadcome to develop a more comfortableand permanent base of living at Para-nal, outside the telescope area.

A Unique Architectural Concept

The concept for the Paranal Resi-dencia emerged from a widely notedinternational architectural competition,won by Auer and Weber Freie Archi-tekten from Munich (Germany), andwith Dominik Schenkirz as principal de-signer. The interior furnishing and dec-oration was awarded to the Chilean ar-chitect Paula Gutierrez.

The construction began in late 1998.Taking advantage of an existing de-pression in the ground, the architectscreated a unique subterranean con-struction with a single facade openingtowards the Pacific Ocean, far below ata distance of about 12 km. It has thesame colour as the desert and blendsperfectly into the surroundings. TheParanal Residencia is elegant, with ro-bust and inexpensive materials.

Natural daylight is brought into thebuilding through a 35-m wide glass-

A series of woollen,handmade rugs werespecially designed bythe Chilean artist LuzMéndes, with motivesfrom astronomical im-ages, spectra and in-terferogrammes ob-tained at the ParanalObservatory. Theydecorate the commonspace of the building.The one seen on thisphoto displays thePavo interactinggalaxies. Photo:Massimo Tarenghi.

From the Cantine it is possible to observethe preparation of the meals in a modernand well-equipped kitchen. Photo: MassimoTarenghi.

A panorama of the Reception Area with the entry to the Cantine in the background. The essential construction and the warm colour of theconcrete walls are clearly visible and help to give the feeling of being “at home”. Photo: Massimo Tarenghi.

A functional and essential room of 16 m2

with all communication connections allowsstaff and visitor to work and rest. Photo:Massimo Tarenghi.

41

ity to warrant scientific work. The dataare available directly on the ESO web(http://www.eso.org/projects/vlti/instru/vinci/vinci_data_sets.html).

Access to these data is restricted toastronomers in the ESO member coun-tries. ESO welcomes community feed-back on any aspect of the reduction,analysis and interpretation of thesedata. Please contact the VLTI projectscientist (fparesce@eso.org), the VLTIgroup head (aglindem@eso.org) or thehead of the Commissioning team(mschoell@eso.org) with your com-ments and for further information onthis release. F. PARESCE

facility in its first phase of development.In addition to these more technicaltasks, a number of observations arecertainly also useful for scientific pur-poses.

In order to fully involve the ESO com-munity in analysing and understandingthe data and its scientific and technicalimplications, ESO has decided to makethese data available to this communitythrough the archive.

The data were obtained in the periodbetween March 17, 2001 and Decem-ber 5, 2001 and have been deemed bythe commissioning team and the VLTIProject Scientist to be of sufficient qual-

Technical commissioning activities ofthe VLTI with the VINCI test cameraand the 40-cm diameter siderostatswith 16 m baseline and the ANTU andMELIPAL 8-m telescopes with 103 mbaseline have been ongoing since firstfringes on March 17, 2001 with thesiderostats and on October 30, 2001with the 8-m. A number of astronomicaltargets from various object classeshave been observed in these twomodes.

The observations were made to as-sess the compliance of the instrumentswith the technical specifications as wellas to characterise performance of the

OTHER ASTRONOMICAL NEWS AND ANNOUNCEMENTS

Release of Scientific Data from VLTI Commissioning

VLT Science Verification Policy and ProceduresReplicated from ESO web pages

1. Science VerificationObservations

After the conclusion of Commission-ing of a new VLT instrument, and priorto the start of regular operations, a se-ries of Science Verification (SV) ob-servations with such an instrument areconducted. SV observations may alsobe conducted in the case of a major in-strument upgrade.

The equivalent of at least 11 VLT UTnights should be dedicated to SV ob-servations.

SV Observations are conducted dur-ing the dry runs preceding the instru-ment regular operations. At the end ofthe scheduled dry runs, the VLT Pro-gramme Scientist submits to the Direc-tor General a report on the status ofcompletion of the planned SV observa-tions. If the corresponding set of data isjudged insufficient to reach the goals ofSV, the Director General may decidethat further SV observations be execut-ed during the first scheduled regularruns in Service Mode.

All SV Observations are conduct-ed in Service Mode, but one or twomembers of the SV Team may bepresent at Paranal Observatory for aprompt reduction of the data, and theselection of the observations to beexecuted.

2. Goals of Science Verification

The goals of SV are manifold, and in-clude:

• offering to ESO users first science-grade data from a new instrument

• demonstrating the scientific potentialof the VLT+instrument

• fostering an early scientific returnfrom the VLT+instrument

• experimenting any pipeline and re-duction tools that may be available atthe time of SV

• providing feedback to Operation(Paranal and Garching), InstrumentDivision, and Data Flow System, asappropriate

• the involvement of scientists from theESO community in the prompt scien-tific exploitation of the data.

3. Science VerificationProgrammes and Data Policy

The SV Plan of an instrument is de-veloped by a dedicated SV Team.

The PI(s) of the instrument subject toSV and the Instrument Science Teamare involved in the definition of the SVplan.

The SV Programme is presented tothe ESO Faculty for discussion.

The SV Programme is finally submit-ted by the VLT Programme Scientist tothe Director General for approval.

SV observations of targets alreadyincluded in GTO or approved GO pro-grammes with the same instrumentcould be executed only with the agree-ment of the PI.

Raw and calibration SV data passingquality control are made immediately

public via the ESO archive, followingthe “Data Access Policy for ESO Data”.

The SV Team will make efforts to re-lease reduced SV data within twomonths from the conclusion of SV ob-servations.

4. Selection Criteria for SVProgrammes

SV Programmes are selected accord-ing to the following criteria: They should

• have outstanding scientific interest• push the VLT+Instrument close to

their limit• address a scientific issue widely stud-

ied within the ESO Community• result in a sufficiently complete data-

set for its prompt exploitation to bescientifically rewarding

• use the core modes of the instrument• help PIs and Co-Is of approved GO

and GTO programmes to get prompt-ly acquainted with the data from theinstrument

• exploit complementarity with otherpublic datasets (e.g. HDF-S/CDF-S/EIS, etc.), if appropriate.

5. The SV Team

5.1 Composition of the SV Team

A dedicated SV Team is assembledfor each of the various SV phases, in-cluding Garching and Chile staff andfellows (typically up to 8–10 people).

In the selection of the SV Team mem-

42

5. Scientific Exploitation of theSV Data

The scientific exploitation of the SVdata can start as soon as the data arepublicly released.

The formation of groups and teamsfor the scientific exploitation of SVdata is left to the initiative of the indi-viduals.

SV Team members are encouragedto promptly use the data and to stimu-late the participation of scientists fromthe community.

Authors are kindly asked to send toESO (Office of the VLT ProgrammeScientist) at submission time copy ofany paper that may result from the useof SV data, along with a concise tech-nical report on the use of the data,pipeline, etc.

• Development and pre-selection of theSV projects

• Preparation of the OBs, and their de-livery to Paranal Observatory prior tothe instrument dry runs

• Maintenance of SV WEB pages, de-scribing the SV plan well in advanceof the SV Observations, and includinginformative lists of SV data as theybecome public

• Real time assessment of the SV dataat Paranal Observatory (maximum 2SV Team members)

• Reduction of the SV data• Delivery through the SV WEB and the

ESO Archive of the raw, calibration,and calibrated data

• On users request provide informationon the data

• The SV Team can have access to theCommissioning data prior to SV ob-servations.

bers the VLT Programme Scientist willfollow the following criteria:• Strong scientific interest for the spe-

cific capabilities of the instrument• Technical experience with the type of

data being produced by the instru-ment

• Wide coverage of the main scientificareas that the instrument is designedto satisfy

To all activities of the SV Team willalso be invited to participate:• The Instrument PI and Co-PI, or one

person designated by each of them• The ESO Instrument Scientist• The ESO and Consortium Instrument

Pipeline experts

4.2 Duties of the SV Team

The duties of the SV Team include:

News from SantiagoThe spectacular fringes obtained at

VLTI, as well as the intensive ALMApreparatory work in Chile, have sud-denly brought the “world of interferom-etry” to full attention of the astronomicalcommunity in Chile.

Astronomers not yet familiar with thistype of observational technique havestarted to realise the originality, thestrength and the astrophysical potentialof aperture synthesis observations, bothat radio and at optical/IR wavelengths.

On the side of the pioneers who havebeen developing the techniques of in-terferometry for almost 30 years now, itis time to advertise widely their toolsand enrol young researchers in this fas-cinating adventure. On the side of theastronomical community at large, thefantastic improvement in spatial resolu-tion brought by interferometry is veryattractive and opens new avenues forsolving astrophysical problems.

Therefore, the demand has beengrowing in Chile for some basic and

practical information about interferome-try: the principles, the instrumental so-lutions to be adopted according to thewavelength domain, and also the effec-tive achievements of today’s interfero-metric instruments.

The idea of organising an “Interfero-metry Week” at ESO/Santiago was bornalmost two years ago and became areality on 2002 January 14–16. Afterthe traditional welcome (D. Alloin) andintroductory remarks (M. Tarenghi), wecould attend very well prepared and en-lightening lectures on the basics ofaperture synthesis (P. Lena), on inter-ferometry in the radio domain and soonwith ALMA (S. Guilloteau), on the sci-ence performed with millimetre interfer-ometers (A. Dutrey), on optical/IR inter-ferometry (A. Glindemann), on phaseclosure (M. Wittkowski), on the VLTIand its instrumentation (M. Schoeller)and, finally, on the science we candream of with optical/IR interferometers(A. Richichi).

A large audience attended the tutori-al, including a noticeable group of stu-dents from ESO/Chile, PUC and Uni-versidad de Chile. At the request ofthe attendees, and thanks to the gener-ous attention of the speakers, all pres-entations have been made availableon a webpage and can be retrievedfrom: http://www.sc.eso.org/santiago/science/interf2002.html

Once more, we thank the lecturersand the attendees for a very interestingscientific meeting which, for some ofus, has opened the door to new hori-zons in astronomical data and results.Many thanks also to the administrationstaff and to the team of the Office forScience in Santiago, in particular A.Lagarini, who have contributed in theorganisation and the success of this“Interferometry Week”.

We hope that this first contact with in-terferometry (for some of the atten-dees) will be the start of exciting workand beautiful discoveries. D. ALLOIN

ESO Studentship ProgrammeThe European Southern Observatory research student programme aims at providing the opportunities and the facilities to en-

hance the post-graduate programmes of ESO member-state universities by bringing young scientists into close contact with theinstruments, activities, and people at one of the world’s foremost observatories. For more information about ESO’s astronomicalresearch activities please see the ESO Science Activities webpage at URL http://www.eso.org/science/index.html .

Students in the programme work on an advanced research degree under the formal tutelage of their home university and de-partment, but come to either Garching or Vitacura-Santiago for a stay of up to two years to conduct part of their studies under thesupervision of an ESO staff astronomer. Candidates and their national supervisors should agree on a research project togetherwith the potential ESO local supervisor. This research programme should be described in the application and the name of the ESOlocal supervisor should also be mentioned. It is highly recommended that the applicants start their Ph.D. studies at their home in-stitute before continuing their Ph.D work and developing observational expertise at ESO.

The ESO studentship programme comprises about 14 positions, so that each year a total of up to 7 new studentships are avail-able either at the ESO Headquarters in Garching or in Chile at the Vitacura Quarters. These positions are open to students en-rolled in a Ph.D programme in the ESO member states and exceptionally at a university outside the ESO member states.

The closing date for applications is June 15, 2002.Please apply by using the ESO Studentship application form available on-line at URL http://www.eso.org/gen-fac/adm/pers/

vacant/studentship2002.html .European Southern ObservatoryStudentship ProgrammeKarl-Schwarzschild-Str. 2, 85748 Garching bei München, Germanyabeller@eso.org

43

team from Hungary, who created an in-genious game called Entropoly, waschosen among 77 competing groups tobe one of the two bests projects and itwon the trip to Chile.

After a day of recovering from the 17-hour flight to Chile, we went to La Silla.Olivier Hainaut gave an inspired guidedtour to the telescopes. From inside wecould see the 3.6-m and the NTT. Theconcept of active optics was explainedas well as all the different instrumentsincluding their purposes. For the win-ners, who have never before seen atelescope with a mirror larger than 1 min diameter, it was a thrilling moment toexperience such big telescopes moving.

In the night we could see astron-omers working in the New Swiss Tele-scope and the NTT.

At the final event in Geneva, MichelMayor gave a talk about extrasolarplanets. It was very interesting for thewinners to see the New Swiss Tele-scope where the images shown duringthat talk were obtained.

In the NTT we could follow a chang-ing of the instrument from EMMI toSOFI. We did not only see how the im-ages of the objects were taken, but wewere also told why they are so interest-ing and what science the astronomerwill do after data reduction.

“Life in the Universe”, a common ed-ucational project of ESA, ESO andCERN, had its final event during theEuropean Week of Science and Tech-nology 2001 in Geneva. The two firstprizes are a trip to Korou, sponsored byESA, and a visit to the ESO Observa-tories in Chile. In November 2001, the

“Please fasten you seat-belts, we aredescending to Santiago de Chile!” Onboard of the airplane are the win-ners of the “Life in the Universe” con-test, Mihaly Kristof, Katalin Lovei, AdamOrban, Andras Sik and Tamas Simon.All are excited and for one it is the firstflight in his life. What a chance!

The Third NEON Observing SchoolThe Network of European Observatories in the North (NEON) is pleased to announce its third observing school, sponsored by

the European Community, which will take place at Asiago Observatory (Italy) from

September 9 to 22, 2002

The school is organised jointly and alternately by Asiago Observatory (Italy); Calar Alto Observatory (Germany-Spain) andHaute-Provence Observatory (France), with additional tutorial assistance from ESO.

The purpose of the school is to provide opportunity to gain practical observational experience at the telescope, in observatorieswith state-of-the-art instrumentation. To this effect, the school proposes tutorial observations in small groups of 3 students, underthe guidance of an experienced observer, centred around a small research project and going through all steps of a standard ob-serving programme. Introductory lectures are given by experts in the field. The list of topics includes telescope optics, photome-try, spectroscopy, data analysis, etc.… Special emphasis will be given this year on polarimetry. Additional topics cover VLT instru-mentation, virtual observatories, wide-field imaging, etc.…

The school is open to students working on a PhD thesis in Astronomy, or young post-docs without previous observational ex-perience, and which are nationals of a member State or an associated State of the European Union. The working language isEnglish. Up to fifteen participants will be selected by the organising committee and will have their travel and living expenses paidif they satisfy the EC rules (age limit of 35 years at the time of the Euro Summer School).

Applicants are expected to fill in an application form (available on the Web site), with a CV and description of previous obser-vational experience, and to provide a letter of recommendation from a senior scientist familiar with the work of the applicant.

The application deadline is April 20th, 2002. Secretary of the school: Mrs. Françoise WARIN at IAP 98bis, Bd Arago F-75014 PARIS warin@iap.frFurther instructions and full practical details can be found on the school Web site, which is hosted by the European Astronomical

Society at:http://www.iap.fr/eas/schools.htmlYou will also find there a description of activities in the previous schools, hosted in 2000 by the Calar Alto Observatory, and in

2001 by the Haute-Provence Observatory. We expect the next school to be as great a success as were the previous ones! Spread the word around your community.

Michel Dennefeld, Co-ordinator of the NEON school

“Life in the Universe” Winners on La Silla and ParanalA. BACHER, ESO and Institut für Astrophysik, Leopold-Franzens-Universität Innsbruck

La Silla: Olivier Hainaut is explaining the 3.6-m telescope.

44

had a look into the control room, wherethe different purposes of the computersand monitors were illustrated.

After dinner we went again up on themountain and we saw the opening ofthe telescopes. The winners were real-ly excited seeing the “big brother” of theNTT moving. After sunset we stayed fora long time in the control room. We gotexplanations of the instruments mount-ed on each telescope as well as of theobjects imaged that night. The Tele-scope Operators, the Staff Astronom-ers and the Visiting Astronomers kindlyexplained us their work.

Of course, on both observatories wehad the possibility to ask questions,which was very important for the win-ners, not just hearing astronomers talk-ing, but speaking directly to them.

Although the long trip was very ex-hausting, we forgot it completely seeingthe observatories. All are thankful toESO for providing this nice prize andthe opportunity to see the sites, wherereal frontline astronomy is done.

got a visit to UT1, ANTU, by day. Theactive optics system and the mountedinstruments were explained. We also

The next destination was the ParanalObservatory. Humberto Varas was ourguide and showed us the site. First we

In the Footstepsof Scientists – ESA/ESO Astronomy Exercise SeriesA. BACHER1 andL. LINDBERG CHRISTENSEN2

1ESO and Institut für Astrophysik,Leopold-Franzens-UniversitätInnsbruck; 2ST-ECF

The first instalments of the “ESA/ESO Astronomy Exercise Series” hasbeen published, on the web and in print(see also ESO PR 29/01). These exer-cises allow 16–19-year-old students togain exciting hands-on experience inastronomy, making realistic calcula-tions with data obtained by the NASA/ESA Hubble Space Telescope andESO’s Very Large Telescope (VLT).Carefully prepared by astronomers andmedia experts, these exercises enablethe students to measure and calculatefundamental properties like the dis-tances to and the ages of differentkinds of astronomical objects.

Cover of General Introduction. (European Space Agency

and European Southern Observatory).

VLT: The winners and Gerd Hudepohl in front of UT1.

45

“Exercise 4: Measuring a GlobularStar Cluster’s Distance and Age”.

Each of the four exercises beginswith a background text, followed by aseries of questions, measurements andcalculations. The exercises can beused either as texts in a traditionalclassroom format or for independentstudy as part of a project undertaken insmaller groups.

The booklets are sent free-of-chargeto high-school teachers on request andmay be downloaded as PDF files fromthe website. More exercises will followin the future, e.g. measuring the veloc-ity and distance to a transneptunian ob-ject.

Contact: info@astroex.orgWeb: www.astroex.org

struct the basic luminosity-temperaturerelation (the “Hertzsprung-Russell Dia-gram”) providing a superb way to gaininsight into fundamental stellar physics.

Six Booklets

The following booklets have beenpublished:

“General Introduction” (an overviewof the exercise series),

“Toolkits” (explanation of basic astro-nomical and mathematical techniques),

“Exercise 1: Measuring the Distanceto Supernova 1987A”,

“Exercise 2: The Distance to Messier100 as Determined by Cepheid Vari-able Stars”,

“Exercise 3: Measuring the Distanceto the Cat’s Eye Nebula” and

Focus on Basic Themes

The first four exercises focus onmeasurements of distances in theUniverse.

The students apply different methodsto determine the distance of astronom-ical objects such as the supernova SN1987A, the spiral galaxy Messier 100,the Cat’s Eye Planetary Nebula and theglobular cluster Messier 12. With theseresults it is possible to make quite ac-curate estimates of the age of theUniverse and its expansion velocity,without the use of computers or sophis-ticated software.

Students can also perform ‘naked-eye photometry’ by measuring thebrightness of stars on two VLT images(taken through blue and green opticalfilters, respectively). They can then con-

January–February 2002

1454. M.J. Neeser, P.D. Sackett, G. De Marchi, F. Paresce:Detection of a Thick Disk in the edge-on Low SurfaceBrightness Galaxy ESO 342–G017. I. VLT Photometry in Vand R Bands. A&A.

1455. D. Elbaz, C.J. Cesarsky, P. Chanial, H. Aussel, A. Fran-ceschini, D. Fadda and R.R. Charry: The bulk of the cosmicinfrared background resolved by ISOCAM. A&A.

1456. T.-S. Kim, S. Cristiani and S. D’Odorico: The evolution of thephysical state of the IGM. A&A.

International Staff(1 January 2002 – 31 March 2002)

ARRIVALS

EUROPE

DELMOTTE, Nausicaa (F), StudentDEPAGNE, Christophe (F), StudentGUIDOLIN, Ivan Maria (I), AssociateIAITSKOVA, Natalia (RU), Associate PAUFIQUE, Jérôme (F), Engineer Adaptive OpticsSTOLTE, Andrea (D), AssociateTAYLOR, Luke (GB), AssociateWOLFF, Burkhard (D), Astronomical Data Quality Control

Scientist

CHILE

KERVELLA, Pierre (F), VLTI AstronomerLEDOUX, Cédric (F), Operations Staff AstronomerMORELLI, Lorenzo (I), StudentPINTE, Christophe (F), AssociateRABELING, David (NL), AssociateRATHBORNE, Jill (AUS), Associate SEST

DEPARTURES

EUROPE

DEMOULIN-ARP, Marie-Hélène (F), AstronomerDESSAUGES-ZAVADSKY, Miroslava (CH), Student

FARINATO, Jacopo (I), Support EngineerGENNAI, Alberto (I), Control/Hardware EngineerSANNER, Jörg (D), AssociateTRIPICCHIO, Alfredo (I), AssociateWEBER, Ingrid (D), Secretary

CHILE

GARCÍA AGUIAR, Martina (D), Mechanical Engineer

Local Staff(1 December 2001 – 28 February 2002)

ARRIVALS

ESPARZA MORALES CRISTIAN, Telescope InstrumentsOperator, La Silla

FAUNDEZ MORENO LORENA, Telescope InstrumentsOperator, Paranal

LA FUENTE PE A EDUARDO, Telescope InstrumentsOperator, La Silla

PALACIO VALENZUELA JUAN CARLOS, MechanicalEngineer, Paranal

RIVERA MAITA ROBERTO, Temporary Site Testing, ParanalSOTO TRONCOSO RUBEN, Software Engineer, La SillaSTRUNK SANDRA, Executive Bilingual Secretary, ParanalVALENZUELA SOTO JOSE JAVIER, Instrumentation

Engineer, La Silla

DEPARTUREVERA ROJAS ESTEBAN, Electronics Engineer, Paranal

PERSONNEL MOVEMENTS

1457. D. Fadda, H. Flores, G. Hasinger, A. Franceschini, B. Altieri,C.J. Cesarsky, D. Elbaz and Ph. Ferrando: The AGN contri-bution to mid-infrared surveys. X-ray counterparts of the mid-IR sources in the Lockman Hole and HDF. A&A.

1458. Y. Momany, E.V. Held, I. Saviane and L. Rizzi: The Sagittariusdwarf irregular galaxy: metallicity and stellar populations.A&A.

1459. A. Franceschini, D. Fadda, C.J. Cesarsky, D. Elbaz, H. Flores,G.L. Granato: ESO investigates the nature of extremely-redhard X-ray sources responsible for the X-ray background.ApJ.

LIST OF SCIENTIFIC PREPRINTS

46

Scale Survey and its Multi-λ Follow-up 105, 32

W. Gieren, D. Geisler, T. Richtler, G. Pie-trzynski, B. Dirsch: Research in Concep-ción on Globular Cluster Systems andGalaxy Formation, and the ExtragalacticDistance Scale 106, 15

H. Böhringer, P. Schuecker, P. Lynam, Th.Reiprich, C.A. Collins, L. Guzzo, Y. Ikebe,E. Molinari, L. Barone, C. Ambros: Har-vesting the Results from the REFLEXCluster Survey: Following-up on an ESOKey Programme 106, 24

F. Bouchy and F. Carrier: Catching theSounds of Stars. Asteroseismology, theRight Tool to Understand Stellar Interiors106, 32

S. Van Eck, S. Goriely, A. Jorissen and B.Plez: Discovery of Lead Stars with theESO 3.6-m Telescope and CES 106, 37

L.-Å. Nyman, M. Lerner, M. Nielbock, M.Anciaux, K. Brooks, R. Chini, M. Albrecht,R. Lemke, E. Kreysa, R. Zylka, L.E.B.Johansson, L. Bronfman, S. Kontinen, H.Linz and B. Stecklum: Simba Explores theSouthern Sky 106, 40

OTHER ASTRONOMICAL NEWS

EIS Data on the Chandra Deep Field SouthReleased 103, 37

J.R. Walsh, L. Pasquini, S. Zaggia: Reporton the FLAMES Users Workshop (FUW):105, 37

R. Fosbury, J. Bergeron, C. Cesarsky, S.Cristiani, R. Hook, A. Renzini and P. Ro-sati: The Great Observatories OriginsDeep Survey (GOODS) 105, 40

D. Alloin: ESO: Research Facilities in San-tiago 105, 42

U. Grothkopf, A. Treumann, M.E. Gómez:The ESO Libraries: State of the Art 2001105, 45

News from Santiago. S. Ellison and M. Ster-zik: Summaries of 2 Topical Meetings106, 45

L. Wisotzki: The ESO Users Committee106, 46

C. Madsen: ESO Presentation in Brussels106, 48

ANNOUNCEMENTS

ESO Vacancy. Head of the Office forScience 103, 36

ESO Studentship Programme 2001 103, 38Personnel Movements 103, 38ESO Vacancies. A Challenge for Astron-

omers, Engineers in the Field of Soft-ware, Electronics and / or Mechanics103, 39

Scientific Preprints (January – March 2001)103, 40

Preliminary Announcement of an ESO-CERN-ESA Symposium on “Astronomy,Cosmology and Fundamental Physics”104, 38

International Conference on Light Pollution104, 38

Koehler, S. Lévêque, A. Longinotti, A.Lopez, S. Ménardi, S. Morel, F. Paresce,T. Phan Duc, A. Ramirez, A. Richichi, M.Schöller, J. Spyromilio, M. Tarenghi, A.Wallander, R. Wilhelm, M. Wittkowski:First Fringes with ANTU and MELI-PAL 106, 1

P. Ballester, P. Kervella, L.P. Rasmussen, A.Richichi, C. Sabet, M. Schöller, R. Wil-helm, B. Wiseman, M. Wittkowski: TheVLTI Data Flow System: From Observa-tion Preparation to Data Processing106, 2

S. Habraken, P.-A. Blanche, P. Lemaire, N.Legros, H. Dekker and G. Monnet: Vol-ume Phase Holographic Gratings Made inEurope 106, 6

A. Wicenec et al.: ESO’s Next GenerationArchive System 106, 11

News from the 2p2 Team 106, 13

REPORTS FROM OBSERVERS

J. Alves, C. Lada and E. Lada: Seeing theLight Through the Dark 103, 1

F. Comerón and M. Fernández: Strong Ac-cretion and Mass Loss Near the Sub-stellar Limit 103, 21

J.U. Fynbo, P. Møller, B. Thomsen: StarFormation at z = 2–4: Going Below theSpectroscopic Limit with FORS1 103, 24

H. Jones, B. Stappers, B. Gaensler: Dis-covery of a Bow-Shock Nebula Aroundthe Pulsar B0740-28 103, 27

L. Testi, L. Vanzi, F. Massi: Young StellarClusters in the Vela D Molecular Cloud103, 28

P. Amram and G. Östlin: Building LuminousBlue Compact Galaxies by Merging103, 31

L. VANZI, L. Hunt and T. Thuan: SOFI andISAAC Pierce the Obscured Core of SBS0335-052 104, 22

J.W. Sulentic, M. Calvani and P. Marziani:Eigenvector 1: An H-R Diagram forAGN? 104, 25

J.-L. Monin, F. Ménard and N. Peretto: DiskOrientations in PMS Binary SystemsDetermined Through Polarimetric Imag-ing With UT1/FORS 104, 29

S. Bagnulo, T. Szeifert, G.A. Wade, J.D.Landstreet and G. Mathys: DetectingMagnetic Fields of Upper-Main-SequenceStars With FORS1 at ANTU 104, 32

Colour Composite of the Spiral Galaxy M83in the Visible 104, 37

S. Zaggia, Y. Momany, B. Vandame, R.P.Mignani, L. Da Costa, S. Arnouts, M.A.T.Groenewegen, E. Hatziminaoglou, R.Madejsky, C. Rité, M. Schirmer, and R.Slijkhuis: The EIS Pre-FLAMES Survey:Observations of Selected Stellar Fields105, 25

M. Pierre, D. Alloin, B. Altieri, M. Birkinshaw,M. Bremer, H. Böhringer, J. Hjorth, L.Jones, O. Le Fèvre, D. Maccagni, B.McBreen, Y. Mellier, E. Molinari, H. Quin-tana, H. Rottgering, J. Surdej, L. Vigroux,S. White, C. Lonsdale: The XMM Large

SUBJECT INDEX

TELESCOPES AND INSTRUMENTATION

A. Gilliotte: La Silla Telescope Status, a GreatAchievement on Image Quality Per-formances 103, 2

A. Gilliotte: Image Quality Improvement ofthe 2.2-m 103, 4

O. Hainaut and the NTT Team: NEWS fromthe NTT 103, 7

H. Jones: 2p2 Team News 132, 9H. Jones, A. Renzini, P. Rosati and W. Sei-

fert: Tunable Filters and Large Tele-scopes 103, 10

C. Cesarsky: Successful First Light for theVLT Interferometer 104, 1

A. Glindemann, B. Bauvir, F. Delplancke,F. Derie, E. Di Folco, A. Gennai, P. Gitton,N. Housen, A. Huxley, P. Kervella, B.Koehler, S. Lévêque, A. Longinotti, S.Ménardi, S. Morel, F. Paresce, T. PhanDuc, A. Richichi, M. Schöller, M. Tarenghi,A. Wallander, M. Wittkowski, R. Wilhelm:Light at the End of the Tunnel – FirstFringes with the VLT 104, 2

F. Paresce: Scientific Objectives of the VLTInterferometer 104, 5

H. Jones, E. Pompei and the 2p2 Team: 2p2Team News 104, 7

L. Wisotzki, F. Selman and A. Gilliotte: Com-missioning the Spectroscopic Mode of theWFI at the MPG/ESO 2.2-m Telescope atLa Silla 104, 8

C. Urrutia, T. Paz, E. Robledo, F. Gutierrez,D. Baade, F. Selman, F. Sanchez, J.Brewer and M. Scodeggio: VLT StyleObserving with the Wide Field Imager atthe MPG/ESO 2.2-m Telescope at LaSilla 104, 14

J. Manfroid, F. Selman and H. Jones:Achieving 1% Photometric Accuracy withthe ESO Wide Field Imager 104, 16

R. Hook: Scisoft – a Collection of Astro-nomical Software for ESO Users 104, 20

C. Madsen: Latest News: ESO High-LevelPresentation in Porto 104, 21

D. Queloz, M. Mayor et al.: From CORALIEto HARPS. The Way Towards 1 m s–1

Precision Doppler Measurements 105, 1R. Bender, G. Monnet and A. Renzini: Work-

shop on Scientific Drivers for FutureVLT/VLTI Instrumentation – Summaryand First Orientations 105, 8

D. Bonaccini, W. Hackenberg, M. Cullum, E.Brunetto, M. Quattri, E. Allaert, M. Dimm-ler, M. Tarenghi, A. Van Kersteren, C. diChirico, M. Sarazin, B. Buzzoni, P. Gray,R. Tamai, M. Tapia, R. Davies, S. Rabien,T. Ott, S. Hippler: ESO VLT Laser GuideStar Facility 105, 9

D. Silva: Service Mode Scheduling: A Primerfor Users 105, 18

Hunting the Southern Skies with SIMBA105, 24

A. Glindemann, P. Ballester, B. Bauvir,E. Bugueño, M. Cantzler, S. Correia,F. Delplancke, F. Derie, P. Duhoux, E. DiFolco, A. Gennai, B. Gilli, P. Giordano,P. Gitton, S. Guisard, N. Housen, A.Huxley, P. Kervella, M. Kiekebusch, B.

MESSENGER INDEX 2001 (Nos. 103–106)

47

L. Noethe: Book Review: Reflecting Tele-scope Optics II, by Ray Wilson 106, 49

Personnel Movements 106, 50Eurowinter School. Observing with the Very

Large Telescope Interferometer 106, 51List of Scientific Preprints (October–Decem-

ber 2001) 106, 52

Vacancy Note: “A Challenge for Astronom-ers, Engineers in the field of Software,Electronics and/or Mechanics…” 105, 50

Personnel Movements 105, 51Scientific Preprints 105, 51ESO Publications Still Available 105, 51International Max-Planck Research School

on Astrophysics 106, 49

ESO Vacancy “Head of Administration”104, 39

Personnel Movements 104, 39Scientific Preprints (April–June 2001)

104, 39ESO-CERN-ESA Symposium “Astronomy,

Cosmology and Fundamental Physics”105, 49

AUTHOR INDEX

AD. Alloin: ESO: Research Facilities in

Santiago 105, 42J. Alves, C. Lada and E. Lada: Seeing the

Light Through the Dark 103, 1P. Amram and G. Östlin: Building Luminous

Blue Compact Galaxies by Merging103, 31

B

S. Bagnulo, T. Szeifert, G.A. Wade, J.D.Landstreet and G. Mathys: DetectingMagnetic Fields of Upper-Main-SequenceStars With FORS1 at ANTU 104, 32

P. Ballester, P. Kervella, L.P. Rasmussen, A.Richichi, C. Sabet, M. Schöller, R. Wil-helm, B. Wiseman, M. Wittkowski: TheVLTI Data Flow System: From Observa-tion Preparation to Data Processing106, 2

R. Bender, G. Monnet and A. Renzini: Work-shop on Scientific Drivers for FutureVLT/VLTI Instrumentation – Summaryand First Orientations 105, 8

H. Böhringer, P. Schuecker, P. Lynam, Th.Reiprich, C.A. Collins, L. Guzzo, Y. Ikebe,E. Molinari, L. Barone, C. Ambros: Har-vesting the Results from the REFLEXCluster Survey: Following-up on an ESOKey Programme 106, 24

D. Bonaccini, W. Hackenberg, M. Cullum, E.Brunetto, M. Quattri, E. Allaert, M. Dimm-ler, M. Tarenghi, A. Van Kersteren, C. diChirico, M. Sarazin, B. Buzzoni, P. Gray,R. Tamai, M. Tapia, R. Davies, S. Rabien,T. Ott, S. Hippler: ESO VLT Laser GuideStar Facility 105, 9

F. Bouchy and F. Carrier: Catching theSounds of Stars. Asteroseismology, theRight Tool to Understand Stellar Interiors106, 32

C

C. Cesarsky: Successful First Light for theVLT Interferometer 104, 1

F. Comerón and M. Fernández: Strong Ac-cretion and Mass Loss Near the Sub-stellar Limit 103, 21

ES. Ellison and M. Sterzik: News from San-

tiago. Summaries of 2 Topical Meetings106, 45

FR. Fosbury, J. Bergeron, C. Cesarsky, S.

Cristiani, R. Hook, A. Renzini and P.Rosati: The Great Observatories OriginsDeep Survey (GOODS) 105, 40

J.U. Fynbo, P. Møller, B. Thomsen: StarFormation at z = 2–4: Going Belowthe Spectroscopic Limit with FORS1103, 24

G

W. Gieren, D. Geisler, T. Richtler, G. Pie-trzynski, B. Dirsch: Research in Concep-ción on Globular Cluster Systems andGalaxy Formation, and the ExtragalacticDistance Scale 106, 15

A. Gilliotte: La Silla Telescope Status, aGreat Achievement on Image Quality Per-formances 103, 2

A. Gilliotte: Image Quality Improvement ofthe 2.2-m 103, 4

A. Glindemann, B. Bauvir, F. Delplancke,F. Derie, E. Di Folco, A. Gennai, P. Gitton,N. Housen, A. Huxley, P. Kervella, B.Koehler, S. Lévêque, A. Longinotti, S.Ménardi, S. Morel, F. Paresce, T. PhanDuc, A. Richichi, M. Schöller, M. Tarenghi,A. Wallander, M. Wittkowski, R. Wilhelm:Light at the End of the Tunnel – FirstFringes with the VLT 104, 2

A. Glindemann, P. Ballester, B. Bauvir, E.Bugueño, M. Cantzler, S. Correia, F.Delplancke, F. Derie, P. Duhoux, E. DiFolco, A. Gennai, B. Gilli, P. Giordano, P.Gitton, S. Guisard, N. Housen, A. Huxley,P. Kervella, M. Kiekebusch, B. Koehler, S.Lévêque, A. Longinotti, A. Lopez, S. Mé-nardi, S. Morel, F. Paresce, T. Phan Duc,A. Ramirez, A. Richichi, M. Schöller, J.Spyromilio, M. Tarenghi, A. Wallander, R.Wilhelm, M. Wittkowski: First Fringes withANTU and MELIPAL 106, 1

U. Grothkopf, A. Treumann, M.E. Gómez:The ESO Libraries: State of the Art2001 105, 45

HS. Habraken, P.-A. Blanche, P. Lemaire, N.

Legros, H. Dekker and G. Monnet: Vol-ume Phase Holographic Gratings Made inEurope 106, 6

O. Hainaut and the NTT Team: NEWS fromthe NTT 103, 7

R. Hook: Scisoft – a Collection of Astro-nomical Software for ESO Users 104, 20

JH. Jones: 2p2 Team News 132, 9H. Jones, A. Renzini, P. Rosati and W.

Seifert: Tunable Filters and Large Tele-scopes 103, 10

H. Jones, E. Pompei and the 2p2 Team: 2p2Team News 104, 7

H. Jones, B. Stappers, B. Gaensler: Dis-covery of a Bow-Shock Nebula Aroundthe Pulsar B0740-28 103, 27

MC. Madsen: Latest News: ESO High-Level

Presentation in Porto 104, 21C. Madsen: ESO Presentation in Brussels

106, 48J. Manfroid, F. Selman and H. Jones:

Achieving 1% Photometric Accuracy withthe ESO Wide Field Imager 104, 16

J.-L. Monin, F. Ménard and N. Peretto: DiskOrientations in PMS Binary SystemsDetermined Through Polarimetric Imag-ing With UT1/FORS 104, 29

NL. Noethe: Book Review: Reflecting Tele-

scope Optics II, by Ray Wilson 106, 49L.-Å. Nyman, M. Lerner, M. Nielbock, M.

Anciaux, K. Brooks, R. Chini, M. Albrecht,R. Lemke, E. Kreysa, R. Zylka, L.E.B.Johansson, L. Bronfman, S. Kontinen, H.Linz and B. Stecklum: Simba Explores theSouthern Sky 106, 40

PF. Paresce: Scientific Objectives of the VLT

Interferometer 104, 5M. Pierre, D. Alloin, B. Altieri, M. Birkinshaw,

M. Bremer, H. Böhringer, J. Hjorth, L.Jones, O. Le Fèvre, D. Maccagni, B. Mc-Breen, Y. Mellier, E. Molinari, H. Quintana,H. Rottgering, J. Surdej, L. Vigroux, S.White, C. Lonsdale: The XMM Large ScaleSurvey and its Multi-λ Follow-up 105, 32

QD. Queloz, M. Mayor et al.: From CORALIE

to HARPS. The Way Towards 1 m s–1

Precision Doppler Measurements 105, 1

SD. Silva: Service Mode Scheduling: A Primer

for Users 105, 18J.W. Sulentic, M. Calvani and P. Marziani:

Eigenvector 1: An H-R Diagram for AGN?104, 25

T

L. Testi, L. Vanzi, F. Massi: Young Stellar Clus-ters in the Vela D Molecular Cloud 103, 28

UC. Urrutia, T. Paz, E. Robledo, F. Gutierrez, D.

Baade, F. Selman, F. Sanchez, J. Brewerand M. Scodeggio: VLT Style Observ-ing with the Wide Field Imager at theMPG/ESO 2.2-m Telescope at La Silla104, 14

48

L. Wisotzki, F. Selman and A. Gilliotte: Com-missioning the Spectroscopic Mode of theWFI at the MPG/ESO 2.2-m Telescope atLa Silla 104, 8

L. Wisotzki: The ESO Users Commit-tee 106, 46

ZS. Zaggia, Y. Momany, B. Vandame, R.

P. Mignani, L. Da Costa, S. Arnouts, M.A.T.Groenewegen, E. Hatziminaoglou, R. Ma-dejsky, C. Rité, M. Schirmer, and R. Slijk-huis: The EIS Pre-FLAMES Survey: Ob-servations of Selected Stellar Fields105, 25

VS. Van Eck, S. Goriely, A. Jorissen and B.

Plez: Discovery of Lead Stars with theESO 3.6-m Telescope and CES 106, 37

L. VANZI, L. Hunt and T. Thuan: SOFI andISAAC Pierce the Obscured Core of SBS0335-052 104, 22

WJ.R. Walsh, L. Pasquini, S. Zaggia: Report

on the FLAMES Users Workshop (FUW)105, 37

A. Wicenec et al.: ESO’s Next GenerationArchive System 106, 11

ContentsTELESCOPES AND INSTRUMENTATION

W. Brandner et al.: NAOS+CONICA at YEPUN: First VLT OpticsSystem Sees First Light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

R. Kurz, S. Guilloteau, P. Shaver: The Atacama Large Millimetre Array 7L. Germany: News from the 2p2 Team . . . . . . . . . . . . . . . . . . . . . . . . 13

CHILEAN ASTRONOMY

L. Bronfman: A Panorama of Chilean Astronomy . . . . . . . . . . . . . . . . 14A. Reisenegger, H. Quintana, D. Proust, E. Slezak: Dynamics and

Mass of the Shapley Supercluster, the Largest Bound Structure in the Local Universe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

REPORTS FROM OBSERVERS

D. Baade, T. Rivinius, S. Štefl: To Be or Not to Be and a 50-cm Post-Mortem Eulogy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

F. Courbin, G. Letawe, P. Magain, L. Wisotzki, P. Jablonka, D. Alloin, K. Jahnke, B. Kuhlbrodt, G. Meylan, D. Minniti: Spectroscopy ofQuasar Host Galaxies at the VLT: Stellar Populations and Dynamics Down to the Central Kiloparsec . . . . . . . . . . . . . . . . . . 28

J. Sollerman and V. Flyckt: The Crab Pulsar and its Environment . . . . 32L. Vanzi, F. Mannucci, R. Maiolino, M. Della Valle: SOFI Discovers

a Dust Enshrouded Supernova . . . . . . . . . . . . . . . . . . . . . . . . . . . 35G. Rupprecht: A Deep Look at an Active Galaxy . . . . . . . . . . . . . . . . . 38Coming Home at Paranal. Unique “Residencia” Opens at the

VLT Observatory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

OTHER ASTRONOMICAL NEWS AND ANNOUNCEMENTS

F. Paresce: Release of Scientific Data from VLTI Commissioning . . . . 41A. Renzini: VLT Science Verification Policy and Procedures . . . . . . . . 41D. Alloin: News from Santiago . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42ESO Studentship Programme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42M. Dennefeld: The Third NEON Observing School . . . . . . . . . . . . . . . 43A. Bacher: “Life in the Universe” Winners on La Silla and Paranal . . . 43A. Bacher and L. Lindberg Christensen: In the Footsteps of

Scientists – ESA/ESO Astronomy Exercise Series . . . . . . . . . . . . 44Personnel Movements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45List of Scientific Preprints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

MESSENGER INDEX 2001 (Nos. 103–106) . . . . . . . . . . . . . . . . . . . 46

ESO, the European Southern Observa-tory, was created in 1962 to “… establishand operate an astronomical observatoryin the southern hemisphere, equippedwith powerful instruments, with the aim offurthering and organising collaboration inastronomy …” It is supported by ninecountries: Belgium, Denmark, France,Germany, Italy, the Netherlands, Portugal,Sweden and Switzerland. ESO operatesat two sites. It operates the La Silla ob-servatory in the Atacama desert, 600 kmnorth of Santiago de Chile, at 2,400 m al-titude, where several optical telescopeswith diameters up to 3.6 m and a 15-msubmillimetre radio telescope (SEST) arenow in operation. In addition, ESO is in theprocess of building the Very LargeTelescope (VLT) on Paranal, a 2,600 mhigh mountain approximately 130 kmsouth of Antofagasta, in the driest part ofthe Atacama desert. The VLT consists offour 8.2-metre telescopes. These tele-scopes can also be used in combinationas a giant interferometer (VLTI). The firsttwo 8.2-metre telescopes (called ANTUand KUEYEN) are in regular operation,and the other two will follow soon. Over1200 proposals are made each year forthe use of the ESO telescopes. The ESOHeadquarters are located in Garching,near Munich, Germany. This is the scien-tific, technical and administrative centre ofESO where technical development pro-grammes are carried out to provide the LaSilla and Paranal observatories with themost advanced instruments. There arealso extensive astronomical data facilities.In Europe ESO employs about 200 inter-national staff members, Fellows andAssociates; in Chile about 70 and, in ad-dition, about 130 local staff members.

The ESO MESSENGER is published fourtimes a year: normally in March, June,September and December. ESO also pub-lishes Conference Proceedings, Pre-prints, Technical Notes and other materialconnected to its activities. Press Releasesinform the media about particular events.For further information, contact the ESOEducation and Public Relations Depart-ment at the following address:

EUROPEAN SOUTHERN OBSERVATORYKarl-Schwarzschild-Str. 2 D-85748 Garching bei München Germany Tel. (089) 320 06-0 Telefax (089) 3202362ips@eso.org (internet) URL: http://www.eso.orghttp://www.eso.org/gen-fac/pubs/messenger/

The ESO Messenger:Editor: Marie-Hélène DemoulinTechnical editor: Kurt Kjär

Printed by Universitätsdruckerei WOLF & SOHNHeidemannstr. 166D-80939 MünchenGermany

ISSN 0722-6691