= 1200 ñ 2400. R m wavelength range, providing m NEWSLETTER and AAT observations of a newlyñdiscovered pulsar wind nebula (Heath Jones et al.) et Jones (Heath nebula wind pulsar newlyñdiscovered a of observations AAT and NTT Orchiston) (Wayne Telescope Radio ëHoleñinñtheñGroundí Heights Dover The Discovery of a shockñexcited wind in the starñburst galaxy NGC 1482 (Sylvain Veilleux & David Rupke) David & Veilleux (Sylvain 1482 NGC galaxy starñburst the in wind shockñexcited a of Discovery al.) et Baldry (Ivan history starñformation and spectra Cosmic Mauch) (Tom data test ñ 6dFGS the in sources Radio al.) et Bedding (Tim asteroseismology of birth the UCLES: with oscillations stellar of Detection Tadhunter) C. & SolÛrzanoñIÒarrea (C. 265 3C galaxy radio extreme the in cones illumination and Shocks This issue celebrates the commissioning and first science runs of IRIS2, the AAOís new nearñ new AAOís the IRIS2, of runs science first and commissioning the celebrates issue This infrared camera and spectrograph. IRIS2 covers the 1ñ2.5 with spectroscopy longñslit and fieldñofñview, arcmin 7.7 x 7.7 a over imaging we Above, see on working Denis Whittard IRIS2, and the first IRIS2 image ñ the Tarantula 25. to 15 pp see story, full the For LMC. the in Doradus) (30 Nebula IRIS2 hits the skies the hits IRIS2 14 26 6 8 10 12 13
contents
NUMBER NUMBER 99 2002 FEBRUARY ANGLO-AUSTRALIAN OBSERVATORY DIRECTOR'S MESSAGE
DIRECTORíS MESSAGE
Eighteen months ago, the strategies for the AAO over the next decade were mapped out in the ìAAO of the Futureî. Even at that time, it was clear that the strategies were ambitious and that the implementation of the plan would be challenging. The AAO moved forward into a future that promised much, but held many uncertainties. Some of the greatest challenges lay in instrumentation, with the AAO forging new directions in building instruments for other telescopes and setting itself ambitious targets to complete the upgrade of its own instrument suite and infrastructure by 2005.
Perhaps more than anything else over the intervening period, two key achievements in the past three months have demonstrated that the AAO has risen to meet these instrumentation challenges and is thus significantly closer to making the vision encapsulated in ìAAO of the Futureî a reality.
The first of these achievements is the successful commissioning of IRIS2, reported on in detail by Chris Tinney and the IRIS2 team in this Newsletter. With IRIS2, the AAO regains its capability in nearñIR astronomy, once more offering competitive wideñfield infrared imaging and spectroscopy with the AAT. It also brings the AAO a step closer to providing the core instrumentation capability of AAOmega, WFI, IRIS2, UCLES/UHRF and 6dF outlined in the ìAAO of the Futureî. Only AAOmega remains to complete the instrumentation suite, with three out of the five instruments (WFI, IRIS2 and 6dF) having been commissioned over the past year. The success of IRIS2 reflects an Observatoryñwide effort from the design and building of the instrument in Epping to the work on interface issues, helium reticulation and commissioning at Siding Spring.
Equally key to the future success of the AAO have been the achievements of the OzPoz team. In December, OzPoz was formally accepted by ESO in Australia and will have been shipped to Paranal by the time you read this. This also represents the culmination of a huge amount of hard work by the OzPoz team whose outstanding effort was clearly recognised by ESO. Although commissioning still lies ahead, the success of the AAOís first multiñmillion dollar external project augurs well for the future. Working with ESO has also been immensely positive for the AAO and the experience gained in key project disciplines has been invaluable.
Moving towards the ìAAO of the Futureî also means letting go of the past. The announcements in this Newsletter of the decommissioning of many AAO instruments and the termination of routine photography by the UK Schmidt reflect the need to move on to new opportunities. Equally, we need to recognise that against the backdrop of declining operational resources for AAT instrumentation, difficult decisions have to be made. Perhaps the most difficult decision with regard to AAT instrumentation has been the eventual decommissioning of Taurus. As evidenced by no less than three articles in the current newsletter, it is clear that Taurus continues to do excellent and unique science.
However, the ìAAO of the Futureî is about new opportunities such as those provided by IRIS2 and OzPoz. Over the past few months, the AAO has taken its first major steps towards the ìAAO of the Futureî with both confidence and success.
Brian Boyle
ANGLO-AUSTRALIAN OBSERVATORY page 2 NEWSLETTER FEBRUARY 2002 ANNOUNCEMENTS
RUSSELL CANNONíS RETIREMENT
Russell Cannon retires formally from the AAO on Friday, 1 March. For those who know Russell, it will come as no surprise that this ëformalí retirement will simply provide an even greater opportunity for Russell to focus on his various research programs, mostly involving 2dF. As an honorary senior AAO research astronomer, Russell will retain his office at Epping and continue to observe on the AAT and UKST as and when he is awarded time.
I hope Russell will not mind if I also take this opportunity to acknowledge Russellís pivotal contribution to the AAO over the past 15 years. Russell joined the AAO in 1986 and one of his first major achievements was to successfully guide the agreement whereby the AAO took over operational responsibility for the UK Schmidt in 1988. At the same time he successfully co-ordinated the AAOís response to SN 1987A, maximising the scientific output from this unique event. As Director, Russell also oversaw the commissioning of UCLES and Autofib, and the successful development of IRIS, AAOís first imaging infrared camera, commissioned in 1991.
However, it was Russellís long association with survey astronomy and his realisation that the AAO needed to move on, if it was to remain a competitive telescope in the 8ñm era, that provided much of the scientific drive and strategic vision behind the 2ñdegree field. Russellís support and advocacy of 2dF was the critical element in getting this project approved and funded in the early 90s. Equally, it took an immense amount of hard work and tenacity to keep the project going through the early years, culminating in the official opening of 2dF in 1995. That we have 2dF today is testament to the fact that Russell never lost this vision.
Following his second fiveñyear term as Director, Russell moved on to become an AAO senior research astronomer, scientifically exploiting the 2dF instrument he had worked so hard to ensure became a reality. He has many ongoing scientific programs with 2dF (perhaps the astronomer with the greatest number) including the galaxy redshift survey, Magellanic Cloud carbon stars and globular cluster abundances. As a 2dF support astronomer, Russell has also spent many nights (and days) relentlessly tracking down the outstanding throughput and positioning issues with 2dF, in turn providing valuable input to the AAOmega project.
I am sure that everyone would want to wish Russell and Fernanda all the best for the years ahead, and hope that Russell manages to find some time from his research to realise some of his other ambitions!
Brian Boyle
Russell Cannon in front of 2dF at the Opening in 1995.
ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 3 FEBRUARY 2002 THE FUTURE AVAILABILITY OF AAT INSTRUMENTATION ANNOUNCEMENTS The ìAAO of the Futureî Strategic Plan envisages the rationalisation of the AATís instrument suite, in order to maintain a scientifically competitive facility in the reduced budget environment of postñ2005. The AAOís facilities will then be: ï AAOmega (multi-object spectroscopy and IFU spectroscopy with the SPIRAL IFU) ï UCLES/UHRF ï IRIS2 ï WFI ï Detector Control via new AAO2 controllers ï Instrument Control via new Unixñbased control interfaces ï Upgrade of the Telescope Control System away from the existing Interdata computers ï Any approved visitor instrumentation So that the AAO can meet the deadlines for the development of the many new components required by this ambitious plan, rationalisation of the current instrument suite will need to begin well in advance. The AAO therefore takes this opportunity to make the following announcements:
IRIS1: with the commissioning of IRIS2 in 2001B, IRIS1 is now formally decommissioned.
UNSWIRF: with the decommissioning of IRIS1, UNSWIRF can now no longer be used until upgraded for use with IRIS2, and in its current form it is now formally decommissioned.
Aux CCD: with the implementation of new CCDñbased A&G cameras, the role of this system in the acquisition of faint targets has been superseded. Together with the prohibitive overheads involved in the operation of this system, this has led to the Aux CCD being now formally decommissioned.
LDSS++: Taurus has now been demonstrated to deliver useful multiñobject spectroscopic performance for most applications. Given this, the overheads involved in maintaining and supporting TWO multiñobject instruments can no longer be supported. LDSS will be decommissioned with immediate effect.
SPIRAL: will be kept available to applicants until the end of Semester 2003B, at which point it will need to be decommissioned in order to be reñintegrated with the AAOmega spectrographs.
RGO: much of the functionality of the RGO spectrograph will be superseded by the SPIRAL + AAOmega combination. Moreover, as other facilities, including ESO (for UK observers), SSO 2.3ñm (Aus) and GMOS (UK/Aus), have made RGO uncompetitive, its decommissioning needs to be addressed. We therefore announce that RGO will remain available until the end of Semester 2003B, when we propose to decommission the facility.
Taurus: Taurus has clearly been one of the AAOís most innovative instruments. Unfortunately, the cost to convert Taurus to the new Instrument Control environment lies well outside the currently planned AAT instrumentation budget for 2002-05. The availability of Taurus in the ìAAO of the Futureî postñ2005 is therefore not foreseen at present. However, the AAO will continue to offer Taurus as an AAT instrument at least until the end of Semester 2003B and, as resources permit, until the end of 2005B. Further, the AAO would be very interested in receiving proposals for external funding which would see us able to continue to offer Taurus in some form beyond 2005.
Visiting instruments will be still be supported within the current guidelines (http://www.aao.gov.au/visitor_inst/ wwwvisitor.html).
Users are also reminded of the revised scheduling constraints on runs of less than 3 nights and the need to include a weather allowance in all AAT applications. Details of these constraints along with all other scheduling/instrument policies are available from http://www.aao.gov.au/astro/policies.html.
ANGLO-AUSTRALIAN OBSERVATORY page 4 NEWSLETTER FEBRUARY 2002 nonñsurvey 6dF work. They will be scheduled under UKST PHOTOGRAPHIC PROGRAM appropriate ëoverñrideí conditions, but only in cases Brian Boyle where they override lowerñranked 6dF nonñsurvey
proposals. ANNOUNCEMENTS Following the successful commissioning of 6dF and the Over the next few months, the UK Schmidt will also start of the 6dF scheduled observations in midñ2001, give heightened priority (up to 50% of the total time the AAO is looking to expedite the completion of the available) to the photographic surveys; Southern I, photographic program on the UK Schmidt telescope so Halpha/short red and the Northern POSSñII I. The goal it may focus on the new scientific opportunities provided will be to complete all priority 1 fields in the Southern I by 6dF. and Halpha surveys, giving coverage over the full survey Based on advice received from the UK Schmidt TAGs area at grade ëBí or better. The UK Schmidt will also and AAO Usersí Committee, the Director wishes to attempt to complete, as far as possible, all outstanding announce that the forthcoming March 2002 deadline will POSS II I and remaining ëBí grade fields in the Southern I be the last opportunity on which nonñsurvey survey prior to May 2002. Together with nonñsurvey photographic applications will be accepted by the UK work, these will done in time currently not used by 6dF. Schmidt Time Assignment Committees. Photographic Outstanding films required to complete the Halpha applications submitted at this time may be for any survey to A grade will only be taken in brightñofñ-moon field(s) accessible over the semesters 2002B and 2003A. time unusable by 6dF programs until the end of semester 2002A. All applicants with photographic applications currently ëactiveí on the UK Schmidt are asked to reñsubmit an In the longer term the AAO plans retain the option to application at this deadline if they wish their program to carry out photographic work, but only under exceptional be considered for further time. All nonñsurvey circumstances (i.e. Target of Opportunity). After the applications submitted at this time will be deemed to March 2002 deadline, such applications for photographic have a oneñyear lifetime i.e. they will expire at the end work should be submitted to the AAO Director, who will of July 2003. Applications for nonñsurvey photographic award time on an override basis only to those programs work will be graded in the same way as applications for that can make a compelling case for their exceptional and unique nature.
Next Issue: We are now producing AAO Newsletters three times a year ñ February, June and October. Our June issue this year will be an EXTRAñspecial one since it is Issue 100! As always, please send in articles on recent AAT and UKST observations. However, we will also be including material of a more reminiscent nature. We want to hear from past staff members and users of the telescopes. Do you have a favourite story or observation, or can you update us on whatís been happening since you left? Contributions and ideas are needed! Send material to: [email protected] Peter McGauran, the Australian Federal Minister for Deadlines: Science, is shown here with the Echidna test rig during his visit to the AAO on February 15. Science articles: 1 May 2002 Other articles: 15 May 2002
ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 5 FEBRUARY 2002 DISCOVERY OF A SHOCK-EXCITED extending along the minor axis of the galaxy ~ 1.5 kpc
SCIENCE HIGHLIGHTS SCIENCE WIND IN THE STARBURST GALAXY above and below the galactic plane. This structure is NGC 1482 more visible in [NII]l6583 than at Ha. This is particularly apparent in the lower left panel of Figure 1, where we Sylvain Veilleux (Maryland, Cottrell) and present a [NII]/Ha ratio map of this object. The David Rupke (Maryland) [NII]l6583/Ha ratios measured in the disk of the galaxy are typical of photoionization by stars in HII regions, In an effort to better constrain the morphology, but the ratios in the hourglass structure are an order of kinematics, and origin of the warm ionized gas on the magnitude higher. This ratio of a collisionally excited outskirts of galaxies we have obtained deep emissionñ line to a recombination line is fundamentally a measure line images of several nearby starburst and active of the relative importance of heating and ionization. The galaxies using the Taurus Tunable Filter (TTF) on the highñ[NII]/Ha structure is highly suggestive of a shockñ AAT (see Veilleux 2001 for more detail). In the course of excited galactic wind powered by the central starburst. this study, we discovered a remarkable emissionñline structure in the earlyñtype spiral galaxy NGC 1482. This Longñslit spectroscopy obtained on September 18ñ20, galaxy has so far received relatively little attention in 2001 with the DualñBeam Spectrograph (DBS) on the the literature. It is classified as a ìpeculiarî SA0/a in the MSSSO 2.3ñm telescope confirms the existence of a Revised 3rd Catalogue (De Vaucouleurs et al. 1991) large-scale shockñexcited outflow in this galaxy. Figure 2 based primarily on the presence of a dust lane across shows the disturbed kinematics in the extraplanar the disk of the galaxy. Located at a distance of 19.6 material. Line splitting of up to ~250 km s-1 is detected Mpc (Tully 1988), it is a luminous infrared galaxy with along the axis of the hourglass structure out to at least 16 arcsec (1.5 kpc) above and below the galaxy disk. log[LIR/LA] = 10.6 (e.g. Soifer et al. 1989; Sanders, Scoville & Soifer 1991) which is rich in molecular gas Normal galactic rotation dominates the kinematics of and dust (e.g. Sanders et al. 1991; Young et al. 1995; the gas within 5 ñ 6 arcsec (~500 pc) from the disk. Chini et al. 1996) and is undergoing vigorous star Maximum line splitting often coincides with regions of formation (e.g. Moshir et al. 1990; Devereux & Hameed low emissionñline surface brightness. These results can 1997; Thornley et al. 2000). In a recent emissionñline be explained if the extraplanar emissionñline material imaging survey of earlyñtype spirals, Hameed & forms a biconical edgeñbrightened structure which is Devereux (1999) noticed the presence in NGC 1482 of undergoing outward motion away from the central disk. ìfilaments and/or chimneys of ionized gas extending This is the first time to our knowledge that a galactic perpendicular to the disk.î The present study expands wind has been identified using excitation maps before on the results of Hameed & Devereux, using deeper knowledge of the gas kinematics. The traditional method emission-line maps at Ha and [NII]l6583 and of identifying galaxyñscale winds in starburst galaxies complementary long-slit spectra. is to look for the kinematic signature (e.g. line splitting) NGC 1482 was observed on the night of December 16, of the wind along the minor axis of the host galaxy disk. 2000 using the blue TTF in the charge shuffling / This method is timeñconsuming since it requires deep frequency switching mode to maximize sensitivity to spectroscopy of each candidate wind galaxy with faint flux levels. Four sets of observations were obtained spectral resolution better than ~100 km s-1. Line ratio of NGC 1482: two centered on redshifted Ha for a total maps such as the one shown in Figure 1 represent a on-band integration time of 32 minutes, and two centered promising new method to detect galactic winds in on redshifted [NII]l6583 for the same duration. The starburst galaxies. continuum images for each set of observations were The line ratio method of looking for starburstñdriven winds obtained in a straddle mode, where the off-band image only requires taking narrowñband images of candidate is made up of a pair of images that ìstraddleî the on- wind galaxies centered on two key diagnostic emission band image in wavelength; this greatly improves the lines which emphasize the contrast in the excitation accuracy of the continuum removal since it corrects for properties between the wind material and the starñ slopes in the continuum, underlying absorption features, forming disk of the host galaxy. The spatial resolution etc. of these images has to be sufficient to distinguish the Figure 1 (see the back page) shows the distribution of galaxy disk from the wind material. Using Figure 1 as a the Ha and [NII]l6583 emission in NGC 1482. Strong starting point, we find that highñ[NII]/Ha winds in edge- Ha and [NII] emission is detected along the plane of the on starburst galaxies are detectable out to a distance host galaxy (P.A.~100o). In addition, an hourglassñ of ~100 Mpc under 1 arcsec resolution. Imagers shaped structure is seen in both Ha and [NII]l6583, equipped with adaptive optics systems should be able
ANGLO-AUSTRALIAN OBSERVATORY page 6 NEWSLETTER FEBRUARY 2002 to extend the range of these searches by an order of magnitude. The line ratio method could therefore HIGHLIGHTS SCIENCE be used in the future to more efficiently identify distant galaxies hosting powerful starburstñdriven winds.
The authors thank R. B. Tully for first bringing to our attention the peculiar properties of NGC 1482. We also thank J. BlandñHawthorn for help in using the Taurus Tunable Filter and for entertaining discussions. We acknowledge partial support of this research by a Cottrell Scholarship awarded by the Research Corporation, NASA/LTSA grant NAG 56547, and NSF/CAREER grant AST-9874973.
References
Chini, R., Kruegel, E., & Lemke, R. 1996, A&AS, 118, 47 Devereux, N. A., & Hameed, S. 1997, AJ, 113, 559 Hameed, S., & Devereux, N. 1999, AJ, 118, 730 Moshir, M., et al. 1990, The Faint Source Catalog, Version 2.0 Sanders, D. B., Scoville, N. Z., & Soifer, B. T. 1991, ApJ, 370, 158 Soifer, B. T., Boehmer, L., Neugebauer, G., Sanders, D. B. 1989, AJ, 98, 766 Thornley, M. D., Fˆrster Schreiber, N. M., Lutz, D., Genzel, R., Spoon, H. W. W., & Kunze, D. 2000, ApJ, 539, 641 Tully, R. B. 1988, Nearby Galaxies Catalog (Cambridge: Cambridge University Press) Vaucouleurs, G., de Vaucouleurs, A., Corwin, H. G. Jr., Buta, R.J., Paturel, G., & FouquÈ, P. 1991, Third Reference Catalogue of Bright Galaxies (New York, Springer-Verlag) Veilleux, S. 2001, ìWarm Ionized Material on the Outskirts of Nearby Galaxiesî, proceedings of the Figure 2. Skyñsubtracted long-slit spectra obtained parallel to the galactic workshop on ìExtragalactic Gas at disk (P.A. ~ 100o). The spectra displayed on the left are offset by 0" (bottom Low Redshiftî held in Pasadena in panel), 10, 12, and 14" (top panel) northñeast from the major axis of the April 2001, in press (astroñph/ host galaxy. The spectra displayed on the right are offset by the same quantity in the southñwest direction. The presence of line splitting above 0108184) and below the disk confirms the presence of a largeñscale wind in this Young, J. S., et al. 1995, ApJS, 98, 219 galaxy.
ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 7 FEBRUARY 2002 Popular tracers of SFR include: the UV continuum from COSMIC SPECTRA AND STARñ shortñlived OB stars; the radio continuum from SCIENCE HIGHLIGHTS SCIENCE FORMATION HISTORY supernovae; and line emission from nebular regions. Ivan Baldry, Karl Glazebrook (JHU) & the 2dFGRS Team Other than direct tracers of the SFR, cosmic starñ formation history (SFH) can be constrained by using Introduction the fossil record of stars visible today. For example, the SFH of the Local Group (Hopkins et al. 2001) has been The analysis of the comoving starñformation rate density shown to agree approximately with the cosmic SFH as a function of redshift, ìcosmic starñformation historyî, derived from UV measurements. However, the SFH of has been the subject of much recent work. The onset of the Local Group is not representative of the low redshift large redshift surveys out to z~4.5 has allowed the cosmos, being dominated by two galaxies ñ the Milky volumetric emission of luminosity in different bands to Way and M31. The data from the 2dF Galaxy Redshift be traced with redshift. In particular, from these studies Survey (Colless et al. 2001) does provide a highly there is now good evidence for a rise in starñformation representative view of the cosmos. Individual stars or rate by a factor of about eight between z=0 and z=1 star clusters are not resolved but the starñformation (Hogg 2002) and either a z>1 decline or plateau (Madau history of the galaxies can be constrained from the et al. 1996; Pettini et al. 1998). spectra using evolutionary, or population, synthesis.
Most measurements of the starñformation rate (SFR) Cosmic spectra to date have been based on some type of luminosity We present constraints on the history of star formation density which is thought on theoretical and/or empirical based on the ìcosmic spectraî of relatively nearby grounds to trace the SFR. This use of luminosity per galaxies (z<0.3). The concept is to use the average unit volume reflects an attempt to decouple the stellar spectrum of nearby galaxies to constrain the earlier history of the Universe from its dynamical history. history of star formation leading up to that stellar population. The cosmic spectra are constructed in redshift slices by coadding the individual spectra after applying an instrumentñresponse correction, deñ
redshifting and weighting according to the galaxyís bJ luminosity. Examples of cosmic spectra1 are shown in Figure 1. Each represents the spectral emission per unit volume in a z i z + Dz interval down to the limiting magnitude of the survey. Stellar continuum/absorption and nebular line emission are the principal components. Accretionñdisk emission should contribute only a small ìcontaminationî. Cosmic Star-Formation History
In order to constrain the cosmic SFH, we parameterized the variation of SFR with time/redshift (incorporating chemical evolution). Given a grid of starñformation histories, we then computed model spectra using the PEGASE evolutionary synthesis code (Fioc & Roccañ Volmerange 1997) and fitted the model spectra to the measured cosmic spectra.
1 Given a cosmic spectrum at z~0.1, what would be the perceived colour? By integrating standard colour matching Figure 1. Cosmic spectra from various redshift bins functions through the spectrum, we have computed RGB measured using 2dFGRS data. The spectra were values of 0.269, 0.388 and 0.342. Thus, the colour of the normalized and offset by unity from each other. Each is a weighted mean spectrum of over 10 000 galaxies. The universe is quite close to the standard shade ìPale features look remarkably similar other than for an increase Turquoiseî although it is a few per cent greener. See in nebular emission at low redshift due to luminosityñ selection effects. http://www.pha.jhu.edu/~kgb/cosspec.
ANGLO-AUSTRALIAN OBSERVATORY page 8 NEWSLETTER FEBRUARY 2002 We first tested a physical parameterization which allows for a rise and then a fall in the SFR, or a rise in the SFR, to the present day. HIGHLIGHTS SCIENCE Despite significant ageñage and ageñ metallicity degeneracies for the bestñfitting models, all showed a peak in comoving SFR density in the past of at least three times the current rate (using the Salpeter IMFand limiting the universe age to <20 Gyr). Some bestñfitting starñformation scenarios are shown in Figure 2. In order to compare our cosmic SFH constraints with luminosityñdensity methods, we also tested an empirical SFR parameterization. In particular, we made the star-formation rate between z=5 (our choice of formation redshift) and z=1 proportional to (1+z)a and between z=1 and z=0 proportional to (1+z)b, with matched SFR at z=1. There is Figure 2. Example set of star-formation histories that fit the measured a strong upper limit of b < 5 for the best-fitting cosmic spectra using the physical parameterization. The formation redshifts shown are for the standard cosmology with (H , W , W ) models, allowing for a large range of metallicity 0 mo Lo = (70,0.3,0.7). In these scenarios, 75-85% of stars formed prior to 4 with the Salpeter or Kennicutt IMF. The upper Gyr ago. limit on b is lowered if there is any significant quantity of star formation prior to z=1 (e.g., b<4 if a>-3).
We compared our results with two different compilations of cosmic SFH based on the restñ frame UV luminosityñdensity technique and with a compilation of various other techniques. The añb empirical parameterization was fitted to the data sets after converting to the standard cosmology. Various confidenceñlimit contours are shown in Figure 3.
The difference in the confidence limits of the two UV compilations reflects a recent debate over the value of b (1.5 versus 4). However, given the errors used in our analysis (minimum of 15% in each luminosity density), they are in agreement at the 3s level and our results are independently consistent with that intersection (1.8 For more details, see Baldry et al. 2002. References Baldry et al. 2002, ApJ, in press (astroñph/ Figure 3. Comparison between different cosmic SFH studies: confidence limits in a versus b. The solid contours represent the 2s 0110676) and 3s confidence limits from 2dFGRS measured cosmic spectra. Colless et al. 2001, MNRAS, 328, 1039 The dashed contours represent the limits from a compilation of UV Cowie et al. 1999, AJ, 118, 603 luminosityñdensity measurements with a z range of 0.2ñ4.5 (Steidel Fioc & RoccañVolmerange 1997, A&A, 326, 950 et al. 1999). The lower and upper contours are for ìno extinctionî Hogg 2002, in preparation (astro-ph/0105280) and ìextinction correctedî. The dotted contours show the limits from a UV compilation with a z range of 0ñ1.5 (Cowie et al. 1999). The Hopkins et al. 2001, ApJ, 558, L31 diagonally shaded region represents the 1s bootstrap confidence Madau et al. 1996, MNRAS, 283, 1388 on b from a compilation of various techniques including nebular- Pettini et al. 1998, ApJ, 508, 539 line emission, far infrared and radio continuum measurements with Steidel et al. 1999, ApJ, 519, 1 a z range of 0ñ1 (Hogg 2002). ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 9 FEBRUARY 2002 RADIO SOURCES IN THE 6DFGS ñ the main redshift survey. We expect a total of about 20,000 radioñsource identifications over the entire 6dF SCIENCE HIGHLIGHTS SCIENCE TEST DATA survey area. This article describes some results from a Tom Mauch (Sydney) preliminary observation of NVSS radio sources made during 6dF commissioning time. The current generation of radio imaging surveys (SUMSS, Bock et al. 1999 ; NVSS, Condon et al. 1998) The objects observed were 105 stars and galaxies are covering the whole sky to sensitivities of a few mJy selected from SuperCOSMOS in a circle of 6 degrees or less. These surveys probe a population of both AGN diameter centered on a =14h, d =ñ30o, with a radio and star-forming galaxies over a wide range of redshift, identification in NVSS not more than 10 arcsec from however radio data alone are inadequate for classifying the optical position for galaxies and stellar objects. An galaxies and determining redshifts. Using these surveys optimal separation of 10 arcsec was determined by in conjunction with optical spectroscopic surveys is vital MonteñCarlo tests on a large sample of NVSSñ to determine if an individual source is an AGN or SuperCOSMOS matches. On the basis of these tests, starburst and to measure its redshift. the probability of a chance coincidence rose to about 25% for a radio-optical separation of 10 arcsec. Objects The 6dF Galaxy Survey is an excellent tool for studying fainter than bJ=19 were removed from the list. This cutoff the local population of radio sources. In my thesis I was set to be fainter than that of the survey in order to plan to cross-match the SUMSS d <ñ30 and NVSS determine a realistic magnitude limit for our list of spare d >ñ40) surveys with galaxies in the 6dF Galaxy Survey. fibre targets. We expect to detect radio emission from about 16% of galaxies from the main 2MASS selected redshift survey. The data were reduced using the IRAF task DOFIBRES We are also providing a list of targets for spare fibres which combines all of the steps of fibre reduction into which will contain radio identifications with stellar objects one package. The flatfield image was used for the fibre and galaxies in SuperCOSMOS which are not part of throughput correction. The spectra had a typical Figure 1. Some examples of reduced spectra. Residual sky lines at 5500≈ and 6000≈ have been removed manually. ANGLO-AUSTRALIAN OBSERVATORY page 10 NEWSLETTER FEBRUARY 2002 Table 1: Classification of the 105 objects in the sample. calculation of redshift possible even at very low signal- to-noise in the continuum. CEC HIGHLIGHTS SCIENCE Classification N This test data provides a very encouraging result for future work on radio sources in the 6dFGS. Most of the Stars 22 spectra out to bJ=18 had sufficient signal-to-noise to Galaxies 39 determine a redshift, and this has allowed us to push our spare fibre target list out to this magnitude. In my Low S/N or R 42 PhD thesis I expect to use about 10,000 of these identifications to investigate the local space density of QSO 2 radio sources. With such large numbers it should be possible to determine the local radio luminosity function for both starñforming and AGN type galaxies at low radio luminosity. From the list of spare fibre targets I expect resolution of 2.9 ≈pixel-1. Sky subtraction using this to find a population of QSOs and BLñLacs at low radio package needed careful attention but recent results with flux density as well as nearby star forming galaxies not the pipeline reduction package 6dFDR show improved part of the main 2MASS sample. Results from this work sky subtraction. will provide a benchmark for studies of the cosmological Redshifts were measured with the RVSAO package in evolution of these radio sources at higher redshift. IRAF. This package uses both cross correlation with template spectra and line fitting to obtain redshifts. The Acknowledgements template spectra used in the 2dF Galaxy Redshift Survey Iíd like to thank Will Saunders for all his help up at the were used for cross correlation. A correlation coefficient Schmidt Telescope and Matthew Colless for providing of R=5 with 2dF spectra was deemed to be enough to 2dF template spectra. produce a reliable redshift, however all spectra with coefficients between R=4 and R=6 were inspected References visually to verify the results of RVSAO. Figure 1 shows Bock, D.C-J., Large, M.I. & Sadler, E.M. 1999, AJ, 117, 1593 some examples of reduced spectra from this Condon, J.J. et al. 1998, AJ, 115, 1693 observation. www.aao.gov.au/ukst/6df.html Table 1 shows the classifications made from the 105 spectra obtained. Twenty percent of the galaxies were not classifiable due to low signal-to-noise. Most of these were fainter (bJ>18) objects well below the intended magnitude limit of the 6dF survey. No quasar template was used in RVSAO, so the spectra with low R were inspected visually to find QSOs. This search turned up 8 two candidate quasar spectra ñ one at z=0.25 and the other at z=0.8. The higher fraction of stars shown in Table 1 is probably because of the fields proximity to 6 the galactic plane (b=25o). These results indicate that a more conservative cutoff in galactic latitude and Radioñ Optical separation may be necessary for stellar Number 4 identifications. Many of the spectra in the sample had reasonable signal-to-noise but no distinguishing features for cross correlation, most of these were fainter objects 2 (bJ>18). This might be due to bad sky subtraction or could indicate a population of BLñLacs in the sample. 0 Figure 2 is a histogram of the redshifts measured from 0 0.1 0.2 0.3 the galaxy spectra. The mean redshift obtained was Redshift z z=0.05 which is in good agreement with the expected value for the survey. The small number of galaxies with Figure 2. A histogram of the redshifts obtained for the z>0.2 all had very strong emission lines making galaxies in the sample. ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 11 FEBRUARY 2002 DETECTION OF SCIENCE HIGHLIGHTS SCIENCE STELLAR OSCILLATIONS WITH UCLES: THE BIRTH OF ASTEROSEISMOLOGY Tim Bedding, Simon OíToole (Sydney), The observations were made over five nights. The Ivan Baldry, Chris Tinney (AAO), Paul observing method and data processing are the same Butler (Carnegie), Francois Bouchy, as used for the planet search program. Highñprecision Fabien Carrier, Francesco Kienzle velocity measurements were obtained by observing the (Geneva), Hans Kjeldsen (Aarhus) & Geoff Marcy (Berkeley) star through an iodine absorption cell that was mounted directly in the telescope beam, immediately behind the spectrograph entrance slit. The cell is temperatureñ The tremendous success of helioseismology in using stabilized at 55±0.1oC and imprints a rich forest of the 5ñminute solar oscillations to probe the interior of molecular iodine absorption lines from 500nm to 600nm the Sun has prompted many searches for similar directly on the incident starlight. oscillations in other solarñlike stars. The potential rewards are great, since stars like the Sun are expected Exposure times were typically 60s, with a deadñtime of to oscillate in many modes simultaneously, with the 55s between exposures (using ìFASTî readout). The frequencies of the various modes being sensitive to the signalñtoñnoise ratio for most spectra was in the range internal sound speed in subtly different ways. 200 to 400, depending on the seeing and extinction. In total, almost 1200 spectra were collected over 3.5 clear The difficulty in detecting such oscillations lies in their nights. tiny amplitudes. The strongest modes of the solar 5ñ minute oscillations have amplitudes of about 25 cm/s. Detection of oscillations Somewhat higher amplitudes are expected for more evolved stars, and so most attempts have focussed on The power spectrum of the velocity measurements is a handful of bright Fñ and Gñtype subgiants. shown in the Figure. There is a striking excess of power around 1 mHz (periods around 17 minutes) which is the After many years of trying, the necessary Doppler clear signature of solar-like oscillations. The frequency precision is finally achievable, thanks mainly to efforts and amplitude of this excess power are in excellent to detect planets around other stars: the techniques agreement with expectations. Furthermore, observations required to measure the tiny wobbles in stellar spectra from Chile with the CORALIE spectrograph on the 1.2m induced by the gravitational tug of an unseen companion Leonard Euler Swiss telescope show a very similar are virtually identical to those needed to measure power spectrum (Carrier et al., 2001). There is no doubt oscillations. Precision Doppler techniques can now that this excess power is coming from the star. We are produce a precision on short timescales of better than now analysing the combined datasets, which partly 3m/s per measurement. To detect stellar oscillations, overlap, to identify individual oscillation frequencies. all we need are a thousand or so (!) measurements of a single star. Meanwhile, we are also processing UCLES observations of a Cen A, which we also observed simultaneously We have begun a program to measure stellar oscillations with the UVES at the VLT. The combined data set should with UCLES. Our first attempt resulted in the clear confirm the recent detection of oscillations in this star detection of oscillations in the G2 subgiant b Hydri by Bouchy & Carrier (2001) using CORALIE, and the (Bedding et al., 2001). This result was reviewed by combined dataset will allow a detailed asteroseismic Douglas Gough (2001), one of the founders of investigation of the nearest star to our solar system. helioseismology, under the title ìThe Birth of Asteroseismology.î After a long wait, it seems that asteroseismology of solarñlike stars has finally become a reality. Observations of b Hydri References The star b Hydri is the closest G-type subgiant, with luminosity 3.5 L , mass 1.1 M and age about 6.7Gy. Bedding, T.R. et al., 2001, ApJ, 549, L105 A A Bouchy, F. & Carrier, F., 2001, A&A, 374, L5 This star therefore gives us a good idea of what our Sun Carrier, F. et al., 2001, A&A, 378, 142 will look like in a few billion years. Gough, D.O., 2001, Science, 291, 2325 ANGLO-AUSTRALIAN OBSERVATORY page 12 NEWSLETTER FEBRUARY 2002 centre. At radii smaller than 7 arcsec (58 kpc) from the SHOCKS AND ILLUMINATION centre, the ionization structure has a ëbutterflyí shape, CEC HIGHLIGHTS SCIENCE CONES IN THE EXTREME RADIO which is consistent with a bicone of 60o ±10o halfñopening GALAXY 3C 265 angle, predicted by the unified schemes (Barthel 1989). C. SolÛrzano-IÒarrea (IfA Edinburgh / By contrast, moving to larger distances from the nucleus Sheffield), C. N. Tadhunter (Sheffield) (>10 arcsec), the emissionñline gas is more tightly aligned along the radio axis, which is not expected on Powerful radio galaxies possess luminous extended the basis of illumination by the broad radiation cones. emission-line regions (EELR), which can reach large However, the lowñionization state of these extended distances from the nucleus, up to ~150 kpc. The structures, together with their close alignment with the properties of these EELR can provide important clues radio axis, is in agreement with the jetñcloud interaction to understand both the origin of the AGN activity and model. the origin of the extended gas. However, these are still In Figure 1 (right) we show the ratio between the filter unresolved issues because the balance between the onñgoing physical processes (jetñcloud interactions and images of the two split velocity components for the [OIII] emission line. The highñvelocity gas is represented by AGN photoionization) affecting the EELR properties is not fully understood. 3C 265 is a large (78 arcsec; white, and black indicates where the ëunshiftedí (lowñ velocity) component dominates. We find that the 643 kpc) radio source, with a redshift z=0.811. Its extreme emissionñline luminosity ñ an order of high-velocity cloud has a spatial extent of approximately 14 x 18 kpc2, is centred at ~2.5 arcsec SE of the magnitude brighter than other radio galaxies at the same redshift ñ makes 3C 265 unique for the study of continuum centroid of the galaxy, and is located along the radio axis. The large velocity (~1000 kms-1) of the emissionñline properties in distant radio galaxies. fastñmoving cloud with respect to that of the surrounding Deep emissionñline and continuum images of 3C 265 gas cannot be caused by gravity of a single galaxy, but were taken on the night 22/01/98 using the TTF on the is easily explained in terms of acceleration by the WHT. Use of the f/2 camera of TAURUS and the Tek5 interactions with the radio structures. This is strongly CCD detector resulted in a pixel scale of 0.557 supported by the close alignment between the highñ arcsec/pix. The etalon was tuned to the redshifted velocity gas and the radio axis. wavelengths of the luminous [OII]l3727 and [OIII]l5007 emission lines. Further, since a line splitting of ~1000 These results indicate that the ambient gas in 3C 265 is mainly ionized by the central AGN at relatively small kms-1 in [OIII] had been detected in the EELR of 3C 265 distances 7arcsec) from the nucleus. In contrast, at (Tadhunter 1991), the etalon was also tuned to observe O the highñvelocity gas. Details of the TTF observations larger distances from the nucleus, jetñcloud interactions can be found in SolÛrzanoñIÒarrea et al. (2002). may become the dominant mechanism of the lineñ emitting gas in 3C 265. Moreover, the presence of An ionization map of 3C 265 is presented in Figure 1 highñvelocity gas near the nucleus, close to the radio (left). This image is the ratio between the [OII] and [OIII] axis, indicates that even close to the nucleus jetñinduced images. White indicates high ionization, and black shocks have an important kinematic effect. Overall, our indicates low ionization. The ionization state of the results underline the need for a variety of mechanisms nucleus is high, and decreases with radius from the to explain the properties of the EELR in radio galaxies. 10" References Barthel P. D., 1989, ApJ, 336, 606. X X SolÛrzanoñIÒarrea C. et al. 2002, MNRAS, in press. Tadhunter C., 1991, MNRAS, 251, 46P. N N E E 10" Figure 1. Images of 3C 265. Left: [OII]/[OIII] line ratio. White indicates high ionization, and black indicates low ionization. Right: Ratio between the ëunshiftedí and the high-velocity components to the [OIII] emission line. The high-velocity gas is represented by white. Note the close alignment between the high-velocity cloud to the east of the nucleus, and the radio axis. In both images the continuum centroid is indicated by an ëxí, and the solid line represents the radio axis. ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 13 FEBRUARY 2002 NTT AND AAT OBSERVATIONS OF The images in Figure 1 show a distinctive keyñhole SCIENCE HIGHLIGHTS SCIENCE A NEWLYñDISCOVERED PULSAR shape which can be divided into three main regions. WIND NEBULA The most western component or ìheadî, has an almost circular shape, with the pulsar located at the leading Heath Jones (ESO Chile/Universidad de edge. Approximately 20 arcsec from the front of the Chile), Ben Stappers (Dwingeloo/Instituut shock, the nebula broadens into a fanñshaped tail and ëAnton Pannekoekí) & Bryan Gaensler a bridge of emission connecting the two sides of the (CfA) nebula can be seen. At the eastern limit of emission, some 45 arcsec from the apex, there is evidence in During the first two nights of a campaign to find optical both the NTT and AAT images that the tail section pulsar wind nebulae, we discovered a nebula associated closes. Figure 1b shows no emission within the interior with the radio pulsar B0740ñ28. This is the fourth such of the head although deeper images are needed to nebula found to be associated with a radio pulsar. It confirm this. Variations in the surface brightness across was discovered in narrowband Ha frames taken through the extent of the nebula are seen in Figure 1a and Figure the 656/7 nm filter of SUSIñ2 at the 3.5m New 1b. The brightest regions have surface brightness a factor Technology Telescope (NTT) at La Silla (Figure1a). A of two brighter than the mean over the entire nebula. separate continuumñsubtracted narrowband frame was derived from followñup imaging with the Taurus Tunable More details on PSR B0740ñ28 can be found in Jones, Filter (TTF) at the AngloñAustralian Telescope (AAT). Stappers & Gaensler (ESO Messenger 103, 27) and a Figures 1b and c show the continuumñsubtracted image forthcoming paper. Work is continuing with our survey and its isophotal map. to locate other pulsars with optical nebulae. (a) 20" (c) A B 40" C −28 o 23’ 00" 52s 07h 42m 50s 48s (b) (d) Figure 1. a) SUSIñ2/NTT discovery image (Ha + continuum) of the nebula associated with PSR B0740ñ28. The arrow indicates the direction and distance travelled by the pulsar over 500 yrs. Coordinate epoch is J2000. b) Continuumñ subtracted Ha image of the same field taken with TTF on the AAT. Some residuals due to saturated stars and in-focus ghost images are present. c) Isophotal map of the image in b). Panels a) ó c) are 1 arcmin on a side with north up, east left, with the pulsar position indicated by a circle. d) Simple surface of rotation based on analytic models of the bowñshock shape, assuming the pulsar is moving 20o to the plane of the sky. ANGLO-AUSTRALIAN OBSERVATORY page 14 NEWSLETTER FEBRUARY 2002 FIRST RESULTS WITH IRIS2 compiled by Stuart Ryder IRIS2 Catches the Afterglow of a Gamma Ray Burst! IRIS2SPECIAL On only its second outing, IRIS2 helped play a major role in monitoring the fading afterglow of a Gamma Ray Burst (GRB). The night of Thursday November 22, 2001 Figure 1a (left) is from a Ks band image of GRB011121 was scheduled to be a Directorís night for ìreñ taken on 22/11/01 with IRIS2 made up of 27 dithered images, each an average of 12 x 5 sec. exposures. Figure commissioningî of IRIS2 with its new reñaligned optics 1b (right) shows the same field 5 nights later, by which (see accompanying article by Chris Tinney). As time the faint underlying host galaxy has become apparent. preparations were made that afternoon for a night of observing photometric standard stars and numerous Cluster Surveys other tests, we were contacted by Paul Price, on behalf The inaugural observers with IRIS2 following the first of the CaltechñNRAOñCARA collaboration: was IRIS2 commissioning run in Oct/Nov 2001 were Warrick Couch ready to allow them to trigger their preñapproved ATAC (UNSW) and Mark Sullivan (Durham). They were out to override program to obtain Ks imaging of a newlyñ acquire deep Ks imaging of two rich clusters at z~0.3, discovered Gamma Ray Burst source, GRB011121? AC114 and AC118, to complement their HST optical Cold tests in the lab had already demonstrated that images, ATCA 1.4 GHz maps, and LDSS++ Ha surveys. IRIS2 could now deliver highñquality images over the The aim is to test recent claims for a (previously full 7.7' x 7.7' field, so we felt pretty confident it could do unidentified) obscuredñstarburst population of galaxies the job. within these clusters. These galaxies show strong radio Undeterred by the fact that the source coordinates of synchrotron emission ñ indicative of ongoing star- (11h 34m, -76deg) meant that it would have to be formation ñ but also possess ìclassicalî post-starburst observed below the pole for most of the night, we swung optical spectra, with no Ha or [OII] emission, both into action, and within a short period of time, had traditional tracers of starñformation activity. The obtained confirmation images of the afterglow (Fig 1a). suggestion is that much of the starñformation could be This was further complicated by the source starting off obscured in these objects. By adding in these new IRIS2 at a zenith distance of 74 degrees, so that it was partly data, they will be able to trace the photometric signatures vignetted by the windscreen! In addition, we discovered of dust in the radioñselected samples. Ultimately, this a bug in the new telescope control system when working will allow them to construct the evolutionary starburst below the pole, but sterling efforts by Jeremy Bailey cycle for these galaxies within representative regions of soon had this rectified. In the meantime, the night distant clusters, as well as searching for radial variations assistant Frank Freeman patiently fed in offsets for characteristic of a triggering mechanism. Despite the dithering by hand. restricted useable imaging field after the first commissioning, the coverage of AC114 is around 15 Recognising the rare opportunity presented here to times larger than their existing dataset from the original follow the rapid decay of an afterglow in the nearñinfrared, IRIS instrument, and reaches a comparable depth. the Director waived the usual override restrictions and recommended that the entire night be dedicated to GRB011121. Over the next 7 hours, the source got higher in the sky and the seeing improved, but GRB011121 continued to fade. Unfortunately, the GRB was already too faint in the J and H bands to permit an Italian collaboration to trigger their own IRIS2 spectroscopic override. Inclement weather precluded any followñup observations on subsequent nights, and it was not until November 27 that we were able to secure calibration images, by which time only the host galaxy of the GRB was visible (Fig 1b). Nevertheless, the opportunity to provide extended photometric coverage of a still fairly unique class of object as part of a worldwide coordinated Figure 2: Part of the AC114 cluster of galaxies at a redshift network of optical, infrared, radio and satelliteñbased z=0.31 seen in the Ks band with IRIS2. Each section of the image is the coñaddition of as many as 900 x 8 second observatories shows that IRIS2 is well on the way to exposures, and reaches a limiting magnitude Ks~20. Note making a significant international scientific impact. the pair of interacting galaxies to the right. ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 15 FEBRUARY 2002 IRIS2 HITS THE SKIES Chris Tinney for the IRIS2 SPECIAL IRIS2 IRIS2 Team In 1995, ACIAAT1 recommended to the AAT Board, that the development of a new nearñ infrared wideñfield imager and spectrograph be made the Observatoryís highest priority project, following the completion of 2dF. IRIS2 was clearly going to represent a major undertaking, and offered several major challenges in areas outside the Observatoryís traditional strengths. As the years rolled by and 2dF never quite got completed, some of us at the Whatís in the Box? AAO (and Iím sure some of you, gentle readers) must But before the telling the story of IRIS2ís first have wondered if IRIS2 would ever really get delivered. commissioning runs, what is IRIS2? What makes it Now, fast forward to a wet and foggy afternoon of special? Wednesday, 23 October 2001. If youíd been on the IRIS2 is intended to fill a role as the AAOís ìworkhorseî ground floor of the AAT, youíd have found yourself infrared imager and spectrograph, based around a surrounded by a crowd of Epping and Coonabarabran 1024x1024 pixel HgCdTe HAWAII detector. In imaging staff, all busily running around a small truck like mode, it has a 7.7íx7.7í fieldñofñview at the f/8 headless chickens. But all with slightly silly grins on Cassegrain focus of the AAT, at an image scale of 0.449î/ their faces. IRIS2 had finally arrived. It was really going pixel. Finer pixel scales can be obtained with the f/15 to go on the telescope! and f/36 topñends. However it is expected that IRIS2 All of which makes it sound like nothing happened in will do most of its observing at f/8, enabling wideñfield, the intervening period, but a bit of stuffing about and rather than high resolution, science. magic wand waving. Which is, of course, not true at all. In spectroscopic mode, IRIS2 will enable longñslit The intervening 5 years was filled with hard work: from spectroscopy at spectral resolutions of l§Dl=1200ñ2400 the astronomers who tried to define the scientific priorities in wavelength ranges corresponding roughly to the J, H, for IRIS2 and then turn these into a functional and K passbands. specification for the instrument; from the engineers who tried to turn those specifications into something useful, IRIS2ís design (above) is conceptually simple. A and then into designs for new hardware and software; collimatorñcamera combination reproduces (and from the mechanical, electronics and software teams compresses) the telescope focal plane onto the HAWAII who had to build and integrate all these components; detector. Between the collimator and camera is a and from the project managers who had to keep all the collimated space in which lie three wheels. One wheel above happening at the right times and in the right places. is reserved for filters, which select the passbands for I canít speak highly enough of the professionalism with imaging and spectroscopic observation. A second wheel which everyone in the IRIS2 Team approached their is primarily for holding coldñstops (which mask out the tasks ñ and that hard work has shown in the quality of thermally ìhotî parts of the telescope, which would the instrument we saw delivered that day, and which we otherwise add an unacceptable background for were going to find out about over the coming weeks. observations at long wavelengths). And a third wheel holds grisms ñ optical components which (when inserted in the beam) disperse light to produce a spectrum. 1Advisory Committee for Instrumentation on the AAT. ACIAAT was the precursor committee to the current AAO User All these components sit inside the ìmain dewarî ñ an Committee (AAOUC) which advises the AAT Board on evacuated stainless steel cylinder, which looks pretty (among other things) instrumentation priorities. much like 2/3rds of a 44ñgallon drum. The internal ANGLO-AUSTRALIAN OBSERVATORY page 16 NEWSLETTER FEBRUARY 2002 components of the main dewar are maintained at near liquid nitrogen temperature (~77K) by a twoñstage cryoñcooler. A separate dewar which sits directly in front of the main dewar (the ìfore dewarî) contains a fourth wheel, which is used to insert a variety of aperture masks at the focal plane before light enters IRIS2. These allow us to either permit the entire field to pass into SPECIAL IRIS2 the instrument for imaging, or to permit just a slit of light (either 1" wide or 5" wide) to enter for spectroscopy. Eventually we hope to be able to mount specially milled masks in this wheel which will permit multi-object spectroscopy ñ allowing us to obtain hundreds of infrared spectra simultaneously, just as we currently do with LDSS or Taurus. The ësplit dewarí design of IRIS2 was chosen so that once the instrument is fully commissioned, we can thermally cycle the fore dewar on short timescales, without risking damage to the sensitive (and very expensive) components mounted in the main dewar. IRIS2ís development required us to surmount a number of technical challenges. In a separate article in this issue, John Dawson describes the difficulties imposed by the need to have small stepper motors mounted inside the dewars, and turning wheels in a cryogenic environment. Another article by Jeremy Bailey describes the significant software issues IRIS2 has had to face. A further challenge was designing grisms for IRIS2; the science drivers indicated a need for resolutions of at least l§Dl=2500, since resolutions this high are needed to resolve and properly subtract the bright OH lines which dominate the sky background in the infrared. Unfortunately, obtaining these resolutions is challenging. A grism consists of essentially two components ñ a prism which deviates the beam of light inside the spectrograph, and feeds this deviated beam into a grating, which disperses the light. Grisms are designed so that the beam deviation produced by the prism, counteracts the beam deviation produced by the gratingís dispersion, so light leaves the grism onñ axis, but dispersed. The resolution of the grism in IRIS2 is then limited by the amount of beam deviation one can obtain from the prism component, which is itself limited by the refractive index of the prism material. Unfortunately, very few materials can be used which have high throughput in the infrared and high refractive index. And of those, even fewer can have gratings either bonded on to them, or ruled on them. The most commonly used material in the past has been fused silica. Unfortunately, fused silica has a very low refractive index, and can only deliver resolutions of l§Dlª1200. An alternative turns out to be sapphire ñ not the little blue stones used in rings, but commercially manufactured optical grade Al2O3. Though large prisms of this material are expensive, they turn out to not be as expensive as you might think. So the IRIS2 Team decided to pursue two avenues ñ traditional fused silica grisms, and an experiment with sapphire grisms. The latter have actually turned out to deliver very nice performance! IRIS2 Hits the Deck An unkind person might remark that IRIS2 arrived at the telescope in typical Siding Spring observing conditions ñ fog and rain. Fortunately, fog and rain are ideal commissioning conditions ñ who wants to be forced to look at sky when there are exciting flat fields, arcs and flexure tests to do! So, undaunted, the IRIS2 team set to unpack the instrument from its airñride truck and move it to the Coude West preparation area on the fourth floor of the AAT (passing, on the way, the venerable RGO spectrograph). ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 17 FEBRUARY 2002 The very first test tried with IRIS2 was to offer it up to the telescope to see if it would fit. The mounting points for Cassegrain instruments on the AAT are four plates, each with a central ënotchí for locating the instrument when it is IRIS2 SPECIAL IRIS2 attached. Matching these notches are ëpinsí on the mounting points of each instrument. IRIS2 turns out to be the first instrument ever delivered to the AAT with all four pins in place ñ no-one else has ever trusted their dead reckoning quite as much. Iíve never seen a man smile as broadly as John Dawson did when every one of those pins slid smoothly into its notch ñ it fitted perfectly. So now it was time to take it off, and see if we could get it pumped and cooled. By the following day (Thursday) both the fore- and main- dewars of IRIS2 were being pumped down, and by Thursday afternoon cryo-cooling of the main-dewar was begun. Cryoñ coolers for instruments have been used at the AAT for many years now, in both IRIS1 and 2dF. However, the need for two cryoñcoolers to operate IRIS2 almost permanently, alongside 2dF, made it clear that we needed to upgrade our infrastructure somewhat to support these refrigerators. So an entirely new He plumbing system has been designed and installed at the AAT by Site staff and staff from the Australia Telescope National Facility. These include new He compressors on the third floor, new cable wraps to plumb the He reticulation system into the telescope itself, and He access points in the Coude W room on the fourth floor for instrument preparation2 . By Thursday night, both dewars were being cryoñcooled, and the detector controller had been powered up, and was reading images. On Thursday the plan had been to mount IRIS2 on the telescope on Friday, and to continue cooling and testing there, while observing was carried out at Coude with UCLES. And so on Friday, the day crew began work again. IRIS2 was mounted at Cassegrain again, final insertion of the mounting structure pins was completed, He lines were reconnected so it could continue cooling, and testing began of the operation of the instrument itself in situ. The amazing thing about this process was that everything went smoothly. By 7pm that evening the commissioning team was standing around looking at each other, when we realised we werenít ready to observe tomorrow. We were ready to observe now! And so while the Epping mechanical crew headed off to the Imperial for a well earned rest, Roger Haynes, Stuart Ryder and myself took the first image every new instrument has to have ñ ìfirst fuzzî ñ an outñofñfocus image of a bright star. We then started the process of aligning the telescope and instrument pupils (i.e. aligning the instrument with the telescope, so that the coldñstop masks actually block out the hot bits of the telescope. Thanks to a very clever pupil imager designed by Peter Gillingham, this turned out to be trivial. We all just kept looking at each other, thinking that life wasnít meant to be this easy. But it was. So by 11:42pm we were ready to slew to our first imaging target for first light on 30 Doradus in the LMC (featured on the cover). At which appropriate moment the team returned from the Imperial, so it was time to toast IRIS2ís future, with a certain Scottish fluid. ANGLO-AUSTRALIAN OBSERVATORY page 18 NEWSLETTER FEBRUARY 2002 How it works could not get a good focus all over the detector. In fact, we could only get good images on the lower half of the When heading up to the telescope in October before a detector. Unfortunately, this is the half of the detector 2ñweek commissioning and observing run with IRIS2, I where the duff quadrant is, so useful images could only frankly expected it to be two weeks of hell. A new be formed over essentially a 512x512 array, for imaging. SPECIAL IRIS2 instrument ñ new instrument ñ is expected to be any For spectroscopy, we could get good images over a packed full of bugs. Especially in a fully cryogenic strip just above the middle of the detector, and so in instrument, there are so many things to go wrong. this first commissioning and Shared Risks Service Vacuum leaks, thermal problems, optical misalignment, Observing run we concentrated on spectroscopy, and an entirely new detector and detector controller system imaging for single targets which could be done well in a running for the first time not on a VAX, an entirely new 3.8'x3.8' field of view. instrument control system, an entirely new telescope interface Ö I expected everything to go wrong. Even before the end of this first run, however, Head of Instrumentation Roger Haynes had ray traced the matrix But it didnít. Over the following two weeks we found that mask images, and diagnosed the problem ñ a in spite of being brand new in so many ways, the IRIS2 misalignment of the field flattener. This is the last element system was remarkably robust. In fact, I think that during of the optical train and sits just 5mm in front of the the IRIS2 run, the AAT afternoon shift technician had detector. The solution to this was a fairly significant reñ the cushiest ride I can remember in years. Very little design of the mount of this element, but undaunted actually went wrong! Roger headed back to Epping to get the IRIS2 team on Which is not to say that nothing went wrong, of course. the job. Between the end of the October run and the We suffered from two major problems. One was a start of the next run, 16 days were available. Given IRIS2 problem we were aware would probably affect us before takes days to warmñup and coolñdown, this left the the run started ñ the engineering grade detector. In the team with a tiny 1 week window to install a new field weeks prior to delivering IRIS2 to the telescope, numerous tests had been carried out on our science grade device in Epping. The performance of this device was excellent, except for an alarming number of hot Table 1: Filters for IRIS2 pixels. On separate thermal cycles of the detector we Filter l cut-on l cut-off Acquired? were seeing between 4% and 20% of the total pixels (mm) (mm) appearing to be ëhotí (i.e. outside a ±5s cut about the bias level). With no obvious cause for this in sight, we Broadband† decided to proceed with our first observing run using J† 1.164 1.325 yes the engineering device. This device has only three working H† 1.485 1.781 yes quadrants, and a very bright ëhot spotí which introduces K 1.982 2.306 yes some 1/f noise into the rest of the detector. So having a s slightly compromised detector was something we K† 2.028 2.364 yes expected. Z† 0.996 1.069 mid-2002† During the run, we received advice from Peter Moore at Narrowband the ING that they have been having the same problem He I† 1.075 1.091 mid-2002† with their HAWAII device Ö the suggestion was made J continuum† 1.198 1.216 yes that the problem could be made to go away by not Pa Beta† 1.272 1.292 yes powering up the detector as it is cooled. Tests in Epping with our science grade device show this indeed works ñ H cont. 1.558 1.582 yes instead of 20% of pixels outside more than 5s from the [Fe II]† 1.632 1.656 yes mean we now get 0.08%. Indeed, doing the same thing Methane Offband† 1.53 1.63 yes with the engineering device on IRIS2ís next run in Methane Onband† 1.64 1.74 soon† December, completely eliminated not only many faint H nu=1ñ0 S(1)† 2.106 2.138 yes hot spots, but also its very bright hot spot. So we are 2 now confident we can install the science grade device Br Gamma† 2.150 2.182 yes for IRIS2ís next run in March 2002. H2 nu=2ñ1 S(1)† 2.231 2.265 yes What we hadnít expected was that we would have K continuum† 2.253 2.287 yes problems in commissioning with the optics. Tests with CO(2ñ0) band 2.278 2.312 yes a matrix mask in the slit wheel quickly revealed that we head† ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 19 FEBRUARY 2002 Table 2: Broadband Imaging Figures of Merit. Filter JHK Zeropoint Flux for JHK=0 Sky Brightness 5s mag IRIS2 SPECIAL IRIS2 †1 ADU/s at 1 (ADU/s) (mag/arcsec2) in 1 hr (on sky) airmass 1.4" seeing, 3" aperture. J22.489.8e815.921.2 H22.541.0e914.120.4 Ks * (22.13) (7.1e8) 13.1 19.8 K22.137.1e813.419.6 † flattener mount and test it before starting the next IRIS2 Scales : The spatial scale of IRIS2 is 0.449±0.002"/ run. A window I am proud to say the team met. Moreover pixel, giving a field of view of 7.68' x 7.68'. The 1" slit for the new mount works brilliantly delivering 1ñpixel images spectroscopy corresponds to limiting spectroscopic (or better) over the full field of view. resolution of 2.2 pixels on the detector. At this resolution the dispersion of the silica grisms provides R~1400 and A slew of minor software glitches also showed up, but of the sapphire grisms R~2400. the vigilant efforts of Jeremy Bailey meant that most of these got cured as they were found in operation. We Filters : Given in Table 1. were quickly taking preñprogrammed dither pattern Broadband imaging performance : See Table 2. observations, controlling instrument configurations, and (by the second night) implementing random offset Grisms & Spectroscopy dithering from command sequences. In general, the IRIS2 software system was not only quickly found to be There are four grisms in IRIS2, though a third (J-band) reliable, but also quite efficient. A remarkable sapphire grism has been delivered and will be installed achievement for a brand new instrumentís first run at for March 2002. At present all spectroscopy must be the telescope. done using one of the broadband filters for order selection. Order sorting filters which better match the Performance grisms (especially in J) are being procured. The blaze functions for the silica grisms are clearly subñoptimal The most up-to-date information available for IRIS2 can and will be adjusted by tilting the grisms before March be found on the IRIS2 WWW Pages at http:// 2002. At present the sapphire grisms have the best www.aao.gov.au/iris2. However, hereís a brief summary throughput of all the grisms on hand. of what weíve found so far. IRIS2 Grisms are listed in Table 3. Detector : The credits Gain : 5.2e/adu. The astute reader will have noticed a certain recurring Readnoise: 14ñ15e from a doubleñcorrelated sampling theme in all the above. IRIS2 has been delivered as a (DCS) read (i.e. differencing pair of array reads). Multiple more robust, functional and exciting instrument than I Read Mode lowers this to ~8e, though these numbers dreamed was possible. The new AAO2 detector will have to be reñdetermined with the science grade controllers performed flawlessly, as did the IRIS2 device in March 2002. instrument control electronics. The dewar gave us no Linearity: better than 1% below 28000adu. problems, holding its vacuum and temperature for 2 Read time is currently 1.5s for all 1024x1024 pixels weeks on the telescope, on two runs. The new helium (through 4 readout amplifiers). This should be lowered reticulation system is a joy to behold. Right from the in coming runs. word go, the software needed to be reñstarted way less often than I expected. Significant problems have been Residual images: remain following bright exposures identified and rectified in an amazingly rapid fashion. at ~ 0.01% level For a brand new instrument, I think itís fair to say that IRIS2ís performance has exceeded all our expectations. ANGLO-AUSTRALIAN OBSERVATORY page 20 NEWSLETTER FEBRUARY 2002 Table 3: IRIS2 Grisms Grism SñK SñK SñH KH KH HJ HJ Filter (order) K (m=1) J (m=2) H (m=1) K (m=2) H (m=3) H (m=2) J (m=3) IRIS2 SPECIAL IRIS2 Resolution (R) 2450 2400 2500 ~1600 1600 1500 1500 Start l (mm) 2.04 1.155 1.575 ? 1.54 1.60 1.17 End l (mm) 2.38 1.278 1.80 2.37 1.80 1.80 1.32 Disp. (nm/pix) 0.45 0.25 0.33 ~0.74 0.51 0.57 0.39 Pk Rate (e-/pix/s) 158 340 152 133 212 76 140 for JHK=10 star Note: Details as at Feb 2002. Please check the IRIS2 web page for upñtoñdate values. This is directly a result of the skills and efforts of everyone has suffered through every one of the 27 nights of involved at the AAO. Without Roger Haynesí tireless commissioning and observing that IRIS2 has been on efforts I think weíd still be months away from the the telescope so far, and worked out how to do quick telescope. The members of the mechanical section (Vlad look reduction, and wrote half the documentation. Churilov, John Dawson, Denis Whittard, Greg Smith, Chris Evans and Gabriella Frost have been tireless in Neal Schirmer, Dwight Horiuchi, Jurek Brzeski, Mark shepherding IRIS2 towards the telescope. Chris has Hilliard and Shane Tan) have been working 6ñ7 days a been overseeing the project for a long time, but had to week for months to get us here. Electronics (Lew Waller, concentrate on OzPoz in the period just before John Barton, Don Mayfield, Rolf Muller, Brian Hingley, commissioning, when Gabriella took over the reins for a Ed Penny) have delivered components which have while. Throughout that period Gabriella was laid low with performed flawlessly ñ so flawlessly I think thereís sometimes a danger some of us may take their work morning sickness, that lasted morning AND afternoon. Despite this she never stopped chasing down details, for granted! The software group (Jeremy Bailey, Keith Shortridge, Tony Farrell, John Straede) have delivered a focussing our attention on the critical items and getting us to make the tough decisions required to keep IRIS2 system which is much more developed than weíd been told to expect, and which will clearly evolve into a very on track and on time. I am in awe of their abilities. flexible and convenient user interface. Our optical And finally designers (Peter Gillingham and Damien Jones) have delivered another excellent optical system. In October 2001 IRIS2 was on the telescope for four nights of Directorís time and Shared risks service The staff at Site have been busy preparing for IRIS2 for observing, followed by 6 nights of scheduled observing. months ñ the new helium reticulation system has been Almost all of this time was clear with good seeing. The developed almost completely at Coona (Allan following run was less lucky, with less than half of the 2 Lankshear, Brendan Jones, Neville Hopkins, Wayne nights of Directorís and SRSO time, and 12 nights of Clarke, Jonathon Pogson and Chris McCowage) with scheduled time, clear. IRIS2 will next see action in March input and work done by ATNF (Pat Sykes and Mark 2002, when we hope to have the full complement of Wright). And the Site staff really pulled out all the stops grisms and most of the filters installed, the science grade to get us from truck to sky in just 54 hours! I think device operating, the lowñres grism performance pretty much everyone on staff must have turned out to improved, and most of the remaining software glitches see IRIS2 arrive, and then ended up contributing resolved. something as we worked towards getting it cold, and bolted in place. In short, IRIS2 is open for business. The night assistants (Steve Lee, Jonathon Pogson and Frank Freeman) have put up with the new IRIS2 control system trying (and failing) to make them redundant. Stuart Ryder has performed simply inhuman feats ñ he ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 21 FEBRUARY 2002 limit for the Cassegrain mount at the AAT. Unfortunately, THE CHALLENGES ENCOUNTERED aluminium has a very large coefficient of thermal DURING THE DEVELOPMENT OF expansion. This means that the outside diameter of each THE IRIS2 WHEEL DRIVES wheel shrinks by 1.5 mm when cooled. The first IRIS2 SPECIAL IRIS2 challenge, then, was to ensure that a cooled wheel was John Dawson still free to rotate on its support bearing and that its position was still accurately determined (i.e. with no This article describes the development process for the four wheels used in the IRIS2 instrument. For further information free play). Free play at the support axle and hubs was on all aspects of IRIS2 you should visit the website at http:/ eliminated by manufacturing them from 400 series /www.aao.gov.au/AAO/iris2. stainless steel, which has a very similar coefficient of This article has a mechanical engineering flavour but it thermal expansion to the bearing races chosen (axle should be noted that IRIS2 was a team effort from the and hub are clearly visible in the centre of the wheel in Mechanical, Electronic, and Software groups, Fig.1). So the axle, ball bearings and hubs all shrink at coordinated by the Project Management team. I would the same rate, allowing the bearings to be unstressed like to take this opportunity to specifically thank those and free to rotate when cold. The aluminium wheel was members of the Mechanical Engineering Group that designed with significant clearance on the hub so that made IRIS2 a reality. They are Greg Smith and Vlad it shrinks onto the hub when cold, giving a slight Churilov who were the two principal designers, together interference, which ensures there is no freeñplay in this with Denis Whittard, Jurek Brzeski, Mark Hilliard, Dwight part of the system. Horiuchi, Neal Schirmer and vacation scholar Marcus In a similar manner the worm gear was designed and Billerwell. Lastly I would like to add the name of Roger manufactured so that no free play existed at cryogenic Haynes. Although not part of the Mechanical design operating temperatures. The worm was manufactured team he certainly made significant contributions. from aluminium and mounted on deep groove ball The function of the wheels is to position the various bearings. To stop the worm gear shrinking away from optical components (filters, grisms and apertures) along the bearing, when cold, a titanium insert was put inside the optical axis of the instrument. The most demanding the shaft. This ensures that the bearing and shaft positional requirement set was for the slit wheel, which maintain their fit when cold. Initially the worm drives had to have slits positioned to within 30 mm. A worm were designed and manufactured using a circlip to hold and wheel arrangement was chosen to drive the wheels the bearings in position. But early testing proved that as its large reduction ratio allows high accuracy small differences in the circlip groove dimension and positioning via a stepper motor drive and encoding variations in the clips themselves meant that system. It has the additional benefit of being compact, unacceptable free play occurred on the worm assembly. allowing three wheels to be placed in very close To overcome this problem the circlip was replaced by proximity. Fig. 1 shows how the three worm drives and madeñtoñmeasure spacers in each worm assembly to one of the filter wheels (with no optical components fitted) ensure zero free play. These spacers are just behind can be closely packed. A helical gear form was chosen the end cap that can be seen at the leftñhand end of for the worm wheel as this made the positioning Fig. 2. of the worm relative to the wheel less critical. The wormñwheel pair also has the added advantage of locking the wheel in position when not being driven by the stepper motor. The required accuracy for the positioning of the wheels was stringent enough for an instrument operating in normal ambient conditions. However, IRIS2 needs its internal components to be cooled to near liquid nitrogen temperatures. This increased the complexity and challenges to be faced during development several fold. Much of IRIS2 is manufactured from aluminium to ensure that the instrument is under the mass ANGLO-AUSTRALIAN OBSERVATORY page 22 NEWSLETTER FEBRUARY 2002 Initially it was thought that this would be sufficient to allow the gear pair to operate once cooled. The hinge was designed and set to apply a slight force to the worm to keep it in contact with the wheel at all times, and to SPECIAL IRIS2 ensure that once it had stopped at a specified position, it stayed there (i.e. no freeñplay). But no matter how small the applied force, the mechanism would not run smoothly, which meant that the stepper motor kept losing steps, and so couldnít maintain positional accuracy. The next step in the development process was to add a magnetic solenoid that pulled the worm gear slightly away from the wheel prior to driving the worm. This slight movement allowed a small clearance at the gear contact point and allowed the gear The bearings supporting the wheels and worm drives to drive on one active face and proved extremely also provided their share of challenges. The lubricants successful. Once the stepper motor is driven to the typically employed in offñtheñshelf ball bearings freeze required position the solenoid is deactivated and the at cryogenic temperatures, making them unsuitable. worm moves in to detent the wheel at that position. The Initially bearings with balls manufactured from ceramic solenoid is shown in Fig. 2 at the lower left of the figure. were tried. Unfortunately the bearings did not have As the movement required from the solenoid mechanism sufficient clearance and seized at low temperatures. was very small the clearance between the solenoid and However, standard bearings lubricated with molybdenum the end of the worm cage assembly had to be very small. disulfide proved to work effectively ñ though finding During the cooling process this small gap also closes somebody who could offer molybdenum disulphide up pushing the worm assembly into the solenoid and coatings at a reasonable price was not easy. consequently also jamming the worm gear into the Controlling the contact point between the worm and wheel. To prevent damage to the worm and gear teeth wheel proved to be very interesting, as the whole wheel the solenoid was also mounted on a spring mechanism system is housed in an aluminium case that supports to give it compliance during cooling. Once the whole the worm drive assembly. On first inspection one might mechanism is cooled and at equilibrium the small gap think that as the case and wheels are both manufactured required is restored. of aluminium that they would end up being the same Finally, we faced the challenge that the stepper motors size after cooling. Which is true. However, because the themselves did not operate at the desired temperatures outer casing is directly cooled down by contact with without significant modification. This was extremely liquid nitrogen (the cooling coils can be seen on the frustrating, given that they were purchased as cryogenic outer case in Fig. 1), as well as being kept at temperature and vacuum rated! Unfortunately, prolonged discussions by a helium refrigeration system; and the wheels are with the manufacturer proved to be of little help. They cooled predominantly by radiation, the outer case cools kept the design and assembly of their motors shrouded much more rapidly than the wheels and thus contracts, in mystery, and were not prepared to provide us with pushing the worm gear into the wheel. To counter this much assistance (5 motors in a market of millions?). effect the worm drive had to have some form of compliance. This compliance was achieved by mounting As can be seen from Fig.1 the motors are packed quite the worm drive on a spring/hinge mechanism. The spring/ closely together. So unless we wanted to completely hinge is made up of two thin plates that support the reñdesign the whole wheel assembly, we needed motors worm drive, shown to the lower right in Fig. 2. These of exactly the same size. Unfortunately, we could not two plates are stiff where required but allow the worm find alternative suppliers for motors of the same size. assembly to rotate about a point at their theoretical So we decided to go it alone, and try to make the motors intersection. The theoretical intersection was designed we purchased work. Many trials and tribulations to coincide with the Centre of Gravity of the worm (involving long days and sleepless nights) resulted in assembly. Thus regardless of the angle at which the us redesigning the end caps in a different material and instrument is employed, the weight of the worm replacing the bearings. The new design has already assembly does not affect the contact forces at the gear completed two long runs at the AAT and we are confident face. that it should do so for a significant time to come. ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 23 FEBRUARY 2002 Data Acquisition Computer THE NEW DATA ACQUISITION SYSTEM FOR IRIS2 The data acquisition computer is a dual processor VME bus system containing a Sparc processor running the Jeremy Bailey Solaris operating system and a 68040 processor running IRIS2 SPECIAL IRIS2 the VxWorks real time operating system. The 68040 The IRIS2 project has given us more than just a new handles the direct communication with the controller, instrument. It also uses a complete new system for while most of the data acquisition software system runs data acquisition which will, over the next year or two, on the Sparc. The software runs under the AAOís Drama replace the old VAX ìObserverî system for all software environment. IRIS2 is the first AAO instrument instruments. The new system replaces the old CCD to run fully under Drama (2dF also makes extensive controllers, the XMEM, the VAX ADAM software and use of Drama, but uses the old Observer system for the old command line ìObserverî interface. The main data acquisition). components of the system are as follows. User Interface New Detector Controller IRIS2 is controlled from a graphical user interface written The new controller (AAO2 controller) has been developed in Java (see figure). This incorporates a mimic display over the last few years by Lew Waller, John Barton and of the IRIS2 optics as well as sections for controlling Don Mayfield. The controllerís analogue and digital the IRIS2 mechanisms (filter, grism, coldstop and electronics are entirely contained in a small box which aperture wheels) and for taking data. mounts on the side of the dewar. It communicates with the data acquisition computer (in the control room) via The Sequencer a fibre optic link. The new controllers can support A useful feature of the new system is the sequencer detectors with multiple readout amplifiers (4 for IRIS2) which is used to run scripts that control sequences of and much higher data rates than have been possible in operations and can involve taking data, moving the the past. We can read out the 1024 by 1024 IRIS2 array telescope, changing filters etc. Scripts are written in continuously at under 0.6 seconds per frame. The new the Tcl language which gives considerable power and controller has proved to be highly reliable, performing flexibility. The sequencer has already proved very useful flawlessly through the first two IRIS2 runs in October during IRIS2 commissioning, being used, for example, and November. to take sets of images using random dithering patterns. ANGLO-AUSTRALIAN OBSERVATORY page 24 NEWSLETTER FEBRUARY 2002 The Real Time Display News in Brief This provides a display of the incoming data A&G upgrade from the detector and replaces the old XMEM display. Rather than develop our own The new A&G unit at Cassegrain of the AAT has been used IRIS2 SPECIAL IRIS2 software here we have made use of ESOís extensively over the last quarter with great success. Runs involving Real Time Display (Rtd) system. This is regular instrument rotation like spectroñastrometry have seen an extensible software component which substantial improvements in efficiency due to the faster rotator. is the basis of applications such as SkyCat The new CCDñbased guider is also performing well. and Gaia. In the IRIS2 system we use the Taurus++ SkyCat application with our own extensions Taurus++ (i.e. multiñobject spectroscopy using VPH grisms in to communicate with our detector and Taurus) has now been used several times, and will be made telescope systems. available to all proposers in Semester 2002B. Taurus++ allows During an IRIS2 exposure which typically for multislit spectroscopy over a 9 arcmin field with full nod & consists of many reads of the array, the shuffle capability. Mask preparation and manufacture is display shows fully processed data and is straightforward, however this process must be completed well updated as each new frame is received. A before the run. Potential users should consult the Taurus web bias or sky frame can be automatically site at http://www.aao.gov.au/taurus, where a stepñbyñstep guide subtracted from the displayed data. There to preparing for a Taurus++ run will be made available. are facilities for measuring the size of star OzPoz images, looking at statistics of the data and much more. OzPoz, the 2dFñinspired robotic positioner for the VLT, has started its journey to Chile, marking the completion of AAOís The images displayed also include a ìWorld largest external project to date. Commissioning at the telescope Coordinate Systemî (WCS) so that it is will begin in early March. possible to read the RA and Dec of sources from the display. It also makes it possible AAOmega to use the display as a way of controlling AAOmega has passed an important stage of its development the telescope. For example it is possible with the completion of the internal review in early February. The to center the telescope on a star by clicking project is now in the Preliminary Design phase, with an external on its image on the display. It is also review due in late February. AAOmega will complete the possible to select sources by name from instrumentation suite of the AAT, with an extensive upgrade to the various catalogues available in SkyCat 2dF encompassing a doubleñbeam spectrograph, two 2kx4k and then slew the telescope to them. detectors and 392 multiñobject fibres. In addition there will be an optional IFU capability using the 512 fibres of SPIRAL. The 2dF Data Files robot, field plates and retractors will be refurbished as part of the With the new system we now write data in AAOmega project, and the new spectrograph will be benchñ FITS format rather than as the ì.sdfî files of mounted and thermallyñcontrolled. the Observer system. The files also include 2dF Service World Coordinate System data. Note that a correct WCS depends on the telescope At their 2002A meeting, PATT have decided to in future set aside pointing calibration procedures being an equal share of nights with ATAC for 2dF service in the normal carefully followed at the start of the night. AAO service observing program. UK astronomers will now be able to apply for 2dF service observing four times a year. The The Future next service deadline is March 15, for targets to observe on the We have recently initiated the Observatory PATT 2dF service night in July 2002. It is expected that this new Infrastructure Project which includes system should help to increase the efficiency of 2dF usage in adapting our optical CCDs and instruments the postñredshift surveys era. to use the new data acquisition system MITLL3 described here, as well as replacing the The MITLL3 was recently damaged, and is being sent back to old Interdata Telescope Control System. On Lincoln Labs for repair. It will be made available again as soon as current plans most or all of our instruments possible, but unfortunately not for the remainder of Semester should be running this new system by the 2002B. Scheduled observers should contact their support end of 2003. astronomers to discuss a replacement detector. ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 25 FEBRUARY 2002 THE DOVER HEIGHTS ìHOLEñINñ OBSERVATORYNEWS THEñGROUNDî RADIO TELESCOPE Wayne Orchiston (AAO & ATNF) Innovation is a special feature of Australian astronomy, with many examples of new concepts in instrumentation leading to major advances in our understanding of the universe. The Dover Heights ìholeñinñtheñgroundî antenna demonstrates this ingenuity in a telescope which was constructed half a century ago. Initial postñwar developments in nonñsolar radio astronomy were inspired by Hey, Phillips and Parsonís 1946 discovery of an intense source of radio emission Figure 1: The 21.9m antenna in June 1952, with the WWII blockhouse and a 4.9m parabolic dish in the background in Cygnus. In Sydney, John Bolton and Gordon Stanley, (ATNF Historic Photographic Archive: 2763-1). aided by Bruce Slee, used a sea interferometer to investigate this first ëradio starí, and soon discovered months, and when they were finished metal strips from others. By observing these from Dover Heights as they packing cases spaced 30cm apart were stretched rose over the eastern horizon they were able to obtain across the surface and pegged in place, in order to obtain approximate positions for CygnusñA, SagittariusñA, good reflectivity. A mast supported by guy ropes was CentaurusñA, TaurusñA and VirgoñA, and after a New installed at the centre of the dish to carry the dipole Zealand field trip of 1947, refined positions enabled the (Figure 1). The dipole was connected to a modified exñ search for optical counterparts. Through one of the WWII receiver tuned to 160 MHz. This novel radio earliest collaborations between optical and radio telescope had a beam width of 6 degrees. astronomers, it was found that TaurusñA was associated The construction and use of this new antenna was with the Crab Nebula, and CentaurusñA and VirgoñA cloaked in secrecy because the Dover Heights team with extrañgalactic nebulae. These pioneering efforts was meant to be making celestial observations, and led to the first allñsky surveys in England and in Sydney. the head of the group, Joe Pawsey, believed that staff While these techniques gave interesting results, in order should keep religiously to assigned projects. On an to make substantial progress new ways had to be found earlier occasion when Bolton and Slee were unofficially to increase the resolution of the radio telescopes. Small searching for nonñsolar emission, the project was ìÖ parabolic radio antennas were already in use, but at cut short by an unheralded visit from Pawsey, who noted that time engineering constraints prevented the that the aerials were not looking at the Sun. Suffice to construction of large steerable dishes. say that he was not amused and we were both ordered back to the Lab.î It was some time before they were The solution was a ìholeñinñtheñgroundî antenna, the allowed to return to Dover Heights! brainchild of John Bolton. The new antenna would not only offer Bolton and his colleagues increased resolution Observations were carried out with this new antenna in but also a means of escaping the various problems 1951ñ52 , resulting in a map of radio emission along associated with sea interferometers. the galactic plane. It was only after this success that Pawsey was told of the ìholeñinñtheñgroundî antenna. The Division of Radiophysicsí Dover Heights field station To everyoneís surprise he was not angry, but on the occupied a 5ha wartime radar station, and was located contrary, was excited by the research potential of this in the eastern suburban Sydney on a cliff top overlooking new radio telescope. the Tasman Sea, 5km south of the entrance to Sydney Harbour. During the second half of 1951, John Bolton It was then decided to expand the aperture of the and Bruce Slee, with some assistance from Gordon antenna to 24.4m and to use an operating frequency of Stanley and Kevin Westfold, excavated a 21.9m 400 MHz. This involved further excavation and concreting diameter parabolic depression in the sand near the cliff of the new surface of the dish. The upgrade took six top. In this lunchtime project they employed shovels for months, resulting in a transit instrument with a focal the excavation work and a wheelbarrow to dump the length of 12.2m (40ft) and an angular resolution of 2∞. spoil round what would become the rim of the antenna. The position of the aerial could be altered in the Northñ A crude wooden jig was used to approximate the South plane to observe strips of sky adjacent to the parabolic shape. The excavations took about three zenith (Figure 2). ANGLO-AUSTRALIAN OBSERVATORY page 26 NEWSLETTER FEBRUARY 2002 The first research project carried out with the new antenna was an all-sky survey at 400 MHz. As Figure 3 illustrates, the most important OBSERVATORYNEWS outcome of this survey was the clear delineation of SagittariusñA, the strong source which McGee, Slee and Stanley correctly identified with the nucleus of our Galaxy. This result showed more detail and less confusion than previous observations, and consequently the IAU agreed to adopt the Sydney position of SagittariusñA as the location of the Galactic Centre, thereby reñcalibrating the galactic coñ ordinate system. Another important outcome Figure 2: The enlarged concrete-coated 24.4-m antenna in Sept. of the 400 MHz survey was the detection of 14 1953, with Gordon Stanley using a theodolite to record to position of the aerial mast (ATNF Historic Photographic Archive: 3150-2) other discrete radio sources. The final project undertaken with the dish was an unsuccessful search for deuterium emission, and this marked the death knell for radio astronomy at Dover Heights. It was realized, as Bowen puts it, that ìÖ more and more sophisticated instruments would be necessary. The days of hasty improvisation were over ñ more and more planning would be required and much more money would be needed for capital expenditure. Little did we realize how large these sums would become.î Although radio astronomy ceased at Dover Heights at the end of 1954, for a while the Division's Cloud Physics group made use of the site. Finally, in 1959, the field station was Figure 3: 400 MHz contour plots showing the strong emission source associated with the galactic centre (McGee et al. 1955: 356). Note closed, and most of the site was converted into that the conspicuous oblique contour ëbulgesí are artifacts typical of a playing field. From all accounts, the ìholeñ this type of antenna. inñtheñgroundî antenna was simply filled with Acknowledgements soil and then grassed over. Today it lies not far north of the northern soccer goal post in what I am grateful to Bruce Slee and Brian Robinson (ATNF) for providing is now Rodney Reserve. Apart from the few information relevant to this study and to the ATNF for supplying the individuals who might chance upon an Figures. inconspicuous plaque hidden away in one corner of the Reserve, most of those who use Sources this popular recreational facility today would be Bolton, J.G., 1982. PASA, 4, 349 unaware of the amazing scientific contribution Bolton, J. G. et al., 1954. Austr. J. Phys., 7, 96 that this site made in the early days of Bowen, E. G., 1984. in Sullivan, 84 Australian radio astronomy. Goddard, D. E. & Haynes, R. F. (eds), 1994. Austr J. Phys. , Vol 47. Hanbury Brown, R., 1984. in Sullivan, 213 McGee, R. X., & Bolton, J. G., 1954. Nature, 173, 985 Bruce Slee is the only Dover Heights scientist McGee, R. X., Slee, O. B. & Stanley, G. J., 1954. Austr. J. Phys., 8, 347 still actively involved in radio astronomical Orchiston, W., 1993. Southern Stars, 35, 46 research (through his continuing association Slee, B., 1994. in Goddard and Haynes, 517 with ATNF). Recently I asked his overall Stanley, G. J., 1994. in Goddard and Haynes, 507 Stanley, G. J. & Price, R., 1956. Nature, 177, 1221 assessment of the ìholeñinñtheñgroundî Sullivan, W. T., 1984. The Early Years of Radio Astronomy, antenna. After pondering a moment, he was Cambridge University Press moved to reply: ìIt certainly was a lot of hard Sullivan, W. T. 1988. Early days of Australian radio astronomy. In work digging it out ñ all those blisters ñ but the Home, R. (ed.) Australian Science in the Making, Cambridge, Cambridge University Press, 308 research output certainly justified the physical input!î ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 27 FEBRUARY 2002 LETTER FROM Tibet COONABARABRAN Rhonda Martin LOCAL NEWS LOCAL ëThere, isnít that one?í ëUm!í ëIt has to be, itís round and yellow isnít it, or is that the smog?í ëUm!í ëPerhaps itís Mars. What do you think?í ëUm!í ëCome on girl, there has to be a star here somewhere. Is it Mars?í ëUm! Maybe ÖÖanyway, Mars is a planet.í Allan and Anne Lankshear and I were on a, wait for it, Seniors tour (This smugly ñ you canít fool me.) of China and Tibet (and didnít I just love being the youngest!) and Some desperation now appeared in the voice were eager to see a new sky from such a high altitude but that ñ ëWe are on the roof of the world ñ there is aspect of the trip was a bit of a fizzer. hardly any air ñ why canít we see any stars?í ëUm! Are there any? It is the north you know.í Otherwise, we ran out of superlatives about China, and more A withering snort issued forth and I quietly particularly Tibet, after the first week; then we just sat back and subsided, my previous scintillating enjoyed it all to the hilt. Flying into Kunming on the coat tails of a conversation demonstrating complete thunderstorm, we surfed down the tarmac on a foot of water, rooster ignorance of the northern sky. tail streaming behind us. There began our love affair with this beautiful land and its people who treated us throughout with courtesy It is understandable that in China one cannot and more importantly, humour. see stars, the pollution being just terrible, but in the high, thin atmosphere of Tibet we Western China does not get many foreign tourists yet so we were thought we would be overwhelmed by a blaze a novelty and always we could coax a smile and a wave from of light so perhaps I was right ñ it IS the north everyone. Pollution is a big problem and westerners are not geared and coming from the richness of the southern to drink the water without becoming ill so we bought it bottled, skies we were flummoxed by the difference. very cheaply, just about everywhere we went. But then again, Russell Cannon was observing in western Tibet about the same time and I In the wilds of Tibet, where gentlemen went one way and ladies bet he saw some. the other, modesty took a back seat and stayed there. Eastern toilets were an everlasting adventure and the source of giggles and shrieks, more so from the ladiesí side of things. We learned to scream, although not loudly due to the lack of oxygen. We screamed as our bus or fourñwheelñdrive hung over an abyss, the river at the bottom a gleaming thread. We screamed as we passed someone coming the other way on those oneñlane tracks which the locals fondly imagined to be roads and we screamed even more when our driver turned back to look at us, grinning hugely, wondering what the racket was about. We were introduced to magnificent scenery, dust, glacierñmelt rivers, irrigated wheat and barley and the smiling, curious people, a lot of whom had rarely seen a Westerner, but no stars. Tibet was the crown jewel of our tour. China proper was beautiful and we loved it, but the high country was all we expected ñ monasteries, monks in crimson robes and sneakers, yaks, prayer flags, prayer wheels and the beautiful, blackñeyed children. The few remaining monks were lovely, shaking our hands and blessing ANGLO-AUSTRALIAN OBSERVATORY page 28 NEWSLETTER FEBRUARY 2002 us. Flying into Lhasa we watched entranced as the earth rushed up to meet us ñ brilliantly coloured lakes EPPING EVENTS fed by glaciers, the snowñcapped Himalayas and the Ray Stathakis wonderful bare bones of the country ñ the hills and mountains of brilliant purples, reds and browns, so sharp Since our last issue came out in September, we have of edge they resembled cardboard cutñouts. plenty of events to record! In October we sadly bid NEWS LOCAL farewell to John Barton, after nearly 27 years at the Everest was sheer magic, a religious experience to our AAO. He and his wife have returned to New Zealand to Tibetan drivers, and also, a little, to us. This mountain, enjoy the pastoral life. We thank John for his the roof of the world, commanded our silence and respect unsurpassed work on detectors, in particular his last and got it. On a cloudless day, when the very air sang, two projects ñ the EEV CCD and IRIS2. We are now we ate our picnic lunch in its shadow and understood about to see another major figure, Russell Cannon, retire why mountains inspire reverence. (see p3), but Russell assures us we will scarcely notice Leaving Tibet was a wrench but our time there was worth the difference as he continues his research programs. every breathtaking moment. AAO has seen a number of visitors pass through. Hagai Back in China proper, we saw Pandas who reminded Netzer from Wise Observatory has visited from us strongly of koalas. Koalas, we think, are definitely December to February, working on the physics of AGN brighter, though not much! We guessed that if all we lines with Brian Boyle, Scott Croom and Elizabeth had to eat was bamboo or eucalyptus leaves, our thought Corbett. Claire Halliday from Liverpool John Moores processes would also be a little, um, slow, as well! University came in November to work with Terry Bridges on Galaxy Dynamics, and Bill and Gretchen Harris, from We saw and did so much ñ visited cities that never McMaster and Waterloo Universities respectively, visited slept, went to Chinese opera and found we liked it (well, Terry to work on stellar populations in Globular Clusters. most of it), travelled by train over vast distances watching the terracing and superb farming on one hand and hitting Our usual Melbourne Cup celebration was enlivened by cockroaches in our sleeping cabin very hard on the other. a fascinating display of Martial Arts Sword routines by We ate food, the origin of which we were never sure and Sonia Cianci. And we congratulate Chris Tinney on his enjoyed it although we did end up longing for something marriage in November. Congratulations also to Geraint sharp in flavour, like the ubiquitous Vegemite. Lewis, on the safe arrival of his son Dylan Francis on Somewhere along the way we think we might have eaten Christmas Eve. sweet and sour canine ñ Guiyang was noticeably short We have enjoyed the visits of our summer students. of dogs but very rich in excellent fruit and vegetables. Troy MacKay has worked with Roger Haynes and Anna We enjoyed ourselves thoroughly. Moore on investigating optical fibres for AAOmega. Kelly So all in all, it didnít matter that we couldnít see any Kranz has worked with Stuart Ryder on an Infrared survey stars in a Chinese sky ñ the real stars of this wonderful of starburst and Seyfert galaxies. Anju Thirunavukarasu country were its people. has worked with Joss BlandñHawthorn on abundances of cool stars, and Kirsty Rhook has been creating We returned to find two new staff members at the AAT ñ composite spectra from 2QZ with Scott Croom. We John Collins is an electronics technician and comes to wish them all well in their future studies. us fresh from Norfolk Island although he is originally Our personnel officer, Marilyn Campbell, has left us to almost a local, and Mick Kanonczuk, who, as a poachee enjoy the tropical pleasures of Queensland, but not before from the ANU, knows our mechanical workshop very winning glory for herself and the AAO with her poem well. Shaun James joined us in January, also in ìThe Ballad of Dusty Maryî (see p31). We welcome Greta Electronics.. With Johnís little girl and Shaunís three Simms back to her old job on a part-time basis. boys, Santa will have someone for his gifts next Christmas. We noticed this year he was looking very Our other new arrival is Andrew McGrath. Andrew arrived woebegone with only one small boy, Fred Watsonís, to in January to take up a position as an instrument receive a pressie. Welcome to all of you and your scientist. He comes to us from the Met Office in the UK families. where he worked as a microwave instrument scientist, maintaining and upgrading the scanning microwave And we all know about IRIS2 ñ she is a goer! radiometers aboard the Met Office Cñ130 Hercules Congratulations. atmospheric research aircraft. Here at the AAO he will make use of his very broad range of skills to work on DAZLE. ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 29 FEBRUARY 2002 GETTING FROM A TO B & NEW ARRIVALS AT THE TELESCOPES Chris McCowage Living in rural Australia with its long distances and fine climate makes personal transportation both important and at times enjoyable. Here are some of the site staff and their preferred method of transport. As there have also been NEWS W E N L A C O L a number of new arrivals at the telescope, we also thought that it might be a good opportunity for readers to see some of the new faces before their next trip to the telescope. † Brendan Jones (on right) supervises the mechanical workshop at the telescope. Brendan enjoys fine engineering and motor cycle touring with his family. Marque: Suzuki Model: Hayabusa Capacity: 1298cc Liquid-cooled inline four cylinder, 16 valves DOHC Power: 129 kW (173 HP) @ 9800 rpm Guy Andrews (on left) is a contract electrician who has done a considerable amount of work at the AAO over the past year. This has included work on the 6dF enclosure and Helium Reticulation. Guy enjoys playing cricket, waterñskiing and motorcycle touring. Marque: Yamaha Model: YZF Capacity: 599cc Liquidñcooled inline four cylinder, 16 valves DOHC Power: 73.6 kW (100 HP) @ 11,500 rpm Michael Kanonczuk has recently joined the mechanical section at site and is a fitter machinist. Mick and his wife enjoy motorcycle touring and bushwalking. Marque: Honda Model: ST1100 Capacity: 1084 cc Liquidñcooled longitudinal 90ñdegree, Vee four, 16 valves DOHC Power:74kW (100 HP) @ 7,500 rpm Wayne Clarke is a technical assistant in the mechanical section. Wayne produces a fine local wine from his small vineyard. He is also an active member of the Rural Fire Service. Wayne enjoys socialising, waterñskiing and motorcycle touring with his wife. Marque: Harley Davidson Model: Wide Glide with Shovel Motor Capacity: 80 Cubic Inches air cooled VñTwin 4 pushrod operated valves Power: Adequate Dionne James manages the archive and reduces 2dF data. Dionne enjoys horse riding and competing in Dressage. Marque: Quarter Horse Model: Phobos Deimos Capacity: ? Power: 746 watts or approx. one horsepower. ANGLO-AUSTRALIAN OBSERVATORY page 30 NEWSLETTER FEBRUARY 2002 Robert Dean is the Computer Manager at site. Robert is very active in the Coonabarabran community and has a special interest in Community Radio Broadcasting. His other interests include private pilot flying. LOCAL NEWS Marque: Piper Model: Warrior II PA28ñ161 Operated by the Coonabarabran Aero Club Capacity: 320 cubic inches (5245 cc) Four cylinder air cooled horizontally opposed, eight pushrod operated overhead valves. Power: 160 horsepower (119 kW) @ 2700 rpm. John Collins has recently joined the Electronics Section and is shown here preparing to load up the Etalon TAURUS. John has moved with his family to Coonabarabran after completing a tour on Norfolk Island with a Telecommunication Company. John enjoys Bushwalking with his family, abseiling and karate. Shaun James has recently joined the Electronics Section and is shown here preparing to install the MIT/LL2 detector on TAURUS. Shaun is no stranger to the astronomical environment having formerly been with the CSIRO ATNF at Narrabri. Shaun enjoys Bushwalking, electronics, restoring classic cars and spending time with his three sons. POETRY CORNER In October 2001, the Crooked Tongue Bush Poetry Competition was launched as part of the annual Coonabarabran ìFestival of the Starsî. Marilyn Campbell won 1st place for Best Original Poem for ìThe Ballad of Dusty Maryî, which was recited by the gifted bush narrator, Llew Denning. A narrative ballad, the poem is about the legendary outback heroine, Dusty Mary. It is still unclear in the minds of most readers whether she is in fact fictitious or perhaps a woman who actually lived. But, no matter, for she was an adventurer ñ as the following stanzas, from different parts of her story, will reveal: ìShe saddled up Canary, ítwas like a girlhood dream, To light out for the territory, to traverse field and stream. And everywhere that Mary went, the dust was sure to follow, It rose in clouds behind her horse, oíer hill and dale and hollow. And that is how it happened, how the wife of Bill McLairy Became known around the district by the name of Dusty Mary.î ìMaryís reputation grew and grew as dust clouds they got bigger, And nothing held her back for she was filled with vim and vigour. Mistress of the stockwhip, she became the boundary rider Roundiní up the bangtails on the hot backblocks of Bunga. 3,000 head oí cattle, enough to make her shiver, Graziní oíer the hills and plains north of the Western River. Ask anyone who lived there then, theyíll tell you folks were wary Of the finest female rouseabout, Mrs ìDusty M.î McLairy!î ANGLO-AUSTRALIAN OBSERVATORY NEWSLETTER page 31 FEBRUARY 2002 6583 line emission, l ] II ratio is well below unity a 6583/H l ] II .au .au > line emission, [N emission, line a @aaoepp.aao.gov @aaocbn.aao.gov.au > @aaocbn.aao.gov.au < user < user RICKETTS Australia Fax +61 2 9372 4880 email 1710 1710 .au .au > ratio. North is at the top and east to the left. The positions of the continuum peaks editorial assistant SANDRA 4800 4800 a .aao.gov HAKIS HAKIS ANGLO-AUSTRALIAN OBSERVATORY 6583/H l ] II and [N and Sylvain Veilleux (Maryland, Sylvain Cottrell) Veilleux and David Rupke (Maryland) have discovered a shockñexcited the with obtained were 1482 NGC of images Narrowñband p6). article (full galaxy starburst a in wind TTF: (clockwise from upper left corner) red continuum, H in the galaxy disk but larger than unity in the hourglassñshaped nebula above and below the disk. This disk. the below and above nebula hourglassñshaped the in unity than larger but disk galaxy the in wind. galactic a of suggestive highly is structure 2 kpc for the adopted distance of 19.6 Mpc for NGC 1482. The [N The 1482. NGC for Mpc 19.6 of distance adopted the for kpc 2 are indicated in each image by two crosses. The or spatial arcsec, scale, 21 ~ to indicated by corresponds a and horizontal image each bar for same at the is the image, continuum red the of bottom AAT/Schmidt Telephone +61 2 6842 6291 Fax +61 2 6884 2298 email < http://wwwURL editor RAY STAT RAY editor ISSN 0728-5833 Published by PO Box 296 Epping, NSW Epping Lab Telephone +61 2 9372 THE BACK PAGE