ESO Turns 40 CATHERINE CESARSKY, Director General of ESO

No. 109 – September 2002 ESO Turns 40 CATHERINE CESARSKY, Director General of ESO

1962 – 2002… start and promising perspectives for a ESO’s skillful and dedicated staff that Four decades super-giant telescope, ESO has ma- brought these great projects to fruition. that changed tured to become a major player on the I also want to salute the commitment of Europe – and world scene, offering to its community a the members of ESO Committees. I am forty years that unique complement of research facili- in particular grateful for the active sup- changed Euro- ties. Always a persistent driver of front- port over these past three years from pean Astron- line research, it is now the prime serv- the President of Council and from the omy. One full ice organization in its field on this con- Chairs of Finance Committee, Scientific generation of tinent, with thousands of scientists prof- and Technical Committee, VLTI Imple- scientists, a iting from precious data obtained with mentation Committee, Observing Pro- wonderful time telescopes at the ESO sites. In a gramme Committee, Users Committee during which steady and carefully planned process, and Visiting Committee, as well as from many of our new and extremely powerful instru- so many influential members of these dreams, our ments and telescopes are being devel- Committees. And, of course, above all, hopes, and our oped for the ESO community in close I wish to thank the member countries goals have finally come true. collaboration with research institutes for their enthusiastic support of ESO Our ancient science has always and the high-tech industry. Innovative throughout the years. been characterized by broad interna- concepts, like the Astro- tional collaboration. However, it was nomical Virtual Observa- only in the early 1950’s that our illustri- tory and associated data ous predecessors, led by Jan Hendrick archives with tens of Tera- Oort and Walter Baade, embarked bytes contribute to the upon the arduous political process that success of European As- ultimately gave birth to ESO. With great tronomy and Astrophys- foresight and an equal measure of ics. stubborn will, they paved the way for The challenges ahead some of the world’s best telescope and are commensurate with instrument facilities, the solid and the achievements of to- durable ground for a strong and brilliant day. To meet them, ESO, future of European Astronomy and a dynamic organization, is Astrophysics. committed to continuous From the construction of its first ob- progress and ever-in- servatory at La Silla on a remote moun- creasing efficiency. I want tain top in the inhospitable Atacama to take this opportunity to desert, to the momentous Very Large pay tribute to those who Telescope at Paranal, with the con- have contributed to ESO’s struction of the ALMA project about to success. I must start with

1 ESO Council Meeting in London, 8–9 July 2002

To mark the occasion of the the tenth member state joining ESO, the UK invited ESO to hold its 98th Council meeting in London on 8–9 July 2002. The hosts provided a wonderful venue for the meeting in Trinity House, overlooking the Tower of London. An excursion to the historic Greenwich Observatory and a banquet at Lancaster House helped to make it an especially memorable occasion. A landmark decision made at the meeting was the approval for the construction and operation of the Atacama Large Millimeter Array (ALMA). The speeches given at the banquet by Lord Sainsbury, A. Freytag and C. Cesarsky are printed below, followed by an article by Prof. Gerry Gilmore on the British astronomers’ perspective.

Speeches to Mark the Accession of the UK to ESO Lord Sainsbury, UK Science Minister

Good evening Ladies and Gentle- than the provider of all the facilities International collaboration is a key men and welcome to you all. It is a themselves. For this reason, the part of this strategy because the char- great pleasure for me to act as host Science and Innovation White Paper acteristics of these large-scale facilities tonight for this dinner, which is held to ‘Excellence and Opportunity’, which we often make collaboration the most ef- celebrate the UK’s accession to the Eu- produced two years ago, undertook to fective means of provision: They are ropean Southern Observatory (ESO). I develop a ten-year rolling plan for fu- expensive to build and operate; they am delighted too, that we have the op- ture large-scale facilities, taking ac- frequently serve national and interna- portunity to host this ESO Council count of developments in Europe and tional users; and they tend to be multi- meeting, and that we have the ESO elsewhere, to ensure that UK re- disciplinary. Council Members and the principal offi- searchers have access to the best fa- The UK astronomy community was cials of ESO present here tonight. cilities in the world. also asked, a few years ago, to exam- I am sure I speak for the entire UK The Large Facilities Strategic Road- ine their science priorities for the next astronomy community when I say how map, which we produced, represents ten to fifteen years, and also to identify much we are looking forward to partici- the first attempt at a ten to fifteen year the facilities required to address those pating in ESO and taking advantage of map of future facility requirements. The priorities. The message received was its marvellous facilities. I also hope very aim of the document is to provide a clear – joining ESO was the top priority much the UK’s participation will lead to longer-term vision of future require- for astronomy. a strengthening of ESO and a widening ments of the UK Science and Engi- Our astronomers recognized that the of its capabilities for astronomical re- neering Base. It reflects the context of current generation of World-leading tel- search. future European, or in some cases glo- escope facilities are on a scale that can As Minister for Science, I see my role bal, requirements for large-scale facili- only be achieved through international as the provider of access to World- ties in order to assess the most effec- partnerships. This of course has been class facilities for UK scientists rather tive approach for satisfying UK needs. an increasing trend for some years and

Gathered in the historic Octagon Room of the Royal Greenwich Observatory, London, Ian Halliday (CEO, Particle Physics and Astronomy Research Council) stresses the benefits to British astronomers of belonging to the European Southern Observatory. The Panel consisted of (left to right) Roy Clare (Director of the National Maritime Museum), Arno Freytag (President of the ESO Council), Lord Sainsbury (Science Minister), Gerry Gilmore (Cambridge University), Ian Halliday, Catherine Cesarsky (Director General of ESO) and Pat Roche (Oxford University). Courtesy PPARC.

2 will no doubt become more pronounced The UK now embarks on a new jour- We believe, therefore, that strong inter- for future telescopes under considera- ney with the joining of ESO. There are national relationships are essential; any tion. some exciting opportunities ahead and society that is closed, inward looking The Government responded to the I am aware in particular of the ALMA and defensive will not long remain at wishes of the astronomy community in project. the forefront of science because it can- the Government’s 2000 Spending This global project, with Europe, not take part in global collaboration. I Review, when it made a special contri- North America, and possibly Japan, all believe that the UK is stronger when it bution to PPARC of £100 m over the working together, promises, once com- collaborates internationally and I want next 10 years specifically to allow the pleted, hopefully in 2009, to be the the UK to be a key player in European UK to join ESO. We are of course very largest ground-based astronomy facility and global science. pleased that the UK is now at last a ever constructed. The UK is very en- The message I want to convey to you member of ESO. The UK has joined thusiastic about becoming involved in is very simple. We are very pleased probably the World’s leading observa- ALMA. I know ESO Council has been that the UK has finally joined ESO, we tory and UK astronomers will gain ac- considering this subject carefully and are excited by the opportunities that lie cess to some of the World’s most ad- we look forward to its decision on par- ahead, and we hope UK participation vanced telescopes including ESO’s ticipation. will serve to strengthen this renowned Very Large Telescope. Joining ESO also The UK currently funds about 5% of international organization. integrates the UK astronomical com- World science. This means that over Thank you. munity with that of continental Europe. 95% of science is funded elsewhere.

Dr. Arno Freytag, President of the ESO Council

Lord Sainsbury, distinguished guests, enormous progress in the past few bution of Vista, a uniquely powerful in- ladies and gentlemen; decades. The outstanding recent tele- frared survey telescope that will consid- Thank you for your warm words of scopes in space and on the ground are erably enhance the already exceptional welcome and for inviting us to dinner in allowing us to accelerate the pace of capabilities of our Paranal observatory. these magnificent surroundings, which that progress. Europe has to work to- I can say, without any false modesty, I must say compare rather favourably gether in astronomy – as it has demon- that ESO has become the leading as- with the facilities in Garching! strated it can do in other fields, such as tronomical observatory in the world. This is a historic occasion. We are all particle physics – if it is to exploit these This is due to our clarity of vision, the privileged to be a part of it, no-one more wonderful instruments. But it is even dedication and skill of our staff, the than me. It is indeed an honour to be more important that we work together strength of our community, and the President of ESO Council, and, on be- to prepare for what is to follow. support of our member states. Now that half of all the member states and the That is why today is so significant. the United Kingdom has joined, we will staff of ESO, to welcome the United For today the ESO Council dis- be stronger and better prepared for the Kingdom into our midst. cussed European participation in future. We know you share our vision, We have had a most enjoyable after- ALMA. This is a truly international proj- we know of your skills and dedication, noon which has served to remind us of ect with a good prospect of turning into we never doubted the strength of your Britain’s long and distinguished contri- a global project. We know that ALMA community, and we now know we can bution to astronomy. So now the nation was a major force behind the United count on your support. I look forward of Newton and Herschel joins the na- Kingdom decision to join ESO, and we confidently to an outstanding future for tions of Galileo, Kepler, Brahe, Cassini, recognize the mutual benefit, for, with- ESO and for European astronomy. Messier, and many others who paved out the United Kingdom, ESO could not I turn once again to you, Lord Sains- the way to where we are now. take up a half share. That would have bury, and thank you, and everyone else But today, of all days, we look to the been a disaster for European astrono- in the United Kingdom who made it future. World astronomy has made my. We also look forward to the contri- possible.

Dr. Catherine Cesarsky, ESO Director General

I would like to thank you, Lord Sains- lar system. But on Earth, a full genera- of what has been and what can be bury, for your hospitality here and for tion of astronomers. achieved by working together. So what your kind words. It is also the age of ESO this year. happens in the next 40 years? Our flag- Forty years ago many of us were still And the time needed for us to prove ship projects – VLT – ALMA – OWL... in school or at the beginning of our ca- that we are the best in the world in our We cannot promise to find that first reers – not able to imagine the incredi- field, good enough for the UK to join af- exoplanet with exo-life, but we will have ble developments going to happen to ter 40 years of hesitation! the means to look for it. We cannot us – to science – to Europe – to the Astronomy is the international sci- promise that we will understand what world. But what is 40 years in astro- ence, since the earliest times. The the enormous amount of dark matter nomical terms? 40 revolutions of the heavens know no borders. There are and dark energy in the Universe is Earth around the Sun – a little more megalithic observatories in your coun- made of, but we will search for it. We than half a revolution of Comet Halley – try and also in my country; surely the cannot promise to discover the ultimate the orbit first calculated by famous master builders talked to each other secret of the world in which we live, but British astronomer Edmond Halley in also in those ancient days. we will certainly know much more about 1705 – a little less than half a revolution Astronomy demonstrates to all of it and our own position. of planet Uranus, discovered in 1783 by Europe the benefit of pooling forces – Astronomy has an enormous poten- British astronomer William Herschel. by doing so, we can do better than any- tial for exciting discoveries that will fas- One 100 millionth of the age of the so- body else. Let us be honest and proud cinate the public and it will continue to

3 attract the most clever minds among fu- welcome it to ESO. Together we have mosphere. Do sit down at the telescope ture generations. an enormous potential for new break- controls and let us look together to- The UK has a long and successful throughs. wards the end of the universe and the history in our science, with many trail- Minister Sainsbury, we would be very beginnings of time! blazing results by theoreticians and ob- happy to welcome you at Paranal. Do servers, and we are proud and happy to come and experience that unique at-

ESO AND THE UK Why Does the UK Need More Astronomy? GERRY GILMORE, Professor of Experimental Philosophy, Institute of Astronomy, Cambridge University, UK

“What was God doing before he and image quality. Each modern large plemented by the APM (Cambridge) made heaven and earth? … He was telescope is both vastly more sensitive, and COSMOS (Edinburgh) measuring preparing hell for those who would pry and vastly more efficient, than were 4- machines, the UK InfraRed Telescope into such profound mysteries.’’1 This metre-class telescopes 20 years ago. (UKIRT) and the beginnings of the joke was already venerable when quot- It is this huge increase in generation JCMT sub-mm telescope on Hawaii, ed by Augustine, in his analysis of the of high-quality data which drives cur- and the Isaac Newton Group on La ancient and still modern problem, time. rent progress in astronomy. Conse- Palma were world-quality facilities quite Understanding the origin(s), mean- quently, the community with the best sufficient to challenge those of us fortu- ing(s), future(s), and significance(s) of technology has the best opportunity to nate enough to be let loose on them. time, space, existence, mass, matter, discover the new, and has a head-start These observatories were (mostly) geometry, of origins and endings, of in attracting bright young people to sci- international partnerships, with the UK what and where, remains one of the ence. But it is not just a question of the largest partner. Next came Gemini, greatest intellectual endeavours of the wealth buying power: the huge techno- two superb 8-m telescopes, with the UK human mind. From the caves of Las- logical investment of Tycho and Kepler as a 25 per cent partner. And most re- caux, through the megaliths of Stone- reached its scientific fruition with cently ALMA, with the UK as (roughly) henge to the dreamtime of Australia, Newton. Real scientific progress, as 20 per cent partner inside the Euro- mankind has striven to understand his that example reminds us, requires both pean-wide 50 per cent share. Why the origins and future. Our generation has technology and people, complementary systematic decrease in share? Why is the exceptional good fortune to be liv- approaches, and trans-national collab- Gemini on-line so long after Keck? ing through the greatest increase in orations. And it works best with a spice Simple: money. knowledge relevant to these fundamen- of competition. Sometime around 1990 optical/IR tal questions since someone first Considerations like those above led astronomy became too expensive for looked up at night. We are also in- to the formation of ESO (cf. ESO’s one country, even one as large as the creasing understanding, while realizing Early History. A. Blaauw) and the for- UK. But something else more funda- how much more there is in the Universe mation of La Silla Observatory, and led mental changed too. ‘International as- still be learned and understood. the UK to found collaborative observa- tronomy’ began to mean more to UK Even more wonderful (sic) for us, our tories in Australia, South Africa, the astronomers than ‘astronomy in the for- rate of progress in knowledge is accel- Canary Islands, Hawaii and Chile. mer British Empire’, or ‘trans-atlantic erating, as the technological advances (Radio and space astronomy have their astronomy’. Routine collaboration be- resulting from research into basic sci- own history and set of personalities, tween institutes in the UK and in conti- ence feed back positively in turn to ad- and are not considered in this article.) nental Europe was less common than vance basic knowledge more rapidly. A significant motivation in develop- was collaboration with the US. But this This is truly a golden age of discovery ment of these observatories was an at- began to change. in astronomy, with almost every class of tempt to regain international research Of course, many European countries object we study having been discov- leadership in astronomy. For whatever besides the UK had close scientific ered in our working lifetimes. mix of reasons, Europe, including the links across the Atlantic: the effect of Why is it so? There are two dominant UK, fared much less well relative to the the Netherlands on US astronomy is a reasons: technology and people, but US in astrophysics research in the ear- famous exemplum. The European only one explanation: efficiency. The ly 20th century than it did in, for exam- (largely Italian) diaspora who made the astronomical community is at most one ple, quantum theory and relativity. Space Telescope Science Institute in order of magnitude larger by number Baltimore so much more than just an- than it was a generation ago: a signifi- UK and European astronomy: other NASA center is a major example cant, but not huge advance. Astronom- a micro-history of the happy internationalization of as- ical telescopes today provide the real tronomy. Cheap and easy travel was of advance, with not only a very consider- I am not aware of the factors consid- course another factor. As was the les- able increase in mirror collecting area, ered when the UK decided to develop son from space science and radio as- but a vast increase in detector area, de- its astronomy independently from ESO, tronomy, which had much earlier tector quantum efficiency/sensitivity, through bilateral partnerships, but by crossed the ‘unaffordable by one coun- 1980, when I arrived in the UK, it was try’ barrier. All these factors changed 1‘Quod faciebat Deus, antequam faceret caelum obviously a successful policy. The the assumption, and encouraged UK et terram?’ Respondeo non illud quod quidam re- Anglo-Australian telescope, with its astronomers to look more widely for spondisse perhibetur, ioculariter eludens quaestio- nis violentiam: ‘Alta’ inquit ‘scrutantibus gehennas marvellous IPCS photon-counting sys- competition, and for colleagues. parabat’. Augustine, Confessions XI xii 14. tem, the UK Schmidt Telescope, com- And what did we see happening in

4 An illustrative view of the development of astronomical telescopes. From the top left, the telescopes of Galileo, Newton and Birr Castle, reflecting European technological innovation and dominance in astronomy until the twentieth century. Mt Wilson and the two Keck telescopes are typical of the dominance of telescope technology by the private US observatories through the twentieth century. Finally, the VLT sets the standard of excellence at the start of the 21st century. continental Europe: by the late 1990s and continental Europe. In 1992, the Europe-wide organization to build and European astronomy was not only not IoA had 50 postdoctoral fellows, about operate it, so that some sort of a part- ignorable, it was seriously good, and one-half from outside the UK, of whom nership between the UK and ESO about to become outstanding. This was 4 were from Western Europe (one each would happen. Fortunately, the UK is only in part a technological change. from Greece, Italy, Norway and Spain). currently in a period of relative wealth, There was one other structural change, In 2002 the IoA has 70 postdoctoral fel- and has a government supportive of all still only in its earliest stages in some lows, of whom 22 are from Western of excellence, science, and Europe. countries, which perhaps had the Europe. In addition, 6 European-regis- The conditions came into phase ideally, largest positive effect: the move away tered PhD students (on an EARA/Marie and here we are in ESO! from tenured positions on completion of Curie EU-funded programme) are visit- a PhD to an assumption of a postdoc- ing. The change is dramatic, from 8 per Will the UK change ESO? toral position, or several, in different in- cent to 30 per cent, and UK astronomy stitutes and countries, between degree is very much better for it. This question has been raised a few and job. It is postdocs who really move With this background, one can now times! My answer is purely personal. I around, who naturally, through re-loca- answer the question: why did the UK think the UK will change ESO: the last tion, become part of multi-Institute col- join ESO. As shown above, over a big European country is in, this is the laborations, and who really link com- decade close and real scientific part- biggest change ESO will experience in munities. nerships were developed. The un- the forseeable future. Much of the Also important was the effort made known was replaced by mutual respect. change will be cultural. UK astronom- by key individuals: for example, Simon This was a necessary but not sufficient ers have a somewhat more aggressive White (then in Cambridge), Alain condition. Then something much more attitude to publishing than do some oth- Omont and George Miley founded important happened: the VLT. er communities. There is in the UK a EARA, the European Association for significantly larger bias than in some Research in Astronomy, a formal link The VLT changed everything countries towards studies of the poorly between Cambridge, Paris and Leiden known: dark matter, inflation, galaxy (now extended to include MPA As it became clear that ESO really formation… rather than more detailed Garching and IAC Tenerife). Specific was delivering the world’s finest large studies of known objects. UK as- initiatives such as EARA, together with telescopes, UK astronomers realized tronomers tend to question extant the sociological change which forced they needed to be part of ESO. This un- structures and priorities rather more young astronomers to move around, funded ambition was complemented by than do some other communities. For had a big and positive impact. The development of ALMA, in which the UK example, some have asked if VLTI de- Institute of Astronomy in Cambridge was an active participant. It was always velopments are proceeding on a provides one very clear illustration of clear that ALMA would be a world-scale timescale and scientific cost-benefit ba- the changed balance between the UK facility, and that ESO was the natural sis which is maximally appropriate to

5 today’s financial and facility situation. I Gregory’s ‘Pastoral Care’, tr. M. new ideas. Gildas, in his subtly-titled know many UK astronomers want the Lapidge and S. Keynes.] De Excidio Britanniae (On the ruin of next-generation European Large Tele- There is a school of thought which Britain), writing c. 540, at the time scope sooner rather than later, even at asserts that Britain’s occasional drifts Arthur is frequently supposed to have the cost of other priorities. “We didn’t into barbarity and ignorance corre- existed, describes the coexistence of join ESO to let the US leave us behind spond to isolation from Europe: Saxons and Britons, leading to the rise again’’, is a common refrain. ALMA Romans civilizing, post Roman Dark of Anglo-Saxon England. must be made a success. But, most of Ages; Vikings exciting, later Alfred’s We look forward to the next stage of all, the astounding VLT must be used to lament; Normans enlivening, medieval coexistence: astronomers across deliver the exciting science for which it black death. Even the quintessential Europe, now including the UK, uniting was built. On that, I am sure all of us in British hero, King Arthur, is associated in progress, and working together for ESO agree. with Saxon and Angle introductions of the future across a whole continent.

TABLE: Who does astronomy in the UK. Astronomy research groups exist in many UK uni- What does Europe get versities. An approximate identification list, with a crude indicator of size, can be found by not- from the UK ing which groups are supported by PPARC (the sole national UK funding agency for astronomy). The table lists all groups funded by PPARC at present, and the number of associated In the preface to his translation of St grants. The number of grants is a very crude indicator of group size, but it must be noted that Gregory’s ‘Pastoral Care’, King Alfred this list includes space hardware groups, solar system research, and some upper-atmos- (c. 890) commented “Learning had de- pheric physics. More specific information can usually be found on www pages. clined so thoroughly in England that there were few men on this side of the Organization Number of grants Organization Number of grants Humber who [could] even translate a single letter from Latin into English. Armagh Observatory 6 Nottingham University 10 There were so few [men of learning] Bath University 1 Open University 9 that I cannot recollect even a single one Birmingham University 12 Oxford University 12 south of the Thames… I recollected Bristol University 7 Portsmouth University 2 Cambridge University 35 Queen Mary and Westfield College 12 how – before everything was ran- CCLRC (Rutherford Laboratories) 2 Queen’s University of Belfast 8 sacked and burned – the churches Durham University 17 Reading University 1 throughout England stood filled with Edinburgh University 10 Sheffield University 8 treasures and books. … And they de- Exeter University 4 Sheffield Hallam University 1 rived very little benefit from them be- Glasgow University 5 Southampton University 15 cause they could understand nothing of Hertfordshire University 5 St Andrews University 15 them, since they were not written in Imperial College 20 Surrey University 1 their own language. I wondered ex- Keele University 2 Sussex University 3 ceedingly why the good wise men who Kent University 7 UK Astronomy Technology Centre 3 Lancaster University 4 UMIST 4 were formerly found throughout Eng- Leeds University 9 Univ of Central Lancashire 5 land, and who had thoroughly studied Leicester University 19 University College London 40 all those books, did not wish to trans- Liverpool John Moores Univ 8 University of Wales Cardiff 17 late any part of them into their own lan- Manchester University 8 Univ of Wales, Aberystwyth 2 guage. But I immediately answered my- Natural History Museum 2 Warwick University 2 self, and said: ‘they did not think that Newcastle University 1 York University 3 men would ever become so careless, and that learning would decay like Note: ‘Number’ is the number of current grants at the institution; taken from the PPARC webpage this.’ ” [Ref. King Alfred’s Preface to http://www.pparc.ac.uk

ALMA: the next major ESO project.

6 ESO TURNS 40

Perspectives from the Directors General, Past and Present ADRIAAN BLAAUW, ESO Director General, 1970–1974

Reflections on ESO, 1957–2002

Nearly half a century ago, I witnessed Walter Baade and Jan Oort dreaming of a joint enterprise which would lift observational astronomy in Europe from the level of their modest national efforts to that of the leading observatories in the United States. I have been privileged to see, and to have been able to contribute to, the realization of that dream. This half century has left a wealth of recollections and senti- ments from which it is difficult to select for this occasion. My direct involvement with ESO began in 1958, upon my return from the US where I had lived in the years 1953–1957. Seventeen years later, in December 1974, I concluded my five- year term as Director General. I was In search of a site, 1963. slightly involved as a Council member for the Netherlands in the late 1970’s and early 1980’s, but became pretty duced on pages 2 and 3 of my book). mittee (the precursor of Council) and in deeply involved again when I started However, by 1957, little progress had this capacity became deeply involved in writing ESO’s history, which first ap- been made, mainly due to the great dif- the organization of ESO’s site testing peared as installments in the Mes- ficulties encountered in obtaining the expeditions, first for several years in the senger in the years 1988–1991 and governments’ agreement and financial South African desert, and then briefly in then as my book ESO’s Early History of support. These efforts continued and Chile until, in November 1963, ESO re- 1991. led to the signing of the Convention in solved to settle in the Andes. Satisfac- When, in 1953, I left for the States, I September 1962. (By that time the UK tion about this excellent choice by ESO had earlier that year witnessed the first had dropped out, Denmark was about is mixed with the recollection of the de- moves toward establishing a joint to join.) But, behind this simple state- votion to the cause of ESO on the part European observatory at the occasion ment lie that immense patience and of all those who in South Africa, so re- of IAU Symposium No.1. This led to the perseverance of ESO’s founding fa- mote from home and European culture, “declaration of intent” signed in January thers. It should not be forgotten by to- devoted years of effort and time to our 1954 by astronomers from Belgium, day’s students of astronomy. cause. France, Germany, Sweden, and the Meanwhile, I had become a sort of My involvement was renewed when, United Kingdom (it has been repro- Executive Secretary of the ESO Com- from January 1968, I became Scientific

La Silla before… … and after.

7 The 1-metre Photometric Telescope started that these two essential parts of the operating in 1966. ESO programme made such excellent progress during my directorate. But perhaps even more fundamental, I be- fortunate we were to be able to engage lieve that by the end of my term, those in a collaboration with CERN for our gnawing doubts that marked its begin- Telescope Project, on the CERN prem- ning had been removed and ESO had ises near Geneva. By the time I hand- won the full confidence of the funding ed ESO over to my successor Lodewijk governments. Woltjer, the 3.6-m telescope was near- Of course, many more recollections ing completion. In another respect our come to my mind, too many to dwell collaboration with CERN was equally upon within this limited space. I feel successful. After ESO’s Schmidt tele- happy to have contributed, through scope project had also been reorgan- ESO’s status and its administrative ized and successfully put into operation services, to the creation of Astronomy – an accomplishment inconceivable and Astrophysics, a European Journal, without the perseverance and patience in 1969. And last, but not least, I feel of some of my close collaborators – we proud to have initiated half a year be- could establish on CERN premises our fore my retirement as DG, the ESO unique Photographic Laboratory, capa- Messenger – at that time meant as a ble of undertaking the extremely de- means to promote communication be- manding job of producing the Sky Atlas tween ESO’s various departments – for ESO and for the UK Schmidt. It is, now serving the astronomical commu- to me, a source of great satisfaction nity at large.

Director of ESO, formally for half of my time, but in practice soon for a larger share. While ESO’s first General Direc- tor, Otto Heckmann continued his efforts to complete ESO’s instrumentation programme as outlined in the Con- vention and with administrative and personnel matters, my task was to initiate the scientific work, i.e. the observational programmes with the telescopes that had become operational. Principal among these were the 1,52-m “Spec- trographic telescope” and the 1-m “Photometric Telescope”. In March, 1969 ESO dedicated, on La Silla, the completion of this “First Phase”. It crowned an effort to which both Chilean and European staff in Chile had essentially contributed, for some of them not without considerable personal sacrifice under very demanding conditions. When, two years later, Heckmann re- tired, I was appointed his successor for a term of five years. There was no mis- take about my principal assignment: realizing the main telescope project and the Schmidt telescope. These two projects, unfortunately, had been lagging far behind schedule. Whereas Heck- mann had admirably and successfully laid the foundations for ESO with all its political and logistic aspects, he had not succeeded on these two topics, and se- rious doubts had begun to arise among the supporting governments. In retrospect, we know that the scope of a project of this size was far beyond what col- lective experience of European astronomy had learned to handle. We had to call on those scientists and engineers used to tackling projects of a size comparable to our big telescope in costs and engineering challenge, whatever the nature of the instrument. When I reflect on my years as DG, I realize how The 3.6-metre telecope, completed in 1976.

8 LODEWIJK WOLTJER, ESO Director General, 1975–1987

Reflecting on my thirteen years as The immediate Director General, what gives me most future of ESO is satisfaction are the following: clear: Adaptive op- The enlargement of ESO member- tics at the VLT and ship with Italy and Switzerland, which the VLT interferom- put it on track to be a pan-European eter, the ECF/NGST organization – now still more fully real- +Astrovirtel, ALMA. ized with the adhesion of Portugal and But what comes the UK. thereafter? A Very The realization of the NTT which Very Large Tele- showed that ESO had developed the scope? A very large capability of technological innovation interferometer? And and the organizational structure for what is very large, handling larger projects. in metres and in eu- The idea of the VLT and the comple- ros? Since ultimate- tion of its planning phase and approval, ly all euros come as well as the discovery and acquisition from the same of Paranal as the best site world-wide sources, what oth- for optical astronomy. er European proj- The extension of ESO’s mandate to ects have to be fi- include the ST-ECF in cooperation with nanced? As one ex- ESO Headquarters in Garching. ESA, and SEST in cooperation with ample, many Euro- Sweden. Some discussion was needed pean radio astron- in Council about the fact that HST omers would wish to participate not or three decades from now? Answers to would also look at the southern sky and only in ALMA, but also in another world- such questions are far from obvious, that, after all, radio photons were not wide project, SKA – the square kilome- but perhaps they should receive more fundamentally different from optical tre array. So it is not clear that budgets attention in a broader circle than they ones. So both projects could be fit- for optical facilities can be increased have until now. ted in by appropriate interpretation of much further. And finally, will optical as- Also through its closer collaboration the ESO Convention. And following tronomy have a long-term future on the with ESA, ESO is now more than ever SEST, ESO’s participation in ALMA ap- ground or will most innovative instru- at the centre of European Astronomy. pears now entirely natural. mentation move into space some two Its future looks very bright, indeed.

The ESO Council in session on December 8, 1987, when the VLT project was approved.

9 HARRY VAN DER LAAN, ESO Director General, 1988–1992 From SEST to ALMA, from NTT to OWL: Of Vision, Dreams and Realities

ESO has come a long way since in certed actions tried to break up the Paranal’s number of clear nights and 1987 the first rocks were blasted at main structure contract into at least the amount of superb seeing, ground- the NTT site on La Silla. Those were three pieces. Summer weeks were based optical astronomy’s most pre- exciting days, when SEST came online spent in design reviews of the main cious asset, were without precedent. and soon after the VLT programme structure tenders, an operation whose That building the VLT on ESO’s La Silla was getting up to speed upon its ap- motive was to meet political objections territory had countless logistic, opera- proval in December 1987. It was not an in a technical guise. The exercise was tional and hence financial advantages easy time for staff or management: well worth it, as the performance of the was as clear to me as it was to admin- taking up the role of main contrac- unit telescopes has by now amply istrative Council- and Finance Com- tor for its own design and construc- demonstrated: the affordable Italian bid mittee members. But unlike them, I tion programme rather than finding for realizing the ESO double-track de- could assess the science-added value an industrial consultant to do so was sign prevailed in the end. of going North and it far exceeded the an enormous challenge. It was not ob- Such troubles are, I believe, a normal extra costs and trouble. All powers of vious that it could be done, for more and inevitable feature of major interna- persuasion had to be mustered but in than ninety per cent of ESO’s staff cational projects, although they have a the end science won over short-term pacity was occupied with running La peculiar flavour in European organiza- economy and convenience. Silla, operating Headquarter services tions. Today the Paranal Observatory is a and constructing the NTT. The VLT The site decision was of major signif- towering witness to astronomical per- Blue Book and the bag of money icance and did not come lightly. Before sistence, engineering skills and ESO Council had allocated to its realization coming to ESO, I chaired the Site staff dedication. Europe will be in the were necessary but by no means suffi- Selection Working Group and was con- lead for many decades to come in ex- cient. For the new, formidable task, vinced that the Paranal area, in the ploring the Universe from there, the manpower had to be found and trained, heart of the Atacama Desert, was much finest cosmic discovery base yet de- manpower both reassigned and newly superior to the La Silla region. Both vised by man. recruited. Change inevitably meets resistance in both staff and community. For astronomers in member states the VLT was a faraway dream that could not help current Ph.D. projects or further institute ambitions within their normal timeframe. Reductions, of services, of instrumentation and of telescopes were therefore opposed, now and then vehe- mently. For staff, ends of contracts or reassignments often seemed unfair and misconceived: was their current work not valuable, their normal effort not in demand? The NTT proved crucial for both sorts of objections. It enabled me to introduce the La Silla Key Programmes very early in my term, providing unparalleled opportunities for trailblazing research of a scope until then not possible in Europe. The very positive response to this initiative made inevitable economies on La Silla more palatable; the resistance faded. Technically and contractually the NTT proved a great learning process for the job, thirty times or so bigger, of designing and constructing the VLT and the Paranal Observatory. The entire process of generating the engineering specifications, the contractual conditions and the financial arrangements was developed to a very professional level that withstood critical tests in very competitive circumstances. When we signed the contract for mirror blanks with Schott in September 1988, I was confident that we were up to the challenge. Of course, the troubles ahead, managerial, technical, financial and above all political, were not all antici- pated, but they were resolved as they came along. An example is the summer of 1991. From several directions con- The 15-m Swedish-ESO Submillimetre Telescope (SEST) at La Silla.

10 The VLT, even its VLTI-mode, is not the end of ESO’s journey; rather their quality brightens the prospects for further ambitions that reach for the stars. A key role in ALMA is called for and is bound to unfold in the next twenty years. OWL is a dream as the VLT was twenty years ago. Twenty years from now it shall, in some rendition reminiscent of the current dream, amaze the world once more. Because ‘A vision is a dream with a deadline’. ESO was Jan Oort’s vision fifty years ago. This vision had great power and has propelled our community to a sequence of extraordinary achievements. With ESO, Europe is first to reach for ultimate frontiers. It’s what our political leaders in a recent Lisbon summit called for.

On February 6, 1990, the ESO NTT was of- ficially inaugurated.

RICCARDO GIACCONI, ESO Director General, 1993–1999

I feel privileged in having had the op- ernization of the La Silla Telescopes, Today ESO is busily proceeding in portunity to lead ESO during a period of the introduction of new managerial and the scientific exploitation of the VLT, in great innovation and expansion. scientific methodology, the expansion completing development of VLTI and is Building on thirty years of heritage, of the Education and Public Outreach cooperating on a 50/50 basis with the working together with an extremely programmes and the start of the VLT in- US and Canada on the Atacama Large competent staff and with the full sup- terferometry development. By achiev- Millimeter Array, the largest ground- port and cooperation of the ESO mem- ing success in all these areas we es- based astronomy programme yet un- ber states, we were successful in many tablished ESO as a model for optical dertaken. I am confident that ESO can endeavours. They include the construc- ground-based facilities around the lead an international cooperative effort tion of the Very Large Telescope and world and redefined the role of ESO in on the next-generation overwhelmingly the development of Paranal, the mod- European astronomy. large telescope (OWL).

CATHERINE CESARSKY, Present ESO Director General

I arrived at ESO at a very interesting In parallel, these three years have The past three years have seen the time. I had the privilege of witnessing been filled with work and meetings in emergence of ESO as a major player the first light of Melipal and Yepun, of preparation for the next large project, on the European scientific scene, in overseeing the installation of UVES, ALMA. Wide collaboration with the which role it is actively contributing to NACO, VIMOS and FLAMES at the fo- European millimetre and submillimetre the establishment of the European cus of VLT telescopes, and of celebrat- wave observatories and laboratories, Research Area advocated by Commis- ing the first fringes of VLTI, first with use of all the available expertise and sioner Busquin. The organization has siderostats and then with 8-m tele- pooling of the forces, and a well coordi- acquired two new member states, scopes. The harvest of scientific results nated sharing of tasks with our Ameri- Portugal and the United Kingdom. with the two FORS, ISAAC and UVES can colleagues, have brought about Council has unanimously endorsed a is already impressive, and the efficien- considerable progress of the project long-range plan allowing continuing the cy of the Paranal Observatory is as- during Phase 1. Now, Phase 2 is about deployment of VLT and VLTI while tounding. ISAAC and UVES both have to be launched. Negotiations with the starting the construction of ALMA on an features unequalled at any other tele- USA and Canada, Chile, Spain and equal partnership with North America. scope; with NACO, we have the best Japan are all converging on time. Several other countries are considering adaptive optics instrument ever, nearly Also, faithful to its original purpose, or negotiating adhesion to ESO, and in ready to be offered to our community, ESO is preparing the long-term future the mean time Spain is participating in while VIMOS and FLAMES are show- in ground optical/infrared astronomy, ALMA with the ESO member states. ing their promise in the current com- with the conceptual study of the OWL Contacts and exchanges with six scien- missioning activities. The VLT archive 100-m telescope. All these develop- tific European organizations and with is open and attracts more and more ments – from VLT instruments to VLTI the European Union have been users, a good omen for the Astrophys- to ALMA and in the future studies for strengthened through the creation of ical Virtual Observatory. Meanwhile, the Extremely Large Telescopes – require EIROFORUM; with ESA in particular La Silla Observatory has also been and foster an ever-growing involvement the cooperation has been greatly en- very productive and has undergone of other European groups, who are no hanced in the perspective of a tighter huge improvements, coming closer and longer just users but also full fledged coordination of space- and ground- closer to VLT standards. collaborators. based astronomical research.

11 Some Snippets of History

Richard West (ESO): grammes Committee (SPC) meant to groups and to install these at CERN in Memories of early times at ESO advise the Directorate and the Council Geneva. Thanks to splendid help from on general scientific policy matters, and CERN we soon succeeded to set up My first encounter with ESO was a to evaluate the observing proposals groups for mechanics and electronics meeting for young European as- submitted by the visiting astronomers. and for site, buildings and domes. tronomers, organized in Nijenrode The SPC held its first meeting in May These groups worked for the design Castle (north of Utrecht, The Nether- 1968 at the Bergedorf office of the ESO and construction of the 3.6-m telescope lands) in the summer of 1963. Here, Directorate, in Germany. and other projects in Europe and at La about thirty future astronomers had a The SPC proposed rules of proce- Silla. wonderful opportunity to meet some of dure which were formally adopted by In the optical field, however, CERN ESO’s famous founding fathers and – the ESO Council in July 1968: tele- was not of much help, and we had not the real aim of this event – to become scope time allocation was to be succeeded otherwise in attracting opti- acquainted with each other. I was one arranged for periods of six months; ob- cal technicians. Finally Alfred Behr and of three from Denmark, as a student at serving proposals had to be submitted I agreed to ask Ray Wilson at the Zeiss the Copenhagen University Observa- 6 months before the beginning of these Works, whether he new of any young tory at that time. With the conference periods; final allocation was done by man he could recommend to us. He programme running late, I had to speak the Directorate following the recom- replied: “No, I do not know of any tech- about my work (computer studies of mendations of the SPC. One third of nician for that job, but I can offer myself light curves of eclipsing binary stars) in the observing time was to be allocated to ESO as an optician.” A new situation the evening session, just before Prof. to the ESO staff. According to the ESO indeed. After consultation with Adriaan Marcel Minneart’s closing lecture. It numbering system of the observing se- Blaauw, we invited Ray for a dinner – in was the first such speech I had ever mesters, in which October 1, 2002 – confidence of course – at the restaurant given in English and I remember being April 1, 2003, corresponds to Period 70, Mövenpick in Geneva. It was a long- suitably nervous, but surviving. The the first observing semester (Period 1) lasting dinner, which resulted in the meeting indeed brought together many was November 1, 1968 – May 1, 1969. agreement on his appointment. of those young scientists who later be- In these early days potential appli- Shortly after taking up his duties Ray came involved in ESO and many of us cants were informed that “Observing presented plans for an Optics Group, still recall this initiation to European co- periods granted may range from sever- and according to this Francis Franza, operation with great pleasure. Thanks al weeks to several months”, a some- Maurice Le Luyer, Daniel Enard, and above all to the persistent efforts of my what unusual length for a run nowa- some others for shorter periods, were Professor in Copenhagen, Anders days ..., but were also warned that engaged. Still at the time when the 3.6- Reiz, Denmark was able to join ESO in “Defrayal of travel expenses of accom- m telescope was under construction 1967. panying wives is foreseen to a limited and installation, they started their de- I myself came to ESO at the begin- extent and that only in the case the ob- velopment of new methods for the sup- ning of 1970 as Assistant to the Director servers will have to stay in Chile for a port of big mirrors. The positive impact General, Prof. Adriaan Blaauw. I often period of at least six months.” This last this group and their work has had for travelled to La Silla to perform observa- statement reveals an interesting socio- the NNT, the VLT and for ESO in gen- tions with the various telescopes there logical fact: in the early 1970’s a visiting eral is well known to everybody in and during the following years. In Novem- astronomer was by definition a man! around the organization. ber 1970, John Graham at Tololo found a nova in the LMC. With the ESO Director in Chile, Prof. Bengt Wester- Svend Laustsen (ESO, ret.): Daniel Hofstadt (ESO): lund and another ESO astronomer How ESO got its Optics Group Renata Scotto at La Silla there, Bob Havlen, we decided to obtain slit spectra of the fading 13th mag- In 1970, at a time when ESO still had Twenty years ago Renata Scotto nitude object; as far as I recall, it was its European seat in Hamburg, I was sang Madame Butterfly at the Santiago only the second time this was done on given the task to build up technical Opera House and later on visited La an LMC nova. I spent three nights at the “Chilicass” spectrograph on the ESO 1.52-m telescope, exposing con- tinuously for 4, 5, and 7.5 hours, respectively. To do the visual guiding properly – each photon really counted! – I had to balance most of the time in total darkness, high up on a ladder at the edge of the floor platforms. It was indeed a rewarding feeling when I finally saw a usable spectrum on the small plate in the dim darkroom light at the end of the night. Ten years later, we started using CCD’s and such heroic efforts are now ancient history.

Jacques Breysacher (ESO): Early days of the OPC The history of the OPC goes back to June 1967 when the ESO Council decided to establish a Scientific Pro- Construction of the building for the 3.6-m telescope at La Silla in 1975.

12 Paranal before ...... and after construction of the VLT .

Silla. Most of us were somewhat stiff in the night assistant had been hired just by the ESO Council in 1987. The VLT our welcome in view of her Prima when I arrived, neither he nor I knew was going to become a reality; interfer- Donna reputation. An incident was to anything about the 1-metre telescope, ometry was going to evolve from a break the ice in a most unexpected and we had no common language... not bonus to a driver, and we now start to manner. Our colleague the “Dottore”, a to mention the lodging and eating “fa- see its fantastic potentialities through great opera fan, came to see the Diva cilities”!). I was allocated 8 or so nights the VLTI. and asked her to sign a music record. at the ESO 1.52-m coudé to do spec- The conclusion of the VLT Site For a moment she acted very surprised troscopy of B[e] stars. Having observed Selection Working Group (SSWG) (VLT and then signed a dedicatory with (± discovered) some interesting objects report n° 62, p. 159, Nov. 14, 1990, ed- grace and smiles. The “Dottore” had with IR excess at Las Campanas, I re- ited by Marc Sarazin) stated: “On the approached her with a María Callas quested to use the Cassegrain spectro- basis of scientific considerations, the record! Such an achievement is most graph to take low-dispersion spectra of SSWG unanimously recommends that likely to remain a world premiere. those objects... but this was refused by the Paranal area be chosen for the lo- the ESO Director for Chile: I had to do cation of ESO’s Very Large Telescope”. my “approved programme”, period. So As chairman of that SSWG I had to de- Daniel Hofstadt (ESO): I did, but in “retaliation” I decided to end fend this at the next Council meeting, La Silla vaut bien une Messe my fruitful run by observing HD 45677 and then came the truncation of a at 3 Å mm–1, which required a 3-night beautiful conical mountain in order to Newcomers at La Silla had to learn exposure. This enabled one to show accommodate the VLTI on what was, and face the peculiarities of a world and that, contrary to the sharp-single [FeII] and hopefully will remain, an excellent culture which had developed at La Silla lines, those of FeII exhibit a double site. Once in a while I shiver a bit about over the years. Ingenuousness was not structure, qualitatively explained as all the consequences of the SSWG rec- part of that culture. Newcomers would originating from a ring around the star... ommendation! be quickly baptized with nicknames re- and not from an earthquake that oc- flecting their physical or psychological curred during the second night of expo- traits. Practical jokes were not absent sure! Daniel Enard (EGO, Pisa): either and most of the beginners would I later became involved with the VLT, The early days be sent to the telescopes to attend as successively chairman of the VLT of instrumentation at ESO weird issues or support important visi- Study Group, the VLT Advisory tors who had not shown up. Probably Committee, and the Site Selection To younger people born in the age of the most striking welcome was staged Working Group. The Workshop on Megapixels and computer control, a for a young technician who enquired if ESO’s Very Large Telescope (Cargèse, narration of the (not so) old ESO times Mass was celebrated at La Silla. His May 1983, in which an ESO VLT was may sound like a medieval tale. Yet, the colleagues immediately reassured him presented for the first time to a number experience acquired in this period and invited him for his first Sunday of scientists from the ESO countries, largely contributed to the present ex- Mass, a Mass which was properly and showed full unanimity about the definite tensive ESO expertise. seriously officiated by a member of need for a 16-m (equivalent) telescope In the early 1970s, the largest tele- Team and with extensive attendance to be located on an excellent site. Five scopes built in Europe were between 1 from the staff. working groups and a VLT Advisory and 2 metres diameter. Several 3- to 4- Committee were set up after the m telescopes were being developed Cargèse meeting in order to “define re- (3.6-m, CFH, Calar Alto, AAT) all much Jean-Pierre Swings (IAP, Liège): alistic objectives” and to “assess the inspired by the 5-m Palomar telescope First experience at La Silla, and implication of the specifications (and which was still a reference model. some activities for the VLT thereby the cost!) of a VLT”. Their re- Astronomical instrumentation consisted ports were presented in Venice (2nd VLT largely of conventional spectrographs, Thirty years ago (January–February Workshop, Sept. 1986) and received with images recorded on photographic 1972) I had my first observing run on La an overwhelmingly positive echo. The plates in which sensitivity was boosted Silla, a “luxurious outfit” after 10 nights VLT proposal was then elaborated into through a complex alchemy. The fore- on Las Campanas. (On Las Campanas the “Blue Book” that was endorsed front detectors of the time were image

13 intensifiers, with images recorded on tation programme however did not go An episode will give a hint on the photographic film, and electronograph- beyond talks and minutes of meetings. mood of the time. Around the Casse- ic cameras that recorded photoelec- Faced with the prospect of the largest grain adapter and instruments arose a trons directly on fine grain emulsions. European telescope deprived of instru- fierce debate, in particular on whether These “electronic cameras”, as they mentation, a crash programme was set the astronomer should sit in the Casse- were called, resulted largely from the up by the new DG. As a “first-aid” solu- grain cage to guide the telescope pioneering work of Lallemand and were tion, a single aspheric plate corrector through an eyepiece or whether it was the most sensitive and most linear of was developed and arrived just on time at all thinkable to trust a TV camera and the time. But operation of the early for the telescope first light in 1977. perform the control from the control models also required much delicate Waiting for better instruments able to room! Although an eyepiece was in- and complex manipulation as the pho- fully exploit the capabilities of the tele- cluded, the “modern” school eventually tographic emulsion was placed inside scope, in particular the large field of won, but not before making three pro- the vacuum and a new photocathode view of the prime focus, a number of totypes of the “Cassegrain chair”, an had to be installed before each opera- beautiful pictures were recorded by improbable object somewhere between tion! Later electronic cameras – the Svend Laustsen while the telescope a middle-age torture device and a den- Spectracon and McMullan cameras – was being commissioned. This first ex- tist chair for cosmonauts. avoided these problems but it was still perience with the 3.6-m was also our More pragmatically, Martin Cullum a challenging task to extract the data first direct encounter with a subtle and meanwhile dealt with the adaptation afterwards. devastating devil: seeing degradation. of electronographic and electronic de- I joined ESO in February 1975, by As part of the initial programme there tectors for imaging (Spectracon in coincidence on the same date as Lo was a 1-degree field triplet corrector for 1978 and 40-mm McMullan camera Woltjer, the newly-appointed DG. My the prime focus that was put into oper- in 1979) and for spectroscopy (two initial position was within the optics ation in 1979, and a Boller and Chivens Image Dissector Scanners in 1978 and group led by Ray Wilson which, with the spectrograph, which had the immense 1979). arrival of Guy Ratier, Maurice LeLuyer advantage of being commercially avail- With the first set of basic instruments and Bernard Delabre grew suddenly able. A Cassegrain adapter, providing under way, there was more time to think from 2 (Ray Wilson and Francis field acquisition, guiding and calibration about more ambitious projects. Simulta- Franza) to 6 people. The ESO 3.6-me- facilities, already under development in neously, the designs of the CES (Cou- tre was under test in Europe and well 1975, was completed and installed in dé Echelle Spectrometer), CASPEC on its way to completion. Its instrumen- 1977. (Cassegrain Echelle spectrometer) and, somewhat later, IRSPEC (IR spectrometer) were initiated, while the CAT (1.4-m Coudé Auxiliary Telescope) was being built. This first generation of modern instruments was put into operation in the early 1980s together with the first solid-state detectors (Reticons and CCDs). The realization of this instrumentation programme within a few years by the small and somewhat novice group in Geneva was not only a great achievement but also a defining experience for many of us. Working on the Boller and Chivens spectrograph, I realized the significant light losses occurring within instruments, in particular in the popular solid-Schmidt cameras, due to vignetting, the large central ob- struction and mirror reflections. High efficiency coatings were already available, yet relatively little used in astronomy because of their limited spectral bandwidth. A rather obvious idea was to split the spectrum into blue and red channels within the same instrument so that high efficiency coatings could be used. This not only provided an important throughput gain but also allowed the use of many more optical surfaces without significant losses, hence opening the path to more complex optical solutions. In particular, with the new flu- orine glasses then available, it became possible to design high-quality and efficient transmission optics instead of mirror combinations. This allowed instruments to be designed that could satisfy both imaging and spectrographic requirements by simply removing or ex- changing the dispersive element. On The ESO Council at Ansaldo, with the mechanical structure of one of the VLT 8.2-m tele- the logistic side, one of the many les- scopes. sons learned by the 3.6-m experience

14 The ESO Council at Paranal Observatory, December 1996. was that frequent change-over of in- 300 × 500 pixels and a read-out noise Infrared Astronomer to advise the struments and of telescope configura- of some 80 electrons (plus a lot of fring- Director General on the development of tions (Prime, Cassegrain, IR second- ing). Today, when megapixel image for- infrared instrumentation. I actually only ary, etc.) was a major contributor to tel- mats and quasi photon-counting per- became aware of the advertisement via escope down time. formance are routine, it is difficult to ap- a letter from Franco Pacini, then Head From all these considerations, the preciate just how significant an ad- of the ESO Scientific Division, with a re- idea progressively emerged of a high- vance these early electronic detectors quest that I let him know of any suitable productivity telescope having a single represented. candidates. The surprising end result, configuration and several focal stations The completion of the 3.6-m telescope despite having felt protected by my equipped with fixed multimode instru- and the development of the first modern non-member state nationality, was that ments. This idea inspired first the instruments has been an extraordinary I found myself leaving ESA to take up NTT and became fully mature with learning period and contributed to the duty at ESO in Geneva on October 1st, the VLT, which was conceived around creation of a core team of instrument 1978! this concept. As a forerunner, the multi- builders fully familiar with the problems As it happens, I was fortunate to mode instrument EFOSC was devel- of astronomical observation as well as have had been preceded by Piero oped in 1982 and put into operation in with the latest technical advances. Salinari, who had worked with me to 1983 with great success. The multi- Capitalizing on the progresses in de- build a balloon-borne IR spectrometer mode concept was then fully devel- tectors, optics and computer control at ESTEC but had then been hijacked oped with EMMI, then used in several technologies, several highly advanced to Geneva by Franco on his way back VLT and other large-telescope instru- and successful instruments and tele- to Italy. As I was to do later, Piero had ments. Another conceptual idea which scopes were built in the early 80’s that already discovered that ‘advise on in- directly emerged from the 3.6-m expe- moved ESO to the forefront of astro- frared instrumentation’ could be loosely rience was the use of natural ventilation nomical instrumentation. The interna- translated as ‘build infrared instrumen- to eliminate dome seeing, a concept tional recognition of this competence, tation’. He had thus already comman- fully validated with the NTT and the and the confidence this generated, con- deered a somewhat dilapidated con- VLT. tributed greatly to the enthusiastic en- tainer on wheels, reminiscent of a gyp- This quick glance at the past would dorsement of the VLT programme in sy caravan but converted into an au- not be complete without mentioning the 1987. It belongs now to the new genera- thentic looking infrared laboratory by in- gigantic progress made in detectors in tion of instrumental developers to main- stalling the golden looking cryostats about two decades. Up to the late sev- tain and further develop this capital. and pumps associated with infrared as- enties, image recording was still essen- tronomers in those days. (His later at- tially done with photographic plates, tempt to improve the container by paint- and solid-state arrays were very much Alan Moorwood (ESO): ing it was less successful, at least the laboratory curiosities. The first solid- The early days of infrared idea of drying it by leaving a powerful state detector at ESO was installed on instrumentation at ESO heater on all night which considerably the CES in 1981; this was a then state- changed its shape). Despite that, the first of-the-art Reticon array with a read-out ESO’s commitment to infrared as- ESO-developed infrared photometer noise of 1000 electrons! Our first CCD tronomy was expanded in 1977 by the system was finished and installed at the put into operation in 1982 had about creation of a new staff position for an 3.6-m on La Silla in 1979 (and tested

15 with software written by Daniel Hof- least for most ESO stadt). staff at the time) In parallel, we had been developing when ESO ac- the idea of building a cryogenic infrared quired and intro- array spectrometer for the 3.6-m tele- duced its very first scope (IRSPEC, later transferred to the digital minicomput- NTT) which was subsequently enthusi- er, a Hewlett Pack- astically approved by Lo Woltjer and ard HP-2114B sys- the STC. Unfortunately, this did not win tem. This “work- me many friends amongst the majority horse” computer of ESO astronomers who were mem- had a core memory bers of a committee still deliberating on of 16 kbytes (inter- the choice of the next visible spectro- esting to compare graph! Being a relatively major under- with today’s com- taking I was also subjected to more puters!). management control, starting with a In order to close summons to appear before Lo Woltjer, a technological gap, Ray Wilson and Wolfgang Richter to ESO committed it- outline the resources I would need. For self to employ lead- a young man on a short-term contract ing-edge technolo- this was a somewhat awe-inspiring gy for acquisition, event but one which I believed to have process control, and mastered with bravado by replying that reduction of astro- I wished first to absorb their wisdom as nomical data. This to how best to develop such an instru- first computer sys- ment at ESO. The answer of ‘ if only we tem was selected knew’ was unexpected but at least an to serve as the cen- honest admission that these were still tral control for the pioneering days in the adventure of in- “Grant Machine”, strument (as opposed to telescope) an automated pho- building at ESO. I therefore decided to tographic-plate mea- concentrate first on the problem of find- suring and scan- The NTT at La Silla. ing a larger caravan which was solved ning facility for stel- surprisingly quickly – albeit with the ad- lar-line radial-velo- ditional work involved in transporting city determinations, and for microden- collaboration with a few of the leading our golden cryostats and pumps from sitometry recordings of stellar spectro- staff astronomers (Dr. J. Rickard, Dr. A. Geneva to Munich. grams. Ardeberg, and others) from the ESO Before this so-called automated Santiago Vitacura office. At the end a mode of operation was feasible, a sig- reliable and successful product was Walter Nees (ESO): nificant number of technical modifica- produced. The “ESO Grant machine” ESO’s first step into the world tions and extensions became neces- became for many years a well-known of minicomputers sary to the original Grant Machine, ini- tool in astronomical data reduction and tially conceived for manual operation: was used extensively by many ESO In today’s world of automation, com- the incorporation of an analogue to dig- and visiting astronomers. Eventually it puterization, data-processing, etc., it is ital data-acquisition system, the attach- was transferred to ESO’s Headquarters rather difficult to imagine how it all start- ment of precision rotary digital en- in Garching were it served until its re- ed. The story goes back to early 1970, coders for Grant table X and Y position tirement some years ago. nearly 33 years ago. I had just joined decoding, and the integration of the ESO in the Hamburg-Bergedorf office computer with all peripherals and I/O- when I became witness to a major ESO interfaces. The main tasks of the mini- Ray Wilson (ESO, ret.): event, synonymous to setting the cor- computer were automatic scan control First Astronomical Light nerstone of automation technology at of the table, table position recording, as at the NTT ESO. Unknown to most people at ESO well as digitization and recording of the today, it was the exciting moment (at density or intensity data from the spec- The night beginning on 23rd March tral photographic plate. The required 1989 was the culmination of my career electrical and electronics hardware at ESO and indeed of my work on tele- adaptation on overall system controls scope optics, which started as an ama- had been contracted by ESO Bergedorf teur when I was six and continued pro- to a specialist electronics company in fessionally at Zeiss in 1963. Stockholm, Sweden. Intensive work by many colleagues in The initial installation of the Grant Garching and La Silla had preceded machine and its dedicated computer this great night of first light at the NTT: system at ESO Headquarters in San- above all I would mention Francis tiago was in July 1970. In spite of the Franza, Paul Giordano and Lothar positive acceptance tests in Stockholm, Noethe on the optics and Krister significant technical work was neces- Wirenstrand on the pointing. The active sary until all problems had been re- optics was working only in open loop, solved. The data-acquisition and con- as we had “borrowed” its CCD to record trol software (all written in awkward the test object I had chosen, the globu- Assembler and Fortran code) was de- lar cluster ω Centauri. The night was signed and implemented by ESO’s perfect, a light laminar wind giving ex- chief programmer, Mr. Frank Middel- cellent ventilation and seeing. The re- The ESO Grant machine. burg (deceased November 1985), in sults started to come in and were eval-

16 uated by the astronomers. The best one was evaluated by Jorge Melnick, but he checked it a second time because he couldn’t believe the result, but then confirmed it: FWHM = 0.33 arcsec. Jubilation and amazement in La Silla, also in Garching as expressed by Richard West. A journalist was also present with us: he absolutely wanted to record that this result had occurred on my birthday (23rd March), but it actually occurred about 02.00 hours on 24th March. I didn’t mind this at all, but the journalist did! This best frame of our test night was shown in a beautiful comparison set-up by Richard West, with blown-up sections of photos from the ESO 1-m Schmidt and 3.6-m telescopes, in the next Messenger and is reproduced in my RTO II (p. 293). The foundations of the incredibly successful active optics system of the VLT, based on identical principles, had been laid.

Piero Benvenuti (ST-ECF): Recovery of a historical The first of the VLT 8.2-m telescopes (Antu) saw “First Light” in 1998, the last of the giant tel- document escopes (Yepun) on September 3, 2000. While clearing his office of over a decade of accumulated papers recent- light marked the successful conclusion Andreas Glindemann (ESO) ly, in preparation for an extended stay of the important period which started et al.: at the ST ScI in Baltimore, Richard with the approval of the VLT project by First Fringes with ANTU and Hook knocked on my door and, smiling, the ESO Council in December 1987. Ex- MELIPAL ceptionally for such a complex and ex- handed over a paper with an handwrit- (from The Messenger No. 106, Dec. 2001) ten note on the front page: “An excel- pensive project, the four VLT telescopes lent idea! Sorry it took 13 years for me came into operation ahead of schedule. On October 30, 2001 at about 1 a.m., to reply! Richard”. The VLT was no longer only a project, the two 8-m Unit Telescopes ANTU and The “historical” document was enti- it was now also an Observatory. MELIPAL of Paranal Observatory were tled “A proposal for the astrophysical By virtue of becoming ESO’s Director combined for the first time as a stellar classification of HST targets” and was General at the right time, I had the priv- interferometer observing fringes on the drafted by me in March 1989, in a final ilege of actually being in the observing star Achernar, only seven months and attempt to convince the HST Project to hut of Yepun at the crucial moment, twelve days after the VLTI produced the implement a classification scheme of sharing the excitement of the VLT first fringes with two siderostats. This the observed targets that would facili- Manager, Massimo Tarenghi, of the was the first time that the VLTI was op- tate the browsing through the HST Director of Paranal Observatory, erated as a truly Very Large Telescope Archive. At the time the proposal was Roberto Gilmozzi, and of the members Interferometer. received with interest, but was never of the commissioning team, Jason The night started with tests of the implemented. Spyromilio, Krister Wirenstrand and Coudé Optical Trains and the Relay Perusing the paper today, it still Rodrigo Amestica. It was a cold night, Optics, converting the light from the makes a lot of sense, although one appropriate to the late Chilean winter, Coudé focus to a parallel beam in the would implement its concept differently. and we could hear the wind howling Delay Line Tunnel. Around midnight, Indeed its scientific goal would be bet- outside. We had chosen our first light when the UT team finished the tests ter achieved today as a functionality of target in advance: the planetary nebula and the search for fringes could start, the Virtual Observatory environment, He 2-428. In a few minutes, the guide not everybody on the mountain would correlating data from more than a sin- star was acquired, the position and have bet how quickly the search was gle instrument together with direct link shape of the mirrors were actively cor- successful. to the existing literature. Nonetheless, it rected, and we could see on the com- Barely one hour after we had started, shows some kind of coherence (stub- puter screen the unmistakable shape of the automatic fringe search routine in born-mindedness?) in the ECF! the source, with an image quality limit- VINCI reported ‘flecos en el cielo’, and ed only by the atmospheric seeing (0.9 the fringes appeared on the screen. We arcsec at the time). The rest of the found that the baseline of 102.5 m be- Catherine Cesarsky (ESO): evening was spent in the VLT Control tween ANTU and MELIPAL differed by First Light of UT4 room in the appropriate celebratory only 28 mm from their nominal length. (from The Messenger No. 101, Sept. 2000) manner, taking more images, attending After refinement, fringes were subse- to the PR requirements, and drinking quently found within 0.4 mm of their cal- At 21:44 hours on the night of champagne with the teams observing culated position. September 3, 2000, the test camera at on the other telescopes. With the experience that we had the Cassegrain focus was opened for Everyone present felt the sense of gathered over the last six months of 30 seconds, and the fourth VLT Unit accomplishment, triumph and elation commissioning, ‘routine operation’ with Telescope, Yepun, saw First Light. A that always accompanies the culmina- the 8-m telescopes started almost im- historic event in the life of ESO; this first tion of a great human adventure. mediately.

17 TELESCOPES AND INSTRUMENTATION APEX – The Atacama Pathfinder Experiment L.-Å. NYMAN1,2, P. SCHILKE 3 and R.S. BOOTH 2

1ESO/SEST, La Silla, Chile 2Onsala Space Observatory, Onsala, Sweden 3Max-Planck-Institut für Radioastronomie, Bonn, Germany

1. Introduction observing time will be dedicated to 2.1 Exploring the star-formation Chilean astronomy. The antenna is be- history of the Universe APEX is a collaboration between the ing purchased by MPIfR, OSO and Max-Planck-Institut für Radioastrono- MPIfR will provide instrumentation and Among the fundamental cosmologi- mie (MPIfR) in Bonn (together with ESO operations. cal questions being asked today are: Astronomisches Institut Ruhr-Universi- when did galaxies and massive black tät Bochum, AIRUB), ESO and Onsala 2. Science holes form in the early universe, and Space Observatory in Sweden (OSO). how did they subsequently evolve? The idea is to construct and operate a APEX will be able to make significant Modern telescopes are now detecting 12-m diameter submillimetre telescope contributions to the solution of a num- galaxies out to redshifts beyond 6, on the ALMA site of Llano de Chaj- ber of current astronomical problems close to the “dark ages” where the first nantor in Chile at an altitude of 5000 m. that cannot be, or are insufficiently ad- stars and galaxies may have formed. APEX will operate at submillimetre dressed with currently available tele- Because much of the stellar light wavelengths as well as in the far in- scopes: constraining cosmological emerging from massive star-formation frared (at THz frequencies), which is models, studying star formation in the regions is immediately absorbed by the possible because of the excellent at- early and local universe, stellar evolu- surrounding dusty clouds, even the mospheric transparency that exists on tion, interstellar chemistry at high fre- most luminous starburst galaxies are the site at these wavelengths; it might quencies, and the exploration of the difficult to observe at optical and even be the best site in the world for sub-mil- southern submillimetre sky. At submil- NIR wavelengths. The absorbed radia- limetre astronomy. limetre wavelengths APEX will have a tion is re-emitted by the dust as long- APEX will explore the southern sky, better spatial resolution than space or wavelength infrared radiation which which is virtually unexplored at submil- balloon borne instruments, by virtue of can easily escape the star forming re- limetre wavelengths, and also serve as the larger dish size. Additionally, it will gions – but cannot cross the Earth’s at- a pathfinder for ALMA, both by per- serve as a pathfinder for ALMA in all of mosphere. However, for very distant forming wide-field surveys for later fol- its wavelength ranges. Surveys with objects this radiation is red-shifted to low-up by ALMA, and by obtaining ex- APEX will be an outstandingly efficient submillimetre wavelengths. This makes perience in operations of telescopes at means of finding target sources for it accessible from the ground, at a very the site. ALMA, and for their line and continuum few places such as Chajnantor. The The project is shared between the exploration prior to their detailed inter- large 870-micron bolometer array partners in the ratio 50% MPIfR/AIRUB, ferometric study with greater spatial (LABOCA, see below) at APEX will be 27% ESO and 23% OSO. 10% of the resolution. ideally suited to detect and map the distribution of the earliest, most distant star-forming galaxies in the Universe. Follow-up observations at 350 micron will provide data on their distance and nature. The unprecedented size of its bolometer arrays and the ideal observing conditions all year round will make APEX the most powerful ground-based instrument to explore the star formation history of the Universe.

2.2 Constraining the Universe: the Sunyaev-Zel’dovich effect

Galaxy clusters are the largest col- lapsed structures in the Universe. Measuring their distribution and structure provides crucial information on the history and structure of our Universe. Galaxy clusters are embedded in vast amounts of hot, ionized gas. This gas scatters the passing photons of the Cosmic Microwave Background (CMB) and increases their average energy. The resulting distortion in the CMB is called the Sunyaev-Zel’dovich (SZ) Effect and can be used as a sensitive Figure 1: An artist’s impression of the APEX antenna. probe of cosmological models and clus-

18 ter physics. Planned 2-mm bolometer arrays at APEX will have an ideal spatial resolution and sensitivity to measure the SZ effect toward distant clusters.

2.3 Unbiased searches for protostars

Another important scientific objective APEX will pursue is a search for protostars in heavily obscured star-formation regions in our Galaxy. Understanding the very earliest stages of star-formation ranks as one of the most important questions in astrophysics. Stars and their surrounding planetary systems form from dense condensa- tions within molecular clouds. Before and during their collapse, these dense gas cores, or protostars, remain very cold (10–30 K), and therefore escape detection with infrared instruments such as ISO, IRAS and MSX. APEX on the other hand will detect these objects in the submillimetre continuum and in molecular lines to study the kinematics of the collapsing objects, deepening Figure 2: The APEX antenna seen from the back. Note the Cassegrain focus cabin and the our understanding of the sources dis- two Nasmyth focus cabins. The container below and behind the focus cabins will contain the covered. spectrometers and other electronics.

2.4 Submillimetre spectroscopy of the Milky Way and explored, especially in the southern interstellar clouds, protostars, the cir- external galaxies hemisphere, but the spectral windows cumstellar envelopes of evolved stars, in this range contain low-lying transi- and comets. Important lines are those The frequency bands between 600 tions of many molecules that are of the light hydrides, of particular inter- GHz and 1.5 THz are relatively poorly known, or expected to be abundant in est in astrochemistry, and some fine

Figure 3: The atmospheric transmission curve for Chajnantor with different amounts of precipitable water vapour (PWV). The THz windows open at PWV levels below 0.5 mm.

19 cision performance even with wind speeds up to 9 m/s, and the pointing accuracy is specified to be better than 2 arcsec (absolute). The main modifications to the original ALMA antenna design are the incorporation of Nas- myth focus cabins and a chopping secondary mirror. These modifications are required for single-dish operations of array receivers and bolometers. The antenna will have in total three focus cabins, one at the Cassegrain focus and two at the Nasmyth foci.

4. Instruments

APEX instrumentation will include both wide-band bolometer array receivers for continuum observations and heterodyne receivers for spectral line observations. Some of the instruments will be specifically designed and cus- tom-built for APEX. Instruments in use at other sites may be transferred to APEX, where they are expected to pro- Figure 4: An example of a bolometer array: MAMBO2, the 117-channel bolometer array built vide better data than at their current by the Max-Planck-Institut für Radio-Astronomie for the IRAM 30-m telescope. home. APEX will initially operate with a 300- element bolometer array at 870 mi- structure atomic lines like the CI lines at 3. Telescope crons, the ideal wavelength to search 809 GHz and 492 GHz as well as the for high-redshift dust emission. It is excited nitrogen line [NII] at 1.46 THz, The APEX antenna, built by VERTEX called LABOCA (LArge BOlometer which is very common in the ISM. The Antennentechnik in Germany, is a mod- CAmera) and is being built through a excitation requirements of most atomic ified copy of the ALMA-US prototype Bonn/ Bochum/Jena collaboration. and molecular transitions at THz fre- antenna. It has a diameter of 12 me- Additionally, a 37-element array at 350 quencies select the densest gas near- tres, and the reflector surface will be set microns will be constructed to deter- est to a young stellar object. As a result to an accuracy of 18 micrometer or bet- mine the spectral index of the radiation it is expected that the most intense ra- ter in order to observe beyond 1 THz. and to study sources with higher angu- diation will be concentrated in regions The telescope is designed to give pre- lar resolution over smaller fields. with angular scales of a few arcseconds, corresponding to the beam size of APEX at these frequencies. The luminous star bursts in interacting galaxies also produce intense emission at THz frequencies, also on angular scales of a few arcsec in the nearest regions. Thus, the highest (THz) observing bands which may be reached through the combination of the superior Chajnantor site and the excellent performance of the APEX antenna are ideally suited to the study of chemical evolution, energetics and dynamics of star- forming regions.

2.5. Objects of special interest

APEX will be able to completely map unique objects at submillimetre wavelengths. Some of the most interesting sources in the sky can best (or only) be studied from the southern hemisphere. These include four out of five of the nearest sites of low-mass star formation (within about 150 pc), the Galactic centre (an important prerequisite study for the future understanding of the central regions of other galaxies), the Magellanic Clouds (the nearest galaxies to our own and prototypes of metal- poor galaxies in an earlier stage of evo- Figure 5: Centaurus A, the most nearby active galaxy, observed with the 37-channel bolome- lution), and Centaurus A (the nearest ter array SIMBA (SEST IMaging Bolometer Array) at SEST. Note the emission from the dust galaxy with an active nucleus). lanes as well as the curved jets perpendicular to the dust lanes.

20 Heterodyne instruments will play an carried out since 1995, showing that 7. Time scales important role for observations from the excellent atmospheric conditions on Chajnantor: APEX will be equipped with Cerro Chajnantor will allow observa- The antenna will be erected on the receivers covering all atmospheric win- tions in all submillimetre windows close site in April 2003 by VERTEX Anten- dows from 200 GHz to 1 THz. In addi- to 50% of the time. nentechnik. At this time receivers oper- tion, several experimental receivers ating at 90 GHz will be installed in order covering selected windows above 1 6. Infrastructure and operations to do holography and to set the surface THz – uniquely observable from to 18 microns rms. First-light receivers Chajnantor – will be provided. APEX APEX will be operated as part of will be installed soon after this, consist- will be equipped with autocorrelation the La Silla Observatory. The staff ing of the SEST 1.3-mm receiver and spectrometers. of 18 will include astronomers, opera- perhaps also a single pixel bolometer. tors and engineers/technicians. There The first heterodyne receivers are ex- 5. Site will be a base in San Pedro de Ata- pected to arrive at the end of 2003, and cama (the nearest village at an alti- LABOCA, the 300 pixel bolometer ar- The greatest problem for ground- tude of 2500 m), which will consist ray, in the beginning of 2004. APEX op- based submillimetre astronomy is the of offices, laboratories, control room, erations are expected to start in the be- absorption of incoming radiation by at- cafeteria and dormitories, and the staff ginning of 2004. mospheric lines, mainly by water will sleep at the base. On the high vapour. This is why the submillimetre site, APEX will be operated and main- 8. SEST and APEX region of the spectrum is still relatively tained from a set of oxygenized and unexplored. Ground-based submillime- heated containers. Diesel generators ESO and OSO are presently operat- tre astronomy can only be done from will provide power, both at the base ing SEST on La Silla. In order to pro- sites with extremely dry atmospheres, and at the high site. There will be a vide operational funds for APEX, SEST such as high mountain tops and in high-speed microwave link between the operations are expected to stop at the Antarctica. San Pedro base and the telescope, al- end of June 2003 and SEST will be Llano de Chajnantor is most likely lowing APEX to be operated remotely closed. There is however a possibility the best place for submillimetre astron- from San Pedro in service mode and that SEST may continue to be used af- omy on Earth (possibly rivalled only by with flexible scheduling. There may ter June 2003, by dedicated groups do- the far more inaccessible sites in also be a visitor mode with observa- ing survey work. Antarctica), because of its high altitude tions being done remotely from San More information on APEX can be at 5000 m and also because of its loca- Pedro. Part of the observing time will be found at: http://www.mpifr-bonn.mpg.de/ tion in the dry Chilean Atacama desert. dedicated to more experimental obser- div/mm/apex.html and Long-range monitoring to characterize vations with PI instruments at THz fre- http://www.oso.chalmers.se/oso/apex/ the site for the ALMA project has been quencies. index.html

VIMOS Commissioning on VLT-Melipal O. LE FÈVRE1, D. MANCINI 2, M. SAÏSSE1, S. BRAU-NOGUÉ 3, O. CAPUTI 2, L. CASTINEL1, S. D’ODORICO4, B. GARILLI 5, M. KISSLER4, C. LUCUIX3, G. MANCINI 2, A. PAUGET1, G. SCIARRETTA2, M. SCODEGGIO 5, L. TRESSE1, D. MACCAGNI 5, J.-P. PICAT3, G. VETTOLANI 6

1Laboratoire d’Astrophysique de Marseille, France; 2Osservatorio Astronomico di Capodimonte, Naples, Italy; 3Observatoire Midi-Pyrénées, Tarbes, France; 4European Southern Observatory, Garching, Germany; 5Istituto di Fisica Cosmica e Tecnologie Relative, Milan, Italy; 6 Istituto di Radio Astronomia, Bologna, Italy

Introduction graphs on 4-m-class telescopes have much more than the few thousand been very powerful tools to quantify the galaxies measured today, all surveys In the mid-80s, multi-object spec- evolution of galaxies over more than included. The need to study the distri- troscopy (MOS) appeared as a new half of the age of the universe, up to bution of galaxies in the local universe and powerful technique to perform the redshifts ~1 [2][3]. This because the has prompted two major science and spectroscopy of many objects simulta- density of galaxies to I ~ 22 (reaching instrumentation programmes: the neously. The idea is simple: instead of redshifts ~1 or about half the current Sloan Digital Sky Survey (SDSS), and using a single slit as the input to a spec- age of the universe) projected on the the 2dF Galaxy Redshift Survey. Both trograph, masks are manufactured with sky is high enough that very efficient are acquiring several hundred thou- slits positioned facing the images of tar- spectrographs with high-quality CCDs sands of galaxy spectra with dedicated gets of interest in the entrance focal [4] can efficiently assemble samples of MOS facilities [7][8]. Similarly, the need plane of the spectrograph. The techni- several hundreds of measured spectra to acquire large numbers of spectra/ cal implementation turned out to be and redshifts. The technique was then redshifts over a redshift range 0–5 cov- more tricky, but the first successful ex- applied on the first 10-m Keck with the ering 90% of the current age of the uni- periments were conducted with punch- LRIS spectrograph [5] and produced verse, has been identified. This is re- ing machines, in particular at ESO and most of the Lyman-break galaxies at quired by the necessity to cover sever- CFHT with the PUMA concept [1]. redshifts 3–4 known today [6]. al time/redshift steps, study the evolu- MOS was then quickly identified as However, the study of galaxy evolution of various classes of galaxies in a the tool of choice to conduct deep tion and of their space distribution over wide range of environments, ranging galaxy surveys. Multi-object spectro- most of the age of the universe requires from the low density of voids to very

21 Figure 1: VIMOS field as full detector is used to record spectra. projected on the sky, The slit sampling is set to allow Nyquist each quadrant has a sampling for a 0.5 arcsec slit, with a 2 field 7 × 8 arcmin , for a plate scale of 0.205 arcsec/pix. In addi- total field of 224 arcmin2. tion, the Integral Field Unit (IFU) covers a field 54 × 54 arcesc2, with 6400 resolution elements 0.67 × 0.67 arcsec2, each leading to a spectrum. In all, it is really 4 instruments in one, with a total field of view of 224 arcmin2, each channel being the equivalent of a complete FORS instrument. For each channel, a mask exchange unit (MEU), a filter exchange unit (FEU), and a grism exchange unit (GEU) permits configuration of the instrument in the imaging or MOS modes. Furthermore, special masks can be positioned at the entrance focal plane to configure the instrument in IFU mode. To produce the masks placed at the VIMOS focal plane, a dedicated mask dense cluster cores. As an example, ing the first commissioning periods and manufacturing unit (MMU) is available the measurement of the evolution of the present the general performance of to cut masks with slits at any location, luminosity function of galaxies or of the VIMOS. This article is also intended to with any size and shape. It is fully de- star-formation rate requires 50 galaxies prepare the community to the arrival of scribed elsewhere [11]. The powerful per measured magnitude bin, over 10 this powerful facility. laser machine is capable of cutting magnitudes, for three basic types ~200 typical slits 1 × 12 arcsec each in (colours) of galaxies, in three types of VIMOS concept less than 15 min. The MMU is also environments. Adding the necessity to used to cut masks for the FORS2 MXU probe several fields (i.e. 4) to minimize VIMOS was designed from the outset mode. the impact of cosmic variance, and 7 to maximize the number of spectra ob- time steps leads to a total galaxy sam- served with spectral resolutions R = VIMOS observing modes ple of 50 × 10 × 3 × 3 × 4 × 7 = 126,000 200–2500 (1 arcsec slits) [10]. The galaxies. Very efficient MOS instru- 4-channel concept allows one to maxi- VIMOS has three main observing ments are therefore needed. mize the multiplex gain: the field of view modes: direct imaging, multi-object In 1994, ESO convened a workshop of each channel is 7 × 8 arcmin2 in both spectroscopy with multi-slit masks, and to canvass the community in defining imaging and MOS, projected on the integral field spectroscopy. The main the full instrument complement for all central 2048 × 2350 pixels of a 2048 × characteristics of these modes are list- unit telescopes of the VLT. A wide-field 4096 pixels thin EEV CCD, while the ed in Table 1. multi-object spectrograph appeared as the most important missing instrument Table 1: VIMOS observing modes in a poll of the community present at the meeting. Our team presented the base- Imaging mode line specifications and a tentative concept [9], the result of discussions 2 across the community, in particular in- Field of view 4 × 7 × 8 arcmin cluding the WFIS concept developed at Wavelength range 0.37–1 micron ESO. A feasibility study was then com- Filters U′ BVRIz missioned by ESO to our consortium of Spatial sampling 0.205 arcsec/pixel French and Italian institutes, and conducted over 9 months in 1995–1996. Multi-Object Spectroscopy mode ESO then issued a call for proposals to build a facility instrument, based on a Field of view 4 × 7 × 8 arcmin2 wide-field MOS. The proposal present- Spatial sampling 0.205 arcsec/pixel ed by our consortium was selected by Low resolution R ~ 200 (1 arcsec slit) Grisms: LRBlue the ESO-STC in October 1996. A contract between ESO and the Centre Number of slits ~1000 of length ~ 8 arcsec LROra National de la Recherche Scientifique LRRed of France represented by the then Medium resolution R ~1000 (1 arcsec slit) Grisms: MR Laboratoire d’Astronomie Spatiale in Number of slits ~ 400 of length ~ 8 arcsec Marseille (now Laboratoire d’Astro- High resolution (1 arcsec slit) Grisms: HRBlue physique de Marseille) was signed in Number of slits ~ 200 of length ~ 8 arcsec HROra July 1997, to construct VIMOS, the HRRed Visible Multi-Object Spectrograph, NIR- MOS, the Near-IR Multi-Object Spec- Integral Field Spectroscopy mode trograph, and the MMU, the Mask Manufacturing Machine. 2 After the successful completion of Field of view 54 × 54 arcsec the Preliminary Acceptance Europe, Wavelength range 0.37–1 micron VIMOS was shipped and reassembled Spatial sampling 0.67 arcsec / fiber in Paranal. We describe here the re- Number of resolution elements / spectra 6400 sults of the main tests carried out dur- Spectral resolution R ~ 200–2500

22 Figure 2: Installation of VIMOS on the VLT-UT3. From upper left, clockwise: (a) transportation of VIMOS from the Paranal Observatory integration facility to the telescope, (b) VIMOS being hoisted inside the dome of Melipal to reach the Nasmyth platform, (c) installation on the Nasmyth rotator, (d) VIMOS after cabling and co-rotator installation.

VIMOS integration and tests ment to relieve the adapter from the ex- completing the work on the support leg tra weight as seen in Figure 3. and on flexures adjustment, VIMOS will After completing integration and test- In a first commissioning run in have its third and last commissioning ing at the European integration facility February 2002, the first 2 channels on the sky in September 2002. at Observatoire de Haute-Provence, were extensively tested on the sky. France, VIMOS was completely disas- While the internal image quality was VIMOS performance sembled and shipped in more than 50 measured to conform to specifications crates (a total of 15 tons) at the end of during integration, images on the sky Image quality 2001. The reassembly took place in the have demonstrated the excellent over- integration facility of the Paranal Ob- all image quality of the combined tele- The image quality of the optical train servatory in January-February 2002. scope + instrument. Images as good as was measured for each channel. A grid Optical alignment was checked, all me- 0.4 arcsec FWHM have been recorded. with pinholes 100 microns in diameter chanical motions were tuned and veri- The complex sequence necessary to was produced with the Mask Man- fied over hundreds of cycles, and all place slits at the focal plane in coinci- ufacturing Unit and placed at the en- software components were implement- dence with selected targets, involving a trance focal plane. The optical aligned prior to the installation at the tele- transformation matrix from sky to mask ment was perfected by means of a rel- scope. The instrument was moved to focal plane to detector, has been tested ative X,Y adjustment of the last element Melipal on February 23rd (Figure 2). and validated. of the optical train coupled to the de- VIMOS is now attached to the Nasmyth In a second technical commissioning tector assembly. All channels are fully focus B of the “Melipal” – UT3 tele- in May 2002, the support structure in specification, with 90% of the field scope of the ESO Very Large Tele- to compensate the extra weight was with images better than 0.5 arcsec at scope (Figure 2). installed, and the 4 channels complete- 80% encircled energy as shown in The weight of the instrument turned ly integrated. Due to bad weather, Figure 4. out to be significantly larger than fore- many calibration tests were obtained seen in the original design. At a total on the complete 4-channel configura- Flexure weight of 4 tons, VIMOS is about 1 ton tion but no sky observations could be overweight with respect to the Nasmyth obtained. Image and spectral quality Flexure control for a 4-ton instrument rotator-adapter specification. It was have been confirmed to be within speci- has been a concern from the start of the necessary to implement a support fications. project. The main VIMOS structure was structure at the back end of the instru- After technical activities in July 2002, designed to minimize flexure defined as

23 Figure 3: The fully integrated instrument on the Nasmyth focus, including its dedicated support structure to the right.

motion on the CCDs of a light spot produced at the entrance focal plane. A mechanical support system was implemented at the back of the folding mirrors on the optical train to allow for pas- sive flexure compensation by means of astatic levers. This support system can also be upgraded to an active support using piezo-actuators to apply motion on the folding mirrors to compensate for flexure.

Figure 4: image quality as a function of radius from the field centre, measured for channels 1 to 4 (top to bottom) on a grid of pinholes 300 microns in diameter. The image quality is better than the specification identified by the dashed line.

Uncompensated flexure is measured to be on order ± 2.5 pixels in both X (along slit) and Y (dispersion) for a full 360° rotation of the instrument. Astatic levers have been installed and are being adjusted at the time of this writing. They are expected to cut the flexure by a factor 3, as based on previous measurements taken during integration at Haute-Provence Observatory. This is Figure 5: Flexure shown on the current corrected flexure measurement on behaviour for channel 2, showing flex- channel 2. The ure contained within a one-pixel radius points represent the (Figure 5). Optimization of the other position of a refer- channels is under way. ence spot of light on the detector, as a VlMOS first light function of the rotation angle of the First light was achieved on February Nasmyth rotator over 360°. The 26, 2002, with VIMOS in 2-channel dashed circle repre- mode. Images with excellent image sents a 1-arcsec quality were recorded right away (see radius motion of the e.g. Figure 6 to Figure 8, and images reference spot. on http://www.astrsp-mrs.fr/virmos

24 Table 2: Photometric zero points

Filter Zero point (preliminary, average of 2 channels) Mag = zero point –2.5 log (flux) in ADU) –2.5 log (CCD gain) +2.5 log (exposure time)-kc.sec (air mass)

U 26.3 B 28.0 V 27.7 R 27.8 I 27.0

Imaging performance

The overall throughput of the instrument can be measured in terms of the photometric zero points used to trans- form the observed CCD counts to magnitudes as listed in Table 2. This is better in the UV and blue than the comparable FORS instrument on the VLT and shows equivalent sensitivity in the visual-red, when the additional reflection in the telescope is taken into account.

Multi-object spectroscopy Figure 6: Image of the “Antennae” taken during the first night of VIMOS on the sky. Composite performance of V, R, and I images, 60 seconds each. The image quality measured on stars is 0.6 arcsec Masks have been produced from im- FWHM. ages taken with the instrument. The transformation matrix from CCD to mask was computed using images from The mask design interface allows ed fashion in order to maximize the a uniformly distributed grid of pinholes. one to define a mask in an automated number of objects, and to minimize the This transformation proved to be very fashion from a “pre-image” taken with effect of overlap between orders when accurate: from the first try onward, slits VIMOS. Slits can be placed either on working with several banks of spectra and reference apertures on masks targets selected from a catalogue of ob- along the dispersion direction. have been successfully positioned di- jects detected on the pre-image, or on Performance in spectroscopy mode rectly on top of astronomical objects. objects from a user imported catalogue. is as expected, and follows the compu- Examples of masks observed are The software cross correlates the tations from the Exposure Time Cal- shown in Figure 9. This demonstrates brightest objects detected in the culator (see ESO web page referred to the great multiplex capability of VIMOS, pre-image with objects in the user cat- below). Spectra of extended IAB ≤ 22.5 with several hundred objects being ob- alogue to define a transformation ma- galaxies have been recorded in 3 × 15 served simultaneously. The details pre- trix. This allows one for instance to use min exposures, with S/N ~ 5–10 on the sented in Figure 10 show the high ac- very deep images taken with VIMOS or continuum (Figure 14). curacy of the slit profile, thanks to the another facility to place slits on objects Spectra with the integral field unit high precision of the laser-based mask not visible on the short-exposure pre- have also been obtained as shown in manufacturing unit [11]. images. Slits are placed in an automat- Figure 11.

Figure 7: image of the cluster of galaxies ACO 3341. Figure 8: Central part of the cluster Cl 1008-12. Composite of V, R, and I band exposures, 5 minutes each.

25 Figure 9: Example of multi-slit data taken with VIMOS during the first light in February 2002. In these masks on 2 channels, more than 220 objects were observed simultaneously with a spectral resolution R ~ 250.

Figure 10: Detail of a raw MOS spectra Figure 11: Spectra taken with the VIMOS-IFU. Left: 3200 spectra obtained with 2 channels in frame. One can identify the trace of the con- February 2002. Right: enlarged portion of the IFU spectra showing the emission lines from a tinuum of galaxies in each slit. The slit pro- planetary nebula observed during tests. file is extremely accurate thanks to the high precision of the MMU. Fringing from the detector is visible in the red part of the spectra (towards the top).

Data processing

Spectra have been processed using dedicated Data Reduction Software (DRS). The spectra are corrected for the detector response and the sky lines are subtracted before the 2D spectra are summed. The 1D extraction then follows, with wavelength and flux calibration. Because of the thin substrate of the EEV detectors, fringing from the Figure 12: Processing of the VIMOS MOS data. Left panel: sequence of 3 spectra taken while the object was maintained at the same position in the slit. Three sky-subtracted spectra are strong sky OH emission appears above shown from left to right (about 200 Å around 8200 Å; wavelength increases towards the top), ~ 8200 Å. This fringing can be efficient- together with the combined average of the three spectra on the right. Significant fringing resid- ly removed using a sequence of shifts uals are present. Right panel: in this set of observations, the object was moved along the slit of objects along slits, as shown in Fig- at positions –1, 0, +1 arcsec; the combined average shows that fringes can be corrected to ures 12 and 13. Spectra of extended a high level of accuracy.

26 galaxies with IAB ≤ 22.5 are shown in Figure 13: Effect of Figure 14. sky and fringe subtraction with Summary (bottom) and without (top) shifting objects The wide-field survey instrument along the slit as de- VIMOS is now being commissioned at scribed in Figure 12. the VLT. In each of the three operational modes (imaging, multi-slit spectroscopy and integral-field spectroscopy), VIMOS offers an unprecedented field of view. In multi-slit spectroscopy mode, several hundred spectra can be recorded simultaneously, while in integral-field mode, 6400 spectra are recorded in a field 54 × 54 arcsec2. It is expected that VIMOS guest observations will start in April 2003. Informa- tion needed to prepare observing proposals is available on the web pages http://www.eso.org/instruments/vimos/ index.html

References [1] Fort, B., Mellier, Y., Picat, J.P., Rio, Y., Lelièvre, G., 1986, SPIE, 627, 339. [2] Ellis, R.S., Colless, M.M., Broadhurst, et al., 1996, MNRAS, 280, 235. [3] Lilly, S.J., Le Fèvre, O., Crampton, D., Hammer, F., Tresse L., 1995, Ap.J., 455, 50. [4] Le Fèvre, O., Crampton, D., Felenbok, P., Monnet, G., 1994, A&A., 282, 325. [5] Oke, J.B. et al., 1995, PASP, 107, 3750. [6] Steidel, C.C., Giavalisco, M., Pettini, M., et al., 1996 ApJ, 462, 17. [7] Gunn, G.E., and Weinberg, D.H., 1995, in ‘‘Wide field spectroscopy and the distant universe”, ed. S. Maddox & Aragon Salamanca, World Scientific Singapore, 3. [8] Colless, M.M., et al., 1999, proc. 2nd Igrap conference “Clustering at high redshifts, June 1999, Mazure, Le Figure 14: Fèvre, Le Brun Eds, ASP series. Examples of spectra [9] Le Fèvre, O., et al., 1994, proceedings taken with VIMOS of the meeting Science with the VLT”, J. Walsh, I. Danziger Eds., Springer, p. during the first com- 367. missioning in [10] Le Fèvre, O., et al., 2000, Proc. SPIE February 2002. The Vol. 4008, p. 546–557, Optical and IR redshift is indicated Telescope Instrumentation and Detec- in the upper left cor- tors, Masanori Iye; Alan F. Moorwood; ner of each panel. Eds. These spectra are [11] Conti, G., Mattaini, E., Maccagni, D., the average of 5 ex- Sant’Ambrogio, E., Bottini, D., Garilli, posures 10 minutes B., Le Fèvre, O., Saïsse, M., Voët, C., each with the LR et al., 2001, PASP, 113, 452. grism (R ~ 200).

2.2-m Team L. GERMANY

Welcome to the last (very short) in- telescopes and instruments on La Silla. moving to the 2.2-m telescope at the stallment of 2p2team news from La The folding of the 2p2team heralds end of Period 69 and we expect it to be Silla. This is mostly just a farewell mes- the end of ESO time at both the ESO up and running in its new home by sage, as in October we cease to oper- 1.52 and Danish 1.54-m telescopes. November 2002. The Danish telescope ate as a separate entity and join with The Boller and Chivens spectrograph is will continue to operate after October the old NTT and 3.6 teams under the only available in Brazilian time until the 2002, but only in Danish time. new guise of Sci-Ops. Never fear end of 2002, after which time the in- So farewell from the 2p2team and though, the next Messenger will see strument will be mothballed and the tel- we'll see you next time as Sci-Ops. this section expanded to include all the escope decomissioned. FEROS is

27 REPORTS FROM OBSERVERS Standing on the Shoulder of a Giant: ISAAC, Antu, and Star Formation M. McCAUGHREAN, H. ZINNECKER, M. ANDERSEN, G. MEEUS, and N. LODIEU

Astrophysikalisches Institut Potsdam If I have seen further, it is by standing on the shoulders of giants. In the Beginning Isaac Newton, in a letter to Robert Hooke, 1676

Today’s astronomers spend much of Fortunately, all of these demanding carrying out photometry or spec- their time staring into regions where specifications are being well satisfied troscopy with single apertures, or labo- stars are forming, whether deep out in by the new large telescopes, optimized riously mapping out extended regions extragalactic space and far back in for infrared observing, equipped with one pixel at a time. Of course, many pi- time, to watch as the first galaxies are state-of-the-art infrared array instru- oneering discoveries and advances assembled, or nearer to home, wit- ments, and situated on sites with excel- were made, but the great sea change nessing the fiery creation of new stars lent intrinsic atmospheric qualities. came in 1986, when infrared-sensitive and planetary systems within our own In this article, we hope to illustrate detector arrays made their way out Milky Way. the great qualitative and quantitative from behind the dark curtains of military Crucial to these endeavours are the strides that star-formation studies have secrecy and into open use on those new 8–10-m diameter telescopic taken in the past few years, by looking large astronomical telescopes. With leviathans, equipped with powerful at three highlights from our own work only 62 × 58 pixels, these arrays seem eyes sensitive to near-infrared light. For using the ESO Very Large Telescope pitifully small in hindsight, and yet an in- the high-redshift surveyors, the justifi- UT1, Antu, and its facility near-infrared stantaneous increase of almost 4,000 cation is straightforward: wavelengths camera/spectrograph, ISAAC. In partic- in the number of detectors in the focal grow as (1 + z ) and surface brightness ular, we have chosen examples which plane of a telescope inspired the com- drops as (1 + z )4, making newly-born illustrate a key theme running through munity and incited a true revolution. galaxies extremely faint, infrared our work, namely that of environmental It is a revolution that continues today. sources. On the other hand, the near- impact, both in the effects that the birth- In the 1960s, Gordon Moore formulated est regions of ongoing star formation place of a star can have on its evolu- his now-famous law that the number of are “only” a few hundred parsecs away, tion, and in the back reaction that star transitors on semiconductor chips dou- yet the rationale is equally compelling. formation can have on its surroundings. bles every 12–24 months (Moore Stars are born from, and shrouded in, These are just a selection from the 1965): the same exponential growth in dense clouds of dust and molecular sample of young clusters and protostel- processor “power” has continued into gas, out of whose obscuration visible lar objects we are studying with the VLT the new millennium (Intel 2002). Inter- light can barely escape. The same (see also, e.g., Brandl et al. 1998; Zinn- estingly, Alan Hoffman of Raytheon/ physics also yields substellar objects, ecker et al. 1999, 2002), and, of course, SBRC has found that a similar scaling the brown dwarfs, with masses perhaps only a small subset of the work being relation has tracked the introduction of as little as 1% of the Sun, feebly emit- carried out by the broad and active Euro- progressively larger infrared detector ting their excess gravitational warmth pean star- and planet-formation com- arrays into common astronomical circu- as they cool and contract forever, nev- munity (Alves & McCaughrean 2002). lation, with a doubling of the number of er able to initiate nuclear hydrogen fu- pixels roughly every 18 months (see sion. Young stars are encircled by disks Amomentary digression: Fig. 1). Following SBRC’s InSb 62 × 58 of barely warm dust and gas, where the infrared Moore’s law pixel arrays in 1986, the widespread swirling vortices gather mass to form adoption of the Rockwell NICMOS3 planets, and the stars and disks con- We are by now all very familiar with HgCdTe and SBRC InSb 256 × 256 pix- spire to generate immense supersonic the current generation of large tele- el arrays occurred circa 1992, and that jets of outflowing gas blasting out of the scopes, perhaps pre-eminent among of the presently common generation of cocoon, lighting up great shocks glow- which are the four 8.2-m diameter units Rockwell HAWAII HgCdTe and SBRC ing in the light of molecular hydrogen. of the VLT: this story could certainly not Aladdin InSb 1024 × 1024 pixel arrays Very often, all of this is going on at be told without them. However, photons circa 1998. Indeed, ISAAC was com- once, as stars are born in dense, must not only be collected but also de- missioned in late 1998, and can switch crowded clusters, all interacting and tected, and an equally important issue between one or other of a HAWAII or competing with each other for survival. in the present context is the huge par- Aladdin 1024 × 1024 pixel array. All of which calls for the largest tele- allel progress made in infrared detector So, while the VLT has “only” about scopes operated with sensitive infrared technology over the 25 years or so five times the collecting area of UKIRT, cameras and spectrographs. Wide field since telescopes like the VLT were first ISAAC has a million times the number coverage is necessary to capture the proposed (Woltjer 1978). of pixels of any of UKIRT’s first-genera- whole story in a given region, but si- At that time, the first purpose-built tion instruments. This combined factor of multaneously with enough spatial reso- large infrared telescopes, such as the five million improvement in the through- lution and dynamic range to disentan- 3.8-m UKIRT on Mauna Kea, were put (used in a deliberately loose sense gle the interactions between the many about to go online, with instruments here) available to infrared astronomers sources, and to ensure that we can de- that focussed all of the primary mirror’s over just 20 years is quite dramatic, and tect even the faintest companions, light onto a single element detector. For has driven our understanding of star disks, and planets immediately adja- much of the following decade, infrared formation and early stellar evolution for- cent to their much brighter parents. astronomy continued in the same vein, ward in leaps and bounds.

28 The Trapezium Cluster: towards a lower mass limit

It is well known that the entire life history of a star is almost uniquely determined by its mass, and yet it remains quite unclear how a star arrives at that mass in the first place. In a more general sense, we do not know how to predict the distribution of masses of a population of stars recently born from a molecular cloud, as found in a young cluster, for example, the so-called initial mass function (IMF). In the broadest possible sense, the IMF has two important components: its form and its limits, that is, the shape of the IMF and where it cuts off at high and low mass. By measuring these parameters as a function of environment, including metallicity, cluster density, the presence of massive stars, for example, we can hope to place important constraints on any general theory of star formation that aims to predict the IMF and its variations. The single pow- Figure 1: The growth of astronomical near-infrared array detectors, illustrated using a sam- er-law form of the upper IMF in our ple of 41 infrared cameras and plotting the year they were introduced into service against the galaxy has been known for many years number of pixels (rows × columns) in their main detector. The dates have been randomised (Salpeter 1955), but at lower masses, within the year of commissioning in order to provide a little horizontal separation. Some sys- things become more interesting, with a tems include more than one array, but typically with the same unit size: just the unit size is turndown and peak in the IMF typically plotted for these. A wide range of telescope sizes is implicitly represented, from 1 to 10-m diameter, although there is no obvious tendency for the newest arrays to be introduced at ei- seen somewhere in the range 0.1–0.5 ther the smallest or largest telescopes. The red line represents Hoffman’s version of Moore’s M, that is, just above the stellar/sub- Law, showing a doubling in pixel count roughly every 18 months over the past 20 years. stellar break (Kroupa 2001). While it Extrapolation points to 4096 × 4096 pixel arrays or mosaics (1.7 × 107 pixels) being the work- seems self-evident that the processes horse imaging systems by 2006. of star formation know nothing about the nuclear fusion that later so brutally separates the fates of stars and brown processes, including supersonic turbu- The cluster has proven an excellent dwarfs, the form of the IMF over this lence (Padoan & Nordlund 2002), dy- site for probing the stellar initial mass peak and down into the brown dwarf namical interactions between proto- function (Hillenbrand 1997), and is regime must nevertheless encode im- stars (Bate, Bonnell, & Bromm 2002), known to include many brown dwarfs portant physics. As a result, consider- and feedback due, for example, to (McCaughrean et al. 1995; Luhman et able effort has been invested in exam- strong bipolar outflows (Adams & al. 2000; Hillenbrand & Carpenter 2000; ining the substellar IMF in young clus- Fatuzzo 1996) and ionizing radiation Muench et al. 2001), and recent studies ters, searching for and investigating from massive stars (Palla & Stahler have suggested there may be sources proto-brown dwarfs. 2000). In addition, the existence of down to as low as ~10 MJup (Lucas & Most recently, several groups have such objects with just a few Jupiter Roche 2000; Lucas et al. 2001). been pursuing the IMF downwards in masses would be interesting in itself, as However, an even deeper wide-field search of a possible lower cut-off. The they could provide important insights survey was needed to test this finding theory of hierarchical fragmentation into the very early evolution of giant and to search for any lower mass limit. predicts that a collapsing molecular planets, even if most astronomers We have carried out such a survey cloud will continue to break into ever would agree that free-floating objects using ISAAC over a 7 × 7 arcminute field smaller clumps as long they are able to formed directly from a molecular cloud centred on the well-known Trapezium radiate away their excess energy in core are not true planets like those OB stars, as shown in Figure 2, a true- less than the free-fall time for local col- formed in a circumstellar disk, and thus colour Js, H, Ks composite made from lapse. However, opacity rises with den- do not deserve that special name (cf. our initial data taken in 1999 (see sity, and at some point the gas cannot McCaughrean et al. 2001). McCaughrean et al. 2001). Adding data cool quickly enough, becomes adiabat- In any case, can we find such objects from 2000 and 2002, the final survey ic and pressure-supported, and frag- and any related mass cut-off directly, has 900 seconds integration time pixel mentation ceases (Hoyle 1953). Tra- via observations? Although it might at per filter, and a mean seeing of 0.5 arc- ditionally, this lower cut-off is predicted first seem futile to think of searching for sec FWHM. These data go significantly to lie at 0.005–0.015 M or 5–15 objects with just a few Jupiter masses deeper over a wide field than any pre- Jupiter masses (1 MJup = 0.001 M) at distances of a few hundred parsecs, vious infrared survey, with 3σ peak-pix- (Lynden-Bell & Low 1976; Rees 1976; they are remarkably warm and bright el point source detection limits of Js, H, m m m Silk 1977), although more recent calcu- when young, and deep infrared imaging and Ks of 21 . 3, 20 .0, and 19 . 6, relations suggest that it may be modified with large telescopes can now be used spectively, limits ultimately set by the by the inclusion of magnetic fields, to go in search of them. One of the ob- bright emission from the Orion Nebula. down to perhaps as little as 1 MJup vious targets for such a hunt is the In the Ks band, these limits correspond (Boss 2001). More importantly, how- Trapezium Cluster in Orion, probably roughly to 3 MJup at 450 pc, assuming ever, the whole fragmentation scenario the most populous and densest of the an age of 1 Myr and a typical intraclus- m down at low masses may have to be re- nearby young stellar clusters, with ter reddening of AV ~ 7 , using the placed by a more complex model in- more than a thousand members DUSTY pre-main-sequence models of volving a wide range of physical crammed into its inner cubic parsec. Chabrier et al. (2000).

29 Figure 2: A true-colour near-infrared (1–2.5 µm) image of the Orion Nebula and Trapezium Cluster made using ISAAC on Antu in December 1999. The Js data are shown as blue, H as green, and Ks as red. In this representation, cube root intensities and unsharp masking were used to compress the dynamic range and emphasise point sources, at a cost of some enhancement in the noise. The original version can be seen in McCaughrean et al. (2001). The image covers 7 × 7 arcmin, or 0.9 × 0.9 pc at the 450 pc distance to the nebula. North is up, east left. Total integration time in this subset of our data is 270 seconds per filter, and the seeing is 0.5 arcsec FWHM. For orientation, we have labelled the eponymous Trapezium OB stars at the centre of the image (θ1 Ori A, B, C, & D); the two other well-known OB stars (θ2 Ori A & B) just the south-east of the Bright Bar ionisation front; active star-formation centres embedded in the background molecular cloud, OMC-1S and BN- KL, the latter also being the origin of a massive outflow and the associated broad fan of shocked emission-line fingers to the north-west; and the Dark Bay, a region of high extinction in visible images that is penetrated here at infrared wavelengths. Also marked with a circle is TC193, m m m a member of the Trapezium Cluster. With Js, H, Ks magnitudes of 18 .4, 17 .7, and 17 .2, respectively, this source is roughly 13 magnitudes fainter than θ1 Ori C, illustrating the huge dynamic range that must be faced in such studies. TC193 is part of a sample for which we already have ISAAC spectroscopy: a preliminary estimate of its spectral type is L2, which would yield a mass of roughly 6 MJup assuming an age of 1 Myr, according to the models of Chabrier et al. (2000). Dereddening the near-infrared magnitudes back to the same isochrone would suggest a mass nearer 10 MJup. Keep in mind also that it lies 2–3 magnitudes above our detection limits, and there are a significant number of potentially lower-mass sources to be studied in detail.

There are roughly 1200 sources in be notoriously difficult to convert a vestigate how close we may get to our the survey region, 700 of which are colour-magnitude diagram for a young, goal of finding a lower limit to the IMF, fainter than the saturation limit of Ks ~ embedded cluster such as this into a however, we can carry out a simple m 13 , and the (Js – H) vs. H colour-mag- mass function, and a discussion of the analysis. In Figure 3, we start by as- nitude diagram is shown in Figure 3. details and caveats of the methods suming that the cluster is 1 Myr old, tak- Without the aid of spectroscopy, it can used would fill an article in itself. To ining a pre-main-sequence model iso-

30 chrone for that age and plotting it in the colour-magnitude diagram. The large and variable extinction in the cluster means a source must be individually dereddened back to that isochrone in order to determine its approximate mass, with care taken to ensure that a suitably complete extinction-limited sample is chosen before a mass function is derived. Next, we displace the 1 Myr isochrone by the median cluster m reddening of AV ~ 7 , noting that the reddened 5 MJup point lies more or less at our observational completeness limit. Thus in principle, we can now derive an extinction-limited sample down to that 5 MJup limit. Finally, we divide the brown dwarf regime into just two equally-spaced logarithmic mass bins span- ning the brown dwarf regime, from 5–20 MJup and from 20–80 MJup. Between the unreddened and reddened isochrones, there are roughly 120 sources in the higher mass bin, compared to 30 in the lower bin. These numbers can be converted into a crude Figure 3: The (Js–H) vs. H colour-magnitude diagram derived from our ISAAC imaging sur- two-point mass function for brown vey of the Trapezium Cluster (see Fig. 2). The completeness limit and typical photometric er- dwarfs, with dN/d log M ∝ M +1; this rors for sources just above this limit are shown. The 1 Myr isochrone from the pre-main se- should be contrasted with the classical quence (DUSTY) models of Chabrier et al. (2000) is plotted assuming a distance of 450 pc. Salpeter mass function in the stellar do- The great majority of the sources lie redwards of the isochrone due to intracluster dust ex- main which goes as ∝ M–1.35. tinction of up to ~ 20m and greater. There is a pile-up of sources at H = 12m–13m due to the Thus, the mass function is falling effects of deuterium burning, although sources brighter than this saturated within the 10 sec- steeply through the brown dwarf ond on-chip integration time used for the survey. The number of potential brown dwarfs is large, but not dominant. As described in the text, we have also plotted the 0.005–0.08 M regime, a general result known previ- (5–80 M ) segment of the isochrone reddened by A ~ 7m: by counting sources in the 5–20 ously for the Trapezium Cluster (see, Jup V and 20–80 MJup bins, we can see that the brown dwarf end of the IMF is clearly falling, as e.g., McCaughrean et al. 1995; Hillen- crudely characterised by the form dN/d log M ∝ M+1. brand & Carpenter 2000; Muench et al. 2002), but now extended all the way down well below the so-called deuteri- needed. Also, as there are very bright source. Thus even in the best case, we um-burning limit at 13 MJup. Indeed, a stars in the field, the observations had would require integration times roughly more detailed analysis of the Ks data to be made in visitor mode to avoid pos- 175 times longer than we presently reveals that there are very few sources sible persistence effects impacting oth- have, i.e., almost 44 hours on-source below 5 MJup. Is this evidence for a low- ers in service mode. While trips to integration time per filter per field of er mass cut-off? Such a claim would be Paranal are always interesting, it unfor- view. premature: better mass determinations tunately suffers relatively poor weather A wide field is necessary in order to are required using spectroscopy to de- in peak Orion season, at the height of obtain enough sources to ensure a sta- rive temperatures and surface gravi- austral summer, and in the end, it has tistically robust result, and using ISAAC ties, and thus eliminate uncertainties taken eight nights over three years to to mosaic the cluster would obviously due to differential reddening and complete the survey to the present be prohibitive. A wide-field imaging sys- non-coevality: we have begun this work depth. tem covering the whole cluster in one using ISAAC, but much larger samples How much longer would be needed shot would improve matters greatly, but will be required. Second, a comprehen- to get down from our current limit in even then, once the two filters neces- sive comparison of the data against the the Ks band of ~ 3 MJup to 1 MJup, where, sary plus sky and other overheads are wide range of available pre-main-se- for example, the magnetically-mediated accounted for, at least a month of clear quence evolutionary and atmosphere fragmentation limit may lie? Pre-main- observing nights on Orion would be re- models is required in order to test the sequence models can be used to as- quired. Ultimately then, getting to the robustness of the mass estimates. sess this, although there are many un- very bottom of this crucial question will Third, it is clear that yet deeper imaging certainties and differences between require the NGST, as discussed briefly is required to probe well below the various models in this low-mass, below. present limits, to ensure that we have low-temperature domain. For illustra- delineated any such boundary on a station, we use the well-known COND and The Eagle’s EGGs: tistically sound basis. DUSTY models of Isabelle Baraffe, fertile or sterile? At first sight, this last point might ap- Gilles Chabrier, France Allard, and col- pear trivial, since the present data only laborators, and in these models, a 1 At the opposite end of the mass have 15 minutes integration time per fil- MJup source at 1 Myr is predicted to be spectrum from the sub-10 MJup objects ter. However, in practice, accumulating at least 2.8 magnitudes fainter than we have been searching for in the those 15 minutes was an onerous task. its 3 MJup counterpart in the Ks band. Trapezium Cluster, the eponymous OB First multiply by three filters, then by At least one infrared colour is re- stars at its heart pose a problem for nine on-source and four sky positions, quired in order to provide a reddening their lower-mass neighbours. The mas- a factor of two for detector readout, tel- estimate, and taking the H band, the sive stars yield a prodigious output of escope offset, and standard star over- most optimistic models predict that our ultraviolet photons which not only heads, and a total of ~ 20 hours of clear 1 MJup source should again be 2.8 sculpt and illuminate the Orion Nebula conditions with excellent seeing were magnitudes fainter than the 3 MJup HII region, but also heat and ionise the

31 dense disks of dust and gas which sur- only the combination of the VLT, HH212: the prototypical round the neighbouring young stars ISAAC, and service mode observations protostellar jet (O’Dell, Wen, & Hu 1993; Bally, O’Dell, was able to deliver the deep, wide-field & McCaughrean 2000). Are planetary images with superb seeing necessary In M 16, we have seen the impact systems able to form under this on- to carry out the definitive survey of this massive stars can have on their envi- slaught? As most stars form in clus- astronomical icon. Our data were ob- ronment. However, even low-mass ters, this poses a major quandary in tained in 2001, covering a wide field (10 stars can play an important part in the our attempts to understand the birth × 10 arcmin) with excellent seeing-lim- feedback loop, as we see in our last ex- statistics of the galactic planetary pop- ited resolution (0.35 arcsec FWHM), ample, an enigmatic protostellar source ulation. However, massive stars can yielding 3σ peak pixel point source de- near Orion’s belt. In the mid-1980s, one m m have an impact even earlier in the tections at Js. H, and Ks of 22 .6, 21 .3, of us (HZ) had become interested in star formation process. When they first and 20 m. 4, respectively. Assuming an very young binary systems, very few of ignite, their ionising photons and age of 1 Myr, these limits correspond to which were known at the time. A newly- strong winds collide with any nearby the detection of a 0.08 M source em- discovered cold, dense molecular cloud molecular cloud material, first com- bedded in 30m of visual extinction, and core, however, IRAS 05413-0104, ap- pressing dense cores and then ripping even brown dwarfs should be visible peared to be an ideal target in which to them apart. Can new stars form via ra- through less extinction (McCaughrean go looking for a low-mass protobinary diative implosion before the cores are & Andersen 2002). system. The opportunity came in 1987, destroyed (Larosa 1983; Bertoldi 1989; The resulting colour composite im- during the commissioning of the origi- Lefloch & Lazareff 1994)? How are the age covering just the elephant trunks is nal IRCAM on UKIRT. With only a 62 × properties of any pre-existing proto- shown in Figure 4: the full field of view 58 pixel array, there was an inevitable stars affected by having their parental can be seen in McCaughrean & Ander- trade-off in IRCAM between field of cores blown away before accretion has sen (2001). A detailed examination of view and spatial resolution, the former ended? the 73 EGGs shows that just 11 appear generally favoured over the latter. This A case in point is M 16, the Eagle to harbour infrared sources, as marked was in part because of the lure of final- Nebula, where the famous HST images in Figure 4. Four of these appear to be ly being able to map large regions, in of Hester et al. (1996) delineated in ex- low-mass stars, while the other seven part because the typical seeing at quisite detail three so-called elephant may be brown dwarfs, several of which UKIRT was not thought to be that great, trunks, parsec-long columns of gas and cluster near the tip of the largest ele- and in part because the received wis- dust being ionised by OB stars of the phant trunk, close to a massive (4–10 dom from single-element detectors was adjacent NGC 6611 cluster. Around the M) young protostar (YSO1; see also that small, noisy pixels would make it fringes of the trunks, Hester et al. re- Sugitani et al. 2002; Thompson, Smith, impossible to detect low-surface-bright- solved a population of small, dense & Hester 2002). Although selection ef- ness emission, forgetting however that knots, which they named EGGs for fects and uncertainties abound, there one only had to integrate long enough evaporating gaseous globules. As does indeed appear to be some limited to become background limited. In any strictly correct as the acronym may star formation going on in the elephant case, the IRCAM field of view in its 0.6 have been, it also encapsulated a less trunks. However, a major question re- arcsec/pixel mode was just 37 × 35 arc- obvious proposition, namely that the mains completely unanswered. Did sec, and in retrospect, these parame- EGGs are also eggs, the birthplaces these objects already exist in the ters conspired to lead us completely of new stars. Based on a couple of trunks, only to be revealed now by the astray for several years. A single K- plausible associations between EGGs passage of the NGC 6611 ionisation band image was taken, revealing two, and stars, Hester et al. hypothesised front, or was their formation indeed ini- apparently point sources separated by that the full population of 73 EGGs tiated by that front triggering the radia- 7 arcsec. Without further ado, we took identified in their images might harbour tive implosion of dense cores? If the this to confirm all of our expectations a plethora of young stars about to be former model holds, then we might ex- and preconceptions, and as a conse- exposed by the ionizing flux of the OB pect to find a distribution of young stars quence, IRAS 05413-0104 was written stars, thus terminating accretion and and brown dwarfs embedded within the up as a young binary system (Zinn- helping to define their final masses. densest parts of the trunks, not just at ecker 1989; Zinnecker et al. 1992). Indeed, Hester et al. went further, and the ionised periphery: due to the ex- Everything changed a few years later suggested that if most stars form in tinction however, this is also an experi- during an observing run at the 3-m such an environment, perhaps the form ment which will probably have to await IRTF on Mauna Kea. While defining of the IMF is determined by the impact a combined thermal-infrared and mil- macros to observe one target, a brief of OB stars. limetre survey by the NGST and ALMA, opening in the proceedings made it This far-reaching hypothesis needed respectively. possible to slew to IRAS 05413-0104 testing. The optical HST images were And finally, we must be careful not to for follow-up imaging using the facility unable to probe the interiors of most of develop tunnel vision, mistaking the camera, NSFCAM. With its 256 × 256 the dense EGGs, to assess how many single HST WFPC-2 field of view ver- pixel array, NSFCAM had improved of them truly contained young stars, sion of M16 for a detached, isolated ob- sampling and field compared to IRCAM, and deep infrared observations are ject, as the makers of the film version of 0.3 arcsec/pixel and 77 × 77 arcsec, re- called for in order to penetrate the ex- Carl Sagan’s “Cosmos” did in their oth- spectively. As soon as the first image tinction and make a detailed census. In erwise bravura opening sequence. The arrived, it was clear this was something addition, these observations must be at real action in the region appears to other than a simple binary system: the high spatial resolution, as the median have taken place in NGC 6611, where original two sources were now seen to EGG diameter is just 1000 AU or 0.5 a cluster of thousands of stars has be extended, not point-like, and they arcsec at 2 kpc, and have to span a formed within the past few million appeared to be just the innermost pair wide field of view, in order to cover the years, apparently with a rather normal of a long, linear trail of faint nebular whole elephant-trunk system and make IMF (Hillenbrand et al. 1993). The on- knots. Guessing that this was in fact a a detailed statistical study of contami- going destruction of the adjacent ele- young jet, not a protobinary, we nation by the dense field star popula- phant trunks and the limited star forma- switched to a narrow-band filter at 2.12 tion seen towards and beyond M 16. tion taking place within them may ulti- µm designed to trace emission from hot Early attempts either had inadequate mately prove to be a sideshow in the molecular hydrogen and after some resolution (McCaughrean 1997) or too grander scheme of things, albeit a mosaicing, a spectacular large outflow limited a field (Currie et al. 1997), and beautiful one. was revealed, later to be named HH

32 Figure 4: A true-colour near-infrared (1–2.5 µm) image of the well-known elephant trunks (or columns, C1, C2, C3) in M 16, the Eagle Nebula, made with data taken with ISAAC on Antu in April–May 2001. The Js data are shown as blue, H as green, and Ks as red. The cube root of the intensities was taken to compress the dynamic range before normalizing and combining the three mosaics. The main image covers 2.6 × 3.6 arcmin or 1.5 × 2.0 pc assuming a distance of 1.9 kpc to M 16, and is a subsection of the full 9 × 9 arcmin data set that can be seen in McCaughrean & Andersen (2001). North is up, east left. Total integration time is 1200 seconds in Js, and 300 seconds in each of H and Ks. The seeing is 0.35 arcsec FWHM. The small subimages have been magnified by a factor of 2.9 and each covers 18.5 × 18.5 arcsec (0.17 × 0.17 pc). Labels mark evaporating gaseous globules (EGGs) identified in the optical HST data of Hester et al. (1996) which we find to be associated with low-mass stars and brown dwarfs as described in the text. Also shown are E23, an EGG with no near-infrared point source, but thought to contain an embedded protostar driving a collimated jet; YSO1 and YSO2, massive sources in the tips of C1 and C2, respectively; and HH 216, an optically-visible Herbig-Haro object (Andersen et al. 2002). Due to the large dynamic range, some of the very faintest sources are not easily seen in the subimages, but can be seen in the original data. From McCaughrean & Andersen (2002).

33 212, not at all coincidentally after the ready able to identify velocities in the reflected off the uppermost surfaces of wavelength at which it was emitting. inner knots of 100–200 km s–1, al- the large flattened, rotating molecular Subsequent deeper observations a though the observations must be con- cloud core associated with the central year later using the 256 × 256 pixel tinued in order to extend the sensitivity protostar (Lee et al. 2000; Wiseman et MAGIC camera on the Calar Alto 3.5-m down to 20 km s–1 or so, where the out- al. 2001), as well as the inner edges of telescope fully delineated the extreme er bow shocks sweep up the ambient the cavity being excavated by the out- bipolar symmetry of the flow, with a se- medium. flow in the larger parent core. Alterna- ries of compact knots and bowshocks In the meantime, however, we have tively, it may be that gas along the edge equidistantly spaced on either side of combined the extant data to make a of the cavity is being lightly shocked the centre. It is highly unlikely that in- single extremely deep image as shown and that the emission is in situ, not re- homogeneities in the surrounding am- in Figure 5, which has a total integration flected. bient medium would give rise to such time of 282 minutes in the central half, Finally, close inspection of the VLT symmetry. Rather, the symmetric se- and 141 minutes in the outer parts. This image reveals that a substantial frac- quence of features appears to repre- is by far the deepest infrared image ob- tion of the sources in the surrounding sent a ticker-tape record of the accre- tained of a young protostellar jet, with field are in fact background galaxies, tion history of the very young central 4.7 hours on an 8.2-m telescope to be confirming that the parent core is very protostars, with variations in the infall compared to the typical longest obser- compact and quite detached from the onto the protostar being reflected in se- vations of any other system of 30 min- nearby Orion B molecular cloud com- quences of knots and bowshocks in the utes on a 4-m-class telescope. In addi- plex. This is especially evident at the corresponding outflow (Zinnecker, tion, the coadded spatial resolution is north-east end, where the outer bow- McCaughrean, & Rayner 1998). an excellent 0.34 arcsec FWHM, with shock appears to pass close to a clus- However, the unknown parameter in one data set having 0.2–0.25 arcsec ter of galaxies, and begs the question this chart recorder is how quickly the resolution over its full two hour span, of how there can be any ambient medi- paper is moving, i.e., the timescales in- equal to the diffraction-limited spatial um in the region to be shocked. volved. How often are new knots emit- resolution of the HST at the same Conversely, in one of the largest depar- ted? How fast are they moving when wavelength. Finally, even though the jet tures from symmetry, there is an addi- emitted? How do they decelerate when is indeed seen to move over the 16- tional large bowshock at the south-west they interact with the ambient medium? month baseline, the maximum shifts end, probably formed as the jet prows What is their time-averaged impact on are on the order of a single ISAAC pix- into another cloud core, as signposted the surrounding medium? High-resolu- el (0.148 arcsec), and thus any blurring by the presence of a bright source sur- tion spectroscopy of the two inner knots introduced is invisible at the scale rounded by nebulosity and an appar- in the 2.12-µm line shows almost iden- shown here. ently edge-on disk system. This inter- tical radial velocities, indicating that the While the basic symmetry of the jet is action between one low-mass protostar jet must lie very close to the plane of obviously maintained, the greatly im- and a dense core apparently containing the sky (Zinnecker et al. 1998), and proved sensitivity and resolution reveal another reminds us again that the cu- thus proper motion measurements are important new details. The innermost mulative effects of feedback cannot be required to reveal the velocity structure knots are all resolved into bowshocks, neglected when trying to understand of the jet. Typical peak outflow speeds small counterparts to the larger bows the formation of a population of young measured in young jets are of the order further out, and overall, the flow ap- stars, whether they be in a dense clus- of 100–200 km s–1, equivalent to just pears very similar to simulations made ter or more distributed over a larger mo- 0.05–0.1 arcsec/yr at 400 pc. But while of so-called “hammer jets” (Suttner et lecular cloud. two images separated by 10 years al. 1997; Volker et al. 1999), where a might appear to be sufficient, it is im- time-variable jet sends out a series of Alook to the near future portant to measure the motions as rap- pulses of dense gas leading to small idly as possible, as shocked molecular bowshocks: these subsequently merge The first few years of VLT operation hydrogen cools on timescales of 1–2 into a larger bowshock plowing its way have been a great success, and ISAAC years, and thus over longer timescales, through the ambient medium. However, has been an outstanding workhorse in- there is no guarantee that it is the same the image reveals almost continuous strument, as hopefully witnessed by the packet of gas that is glowing. In addi- emission between the knots following data shown in this article and else- tion, it is important to measure lower apparently sinuous channels, and the where. However, there is a fly in the velocities where the outflowing jet de- origin of this emission is not entirely ointment: all of the projects described celerates into the surrounding gas. clear: why are there continuous shocks in this article relied on mosaicing to Thus we require deep, high spatial between the bows? cover the desired field, which is not only resolution imaging separated by rea- Near the base of the jet, there is a inefficient, but also compromises accu- sonably short intervals. We are observ- pair of diffuse nebulae that broadly re- racy when it comes to PSF-fitting pho- ing HH 212 in the 2.12-µm line with semble the parabolic reflection nebulae tometry, proper-motion monitoring, and ISAAC once per year over a five-year associated with circumstellar disks detection of very extended low-sur- period. Service mode observing almost around young stellar objects, as seen in face-brightness emission. As a result, it guarantees that we will obtain the de- HH 30, IRAS 04302+2247, and Orion is worth taking a brief look at the future sired excellent seeing (< 0.4 arcsec) for 114–426 (McCaughrean, Stapelfeldt, & of wide-field near-infrared imaging at the annual two hours of integration Close 2000). However, the HH 212 ESO, and thus by extension, how our time, and the relatively large ISAAC nebulae are separated by roughly 5.5 studies may be further improved. field means that only a two-position arcsec or 2200 AU, and comparison The standard answer is VISTA, the 4- mosaic is required to cover the whole with continuum data shows that these m survey telescope which will feed a jet, which in turn ensures reasonably nebulae are in fact emitting almost ex- huge infrared camera covering the accurate proper-motion determina- clusively in the 2.12-µm line, which ap- equivalent of 42 × 42 arcmin at 0.31 tions. Adaptive optics imaging is not an parently rules out the central YSO as arcsec/pixel with 16 2048 × 2048 pixel option in this region given the lack of a the source of illumination. Finally, the arrays. It is argued that VISTA elimi- suitably bright guide star, and distortion deep data show that these nebulae ex- nates the need for a wide-field camera problems introduced by AO over wide tend up and along the jet. It seems like- on the VLT, since for large-area sur- fields might also be a concern. To date, ly that the two brightest, innermost veys, it will make up with field of view we have imaged the jet three times knots of the jet are acting as huge spot- what it lacks in collecting area. But there over a 16-month period, and are al- lights hanging in space, their light being are caveats. First, the VISTA camera

34 Figure 5: A deep image of the protostellar jet, HH212, in the 2.122 µm v = 1–0 S(1) line of molecular hydrogen, made using ISAAC on Antu, with data from October 2000, October 2001, and January 2002. North is up, east left. Total integration time in the central half is 282 minutes, yielding a 3σ per pixel limiting surface brightness sensitivity there of 1.2 × 10–19 W m–2 arcsec–2. The image covers 3.2 × 3.9 arcmin or 0.37 × 0.45 pc assuming a distance of 400 pc to HH212. The resolution is seeing limited at 0.34 arcsec FWHM. Intensities have been scaled log- arithmically. Continuum emission has not been subtracted as the jet is known to be emitting almost exclusively in the S(1) line: however, continuum sources such as field stars and galaxies remain visible. As discussed in the text, the nebulosity surrounding the base of HH212 is line emission, while the bipolar nebula at the south-west end of the image is seen in continuum emission, and is most likely a large circumstellar disk. Judging from broad-band data, the point source at the centre of the southwest nebula appears to be an unrelated field star. The two subimages in the corners have been expanded by a factor of 2.5, and show the south-west bipolar nebula (lower left) and the inner knots of HH 212 (upper right) in more detail. From McCaughrean et al., in preparation.

35 will not have the contiguous field which Together, these features will yield ex- Currie, D., Kissell, K., Shaya, E. et al. 1996, is often so important for star-formation traordinary gains in sensitivity from 0.6 The Messenger, 86, 31. targets: mosaicing will still be required. to 28 µm for both imaging and spec- Hester, J.J., Scowen, P.A., Sankrit, R. et al. 1996, AJ, 111, 2349. Second, for observations of a single field troscopy. As examples in the context of Hillenbrand, L.A. 1997, AJ, 113, 1733. trying to reach a given faint flux limit, this article, consider a survey for point Hillenbrand, L.A. & Carpenter, J.M. 2000, the VLT will be four times quicker than sources at 2 µm: the NGST should go ApJ, 540, 236. VISTA. Third, the infrared seeing is fre- more than 6 magnitudes deeper than Hillenbrand, L.A., Massey, P., Strom, S.E., & quently excellent on the VLT, and the ISAAC in the same integration time, Merrill, K.M. 1993, AJ, 106, 1906. VISTA image scale will not take proper and folding in all factors, the NGST will Hoyle, F. 1953, ApJ, 118, 513. 4 Intel 2002, http://www.intel.com/research/ advantage of it: a camera with an im- be roughly 10 times faster than a pres- silicon/mooreslaw.htm age scale similar to ISAAC’s would ent-day 8-m telescope, completing in Larosa, T.N. 1983, ApJ, 274, 815. make it quicker yet for point sources, as just one hour what would take literally Lee, C.-F., Mundy, L.G., Reipurth, B., Ostriker, well as yielding superior resolving pow- years of ground-based observing time. E.C., & Stone, J.M. 2000, ApJ, 542, 925. er for extended sources, and reducing Thus, when the NGST is launched in Lefloch, B. & Lazareff, B. 1994, A&A, 289, confusion in crowded fields. 2010, we will take yet another giant 559. Lucas, P.W. & Roche, P.F. 2000, MNRAS, On the VLT itself, NAOS/CONICA de- leap in the ongoing revolution that is in- 314, 858. livers excellent spatial resolution, but frared astronomy, less than 25 years af- Lucas, P.W., Roche, P.F., Allard, F., & over a smaller field than even ISAAC. ter the first infrared arrays were brought Hauschildt, P. 2001, MNRAS, 326, 695. NIRMOS should include an imaging to UKIRT and 12 years after the open- Luhman, K.L., Rieke, G.H., Young, E.T. et al. mode covering roughly 16 × 14 arcmin ing of the VLT. In tandem with the 2000, ApJ, 540, 1016. at 0.2 arcsec/pixel, but again not con- wide-field, high spatial resolution mil- Low, C. & Lynden-Bell, D. 1976, MNRAS, tiguous, and only covering 1–1.8 µm, limetre and submillimetre imaging to be 176, 367. McCaughrean, M.J. 1997, in Herbig-Haro making it of limited utility for studies of enabled by ALMA in the same time- Flows and the Birth of Low Mass Stars, low-mass or deeply embedded, cool frame, the future for star-formation eds. B. Reipurth & C. Bertout, IAU Symp. sources, or shocked H2 emission asso- studies looks very rosy indeed. Given 182, 551. ciated with outflows. Among the VLT what we have seen in the past decade, McCaughrean, M.J., Alves, J., Zinnecker, H., second-generation instruments under who knows what surprises lie ahead? & Palla, F. 2001, ESO press release consideration, there is a near-infrared (http://www.eso.org/outreach/press-rel/ pr-2001/phot-03-01.html). multiobject spectrograph, KMOS, which Acknowledgements McCaughrean, M.J. & Andersen, M. 2001, would extend to 2.5 µm, but which may ESO press release (http://www.eso.org/ or may not include an imaging mode. In It is a great pleasure to thank our outreach/press-rel/pr-2001/phot-37-01.html) any event, according to present plan- many friends at ESO, on Paranal and in McCaughrean, M.J. & Andersen, M. 2002, ning, it does not appear that any wide- Garching, especially those who built A&A, 389, 513. field imager covering the whole 1–2.5 ISAAC, those who look after it, those McCaughrean, M.J., Reid, I.N., Tinney, C.G. et al., 2001, Science (Letters), 291, 1487. µm region would be in operation on who ensure it works well with the VLT, McCaughrean, M.J., Stapelfeldt, K.R. & Close, Paranal before 2007. and those who have helped us get the L.M. 2000, in Protostars & Planets IV, Hoffman’s version of Moore’s law very best data out of it in both visitor eds. V. Mannings, A.P. Boss, & S.S. Rus- suggests that we should have already and service mode. In alphabetical or- sell (Tucson: Univ. Arizona Press), p. 485. had a 2048 × 2048 pixel infrared array der, these include Fernando Comerón, McCaughrean, M.J., Zinnecker, H., Rayner, camera available in 2001, and indeed, Jean-Gabriel Cuby, Vanessa Doublier, J.T., & Stauffer, J.R. 1995, in The Bottom of the Main Sequence and Beyond, ed. C.G. Gemini-South commissioned such a Gerd Finger, Roberto Gilmozzi, Olivier Tinney, (Berlin: Springer Verlag), p. 209. camera (FLAMINGOS) at that time. In Hainaut, Bruno Leibundgut, Chris Moore, G.E. 1965, Electronics, 38, 8. the 2004–2006 timeframe, 4096 × 4096 Lidman, Gianni Marconi, Alan Moor- Muench, A.A., Lada, E.A., Lada, C.J., & pixel cameras should be shouldering wood, Almudena Prieto, Dave Silva, and Alves, J. 2002, ApJ, 573, 366. most of the load, and indeed, several Jason Spyromilio. MJM would like to O’Dell, C.R., Wen, Z., & Hu, X. 1993, ApJ, ESO member states are developing thank Isabelle Baraffe, Michael Meyer, 410, 696. Padoan, P. & Nordlund, Å. 2002, ApJ, sub- such cameras for their 4-m-class tele- and Francesco Palla for useful input, mitted (astro-ph/0205019). scopes, including WFCAM for UKIRT and the Institute of Astronomy in Cam- Palla, F. & Stahler, S.W. 2000, ApJ, 540, 255. and WIRCAM for the CFHT. Thus it bridge, where much of this paper was Rees, M.J. 1976, MNRAS, 176, 483. seems reasonable to explore the op- written, for their hospitality. This work Salpeter, E.P. 1955, ApJ, 121, 161. tions for developing a wide-field imager was funded in part by the EC Research Silk, J. 1977, ApJ, 214, 152. for the VLT. A straightforward direct im- Training Network “The Formation and Sugitani, K., Tamura, M., Nakajima, Y. et al. 2002, ApJ, 565, L25. ager with a 2 × 2 mosaic of 2048 × 2048 Evolution of Young Stellar Clusters” Suttner, G., Smith, M.D., Yorke, H.W., & pixel arrays would cover a 10 × 10 ar- (HPRN-CT-2000- 00155) and DLR Zinnecker, H. 1997, A&A, 318, 595. cmin field at the same 0.15 arcsec/pix- grants 50-OR-0004/ 9912. Thompson, R.l., Smith, B.A., & Hester, J.J. el scale as ISAAC, sampling the kind of 2002, ApJ, 570, 749. excellent infrared seeing seen in this References Völker, R., Smith, M.D., Suttner, G., & Yorke, H.W. 1999, A&A, 343, 953. article, and would be very well suited to Adams, F.C. & Fatuzzo, M. 1996, ApJ, 464, Wiseman, J., Wootten, A., Zinnecker, H., & deep pointed imaging of embedded 256. McCaughrean, M.J. 2001, ApJ, 550, L87. star clusters, and outflows, as well as Alves, J. & McCaughrean, M.J. 2002, eds. Woltjer, L. 1978, in Optical Telescopes of the nearby galaxies and distant galaxy The Origins of Stars and Planets: The Future, eds. F. Pacini, W. Richter, & R.N. clusters. VLT View (Berlin: Springer Verlag). Wilson (Geneva: ESO), p. 5. Looking even further ahead, the joint Andersen, M., Reipurth, B., Knude, J. et al. Zinnecker, H. 1989, in Low Mass Star For- 2002, A&A, in press. mation and Pre-main Sequence Objects, NASA/ESA/CSA Next Generation Bally, J., O’Dell, C.R., & McCaughrean, M.J. ed. B. Reipurth, (Garching: ESO), p. 447. Space Telescope will combine a large 2000, AJ, 119, 2919. Zinnecker, H., Andersen, M., Brandl, B. et al. primary mirror (in this case, 6.5 m in di- Bate, M.R., Bonnell, l., & Bromm, V. 2002, 2002, in Extragalactic Star Formation, ameter), with diffraction-limited resolu- MNRAS, 332, L65. eds. E. Grebel, D. Geisler, & D. Minniti, tion (0.08 arcsec at 2 µm) over a field of Bertoldi, F. 1989, ApJ, 346, 735. IAU Symposium 207, in press. more than 3 × 3 arcmin, and a location Boss, A.P. 2001, ApJ, 551, L167. Zinnecker, H., Bastien, P., Acoragi, J.-P., & Brandl, B., Brandner, W., Eisenhauer, F., Yorke, H.W. 1992, A&A, 265, 726. at the Sun-Earth L2 point where pas- Moffat, A.F.J., Palla, F., & Zinnecker, H. Zinnecker, H., Krabbe, A., McCaughrean, sive cooling to 50 K will avoid the 1999, A&A, 352, L69. M.J. et al. 1999, A&A, 352, L73. scourge of OH airglow and thermal Chabrier, G., Baraffe, l., Allard, F., & Zinnecker, H., McCaughrean, M.J., & Ray- background fought from the ground. Hauschildt, P. 2000, ApJ, 542, 464. ner, J.T. 1998, Nature, 394, 862.

36 Gamma-Ray Bursts: the Most Powerful Cosmic Explosions L. KAPER1, A. CASTRO-TIRADO 2, A. FRUCHTER 3, J. GREINER4, J. HJORTH 5, E. PIAN 6, M. ANDERSEN7, K. BEUERMANN 8, M. BOER9, I. BURUD 3, A. JAUNSEN 5, B. JENSEN 5, J.M. CASTRO CERÓN10, S. ELLISON11, F. FRONTERA12, J. FYNBO13, N. GEHRELS14, J. GOROSABEL2, J. HEISE15, F. HESSMAN16, K. HURLEY17, S. KLOSE16, C. KOUVELIOTOU18, N. MASETTI12, P. MØLLER13, E. PALAZZI12, H. PEDERSEN 5, L. PIRO19, K. REINSCH 8, J. RHOADS 3, E. ROL1, I. SALAMANCA1, N. TANVIR20, P.M. VREESWIJK1, R.A.M.J. WIJERS1, T. WIKLIND 21, A. ZEH16, E.P.J. VAN DEN HEUVEL1

1Astronomical Institute “Anton Pannekoek”, Amsterdam, The Netherlands; 2IAA-CSIC, Granada, Spain; 3STScI, Baltimore, USA; 4MPE Garching, Germany; 5Copenhagen University Observatory, Denmark; 6Trieste, Italy; 7Potsdam, Germany; 8Universitäts Sternwarte, Göttingen, Germany; 9CESR/CNRS, Toulouse, France; 10Real Instituto y Observatorio de la Armada, Cádiz, Spain; 11ESO Paranal, Chile; 12Bologna, Italy; 13ESO Garching, Germany; 14NASA/GSFC, Greenbelt, USA; 15SRON Utrecht, The Netherlands; 16Thüringer Landessternwarte (Tautenburg, Germany); 17UC Berkeley, Space Sciences Laboratory, USA; 18NASA/MSFC, Huntsville, USA; 19Rome, Italy; 20Hertfordshire, United Kingdom; 21Onsala, Sweden

1. Introduction Metzger et al. 1997), active collabora- States of America. The nodes take tions between many observatories turns for being “on duty” for periods of Gamma-ray bursts (GRBs) are brief around the world has resulted in a time- two weeks. Starting September 2002 flashes of cosmic γ-rays, first detected ly and detailed study of several dozen our collaboration will be supported by a in data from the US military Vela satel- GRBs. From the start ESO has played Research and Training Network funded lites in 1967 that were launched to a very important role in the identifica- by the European Commission for a pe- verify the Nuclear Test Ban Treaty tion and analysis of the optical and in- riod of four years. (Klebesadel et al. 1973). Lacking a dis- frared afterglows. Gamma-ray detec- tance scale, the physical nature of tors onboard spacecraft orbiting the 3. GRBs and their afterglows GRBs remained a mystery for thirty Earth or exploring the solar system pro- years. Their cosmological origin was vide the GRB alerts, which are prompt- GRBs are short flashes of γ-rays, suggested by their isotropic sky distri- ly announced on the Gamma-ray burst with a duration ranging from several bution, demonstrated in the early 1990s Circular Network (GCN); these trigger milliseconds to tens of minutes, and in by the BATSE experiment onboard the immediate follow-up observations at most cases an observed peak energy Compton Gamma-Ray Observatory ground-based observatories. By build- around 100 keV. The daily rate of (Fig. 1). ing up a network of astronomers at ob- GRBs, detectable from Earth, is about However, the definite proof of their servatories all around the world, it be- two (Paciesas et al. 1999). The γ-ray distant, extragalactic nature came from comes possible to quickly (within light curves are extremely diverse, the discovery of their rapidly fading af- hours) respond to a GRB alert (from some very smooth, others with numer- terglows at X-ray, optical, and radio one or both hemispheres) and to locate ous spikes. BATSE data showed that wavelengths in 1997, thanks to the and monitor the afterglow. there are two distinct classes of GRBs: alerts of the Italian-Dutch BeppoSAX Here we report on behalf of the a class with a short duration (less satellite. Absorption and emission lines GRACE consortium, the Gamma-Ray than 2 seconds) and relatively hard in the afterglow spectra provided red- burst Afterglow Collaboration at ESO 1. spectra, and a class of long-duration shifts in the range z = 0.1–4.5, corre- The GRACE consortium was awarded bursts with softer spectra. It is impor- sponding to distances of several billion an ESO Large Programme that started tant to note that only afterglows of the light-years out to the edge of the visible in April 2000 and ended in March 2002. latter population have been observed universe. This made it clear that GRBs So far, the GRACE collaboration has so far: it is not known whether short represent the most powerful explosions identified most of the known GRB opti- bursts produce afterglows at all. The in the universe since the Big Bang. cal counterparts and has measured best limit obtained so far is for the There are strong indications that about two thirds of all known GRB red- short/hard HETE-II burst GRB 020531, GRBs are caused by highly relativistic, shifts. Currently, ESO observing time is for which Salamanca et al. (2002) did collimated outflows powered by the col- allocated to GRACE through normal not detect any afterglow candidate lapse of a massive star or by the merg- Target of Opportunity one-semester brighter than V ~ 25, just 20 hours after er of two compact objects. Their enor- programmes. Our collaboration is also the alert. mous brightness make GRBs powerful involved in GRB follow-up programmes In a previous Messenger paper probes of the distant and early uni- awarded observing time on the Hubble (Pedersen et al. 2000) the first scientif- verse, yielding information on the prop- Space Telescope (HST), the Chandra ic break-throughs in this field were re- erties of their host galaxies and the cos- X-ray observatory, and INTEGRAL. ported. The Italian-Dutch BeppoSAX mic star-formation history. GRACE consists of teams of as- satellite played a crucial role in these tronomers (“nodes”) based in Denmark, discoveries by rapidly determining the 2. The GRACE consortium Germany, Italy, Spain, the Netherlands, position of a GRB with arcminute preci- the United Kingdom and the United sion. Arcminute-sized error boxes Since the discovery of the first GRB match the typical field size of modern afterglow on 28 February 1997 (Costa (optical) detectors, so that it became et al. 1997, Van Paradijs et al. 1997, 1http://zon.wins.uva.nl/grb/grace/ feasible to detect the GRB afterglows,

37 Figure 1: This map shows the locations of a total of 2704 GRBs recorded with BATSE on board NASA’s Compton Gamma-Ray Observatory during its nine-year mission. The projection is in galactic coordinates; the plane of the Milky Way Galaxy is along the horizontal line at the middle of the figure. The burst locations are colour-coded based on the fluence, which is the energy flux of the burst integrated over the total duration of the event. Long-duration, bright bursts appear in red, and short-duration, weak bursts appear in purple (credit BATSE team). which fade quickly, on a timescale of Besides HETE-II, satellites in the afterglow identification pipeline is nec- only a few days (the typical time profiles Interplanetary Network (IPN) provide essary. Our consortium has developed are t–α, with α ~ 1–2). burst positions, though at a low rate. such a pipeline using colour-colour dis- The High Energy Transient Explorer II The BeppoSAX mission was terminat- crimination techniques. The efficiency (HETE-II), launched in October 2000, ed on April 30, 2002, exactly 6 years af- of this procedure was demonstrated by was designed to provide very rapid (< 1 ter its launch. the discovery, at ESO, of the optical minute) and very precise positions (er- With Integral (launch October 2002) and near-infrared counterpart of GRB ror boxes down to arcseconds) of both and Swift (2003) the rate of GRB alerts 001011 (Gorosabel et al. 2002a). long- and short-duration bursts. This will definitely increase again. In prepa- Since the advent of rapid (within a mission has been one of the main ration for these missions, robotic tele- few hours to days) GRB locations2, 39 drivers of our ESO Large Programme, scopes are installed at ESO La Silla to alerts resulted in the detection of an but so far only few accurate HETE-II perform prompt follow-up of GRB alerts. positions have become available. Given the expected high data rate, a fast 2http://www.aip.de/People/Greiner/grbgen.html

Figure 2: Left: VLT spectrum of GRB 990712 taken 12 hours after the burst. Absorption lines of Mg I and Mg II are detected, as well as several emission lines from the underlying bright (V ~ 22) host galaxy (from Vreeswijk et al. 2001). The redshift of the galaxy is z = 0.43. Right: A low-resolution FORS spectrum of the currently most-distant GRB 000131 at z = 4.5 (from Andersen et al. 2000). The redshift is determined from the Lyman break.

38 Figure 3: The fading afterglow of GRB 011121 at near-infrared wavelengths (1.2 µm, J band, Greiner et al. 2002). The field size is 40 × 40 arcseconds, North is up and East is to the left. On November 22, 2001, the afterglow is detected by NTT/SOFI at J = 17.5; 2 days later the afterglow has faded to J = 20.9. On February 9, 2002, the afterglow is not detected anymore (J > 24) in this superb VLT/ISAAC image (seeing 0.4 arcsecond). The light sources near the position of GRB 011121 might represent some bright regions in the host galaxy.

X-ray afterglow; 32 optical afterglows will provide accurate burst positions 5. The origin of GRBs: possible have been found (of which more than within a few minutes, many such bright progenitors half were discovered by our collabora- early afterglows become detectable. tion), and 20 radio afterglows (status From a variety of arguments, such as July 1, 2002). For many GRBs no after- 4. Evidence for collimation their total energy and the evidence for glow is found. Adverse observing con- collimation, the general expectation is ditions can explain many of these Assuming isotropic emission, the that a system consisting of a black hole non-detections. For example, the opti- measured distances imply peak lumi- and a surrounding accretion torus is cal afterglow of GRB 000630 had faded nosities of 1052 erg/s (1045 Watt). Thus powering the GRB. Such a setting, just to an R-band magnitude of 23 just 21 the peak luminosity of each event cor- before the GRB goes off, can occur in hours after the burst (Fynbo et al. responds to about 1% of the luminosity several ways. One way is the merging 2001), and would certainly have re- of the visible universe! The resulting of a binary neutron-star system, like the mained undetected in searches initiat- energy budget is about 1053 erg, which Hulse-Taylor binary pulsar, or a neutron ed a bit later. In some cases, however, is actually comparable to the total star and a black hole (e.g. Lattimer & another explanation is needed. For ex- amount of energy released during a Schramm 1974, Eichler et al. 1989). ample, the extinction by gas and dust in stellar collapse (supernova). The Another popular model involves the the circumburst environment might hin- measured rate of GRBs corresponds to core collapse of a rapidly rotating mas- der the detection of an afterglow at about one per million year per galaxy. sive star, the “collapsar” model (Woos- rest-frame UV and optical wavelengths, There is mounting evidence, howev- ley 1993, Paczynski 1998, MacFadyen or it may be too faint or even absent. er, that γ-ray bursts are collimated into & Woosley 1999). The nature of these dark bursts re- jets, with opening angles of a few de- There are several indications that mains to be resolved. grees only. This evidence comes from the observed population of GRB after- For 24 GRBs the distance has been the interpretation of the occurrence of a glows, i.e. the long-duration bursts, is determined. The GRB spectrum itself is kink in the slope of the afterglow light best explained by the latter model. featureless (consistent with optically curves, and from the detection (in a few The first indication comes from the thin synchrotron emission), but absorp- cases) of polarization (see ESO press models themselves. The collapsar tion and/or emission lines formed in the release 08/99). Also, the total isotropic model naturally produces bursts that GRB host galaxy, or the position of the energy inferred for GRB 990123 is un- have a duration longer than a few Lyman break (912 Å), provide the red- comfortably high (to be explained by a seconds, but cannot make short bursts. shift (Fig. 2). The majority of redshifts stellar-collapse model), but would be On the other hand, the merger model are in the range between 0.5 and 1.5. reduced by a factor of 500 if the energy can produce short bursts, but has prob- The current record holder, achieved were emitted into a cone with an open- lems keeping the engine on for longer with the VLT, is GRB 000131 with z = ing angle of 5 degrees. than a couple of seconds. The clear 4.5, corresponding to a “distance” Frail et al. (2001) determine the jet distinction between short- and long- (look-back time) of 13 billion light-years opening angle of several GRB after- duration bursts suggests that both (Andersen et al. 2000, ESO press re- glows and show that the spread in the progenitor models may be at work in lease 20/00). output energy distribution of their sam- nature. That GRBs can potentially probe the ple becomes much narrower when tak- very distant universe was demonstrat- ing the collimation into account, with a 6. The supernova connection ed by the impressive burst detected in mean energy output of 2 × 1051 erg. January 1999: GRB 990123 (Akerlof et They suggest that this may be the stan- Another indication that long-duration al. 1999). Within the first minutes fol- dard energy reservoir for all GRBs. GRBs are related to the core collapse lowing the burst, its optical afterglow Though speculative, the implications of of a massive star is that some GRBs reached visual magnitude V = 9, i.e. ob- this finding are great if these intrinsical- seem to be associated with a superno- servable with a pair of binoculars. It ly bright GRBs can be used as standard va (SN). The first evidence for a super- was briefly one million times brighter candles at high redshifts, e.g. to meas- nova connection came from GRB than a supernova. This particular burst, ure the expansion rate of the Universe. 980425/SN1998bw (Galama et al. at z = 1.6, would have been detectable Another consequence of the collimation 1998; ESO Press Release 15/98). This (at its maximum, in the K band) with is that the GRB rate also increases by supernova, approximately coincident in the Very Large Telescope up to a a factor of 500, and that the vast ma- time and position with GRB 980425, redshift of about 15. With Swift, which jority of bursts are not visible. was of the rare type Ic, and at radio

39 wavelengths the brightest supernova Figure 4: The light ever detected. Interpretation of the light curve of GRB 011121 curve indicated that during this super- at optical and nova a black hole was formed (Iwamoto near-infrared wavelengths. The blue et al. 1998). However, the amount of dotted line prompt γ-ray emission was very mod- represents the contri- est, which makes GRB 980425, the bution to the light of closest GRB at a redshift of z = 0.0085, the GRB afterglow. a peculiar event. About 10 days after In the mean time, evidence has been the burst the light found that several GRB afterglow light curve shows a bump, curves show a so-called supernova indicative of an bump, i.e. a bump in the light curve at a accompanying supernova,shown in time interval compatible with the rise magenta (Greiner et time of a SN, assuming it has gone off al. 2002). simultaneously with the GRB. The bump would thus represent the SN maximum light. Amongst them is the recent burst GRB 011121 (Fig. 3), which was followed up by our collaboration with ESO telescopes in several wavelength bands (Fig. 4, Greiner et al. 2002). Emission lines produced by the host galaxy indicate a redshift of 0.36. occur in regions where active star for- 150,000 light-years. Apparently, the The late-time light curve shows a mation is taking place (see, e.g., the GRB went off in the outskirts of one of bump, some 10 days after the burst VLT observations of the host of GRB the spiral arms. when the GRB afterglow has faded by 001007, Castro Cerón et al. 2002). For several host galaxies the star-for- a factor of 250. After correcting for the Neutron-star binaries do not neces- mation rate has been determined. The flux contribution and extinction due to sarily reside in star-forming regions. emission lines in the VLT spectrum of the host galaxy, the late-epoch light Due to the kick velocities received the host galaxy of GRB 990712 (Fig. 2) curve is consistent with a SN similar to during the two supernova explosions are produced by H II regions in that SN1998bw (taking into account the dif- forming the neutron stars, such bi- galaxy. The strengths of these lines in- ference in redshift). naries are high-velocity objects. As the dicate an (extinction-corrected) star- –1 merging process of the binary, driven formation rate of about 35 M yr by the emission of gravitational radia- (Vreeswijk et al. 2001). For some host 7. GRB host galaxies tion, can take up to a billion years, the galaxies even higher rates of star for- –1 binary may have travelled several thou- mation are claimed, up to 1000 M yr For practically all GRB afterglows sand light-years before producing a (e.g. Berger et al. 2001). These obser- with an accurate location, a host galaxy GRB. vations show that at least some of the has been detected. In nearly all cases With the Hubble Space Telescope GRB host galaxies belong to the class the burst is located within the optical (HST) the morphology of the GRB host of starburst galaxies. (rest frame UV) extent of the galaxy. galaxies is studied in detail. Figure 5 Thus, the observations of GRB host This, in combination with the blue shows an HST/STIS image of the galaxies support the collapsar model. colours of the galaxies, suggests that galaxy hosting GRB 990705 (Andersen Since these galaxies, due to their dis- GRBs originate in galaxies with a rela- et al. 2002). It is a giant grand-design tance, are often very faint, the bright tively high star-formation rate. The col- spiral at a distance of about 8 billion GRB afterglow provides a unique op- lapsar model predicts that GRBs will light-years with a diameter in excess of portunity to study the gas and dust con- tent of the host galaxy. The metallicity and star-formation rate of these relatively young galaxies can be measured. Figure 5: HST/STIS As part of our host-galaxy programme, image of the host galaxy of GRB the spectral energy distribution (SED) 990705. It is a of the host galaxy of GRB 000210 has grand-design spiral been determined. Fitting the observed at a distance of 7.6 SED with a grid of synthetic templates, billion light-years. the age (0.2 Gyr) of the dominant stel- North is up and East lar population and the galaxy’s photo- is to the right. The metric redshift (z = 0.84) is determined location of the GRB (Gorosabel et al. 2002b). If the collap- is marked with a sar model is right, the GRB rate is cross; the red circle a direct measure of the formation rate gives the 3σ error on its position. Note of massive stars in the early universe, that, due to the red- an important quantity for the study of shift, the image the star-formation rate as a function of shows the redshift. rest-frame UV light emitted by the galaxy (Andersen et 8. Remaining fundamental al. 2002). questions

Much progress has been made in understanding the GRB phenomenon. However, many fundamental questions

40 remain to be addressed. What is the support which our collaboration has re- Gorosabel, J., et al. 2002b, in preparation. origin and nature of the short-duration ceived from the staff at several obser- Greiner, J., Klose, S., Zeh, A., et al. 2001, bursts? Do they produce afterglows, vatories. GCN 1166. like the long bursts? Are all GRBs Iwamoto, K., Mazzali, P.A., Nomoto, K., et al. associated with a supernova, and if 1998, Nature 395, 672. ApJ so, why do we rarely observe it? Is Lattimer, J.M., Schramm, D.N. 1974, References 192, L145. this related to the difference in collima- Klebesadel, R.W., Strong, I.B., Olson, R.A. tion of the γ-rays with respect to the op- Akerlof, C., Balsano, R., Berthelemy, S., et 1973, ApJ 182, L85. al. 1999, Nature 398, 400. tical light? Are GRBs preferentially Kouveliotou, C., Meegan, C.A., Fishman, Andersen, M.I., Hjorth, J., Pedersen, H., et found in galaxies undergoing a star- G.J., et al. 1993, ApJ 413, L101. al. 2000, A&A 364, L54. burst? MacFadyen, A.I., Woosley, S.E. 1999, ApJ Andersen, M.I., Hjorth, J., Gorosabel, J., et 524, 262. The number of GRBs with optical al. 2002, A&A, submitted. Meegan, C.A., Fishman, G.J., Wilson, R.B., counterparts roughly corresponds to Berger, E., Kulkarni, S.R., Frail, D.A. 2001, Nature 355 the number of supernovae observed ApJ 560, 652. et al. 1992, , 143. before 1934, when Baade and Zwicky Castro Cerón, J.M., Castro-Tirado, A.J., Metzger, M.R., Djorgovski, S.G., Kulkarni, suggested that supernovae might be Gorosabel, J., et al. 2002, A&A, in press. S.R., et al. 1997, Nature 387, 878. powered by the gravitational collapse to Costa, E., Frontera, F., Heise, J., et al. 1997, Paciesas, W.S., Meegan, C.A., Pendleton, G.N., et al. 1999, ApJS 122, 465. a neutron star. The collapsar model, a Nature 387, 783. Eichler, D., Livio, M., Piran, T., et al. 1989, Paczynski, B. 1998, ApJ 454, L45. massive star collapsing to a black hole, Pedersen, H., et al. 2000, The Messenger has now become widely accepted to Nature 340, 126. Fynbo, J.U., Jensen, B.L., Gorosabel, J., et 100, p. 32. explain GRBs. But, just as Baade and al. 2001, A&A 369, 373. Salamanca, I., et al. 2002, GCN 1443. Zwicky failed to anticipate Type Ia su- Frail, D.A., Kulkarni, S.R., Sari, R., et al. Van Paradijs, J.A., Groot, P.J., Galama, T., et pernovae, the collapsar model, even if 2001, ApJ 562, L55. al. 1997, Nature 386, 686. correct, may be incomplete. A challeng- Galama, T.J., Vreeswijk, P.M., Van Paradijs, Vreeswijk, P.M., Fruchter, A.S., Kaper, L., et ing future lies ahead. J., et al. 1998, Nature 395, 670. al. 2001, ApJ 546, 672. We would like to acknowledge the Gorosabel, J., et al. 2002a, A&A 384, 11. Woosley, S.E. 1993, ApJ 405, 273.

Cataclysmic Variables: Gladiators in the Arena R.E. MENNICKENT1, C. TAPPERT1, M. DIAZ 2

1Departamento de Física, Universidad de Concepción, Chile 2Instituto Astronomico e Geofisico, Universidade de São Paulo, Brazil

1. Warriors and weapons action. This is the case for cataclysmic cretion disc around the white dwarf. variables (CVs, Fig. 1), consisting of a The WD reacts by emitting X-rays and Life can be different for individuals red dwarf filling its Roche lobe and UV radiation from the region where the growing up alone or closely interacting transferring matter onto a white-dwarf inner disc reaches its surface. As a re- with members of their community. The (WD) companion. If the white dwarf is a sult, part of the red dwarf atmosphere is same is true for stars, which present in- non-magnetic one, the orbiting gas in- irradiated and mildly heated, and possi- teresting phenomena when they are teracts with itself dissipating energy by bly the upper accretion disc layers forced to evolve together in close inter- viscous forces, forming a luminous ac- evaporated. It has been suggested that after a long-time of having accreted Figure 1: Computer- matter, the outer layers of the white generated view of a dwarf eventually undergo a thermonu- cataclysmic variable clear runaway in a so-called nova ex- star by Andrew plosion. Beardmore. The Here we present some results of our donor, usually a recent research in the area of dwarf no- M-type star, trans- vae, a subclass of cataclysmic vari- fers matter onto a ables showing semi-regular outbursts white dwarf. Due to in time scales of days to years with typ- viscous forces, an ical amplitudes of 2–6 mags. The origin accretion disc is of these dwarf-nova outbursts is not a formed. Colours thermonuclear runaway as in the case represent different of a nova outburst, but a sudden jump temperatures, from in disc viscosity and mass transfer rate the higher (white) to as a result of the hydrogen ionization. the lower ones In this article we do not go into deep de- (red-black). tails. Instead, we will illustrate the ap- Irradiation of the plication of some standard techniques donor by the white dwarf and disc in the field of CVs aimed to explore disc shading are also dynamics and also to reveal the nature shown, as well as of the donor star. The latter point is es- the hotspot in the pecially important to constrain theories region where the of CV evolution. gas stream impacts A typical spectrum of a dwarf nova in the disc. quiescence is shown in Figure 2. It is

41 characterized by HI and He I emission lines. Flanking the strong Balmer emission lines we observe the gravity broad- ened absorption profiles typical of a white dwarf. The Hα emission is double peaked, reflecting an origin in a rotating disc, with an equivalent width of –110 Å and half peak separation 560 km s–1, whereas their full width at zero intensity is 2950 km s–1. These velocities are typical for dwarf nova accretion discs and reflect the rotation of the outer and inner disc regions, respectively. The steep emission decrement suggests an origin in an optically thin accretion disc. The fact that we can observe the WD in this system suggests that the disc has a low luminosity, which should be related to a low-mass accretion rate. There is evidence that radial velocities (RVs) of the emission lines in CVs do not reflect the orbital motion too well, so most of the mass determinations for short orbital period dwarf novae available in the literature could be heavily biased. Therefore, the possibility of measuring the WD absorption RVs is very interesting, since it could be the only source Figure 2: Upper panel: Spectrum of 1RXSJ105010.3-140431, a cataclysmic variable with a of reliable stellar masses for these sys- low mass accretion rate (Mennickent et al. 2001). The double emission lines in the Balmer tems. As an example, we show in series are formed in the accretion disc and hotspot. The flux is given in units of 10–16 erg s–1 Figure 2 the radial velocities for the cm–2 Å–1, the horizontal units are in Å. Middle panel: The Hβ (squares) and Hγ (circles) hotspot emission and absorption components velocity folded with the ephemeris given by Mennickent et al. (2001). Bottom panel: The Hγ of 1RXSJ105010.3-140431. It is clear absorption line velocity (squares) obtained by cross correlation and the best sine fit. Note the that, whereas the emission RV reflects different amplitude for the emission and absorption components. the large amplitude motion of the hotspot around the centre of mass, the also illustrates that high resolution low), we should expect many short or- white dwarf RVs have almost zero am- spectrography is needed to measure bital period CVs hosting brown-dwarf plitude. This result has been interpreted the subtle motion of a white dwarf grav- like secondaries. We are currently con- as evidence for an undermassive donor itationally linked to an undermassive ducting a project aimed to determine star, likely in the realm of brown dwarf secondary star. According to the stan- the secondary mass of a sample of stars (Mennickent et al. 2001). Figure 2 dard CV evolution scenario (see be- short orbital period CVs showing white dwarf absorptions. We will obtain radial velocities using cross-correlation in high-resolution UVES spectra.

2. Imaging the invisible arena

In CVs, accretion onto a white dwarf releases a considerable amount of energy. This makes the accretion discs luminous, and visible at large astronomical distances, and they can be studied in some detail. CVs are ideal laboratories to study accretion-related phenomena, since their binarity and the time scales involved (much shorter than in other astronomical objects), make it possible to obtain insights into the accretion processes that drive some of the most energetic objects in the Universe. The angular diameter of a typical CV accretion disc as seen from Earth is of the order of 10–4 arcseconds, too small to be resolved from Earth even with modern interferometers. It is possible however, to use indirect imaging techniques to image the accretion disc and the processes taking place in the CV. The technique of Doppler tomogra- Figure 3: Left: Hα trailed spectra of the cataclysmic variable VW Hyi. The sinusoids followed by the double emission peaks reflect the orbital motion of the disc around the binary centre phy was introduced by Marsh & Horne of mass. The s-wave indicates the presence of an additional emission component in the sys- (1988) to the realm of interacting bina- tem. Right: Reconstructed data from the Doppler map of Figure 4. ries. A nice analogy useful to under-

42 stand Doppler imaging as applied in CVs comes from the field of medical diagnostics. In computer tomography, a 3-D image is reconstructed from 2-D slices obtained at different angular positions around the body. In astronomy, we cannot move the telescope around the CV, but the spinning binary does the job for us. We simply take spectra of the system at several binary phases, and then combine them in a 2-D image of the emitting region in velocity space. This can be done since optically thin accretion discs are strong line emitters, and the emission line contains information on disc emissivity projected into the line of sight. Doppler maps can be obtained for different spectral lines, yielding useful probes for the physical conditions inside the accretion disc. Doppler tomography has been successfully applied to a large number of CVs (e.g. Kaitchuck et al., 1994), and now is widely used in the area of CVs, Algols, X-ray binaries and isolated rotating stars (see a recent review in Marsh 2001). As a reminder, we list the facts regarding the interpretation of the resulting maps: Figure 4: Doppler map of VW Hyi obtained from the data shown in Figure 3. The cross and • The coordinates of the white dwarf the plus signs indicate the position of the white dwarf and the system centre of mass, re- in the map are (vx,vy) = (0,–K1), those of spectively. The accretion disc is revealed in the donut-shaped emissivity region. The central the secondary (0,–K2), with K1 and K2 bright spot in the upper region represents emission from the secondary star. The hotspot is being the respective semi-amplitudes also revealed in the upper left quadrant, as well as the gas stream connecting the donor and of the radial velocity curves. the hotspot. • The velocity image of the accretion disc is inverted with respect to the spa- Doppler map of VW Hyi obtained from has been observed in roughly 32% of tial one, as the material near the pri- the data shown in Figure 3. The accre- the CVs (Tappert & Hanuschik 2001). mary has high rotational velocities, and tion disc is revealed in the donut- Doppler imaging is already well es- material at the outer parts of the disc ro- shaped emissivity region. The central tablished, but it has an even more tates with lower velocities. bright spot in the upper region repre- promising future in CV research. The • A transformation of the resulting sents emission from the secondary method is being improved in several map into a spatial coordinate system is star. The hotspot is detected as the ways. Steeghs (2001) describes a only possible if the valid velocity law is second maximum in the upper left modification which allows orbital vari- known. This is not necessarily always quadrant, as well as the gas stream ability to be included in Doppler recon- the case, since also emission from connecting the donor and the hotspot. structions. Bobinger et al. (1999) de- non-Keplerian sources is possible. This is one of the few maps showing scribe a method to simultaneously fit In this work, the implementation of emission from all of these four compo- spectra and light-curves of emission Spruit (astro-ph/9806141) has been nents. We must keep in mind, however, lines. Skidmore et al. (2000) introduced used to perform the Doppler tomogra- that these emissions are highly vari- the method of ratio maps between re- phy. We have replaced the original IDL able, and different components may be constructions obtained at different routines by a corresponding MIDAS invisible at different times, depending on wavelengths. Using this method, terface, but still use the FORTRAN core the luminosity state of the CV. Mennickent et al. (2001) determined a program (version 2.3.1), to run the Figure 5 shows the Doppler map of steady-state (T ~ r–3/4) accretion disk computation on a Linux PC. another cataclysmic variable star of the mainly emitting in Hα and an optically We show in Figure 3 the Hα trailed dwarf nova type, RZ Leo, which is char- thicker hotspot with a strong contribu- spectra of the cataclysmic variable VW acterized by a relatively long recur- tion to the higher-order Balmer lines Hyi. This is a rather bright southern SU rence time and short orbital period. The and He I 5875 in 1RXSJ105010.3- UMa star with a large photometric data- data presented here have been dis- 140431. base. However, the spectroscopic cussed by Mennickent & Tappert record is rather poor. The outburst re- (2001) in the context of a search for the 3. Looking for signatures of the currence time is about 27 d. We ob- orbital period, but they did not construct red warrior tained 44 spectra with the EMMI spec- Doppler maps from these data. trograph mounted at the ESO 3.5-m Contrary to that seen in VW Hyi, the The determination of the secondary NTT at La Silla Observatory, on August map reveals the existence of two iso- mass in CVs is the key to understand- 29, 1998. The spectra had a wave- lated emission maxima on roughly op- ing the secular evolution of these ob- length range of 4475–7040 Å and a posite sides of the disc. The feature in jects. Cataclysmic variables are found spectral resolution of 2.5 Å. The sinu- the (–vx,±vy) quadrant is the more dom- with orbital periods between 78 minutes soids traced by the Hα double emission inant one and could be related to the and 10 hours, with an abrupt drop in the peaks reflect the orbital motion of the classical hotspot. Up to now, no con- number of systems in the range 2–3 disc around the binary centre of mass. vincing theory can explain isolated hours, the so-called “period gap”. The s-wave indicates the presence of emission-line sources like the one ob- Current theories state that the process an additional emission component in served in the (+vx,+vy,) quadrant of RZ of mass transfer becomes linked with the system. In Figure 4 we show the Leo, although emission in this quadrant the loss of orbital momentum, so the bi-

43 VY Aqr should suggest a system beyond the orbital period minimum, probably containing an undermassive secondary star. This has also been supported by the application of Method 3 (Patterson 2001). We have found direct evidence for this scenario, based on the unambiguous detection of the secondary star in the spectrum of VY Aqr (Mennickent & Diaz 2002). The ISAAC infrared spectrum of VY Aqr shown in Figure 6 reveals Bracket and Paschen emission lines. We also observe the K I doublets at 1.169-1.177 and 1.244-1.253 and the Na I line at 1.141, which are signatures of a cool secondary star, confirming previous indications found by Littlefair et al. (2000). When fitting the spectral energy distribution (SED) we observed that spectral types earlier than M7 fail to repro- duce the depth of KI lines in the J band and the continuum in the K band. On the other hand, spectral types later than L3 do not fit well the H and K band continuum shape. These cool types present a well-defined CO band head at Figure 5: Doppler map of RZ Leo. The spot on the left side could be associated with the 2.29 µm which is not seen in our data. disc-stream interacting region. Our fit with a M9.5 type secondary plus power-law continuum is slightly better than that for M7 and L3 type templates, nary spins faster whereas the second- 2000, Dhillon & Marsh 1995), (3) giving a χ2 parameter about 15% lower. ary becomes less and less massive, “weighting” the secondary star, possi- If such a spectral contribution is in fact eventually being eroded by the ble in systems which develop an ellipti- due to the emission of the secondary process, resulting in a kind of brown cal accretion disc during certain lumi- star in the system one may estimate its dwarf star when the orbital period ap- nosity states (e.g., Patterson 2001), temperature. Using the effective tem- proaches 80 minutes (e.g. Howell et al. and (4) by spectroscopic diagnostic of peratures for L type dwarfs derived by 2001), which is consistent with our find- the stellar masses in systems where Leggett et al. (2001) using structural ing for 1RXSJ105010.3-140431 (see the white dwarf is revealed through its models, we find T2 = 2300 ± 100 K for above). While above the gap CV evolu- optical absorption wings (Mennickent et the secondary star in VY Aqr. The best tion is likely driven by loss of angular al. 2001). fit with a M9.5 companion is shown for momentum due to magnetic-braking While the IR spectra of bright CVs illustration in Figure 7. Representative (MB), below the gap the responsible have been measured and modelled in fits using types between M7 and L3 in- mechanism is thought to be gravitation- the past years, there are only a few dicate that the secondary star may con- al radiation (GR). Since the efficiency spectrophotometric observations of tribute with 45% to 55% to the flux at for removing angular momentum by GR low-luminosity systems in J, H and K 2.17 µm, depending on the spectral is much smaller than for MB, a pile-up band. In order to investigate the rela- type. Later types yield better fits for of systems is expected below the or- tively unexplored infrared region of smaller flux fractions. Using the dis- bital period gap. Some of them should short period CVs, we have initiated a tance values for our templates from M7 be systems with normal secondary programme with ISAAC aimed to find to L2 (LHS3003 and LHS429 by van stars approaching the orbital period evidence for undermassive secondary Altena et al. 1995 and Kelu-1 from minimum at 80 minutes, others should stars. Dahn et al. 2000) and the flux fractions be systems which have “bounced” near VY Aqr is a cataclysmic variable derived from the spectral fitting we Pmin and now are receding towards the showing one of the largest outburst am- were able to derive a distance estimate longer period area. These systems plitudes among dwarf novae (Downes for the system between 80 and 120 pc, should have degenerated secondaries. et al. 2001). Accordingly to current with a most likely value of 100 ± 10 pc. The period gap is understood as an in- models for dwarf nova outburst (Osaki The spectroscopic parallax given terruption of the mass transfer rate 1996), this is consistent with a very low above is in agreement with the distance when the secondary becomes de- mass transfer rate system. Spectros- of 110 pc found by Augusteijn (1994) tached from its Roche-lobe, probably copic studies in the optical region have using the average absolute magnitude due to the onset of convection at M2 ~ revealed the orbital period (0.06309(4) value for dwarf novae in outburst. 0.3 M. d, Thorstensen 1997), but not the stel- We have applied the method outlined At present, four methods have been lar masses, likely due to the distorted above to a sample of cataclysmic vari- used to search for undermassive sec- nature of the emission-line radial ve- ables which are candidates to host ondary stars in cataclysmic variable locities (e.g. Augusteijn 1994). J-band brown-dwarf like secondaries (Men- stars: (1) analysis of the spectral ener- spectroscopy by Littlefair et al. (2000) nickent & Diaz, 2002). The SED fitting gy distribution using multi-wave-band revealed spectral features of the sec- for RZ Leo and CU Vel suggests M5 observations through the ultraviolet, ondary star, but too weak to make an type dwarf companions, and distances optical, and infrared spectral regions estimate of the spectral type. According of 340 ± 110 and 150 ± 50 pc, respec- (Ciardi et al. 1998; Mason 2001), (2) to the current CV evolution scenario tively. We find no evidence of a sec- looking for signatures of the secondary (e.g. Howell et al. 2001), the orbital pe- ondary star in the IR spectra of WZ Sge star (Steeghs et al. 2001, Littlefair et al. riod and the low mass transfer rate of and 1RXSJ105010.3-140431. The in-

44 frared SED in these objects is dominated by the accretion disc, and it can be well modelled by a simple power-law continuum. Figure 8 shows a comparison of the Teff – Porb CV evolutionary tracks near the orbital period minimum with our data and some additional data taken from the literature. From this figure we conclude the following: (1) HV Vir, WZ Sge, EF Eri, WX Cet, LL And and SW UMa seem to be post-orbital period minimum systems. (2) It is difficult to reconcile the positions of VY Aqr in the diagram with the code’s predictions. (3) In the same context, RZ Leo and CU Vel should be evolving toward the orbital period minimum. The fact that the predicted density of short orbital period CVs is at least a factor 10 higher than observed (e.g. Pat- terson 1998) and that the observed orbital period minimum is slightly, but significantly longer than the theoretical one, has motivated the entrance of two new theoretical models. Both explain Figure 6: Comparison of the normalized spectrum of VY Aqr with late-type templates. the absence of the spike at Pmin as an age effect, i.e. that CVs have not yet evolved down to Pmin. While Taam & Spruit (2001) invoke a circumbinary The results from the application of to provide constraints on the proposed disc as braking mechanism for the evo- the spectral fitting procedure described evolutionary models. lutionary process, King & Schenker above suggests that the infrared con- (2002) propose a reduced duration of tinuum shape in short-period cata- Acknowledgements the CE phase, leading to a much longer clysmic variables may be a useful indi- lifetime of the pre-CV state. This ap- cator of the companion spectral type. We thank Sandy Leggett who pro- proach additionally solves the space This point is especially important if we vided the IR digital spectra of low- density problem. In this picture, most consider that, due to the limitation im- mass red objects used in our SED systems around the orbital period mini- posed by the spectrum S/N in such faint models. We also thank Steve Howell mum in Figure 8 could be systems born systems, it is not always possible to de- and Elena Mason for providing digi- with this mass-period configuration, not tect the individual lines of the second- talized versions of the CV evolution the remnants of the evolution of longer- ary star, but nevertheless to determine tracks shown in Figure 8. This work period systems born with more massive the shape of its continuum. Also, the was supported by Grant Fondecyt secondaries. method has the advantage of avoiding 1000324, DI UdeC 99.11.28-1, CNPq the uncertainties as- 301029 and FAPESP 99-06261. We sociated with non- also acknowledge support by grant simultaneous multi- Fundación Andes C-13600/5. wavelength observations, although their References predictive power clearly is inferior to Allen 2000, Allens Astrophysical Quantities, the ideal case of Arthur N. Cox (ed), AIP Press, Springer. modelling of simulta- Augusteijn, T., 1994, A&A 292, 481. neous multi-wave- Bobinger, A. et al., 1999, A&A 348, 145. Ciardi, D.R., Howell, S.B., Hauschildt, P.H., band observations. & Allard, F., 1998, ApJ 504, 450. We are currently an- Dahn, C. et al., 2000, Giant Planets to Cool alyzing ISAAC data Stars, ASP Conf. Ser. C. Griffith and M. of a larger sample of Marley eds. CVs which are can- Dhillon, V.S., Littlefair, S.P., Howell, S.B., didates for harbour- Ciardi, D.R., Harrop-Allin, M.K., & Marsh, ing brown dwarf like T.R., 2000, MNRAS 314, 826. secondaries in order Dhillon, V.S., Marsh, T.R., 1995, MNRAS 275, 89. Downes, R.A., Webbink, R.F., Shara, M.M., et al.,2001 astro-ph/0102302 (see also http://icarus.stsci.edu/ downes/cvcat Figure 7: ISAAC spec- Howell, S.B., Nelson, L.A., & Rappaport, trum of VY Aqr and S., 2001, ApJ 550, 897. the best composite ApJ SED fit (red line). The Howell, S.B., Ciardi, D.R., 2001, 550, L57. individual M 9.5 type Ishioka, R.et al. 2001, PASJ 53, 905. template spectrum Kaitchuck, R.H., Schlegel, E.M., Honeycutt, and power law R.K., et al., 1994, ApJS 93, 519. continuum are shown. King, A.R., Schenker, K., 2002, in The The spectra are Physics of Cataclysmic Variables and normalized to 29.7 × Related Objects, Gansicke et al. eds. 10–17 erg cm–2 s–1 Å–1. ASP Conf. Ser., Vol. 261, p. 233.

45 Figure 8: Cataclysmic vari- Marsh, T.R., Home, K., 1988, MNRAS 235, ables close to the orbital pe- 269. riod minimum. Observations Mason, E., 2001, PhD thesis, Department of are compared with results of Physics & Astronomy, University of the CV population synthesis Wyoming. code by Howell et al. (2001). Mennickent, R.E., & Diaz, M., 2002, MNRAS, Normal and degenerate stars in press (astro-ph/0206343). are represented by green Mennickent, R.E., Tappert, C., 2001, A&A and blue dots respectively. 372, 563. The evolution of a particular Mennickent, R.E., Diaz, M., Skidmore, W., system is from longer peri- Sterken, C., 2001, A&A, 376, 448. ods to shorter periods, even- Osaki, Y., 1996, PASP, 108, 39. tually passing by the mini- Patterson, J., 1998, PASP, 110, 1132. mum around 80 minutes. We Patterson, J., 2001, PASP, 113, 736. have used spectral type – Skidmore, W., Mason, E., Howell, S.B., temperature calibrations Ciardi, D.R., Littlefair, S., & Dhillon, V.S., based on data of M-L dwarfs 2000, MNRAS, 318, 429. by Leggett et. al. (2000, Steeghs, D., Marsh, T., Knigge, C., Maxted, 2001). Data for LL And and P.F.L., Kuulkers, E., & Skidmore, W., EF Eri are from Howell & 2001, ApJ, 562L, 145. Ciardi (2001), for WX Cet, EF Steeghs, D., 2001, in “Astrotomography”, Peg and SW UMa from Lect. Notes on Phys. 573, Eds. H.M.J. Mason (2001) and for WZ Boffin, D. Steeghs and J. Cuypers, Sge (a temperature upper Springer, p. 45. limit) from Ciardi et al. Taam, R.E., Spruit, H.C., 2001, ApJ 561, (1998). All others are from 329. Mennickent & Diaz (2002). Tappert, C., Hanuschik, R.W., 2001, in “Astrotomography”, Lect. Notes on Phys. 573, Eds. H.M.J. Boffin, D. Steeghs and J. Cuypers, Springer, p. 119. Leggett, S.K., Allard, F., Dahn, C., Littlefair, S.P., Dhillon, V.S., Howell, S.B., & Thorstensen, J.R., & Taylor, C.J., 1997, Hauschildt, P.H., Kerr, T.H., & Ryner, J. Ciardi, D.R., 2000, MNRAS, 313, 117. PASP, 109, 1359. 2000, ApJ 535, 965. Marsh, T.R., 2001, in “Astrotomography”, van Altena, W.F., Lee, J.T., & Hoffleit, E.D., Leggett, S.K., Allard, F., Geballe, T.R., Lect. Notes on Phys. 573, Eds. H.M.J. 1994, The General Catalogue of Hauschildt, P.H., & Schweitzer, A. 2001, Boffin, D. Steeghs and J. Cuypers, Trigonometric Parallaxes (New Haven: ApJ 548, 908. Springer, p. 1. Yale University Observatory).

Reproduction of a colour-composite image of the nearby spiral galaxy NGC 300, obtained in 1999 and 2000 with the Wide-Field Imager on the MPG/ESO 2.2-m telescope at the La Silla Observatory. For more details see http://www.eso.org/outreach/press-rel/pr-2002/phot-18-02.html

46 Supernova Polarimetry with the VLT: Lessons from Asymmetry L. WANG1, D. BAADE 2, P. HÖFLICH 3, J.C. WHEELER 3

1Lawrence Berkeley National Laboratory, California, USA 2European Southern Observatory, Garching, Germany 3Department of Astronomy and McDonald Observatory, The University of Texas at Austin, Austin, USA

1. Introduction Fritz Zwicky shortly after the discovery in this sort of supernova resulting from of the neutron in 1932. When the first the spin of the white dwarf, the motion Picture a 55-year-old telescope of neutron stars were discovered by of the orbit, a surrounding accretion modest two-metre aperture and the Jocelyn Bell they were manifested as disk, or the presence of the companion need to take multiple spectral expo- rapidly rotating pulsars with intense star. As for the case of core-collapse sures of the same object to elicit any magnetic fields. Pulsars have space supernovae, this was known, but there signal from the noise with an inexpen- velocities that average several hundred was no compelling observational rea- sive spectrograph jury rigged to do po- kilometres per second. This indicates son to consider departures from spher- larimetry, an effort requiring integra- that they are somehow “kicked” at birth ical symmetry. This, too, is changing. tions on a single object lasting a whole in a manner that requires a departure long winter night with a single observer from both spherical and up/down sym- 2. Polarization of Supernovae and no night assistant, if any transient metry. More recently, NTT and then target is available during the scheduled Hubble Space Telescope images of The question of the shape of super- time. Contrast that with ordering up Supernova 1987A in the Large Magel- novae has undergone a revolution in queue-scheduled target-of-opportunity lanic Cloud showed rings of gas that the last decade. The driving force has observations by a dedicated profes- had been ejected by the progenitor star been a new type of observation: meas- sional staff on an 8 metre telescope before it exploded. This means either urement of the polarization of the light with a state-of-the-art spectrograph and the progenitor star or its surroundings from supernovae. polarimeter. That is the leap that has possessed some sort of asymmetry. Light consists of oscillating electric occurred in our programme to obtain Further observations showed that the and magnetic fields. An ordinary beam spectropolarimetry of all accessible su- debris of the explosion were also asym- of light is a mix of photons with all ori- pernovae. The spectropolarimetry ob- metric. The supernova remnant Cas- entations equally present, a state that tained in a brief exposure on the VLT siopeia A shows signs of a jet and coun- leaves the light unpolarized on aver- with the FORS1 spectrograph is com- terjet that have punched holes in the age. Some processes of producing or parable to the total flux spectrum ob- expanding shell of debris and there are scattering light favour certain orienta- tained on those long nights on the 2.1- numerous other asymmetric supernova tions of the electric and magnetic fields m Struve Telescope at McDonald Ob- remnants. Each of these things has over others. One such process is the servatory where this programme be- been known. The question has been: reflection of light. When light scatters gan. While the data obtained at McDon- are they merely incidental or a vital clue through the expanding debris of a su- ald pointed the way to a revolution to how supernovae work? pernova, it retains information about in the way we think about super- The previous discussion pertained to the orientation of the scattering layers. novae, it is the quality of the data from core-collapse supernovae. These If the supernova is spherically symmet- the VLT that has led the programme to come in a variety of spectral classifica- ric, all orientations will be present flourish. tions, Type II, Type Ib or Type Ic, de- equally and will average out, so there Supernovae have been studied with pending on whether there is abundant will be no net polarization. If, however, modern scientific methods for nearly a hydrogen, helium or neither in the out- the gas shell is not round, a slight net century. During this time, it has been er layers. There is another kind of su- polarization will be imprinted on the traditional to assume that these cata- pernova known as Type Ia. These are light. strophic stellar explosions are, for all thought to be the result of the ther- Since we cannot spatially resolve the practical purposes, spherically sym- monuclear explosion of a white-dwarf average extragalactic supernova, po- metric. There were observational rea- star composed of carbon and oxygen. larization is the most powerful tool we sons for this. The Sun is essentially When the mass of the white dwarf have to judge the shape of the ejecta. spherically symmetric and most stars closely approaches the Chandrasekhar The method used is called spectropo- are thought to be. The assumption that limit of about 1.4 solar masses, the car- larimetry. This technique both spreads stars are round is quite reasonable. bon ignites. The resulting thermonu- the light out into its spectrum of colours Self-gravity will tend to pull any large clear explosion is thought to complete- and determines the net orientation of body into a sphere which is the mini- ly disrupt the star, leaving no compact the electric field at each wavelength. mum-energy configuration. There were remnant. This sort of explosion has re- This way both the overall shape of the also practical reasons. Theoretical ceived prominence recently because emitting region and the shape of re- study of stellar explosions has been dif- they have been the tool to discover the gions composed of particular chemical ficult enough even with the assumption accelerating Universe and the dark en- elements can be determined. We note of spherical symmetry. There has been ergy that drives the expansion. The that the effective spatial resolution at- no commanding observational need to progenitor white dwarf has long been tained by polarimetry of a supernova of abandon that simplifying assumption. treated as basically spherically sym- radius 1015 cm at 10 Mpc is 10 mi- Now there is. metric even though the popular model croarcsec. This is a factor of 100 better The evidence that supernovae may is that the explosion must take place in resolution than VLTI or other compara- depart a little or even drastically from a binary system where the white dwarf ble optical interferometer installations – spherical symmetry has been growing grows to the critical mass by accretion and at a tiny fraction of the cost. for years. Neutron stars were predicted of mass and, inevitably, angular mo- There were systematic and stimulat- to form and to power supernovae by mentum. There could be asymmetries ing observations of the polarization of

47 the light of SN 1987A that are still being studied and interpreted. Another event that was modestly well studied was the hydrogen-depleted event SN 1993J in M81. These two events just illustrated how poor the overall data base of supernova spectropolarimetry was. In 1994 we began a programme to obtain spectropolarimetry of as many supernovae as possible that were visible from McDonald Observatory. At the time, only a handful of events had been examined at all and there were virtually no statistics. The data reduction was tricky, if only because the intervening interstellar medium can impose a polarization signal that has nothing to do with the supernova. The data were also difficult to interpret. There are, in principle, many reasons why the light from a supernova could be polarized. The supernova could be aspherical, it could be spherical but have off-centre sources of light, or other matter in the vicinity could be asymmetrically distributed, blocking part of the scattering surface and yielding a net polarization signal from even a spherical surface. To make matters worse, the first few supernovae our group studied (and those in the previous sparse record like SN 1987A and SN 1993J) were classified as “peculiar” in some way, so we did not know whether we were seeing incidental pe- Figure 1: Polarimetry of the Type II SN 1999em on the Q-U plane. The wavelength of differ- culiarities or something truly significant. ent data points are colour encoded. The data points are rebinned to 100 Å for clarity. The 1999 Nov 8 data (clustered in the lower-right quadrant) are clearly separated from the 2000 As data accumulated, however, this Jan. 9 data (clustered in the upper-right quadrant), suggesting strong polarization evolution. uncertainty was removed, and signifi- The data points of both epochs fall roughly on the line denoted AB. Line AB defines the axis cant new insights were revealed. With of symmetry if the object is axially symmetric. Note that the blue and red dots are well sepa- more data and better statistics, we rated for each epoch with the blue points preferentially located at the lower-right of the data identified the first key trend. In 1996, we cluster of each observations. The circles are the upper limits to the interstellar polarization realized that the data were bi-modal. assuming E(B-V) toward the supernova to be 0.05 (inner circle) and 0.1 (outer circle). The Type Ia supernovae showed little or no approximate location of the component due to interstellar dust is shown as a solid circle. polarization signal (we will talk about some significant exceptions below). Supernovae thought to arise by core and allows us a view deeper inside. We this tendency to follow a fixed orienta- collapse in massive stars – Types II, Ib, found that even in a Type II supernova, tion in space. and Ic – were, by contrast, all signifi- the longer we watched and the deeper These observations of core-collapse cantly polarized. So far there has been inside we could see, the larger the po- supernovae taken together tell us that no exception. Every core-collapse su- larization became. This was illustrated the closer we see to the centre, the pernova for which we or other groups by our VLT observations of the classic larger is the asymmetry. The asymme- have obtained adequate data has been Type II supernova, SN 1999em, that try is not some incidental aspect of the substantially polarized. Core-collapse was characterized by an especially progenitor’s surrounding environment supernovae are definitely not spherical- long plateau, suggesting an especially that is causing the polarization, but ly symmetric. The question is, why? As massive outer hydrogen envelope. something deep in the heart of the ex- data continued to mount, new trends Observations were obtained around plosion. The implication is that the ex- appeared that give critical clues to ad- optical maximum and about 2 months plosion mechanism itself is asymmet- dress that question. past optical maximum. The polarization ric. Normal Type II supernovae explode rose from 0.1 per cent in the early ob- in red-giant stars that retain large, mas- servations to about 1 per cent in the lat- 3. The Cause of Asymmetry in sive outer envelopes of hydrogen. er data. Figure 1 shows the data for SN Core Collapse Types Ib and Ic are thought to happen 1999em on a “Q-U” plot where Q and U in stars that have already shed much or represent different projections of the The fact that the asymmetry of many all of their outer hydrogen layers, so polarization vector. The striking feature core-collapse supernovae is aligned in they allow us to see deeper into the of the SN 1999em data presented in one direction provides an important heart of the exploding star. We noticed this way is that all the data points fall on clue to the engine of the explosion. The that the Type II supernovae, with their a single line. This suggests a well-de- explosion mechanism must impose large blankets of hydrogen, showed rel- fined symmetry axis throughout the some axial symmetry. There must be atively less polarization. The Type Ib ejecta and independent of wavelength. some sustained bi-polar influence be- and Ic that allowed us to peer deeper The explosion of SN 1999em is asym- cause otherwise, as the supernova ex- into the exploding matter had higher metric, but aligned in a significant way. pands, pressure gradients will tend to polarization. In addition, as a given su- Figure 2 shows data from another Type heal irregularities and the ejecta will pernova expands, the debris thins out II event, SN 2001dh, that also shows tend to become more spherical rather

48 Figure 2: A total flux fast rotation and magnetic fields that spectrum (bottom) are intrinsic to pulsars. Progress in un- and polarization derstanding the origin of jets from mag- spectrum (top) of the netized disks around black holes Type II SN 2001dh, obtained on August makes us optimistic that similar 8, 2001 with the VLT. processes will form jets from a newborn In the top panel the pulsar in a stellar core. Recent work in locations of the Austin has shown that the magnetoro- prominent lines of tational instability may play a significant hydrogen and other role to produce strong magnetic fields elements can be in a fraction of a second after core identified by the bounce. These fields may, in turn, pro- emission peaks at mote the flow of energy up the rotation these wavelengths. To the short wave- axis by a combination of hoop stresses length side of the and other pressure anisotropies. emission peaks are Doppler blue-shifted 4. SN 2002ap – Not a absorption troughs Hypernova? corresponding to absorption in the ex- This work may also shed light on the panding atmosphere supernova/gamma-ray burst connec- of the supernova. The average polarization at the epoch of these data (a week or so after dis- tion. Several supernovae have been covery) was about 1 to 2 per cent. The peaks and troughs in the polarization spectrum prove that the polarization comes from the supernova, not the intervening interstellar matter. Some identified as “hypernovae” since they of the features correspond to those in the total flux spectrum, but others correspond to fea- show excessive velocities and luminos- tures that are only clearly revealed by the polarization. ity. The most famous example, SN 1998bw, was apparently associated with the gamma-ray burst of April 25, 1998. We have been concerned that than less. To do what we see, the alone explain the explosion or whether asymmetric explosions could mimic mechanism that drives the supernova it merely supplements the standard some of the effects of “hypernova” ac- must produce energy and momentum neutrino-driven explosion remains to be tivity as interpreted by spherical mod- asymmetrically from the start, then hold seen. If the jets up and down the sym- els. In particular, asymmetric models that special orientation long enough for metry axes are somewhat unequal, could give especially high velocities in its imprint to be permanently frozen into they might also account for the run- some directions, the directions of axial the expanding matter. Appropriate out- away velocities of pulsars. jets, and be brighter in some directions flows might be caused by MHD jets, by If such jets produce the asymmetries, than others because of the resulting accretion flow around the central neu- the most likely cause of the jets are the asymmetric flux distribution. This could tron star, by asymmetric neutrino emission, or by some combination of those mechanisms. The light we see from a supernova comes substantially from the decay of short-lived radioactive elements, nickel-56, cobalt-56 and later titani- um-44 in the debris. If this material is ejected in a bi-polar fashion, then the overall debris shell could be nearly spherical, while the asymmetric source of illumination leads to a net polarization. This mechanism may be at work in the early phases of Type II supernovae such as SN 1999em and SN 2001dh. By injecting jets of mass and energy up and down along a common axis deep within a model of an evolved star, we have shown that typical asymmetric configurations emerge. As shown in Figure 3, bow shocks form at the heads of the jets as they plow through the core, and a significant portion of the star’s matter bursts through the core along the jet axis. The bow shocks also drive “transverse” shocks sideways through the star. These shocks proceed away from the axis, converge toward the star’s equator, and collide in the equatorial plane. From there, matter is compressed and ejected in an equato- Figure 3: Three-dimensional computation of a jet-induced supernova explosion in a helium rial torus perpendicular to the jets. core. The frames show the density in the plane parallel to the jet axis that passes through the centre of computational domain. The time since the beginning of the simulation is given in the These models have shown that suffi- upper left corner of each frame. The scale expands in each frame with the left frame being ciently energetic jets can both cause about 1010 cm, the middle frame about 5 × 1010 cm and the right frame about 1011 cm. Note the explosion and imprint the observed the transverse shock waves that converge on the equator. From Khokhlov et al. 1999, asymmetries. Whether this process can Astrophys. J., 524, L107.

49 be especially true for Type Ic events tion in terms of the impact of a bi-polar where the lack of hydrogen and helium flow from the core that is stopped with- envelopes give a close view of the in the outer envelope of a carbon/oxy- asymmetries of the inner explosion. gen core. Although the symmetry axis A particular case in point is the recent remains fixed, as the photosphere re- Type Ic event SN 2002ap. This event treats by different amounts in different showed high velocities, but none of the directions due to the asymmetric veloc- other characteristics of a “hypernova,” ity flow and density distribution, geo- neither a strong relativistic radio source, metrical blocking effects in deeper, nor excessive brightness. High-quality Ca-rich layers can lead to a different spectropolarimetric data of SN 2002ap dominant axis in the Q-U plane. The were obtained with the VLT Melipal and features that characterize SN 2002ap, the FORS1 spectrograph at 3 epochs specifically its high velocity, can be ac- that correspond to –6, –2, and +1 days counted for in an asymmetric model for a V maximum of 9 Feb 2002. A sam- with a larger ejecta mass than the ple of the data is presented in Figure 4. well-studied Type Ic SN 19941 such The polarization spectra show three that the photosphere remains longer in distinct broad (~ 100 nm) features at ~ higher velocity material. 400, 550, and 750 nm that evolve in We conclude that the characteristics shape, amplitude and orientation in the of “hypernovae” may be the result of Q-U plane. The continuum polarization orientation effects in a mildly inhomo- grows from nearly zero to ~ 0.2 per geneous set of progenitors, rather than cent. The 750 nm feature is polarized at requiring an excessive total energy or a level 1 per cent. We identify the luminosity. In the analysis of asymmet- 550 and 750 nm features as Na I D and ric events with spherically symmetric OI λ 777.4 moving at about 20,000 km models, it is probably advisable to refer s–1. The blue feature may be Fe II. to “isotropic equivalent” energy, lumi- We interpret the polarization evolu- nosity, ejected mass, and nickel mass.

Figure 4: Spectropolarimetry of the Type Ic SN 2002ap on 2002 3 Feb, 6 days before V maximum. The Stokes parameters Q and U are rebinned into 15 Å bins. An interstellar polarization component is subtracted from the observed Stokes parameters so that the data points represent intrinsic polarization due to the supernova. The assumed interstellar polarization is shown as the solid dot in the Q–U plot (top panel). Without subtraction of the interstellar component, the origin Figure 5: Spectropolarimetry of the Type Ia SN 2001el on 2001 Sept. 26, 7 days before max- of the coordinates would be centred at this imum. The Stokes parameters Q and U are rebinned into 15 Å bins. An interstellar polariza- solid dot. The solid line represents the axis tion component is subtracted from the observed Stokes parameters so that the data points from the origin through the value of the ISP. represent intrinsic polarization due to the supernova. The assumed interstellar polarization is The dashed line illustrates the locus of the shown as a solid dot in the Q–U plot (panel a, upper left). Without subtraction of the inter- OI feature in the Q–U plot. The polarization stellar component, the origin of the coordinates would be centred at this solid dot. The straight spectra (middle panel) and polarized flux line illustrates the dominant axis shifted to the origin of the Q–U plot. The Q (panel b, upper (bottom panel) show conspicuously polar- right) and U (panel d, lower right) spectra show conspicuously polarized spectral features. ized spectral features corresponding to Fe II, The degree of polarization is shown as the thin line in panel c (lower left) with the flux spec- Na I D, and O I 777.4 nm. The wavelength trum (panel c, lower left, thick line) overplotted to show the correlations of the degree of po- colour code is presented at the bottom of the larization and the spectral features. The wavelength colour code is presented at the bottom top panel. of panel a.

50 5. Asymmetries in Type Ia due to the Ca II IR triplet at very high that all core-collapse supernovae are Supernovae velocities (20,000–26,000 kms–1). The substantially asymmetric. They explode 800 nm feature is distinct in velocity by means of bi-polar flow associated Most Type Ia supernovae are not sub- space from the photospheric Ca II IR with the newborn neutron stars. This stantially polarized at the epochs that triplet and has a significantly higher de- discovery may, in turn, give new in- have been observed. This suggests that, gree of polarization (= 0.7%), and dif- sights into more exotic jet-induced despite occurring in binary systems, the ferent polarization angle than the con- events like gamma-ray bursts. While explosions are essentially spherically tinuum. Taken together, these aspects most Type Ia supernovae have been symmetric. There are some interesting suggest that this high velocity calcium found to be little polarized, the number exceptions to this, however. SN 1999by is a kinematically distinct feature with of exceptions is growing. The asym- was one of the class of subluminous, the matter distributed in a filament, metries observed in Type Ia may fi- rapidly declining Type Ia events. It was torus, or array of “blobs” almost nally yield direct observational evi- substantially polarized and hence edge-on to the line of sight. This feature dence that they occur in binary sys- asymmetric in some way. We do not yet could thus be an important clue to the tems, as long assumed, and clues to the know whether this was characteristic of binary nature of SN Ia, perhaps associ- combustion mechanism. Understand- subluminous Type Ia, or whether SN ated with an accretion disk, or to the na- ing these asymmetries may be neces- 1999by was odd in this regard. ture of the thermonuclear burning, per- sary to properly interpret future data on In this context, it is important to obtain haps representing a stream of material cosmologically distant Type Ia’s. spectropolarimetry of “normal” Type Ia ballistically ejected from the site of the The authors are grateful to the Euro- supernovae. A step in this direction was deflagration to detonation transition. pean Southern Observatory for the gen- taken with our observations of the Type If modelled in terms of an oblate erous allocation of observing time. We Ia SN 2001el. High-quality spectropo- spheroid, the continuum polarization are also anxious to acknowledge that, larimetry of the SN 2001el was also ob- implies a minor to major axis ratio of contrary to the impression perhaps giv- tained with VLT Melipal and FORS1 at 5 around 0.9 if seen equator-on; this lev- en in the Introduction, requests for epochs. Some of these data are shown el of asymmetry would produce an ab- service-mode observations with the in Figure 5. The spectra a week before solute luminosity dispersion of about 0.1 VLT are much different than orders to a and around maximum indicate photos- mag when viewed at different view- pizza home-delivery service: The Para- pheric expansion velocities of about ing angles. If typical for SNe Ia, this nal Science Operations staff and the 10,000 km s–1. Prior to optical maxi- would create an RMS scatter of sever- User Support Group in Garching have mum, the linear polarization of the con- al hundredths of a magnitude around gone to considerable effort to augment tinuum was 0.2–0.3% with a constant the mean brightness-decline relation. our proposal with their full range of position angle, showing that SN 2001el This scatter might have implications for expertise. We recognize that accom- has a well-defined axis of symmetry. the high precision measurements re- modating our target-of-opportunity ob- The polarization was nearly unde- quired to determine the cosmological servations in an already busy observing tectable a week after optical maximum. equation of state of the “dark energy.” and work schedule often poses a spe- The spectra of SN 2001el are similar cial extra challenge. Only this sym- to those of the normally-bright SN 6. Conclusions biosis enables the ongoing success of 1994D with the exception of a strong this project. We are especially grate- double-troughed absorption feature The acquisition of systematic super- ful for that. This work was supported in seen around 800 nm (FWHM about 22 nova polarization data has led to re- part by NASA Grant NAG5-7937 to PAH nm). The 800 nm feature is probably markable new insights. It seems likely and NSF Grant AST 0098644 to JCW.

OTHER ASTRONOMICAL NEWS An Exciting Working Session on Cataclysmic Variables at ESO/Santiago E. MASON (ESO/Chile, fellow) and S. HOWELL (ESO/Chile, visiting scientist) An intensive working session on We hope that the wealth of ideas and etry and (v) the CVs accretion disks and Cataclysmic Variables (CVs) was held projects discussed during the working current analysis of their emission lines. at ESO/Santiago on August 14, 2002. session, will trigger regular CV work- The afternoon talks were more The workshop was organized on the shops here in Chile, possibly involving specifically focused on the observation occasion of the presence in Santiago a larger number of participants and in- of particular objects or on some as- of Dr. S. Howell, from the Planetary vited speakers. The workshop was or- pects of theoretical modelling. Science Institute in Tucson, thanks to ganized in a morning review session on The participants really benefited from the ESO/Chile visiting scientist pro- CVs, both on observations and on the- being in a fairly small but highly moti- gramme. ory, and an afternoon discussion ses- vated group and could present, dis- The goal of the workshop was to sion. cuss, and confront various problems gather all astronomers in Chile working Reviews were about: (i) the photo- and results of their current research on CVs, for exchanges and fruitful dis- metric behaviour of dwarf novae (DNs) programmes. In particular, the discus- cussions. The participants were from during cycles and super-cycles, (ii) the sion of unsolved problems turned out to the University of Concepción and from spectroscopic characteristics of CVs in be important and fruitful as it triggered ESO/Chile, and we could also welcome the wavelength range UV-IR, (iii) the evo- the submission of new proposals (on Dr. N. Vogt, from Heidelberg, who had lution of CVs – theory vs. observations, ESO telescopes!), as well as the devel- organized the first workshop on CVs (iv) radial velocity measurements as a opment of new research projects and ever held in Chile (Viña del Mar, 1992). diagnostic for the binary system geom- collaborations.

51 CONFERENCE SUMMARY From Twilight to Highlight: the Physics of Supernovae ESO/MPA/MPE Summer Workshop 2002

(http://www.mpa-garching.mpg.de/~supnov/)

W. HILLEBRANDT (MPA) and B. LEIBUNDGUT (ESO)

This year’s joint workshop between many constraints on the progenitors of able supernova in the Milky Way. ESO, the Max-Planck-Institut für thermonuclear supernovae (Type Ia Neutrino oscillations can be measured Astrophysik and the Max-Planck Institut Supernovae) and their evolution to ex- from these supernovae and will provide für Extraterrestrische Physik, already plosion were presented in a coherent important diagnostics for neutrino the fifth in this series, was dedicated to picture (Nomoto). This has also im- physics and the neutrino masses the physics of supernovae. With active plications on the relative supernova (Sato). The models of thermonuclear groups at all three institutions (and at rates as a function of look-back time, supernovae have progressed tremen- both of ESO’s scientific centres in i.e. redshift. Nevertheless, there remain dously during the past decade as well. Garching and Vitacura) this topic was many open questions and the debate 3-dimensional calculations of the turbu- ideal. Over 100 experts came to whether double-degenerate models (a lent nuclear flames in deflagration mod- Garching during the last three days of binary consisting of two white dwarfs) els appear now to solidly produce ex- July to discuss the progress made in or the single-degenerates (a white plosions and about the right masses of the explosion physics, the current ob- dwarf with a main sequence or giant 56Ni. With increasing resolution in the servational status, the astrophysical re- star as companion) are the progenitor simulations more energy is released lation of supernovae and their environ- systems was continued at this con- and explanations for regular SN Ia dy- ment, and the most energetic explo- ference. From the cosmic SN Ia rate namics appear within reach even with sions known. the double-degenerate systems appear pure deflagration, i.e. subsonic, explo- With increased attention to super- less favoured due to their long evo- sions (Niemeyer). Should a transition to novae for their cosmological applica- lutionary time (Canal) and the recur- detonations occur then the explosion tion and their possible connection to rent novae, in particular in super-soft can be further strengthened (Garcia- the Gamma-Ray Bursts (GRB) during X-ray sources, appear more promising Senz). the past few years these stellar ex- candidates (Starrfield). An especially There is a bonanza of new observa- plosions have entered into the lime- attractive feature of the super-soft tions being assembled on supernovae. light of the astronomical stage. How- sources is that the measured white- Most of the second day of the workshop ever, our understanding of the underly- dwarf masses are already very close to was dedicated to presentations of new ing physics advances more slowly and the Chandrasekhar limits. data. The Berkeley group is not only its progress has to be evaluated regu- A systematic survey of white dwarfs currently finding most of the nearby su- larly. It was the goal of this meeting to for radial velocity changes with UVES pernovae, they are also assembling a focus on the basics of the explosions has so far yielded quite a high fraction large sample of light curves and spec- and the current knowledge of these of binary white dwarfs (16%). Among tra (Filippenko). Among the recent su- cosmic highlights. the best candidates are three objects pernovae there are many peculiar ones The meeting was structured with re- with a combined mass very close to the and with SN 2002cx a truly strange SN views opening the sessions on the Chandrasekhar limit and two of these Ia has been added to the zoo. The lat- individual topics. The first day started will merge in about 4 Gyr (Napiwotzki). est news on SN 1987A was presented with the evolution of stars towards the The search for the companion stars af- in two presentations. The shock inter- final core collapse or the ignition of a ter they have lived through the violent action with the inner ring has become thermonuclear flame. The outcome of explosion in their vicinity has so far not quite apparent during the last few years massive star evolution depends on the yielded positive detections (Ruiz- in HST imaging (Kirschner). The light initial metallicity, which governs the Lapuente). In particular, Tycho’s super- curve is dominated by the ring emission stellar mass loss and hence the mass nova (SN 1572) and SN 1006 have and extracting the fading ejecta re- at the explosion, and during the early been investigated. The physics of the quires HST imaging (Suntzeff). The su- universe even very massive stars core collapse in massive stars was pernova itself is fading now very slowly. (up to 250 M) can explode as super- beautifully introduced (Janka). There The first tentative spectral identification novae (Heger). Direct observational might be light at the end of the tunnel of a supernova (SN 2001ek) in a evidence of the progenitor stars of with the first multi-dimensional calcula- Gamma-Ray Burst afterglow (GRB core-collapse supernovae is still hard tions yielding explosions when general 011121) triggered some discussion on to come by and, with the notable ex- relativity and new neutrino opacities are data quality and wishes for longer inte- ceptions of SN 1987A and SN 1993J, used in the models. The treatment of gration times (Kirschner). Densely only upper limits can be set on the the neutrino transport is key in these sampled light curves in UBVRI and the progenitor luminosity and initial mass models, a view also echoed by other infrared JHK bands for several nearby (Smartt). The finer details of the pro- experts (Burrows, Mezzacappa) but supernovae are now becoming avail- genitor evolution for SN 1987A still even more sophisticated models may able. With these excellent observations have to be worked out. Almost every- be needed to confirm the first positive the construction of bolometric light body agrees that there must have been results. Gross asymmetries appear in curves becomes a lot easier and the a binary system (Podsiadlowski). The some model calculations. The copious global properties of SNe Ia can be as- binary may have fully merged shortly production of neutrinos in the core col- sessed (Suntzeff). Core-collapse su- (104 years) before the explosion. The lapse might be observable with a suit- pernovae with very weak explosions

52 are observed more often and their gen- this circumstellar interaction of the su- guiding theme of the following presen- eral appearance can be investigated pernova shock, but optical spec- tations. The very first massive stars (Turatto, Pastorello). The ejecta mass troscopy can reveal the emission sites should explode in truly gigantic explo- of the weak explosions is still debated, through line shapes as well (Chevalier). sions even on the supernova scale. but higher masses appear to be Most regular SNe IIP show very little The chemical composition of the met- favoured (Zampieri). For these weak sign of interaction consistent with pro- al-poor stars might be dominated by objects, it might be possible to directly genitor masses in the range of 10 to 20 these explosions of the first generation see emission from the infall of matter M. Many other objects, especially of stars (Limongi). The observed abun- into a black hole at very late phases. stripped core-collapse supernovae (IIL, dances cannot be matched with the With large telescopes, including the Ib/c), are enshrouded in a dense envi- current models, but progress could be VLT, some supernovae can now be fol- ronment, which most likely resulted reported. The nucleosynthesis can also lowed for over a decade enabling one from the evolution in a close binary sys- be altered, if the explosion is asymmet- to look deep into the ejecta and explore tem. ric as the conditions for the burning are the heating mechanisms working in To date only 15 supernovae have not isotropic any longer (Maeda). Direct these cinders (Turatto). An interesting been observed in X-rays (Aschen- detection of the γ radiation from the ra- transition object between the subclass- bach). All of them are core-collapse dioactive decays would constraint the es Ib and Ic has been observed in a supernovae interacting with their cir- explosion models considerably. With Large Programme at the VLT (Hamuy). cumstellar material. Recently the first the past satellites this has been very The early adiabatic cooling from the X-ray spectra have been obtained difficult but the situation will improve af- shock breakout and the exquisite IR showing high ionization lines of oxygen ter the launch of INTEGRAL this fall spectroscopy firmly establish the core- through iron. The flux from SN1987A (Diehl). Decay lines from 56Ni and 57Co collapse nature of this object. The in- is increasing as the shock is inter- were observed from SN 1987A and a frared wavelength regime will draw acting inside the ring with dense, ion- tentative detection of the 56Co decay in much more attention in the near future. ized material. SNe Ia, on the other the SN Ia SN 1991T was reported. It offers the possibility to observe su- hand, are postulated to explode in a Other decays have been observed in pernovae with much less influence from rarefied environment and no interaction supernova remnants (44Ti in Cas A) and host galaxy absorption. The possibility is expected. However, in some progen- distributed throughout the Galaxy of an H-band Hubble diagram of SNe Ia itor models the presence of hydrogen (26Al). The connection of Gamma-Ray would make this fundamental measure- or helium has to be expected. Finding Bursts and supernovae may be inti- ment less dependent on assumptions the traces of such material has been mate, it certainly is intricate (Woosley). about absorption in the host galaxies the goal of a UVES programme (Lund- With sufficient angular momentum the (Phillips). At the same time, many new qvist). So far, only upper limits for the formation of a black hole can trigger the SNe Ia show light curves and spectral mass loss can be derived from the ab- formation of a disc formed of nucleons, evolutions that can not be fit into the sence of any detection of hydrogen or which can reverse the infall along the one-parameter relations used in the helium lines. On a grander scale, the poles into a strong jet explosion. past. The infrared also provides access supernova light can scatter off dust Depending on the viewing angle the to phenomena which are not easily ob- grains between the explosion and the observer will see a GRB, an inter- servable in the optical. The core-col- observer. There are now two SNe Ia mediate object like GRB 980425/SN lapse supernova SN 1998S is an in- and two core-collapse SNe (SN 1987A 1998bw or an unusual supernova. triguing case where the careful optical and SN 1993J) with observed light Depending on the conditions outside and infrared monitoring not only re- echos (Patat). Since supernovae are the disc the nucleons can form nuclei vealed very strong interaction with cir- the main producers of heavy elements, up to Ni and hence trigger a supernova cumstellar material, i.e. the remnant of they are of special interest for models (MacFadyen). The disc itself could be the wind of the progenitor star, but also of the chemical evolution of the Galaxy the site of an reprocess. The problem of an infrared excess (Meikle). Dust for- (Thielemann). this scenario is to produce stellar cores mation in the cool shell between for- The relative contributions of ther- with sufficient angular momentum for ward and reverse shock has been de- monuclear and core-collapse super- these models. A possibility is to have ei- duced from these observations. novae should sum up to the abun- ther a merger event or mass accretion To bridge the gap from explosion dances measured in the Sun. The de- onto a massive star (Joss). The obser- models to the observations the radia- pendences on various parameters in vational parameters of some of these tion transport has to be calculated. The the explosions (density, temperature, energetic explosions can vary from ob- comparison of the latest explosion and metallicity) are still not fully ex- ject to object (Mazzali). A range of ex- models from the group at the MPA with plored. It has been claimed recently plosion energies and masses has been observed light curves are very encour- that the chemical composition of the observed until now. The spectral signa- aging (Blinnikov). Indeed there appears most metal-poor stars could be domi- tures of wide, high-velocity lines of what to be enough energy in the deflagration nated by a single supernova. Hence, otherwise might appear as a SN Ic are models to explain the ejecta velocity of these most pristine stars are of particu- quite obvious. the Tycho supernova remnant. During a lar interest for the explosion models. Despite an intense programme there poster session the 28 posters on a wide The comparison of the evolution of in- was plenty of time for discussions. The range of topics were discussed. dividual elements in stars yields hints workshop was opened with a buffet din- Supernova searches and recent obser- about the contribution of different su- ner and a speech by the Mayor of vations of specific objects were pre- pernova types (Primas). Some of these Garching in the town hall. The first day sented. The latest ideas and details on stars may be the first ones formed in ended with a relaxed ‘Beer and modelling the explosions were dis- the galaxy, the putative Population III. Bretzen’ party outside the MPA. The played as well as plans for future proj- But also celestial objects picked up conference dinner took place at a typi- ects. from the ground can provide informa- cal Bavarian Biergarten outside Munich The last day was devoted to the in- tion on supernova enrichment. Inclu- during a wonderful summer evening. All teractions between other astronomical sions in meteorites give hints on the occasions to informally catch up on the fields and supernovae. Not all super- composition of the early solar system latest news and developments. novae explode in isolation. Some of and material ejected from nearby su- The proceedings of this conference them strongly interact with their close pernovae (Ott). will appear in the ESO Astrophysics environment. The radio and X-ray A discussion on the nomenclature of Symposia Series published by Springer emission are the most direct tracers of the most energetic explosions was the Verlag.

53 Developing 3D Spectroscopy in Europe J.R. WALSH, ESO M.M. ROTH, Astrophysikalisches Institut Potsdam

1. Background es nor the problems associated with at- the 3D observations of the stellar kine- mospheric refraction which plague con- matics of the nuclear region of the near- One of the inherent preoccupations ventional spectroscopy. Most impor- by galaxy M31. Whilst the distribution of of astronomy is to obtain a three-di- tantly, there is, by definition, no preju- light shows a double nucleus, the ve- mensional view of the Universe and its dice as to the selection of the region of locity map reveals a structure that is not components. Except for Solar System interest: thus a particular slit pointing aligned with the two nuclei and the objects, we are always presented with and orientation may easily miss a phys- peak in the velocity dispersion, as- a two-dimensional view of celestial ob- ically important structure which ap- sumed to mark the location of the su- jects. Spiral galaxies for example could pears to be unassuming on a broad- permassive black hole, is also offset be considered only as flat structures if band image. Integral-field observation from one of the peaks (Bacon et al. it were not for the rotational velocity can unveil the complete spectroscopic 2001a). Orientation of a long slit over which shows them to be spinning three- signature within the 2-dimensional field the obviously visible features would dimensional entities. The distance of view of the wavelength coverage of have missed the essentials of the ve- scale is another fundamental aspect of the 3D instrument. A good example is locity structure. this question – placing astronomical sources in the third dimension. Once their distance is determined, physical parameters can follow such as luminosity, radius and mass. In order to determine this information one could ideally imagine a “maximal spectrograph’’ which produced the spectrum of the whole sky at some desired spectral resolution and spatial sampling on the sky. The complexity of such an instrument is obviously beyond current technological means and the sheer size of the resulting data set would be prohibitively large. Nevertheless, a small, but significant, step towards this goal is to obtain the spectrum of an area of sky and this is what 3D spectroscopy achieves. With advances in technology the sampled area is becoming bigger. 3D spectroscopy is called integral field (IFS) or area spectroscopy and the principle is summarized in Figure 1. The resulting data have three dimensions – two spatial and one spectral. The spectrum at a spatial pixel (dubbed a “spaxel’’), or in an aperture of any desired shape over a substructure of interest, can be extracted or an image over a spectral range can be formed by summing in the spectral direction. A long-slit spectrum can be formed by slicing the 3D data in one spatial and the spectral direction. Such data have very powerful advantages over aperture or long-slit spectra which sample pre-defined spatial regions. With 3D spectroscopy not only can spectra of a whole extended object be obtained, such as a nearby elliptical galaxy to investigate its velocity field, but areas of the sky can be searched for objects which are difficult to detect in wide- band imaging, such as emission line sources with a few, even only one, vis- Figure 1: The principle of 3D spectroscopy is illustrated. An area of interest on the sky is sam- ible line over the wavelength range of pled on a discrete grid of spatial elements, using techniques such as a fibre bundle, lenslet the instrument (such as a search for array, slicer, etc. The light collected in each spatial element is dispersed into a spectrum. After Lyman-alpha emission from very high- extracting the family of individual spectra from the detector image, they can be rearranged to redshift galaxies). form a data cube. Spectra of individual spatial elements, or defined regions on the sky, can Since there is no slit in the conven- be extracted from the data cube. Slicing the data cube over a range of wavelengths allows a tional sense, there are neither slit loss- narrow-band or broad-band image of the sky to be formed.

54 The process of data-taking sounds eraged. The TAURUS instrument, used some of them are still in use today. The simple – point at a target of interest at many 4-m telescopes, was the most application of microlens arrays to as- (high-precision pointing is not re- advanced realization (Atherton et al., tronomy brought a revolution in this quired), obtain spectra at many spatial 1982) and emission line maps of many field. An area of sky could be divided up positions (currently hundreds to thou- extended targets were observed. by a monolithic microlens array. The sands). The removal of the instrument Photon-counting detectors could beams from the microlenses could then signature and the assembly of the data also be employed in rapid slit-scanning be fed to a spectrometer and many into a 3D data cube proceeds similarly techniques where the positioning of a spectra recorded on the same detector. to spectroscopic reduction with long long slit on the sky was synchronized The spectrometer design can ensure slits except for the much larger volume with the readout of the detector. The that the many individual spectra are of data. However, it is the analysis of ASPECT system at the AAT (Clark et al. packed on the detector so that there is those thousands of spectra which pro- 1984) using the IPCS (Boksenberg & minimal overlap. The Tiger, subse- vides the greatest hurdle. Integral-field Burgess 1973) was successfully used quently Oasis, instruments used for spectrometers in various forms have for a number of projects from kinematic many years on the CFHT was the most been available for decades but the pub- mapping of elliptical galaxies to spatial successful example of this design prin- lications resulting have in no way been abundance mapping of spiral and star- ciple and much science was achieved proportional to their data volume, or al- burst galaxies. The data volumes were from resolving the kinematic compo- located telescope time. The sheer modest with typically ten long slit posi- nents of galaxy nuclei to the jet struc- scale of the data analysis and the need tions. Scanning techniques suffer from ture of PMS stars (Bacon et al., 1995). to do justice to the quantity of spectra changing seeing and transparency, Using the micropupil principle, the cou- has deterred many, and even the 3D which also produce line profile varia- pling of lens arrays with fibre bundles spectroscopy pundits have to admit tions for Fabry-Perot spectrometers. allowed more flexible designs even with that they cannot analyse their data The first attempts to measure simul- several spectrometers. The integral cubes fast enough. The lack of ade- taneously spectra over a 2-dimensional field mode of the Gemini GMOS instru- quate data-analysis tools is becoming field were made in the 1980’s with fibre ment uses this design, as does the VLT more acute with the installation of bundles packed into an area at the tel- FLAMES facility; in VIMOS, currently new common-user instruments offer- escope focal plane and aligned onto a the largest IFU unit in operation (80 × ing IFS modes on 8–10-m-class tele- common “pseudo-slit’’ of a convention- 80 elements), the fibres feed four spec- scopes, such as VIMOS, FLAMES and al spectrometer. Each fibre generated a trometers. There is no reason in princi- SINFONI at the VLT, and GMOS at single spectrum on the detector of one ple for not extending the number of GEMINI. position on the sky. Several prototype spatial elements towards that of the In order to try to ease this “data jam’’, instruments have been developed, and maximal spectrometer and two propos- all the European groups working in 3D spectroscopy came together in a working group launched by OPTICON – the Optical and Infrared Coordination Network for Astronomy. A proposal for a OBITUARY GUILLERMO DELGADO Research Training Network (RTN) in the 5th Framework of the European (1961–2002) Commission was made in which young post-docs would be enabled to work on Guillermo started his career in radio astronomy in 1983 when he participat- science projects with 3D spectroscopy. ed in the activities and development of the Maipu Radio Astronomy User tools would be developed and Observatory. shared to increase the scientific ex- In 1986, at the time when Onsala Space Observatory and ESO embarked ploitation and productivity of the data. on the SEST project, Guillermo went to Sweden to become involved in the proj- The RTN, entitled “Promoting 3D ect. Upon his return to Chile he took an active part in the commissioning of the Spectroscopy in Europe” was awarded SEST. His contribution helped greatly in the early readiness of the SEST op- and began on 2002 July 1. Post-docs erations. Later on, he was responsible for the instrument maintenance and up- are now being sought in ten European grades, including the installation of receivers and the design of their control institutes. This article provides a brief system. overview of the 3D spectroscopy and a In 1989 Guillermo graduated as Electrical Engineer from the Universidad de flavour of what can be expected from Chile. In 1992 he went to Sweden to work on his doctorate at Chalmers the RTN over the next few years. University of Technology. He returned to Chile and the SEST in 1995 where he completed his Ph.D. thesis. 2. Growth of 3D Spectroscopy In 1997, health problems obliged him to be based in Santiago, where he was, essentially, dedicated to support the ALMA site testing and development cam- The first attempts at imaging spec- paign, including the maintenance of its equipment, the interpretation of the at- troscopy used scanned Fabry-Perot in- mospheric transparency data from Chajnantor and the modelling of phase cor- terferometers to observe the velocity rection methods. His efficient work under difficult health conditions was im- fields of emission lines in gaseous neb- pressive. Above all, Guillermo had an extraordinary capacity to handle a large ulae. Groups at Marseille and Man- spectrum of skills, ranging from electronic designs and opto-mechanics to soft- chester used photographic and image- ware development and data analysis. He taught at the Universidad de Chile tube recorders to obtain multiple nar- and encouraged students towards the world of astronomy. ESO and Onsala row spectral band maps which, when Space Observatory are deeply grateful for his contribution to ALMA and the stacked, allow the line profiles over an SEST. area to be mapped. With the advent of Guillermo was a first-rate colleague whose generosity and dedication was piezo-scanning Fabry-Perot spectrom- highly appreciated by all who worked with him. Until his very last weeks at the eters coupled with photon-counting de- hospital he was actively involved in his tasks with an energy and will power only tectors, rapid sampling of the spectral to be defeated by his physical condition. range could be achieved. The effect of Our expression of condolence goes to his wife, Alejandra, and their sons. transparency variations in the atmosphere would be reduced by the fast DANIEL HOFSTADT scanning and many scans could be av- LARS-ÅKE NYMAN

55 als for VLT wide-field (1 × 1′) 3D optical and many of the instruments currently Max-Planck-Institut für Extraterrestri- spectrometers are under consideration in use, or planned, are for telescopes in sche Physik, Garching, Leiden Ob- (see Monnet, 2002). which European institutes, including servatory, CRAL Observatoire de Lyon, One other technique has found ap- ESO, have strong participation. The Laboratoire d’Astrophysique de Mar- plication for dividing up the field subse- need to foster good communication and seille, Istituto de Fisica Cosmica “G. quent to feeding the spatial elements to interchange between these groups, Occhialini” of the Italian CNR in Milan, a spectrometer and that is a develop- which represent all the different 3D Observatoire de Paris section de ment of the image slicer. Originally im- methods sketched above, led to the for- Meudon, Instituto de Astrofísica de age slicers were used to increase the mation of an OPTICON 3D Spec- Canarias, ESO – all of which have throughput of slit spectrometers for a troscopy Working Group. This group active involvement in 3D spectros- point source by stacking slices along a identified that, whilst individual instru- copy projects. Full details of the RTN narrow slit. Applied in two dimensions ments are diverse and the responsibili- are available on the Web at: mirrors can be used to reformat a ty for removal of the instrument signa- http://www.aip.de/Euro3D/ and there square field onto a long slit which is ture must rest with the instrument are also links to detailed descriptions of then packed on to the detector (e.g. for builders, there existed a lack of instru- the 11 3D instruments with which the the MPA instrument 3D, Weitzel et al., ment-independent data analysis soft- RTN members are involved. The coor- 1996, which is the fore-runner of the ware. The Euro3D RTN was proposed dinator of the network is Martin Roth at VLT instrument SPIFFI). In common and planned by this group. AIP Potsdam ([email protected]) and with all methods the limitation is detec- A 3D data format for the exchange of questions about participation or interest tor area, and as CCDs have grown larg- 3D data and a software platform for the in the scientific or software activities er so have the areas encompassed by development of analysis tools form two should be directed to him. 3D instruments, whilst the sampled size of the cornerstones of the Euro3D ef- on the sky has remained relatively con- fort. A draft format for a Euro3D format References stant. A survey of the commissioned has been issued and the essence of the and planned 3D instruments, or instru- format, which is FITS, is a stacked Atherton, P.D., Taylor, K., Pike, C.D. et al., 1982, MNRAS, 201, 661. ments with an integral-field capability, spectrum image with a table to refer- Bacon, R., Emsellem, E., Combes, F., et al., around the world showed the astonish- ence each spectrum to its position on 2001a, A&A, 371, 409. ing number of 26. Truly, this is a bur- the sky plane. For the data analysis Bacon, R., Copin, Y., Monnet, G., et al. (The geoning field and many integral field in- tools, it was decided to write individual Sauron Project), 2001b, MNRAS, 326, 23. struments are planned for the large tel- applications in C and to use a scripting Bacon, R., Adam, G., Baranne, A., et al., escopes, and for NGST. Within two language such as Python, Tcl/Tk or IDL 1995, A&AS, 113, 347. years there will be three IFU-capable for analysis scripts. The I/O library Boksenberg, A. Burgess, D. E., 1973, in instruments on the VLT – VIMOS, would be adapted from the extensive Proc. Symp. TV type sensors, eds. FLAMES and SINFONI. Lyon Oasis libraries for the Euro3D for- Glaspey, J. W., Walker, G.A.H., 21. Clark, D., Wallace, P., Fosbury, R. F., Wood, mat. R., 1984, QJRAS, 25, 114. 3. The Euro3D RTN The RTN consists of a network of Monnet, G., 2002. The Messenger, No. eleven institutes – Astrophysikalisches 108, 2. Europe currently has the lead in the Institut Potsdam, Institute of Astronomy Weitzel, L., Krabbe, A., Kroker, H., et al., development of integral-field devices Cambridge, University of Durham, 1996, A&AS, 119, 531.

Forty Years ESO – Public Anniversary Activities C. MADSEN, ESO

Readers of The Messenger will be taria to ensure that the 40th anniver- well aware of ESO’s 40th anniversary sary will also become a visible event in in October this year. This is most cer- the public arena. tainly a reason for ESO’s friends to cel- Among the activities and products ebrate. Beyond the professional as- that will be ready in the autumn are a tronomers, engineers and other people major book, a CD-ROM, a video and a with direct links to the organization, this planetarium show. includes many people all over Europe, Geheimnisvolles Universum – Euro- is a forward-looking book, dealing with e.g. amateur astronomers, science pas Astronomen entschleiern das Welt- some of the hottest topics of contem- teachers, and people with a general in- all is the title of an ‘anniversary book’, porary astronomy. It looks at the tech- terest in science. At the same time, the published by the German Kosmos nology and, by means of interviews with European Intergovernmental Research Verlag. Comprising 208 large-format both ESO staff and visiting observers, Organizations constitute fine examples pages with more than 150 colour pic- offers a very human account of the peo- of how, through collaboration, European tures, the book is written by science ple now engaged in front-line astro- countries can interact and achieve am- journalist Dirk Lorenzen, known from nomical research. A chapter describes bitious goals. Thus the success of ESO German Radio and author of Deep an observing night at Paranal and pro- reaches beyond the confines of profes- Space and Raumsonde Galileo. The vides an outlook to the future, in par- sional Astronomy. original version of the book is published ticular the ALMA and the OWL projects. Taking account of this, ESO’s Edu- in German, but other language versions For readers who are interested in the cation and Public Relations Depart- are planned. past, the book includes a chapter with ment has worked intensively with part- Although published on the occasion some historical aspects and milestones ners in the publishing world and plane- of the 40th anniversary of ESO, the book in ESO’s development, and a fascinat-

56 video clips from Paranal, the CD-ROM also contains spectacular 3D images of astronomical objects, thanks to a unique rendering software developed by Planetary Visions. The CD-ROM will initially be available in English and German, with a French version under preparation. On the background of the status of post-World War II astronomy in Europe, the video Europe Reaches for the Stars – Forty Years ESO traces the evolution of ESO, from the humble beginnings until today – with the VLT in full scientific operation and VLTI in the development stage – and projects the current successes into the future. The 50-minute film includes rich historic footage not shown before and interviews with the past directors general. Scientific highlights are exposed, both in the context of the general development of Astronomy and the research activities at ESO. The film is produced by ESO’s EPR department and will initially be available with English narration. Finally, a planetarium show with the title Les mystères du ciel austral has been developed in collaboration with APLF, the association of French-language planetaria and with the help of its sister association in the German- speaking countries (ADP). Prof. Agnès Acker from the Louis Pasteur University of Strasbourg and Marc Moutin, head of the planetarium at the Cité de l’Espace in Toulouse, have been the driving forces behind the project, with the technical preparation of the show being ex- ecuted by Master Image Group of France. This show, which focuses strongly on ing description of the early search for Dirk Lorenzen is also the author of the VLT and the recent scientific re- sites by the German astronomer Jür- the CD-ROM entitled 3D Atlas of the sults, is initially produced in two ver- gen Stock, whose work played a key Universe, which is produced in a col- sions, customized for France and role as the ESO Council took its initial laboration between ESO, USM (Mu- Germany, and other language versions decision about where to locate the ob- nich) and Planetary Visions (University may be produced. servatory. College, London). Apart from audio and

First Teachers Training Course at ESO HQ was a Great Success A. BACHER, R.M. WEST, ESO

On August 20–24, 2002, School teachers went through three of the four Planets. Results taken at the Leonhard Teachers from a dozen different Euro- ESA/ESO Astronomy Exercise Series Euler Swiss Telescope at La Silla by the pean countries (including eastern coun- (see The Messenger No. 107, March group of Michel Mayor were elaborated tries) came to ESO HQ to learn about 2002), trying out different methods of in ways that students of different ages recent developments at ESO. The determining astronomical distances. can understand. This included the de- training course called FAST2002 (Fron- The fourth, major workshop aimed at termination of planetary parameters tline Astrophysics for School Teachers) creating new exercises. Two different and how to judge if life would be possi- consisted of several lectures and work- topics were discussed in great detail. ble on the planet (concept of “habitable shops. One was to determine properties of a zone”). The lectures were given by ESO Transneptunian Object using six differ- In addition there was a poster ses- astronomers and dealt with ESO in ent images, kindly provided by Olivier sion, where the participants presented general, VLTI, OWL, and Science at ESO. Hainaut (ESO La Silla). projects about their own educational During the first three workshops, the The other topic was about Extrasolar work.

57 Veselka Radeva from Bulgaria made the following statement, when she was asked about her impression of this course: “Excellent organization, wonderful presentation of the observational possibilities of ESO, excellent work on the existing exercises and efficient cre- ative work for the invention of new exercises by an excellent group of teachers. Thank you very much!” After this good start, the ESO Edu- cational Office now looks forward to organizing more teacher training courses in the next years.

Participants in front of the ESO Headquarters building in Garching. Photo taken by Gian Nicola Cabizza.

ANNOUNCEMENTS

STRUCTURE EVOLUTION STELLAR CANDLES FOR AND COSMOLOGY: THE EXTRAGALACTIC New synergy between ground-based DISTANCE SCALE observations, space observations An international Workshop to be held at the and theory Universidad de Concepción, Chile, on An international workshop to be held at December 9–11, 2002 ESO/Santiago, Chile, on October 28–31, 2002 Sponsoring Organizations: Sponsoring Organizations: CONICYT/FONDAP Institute for Astrophysics, Chile; European European Southern Observatory (ESO); Centre National d’Etudes Southern Observatory; Fundación Andes; Universidad de Spatiales (CNES); Commissariat a l’Energie Atomique CEA); Concepción, Chile DAPNIA/Service d’Astrophysique (SAp) Organizing Committee: Scientific Rationale: D. Alloin, ESO (Co-chair); P. Fouqué, Paris; With the upcoming of the new generation of powerful wide-field in- D. Geisler, Concepción; W. Gieren, Concepción (Co-chair); struments (XMM, Megacam, VIRMOS, Integral, SIRTF, GALEX, G. Pietrzynski, Concepción; T. Richtler, Concepción VLA, Omegacam/VST, VISTA...), the first decade of the XXIst century is to open a decisive era in the study of large-scale structure for- Rationale of the workshop: mation. The past decade has seen a huge effort to improve the cali- These observational developments are being complemented by bration of the extragalactic distance scale. Stellar methods of considerable numerical and semi-analytical advances. The work- distance determination are used to measure the distances to shop aims to bring together groups closely involved in carrying out nearby galaxies, setting the zero point of the extragalactic dis- and coordinating ground-based and space surveys with efforts made tance scale. Yet, comparison of the results from a variety of in modeling the formation of structures. An important point will be the stellar standard candles shows that there are significant sys- optimization of observing strategies and science returns in the con- tematic uncertainties attached to most, if not all stellar methods text of th forthcoming Virtual Observatory. First results from various of distance measurement, preventing a truly accurate calibra- on-going programmes will be presented. Attendance by young re- tion of the distance scale. This workshop will bring together searchers (students and postdocs) is most welcome. In this respect, leading experts on the most prominent stellar standard candles a half-day cosmology introductory session will be given. including Cepheid variables, RR Lyrae stars, Type Ia supernovae, blue supergiants, planetary nebulae, novae and globu- Scientific Organizing Committee: lar clusters to explore their current usefulness for the calibra- M. Birkinshaw (Bristol), R. Ellis (Caltech), M. Kamionkowski tion of the distance scale, and for putting constraints on the (Caltech), C. Lonsdale (Caltech/IPAC), M. Pierre (CEA), A. Hubble constant as a fundamental cosmological parameter. Refregier (Cambridge), J. Silk (Oxford), S. White (MPA). Special attention will be given to improve our understanding of systematic uncertainties in the various methods of distance Local Organizing Committee: measurement, and in designing strategies to reduce these uncertainties in the near future. D. Alloin (ESO), R. Cabanac (ESO), H. Quintana (PUC), J. Willis (PUC). More details can be found at: [email protected] More details are available at: http://www.eso.org/cosmology2002 http://cluster.cfm.udec.cl

58 A meeting on ESO VACANCY The Education and Public Relations Department (EPR) in the Science Operations Office of the Director General at the ESO Headquarters in Gar- with the Atacama Large ching near Munich, Germany, offers the following job opportunity. EDITOR (EDG 604) Millimeter Array Assignment: Within the ESO Education and Public Relations De- will take place at ESO, Garching bei München, partment team, your main tasks and responsibilities will comprise: • Development, update and maintenance of the comprehensive on Friday, November 8, 2002, ESO Outreach website in its new look, including preparation of from 9 a.m. to 4 p.m. related material (texts, images, etc.) to be displayed; The Atacama Large Millimeter Array (ALMA) project is an interna- • Design, layout and production of the ESO quarterly journal “The tional collaboration between Europe and North America to build Messenger” (e.g. image selection and processing, technical ed- an array of telescopes that will operate at millimetre and submil- iting, etc.), in close collaboration with the Messenger editor; limetre wavelengths at the high-altitude (5000 m) Chajnantor site • Conception and production of promotional brochures, posters and in Chile. It reached a critical milestone this summer when on July other EPR products, in close collaboration with the Head of the 9, the ESO Council approved the European participation through Education and Public Relations Dept. of ESO Press Releases and ESO in the bilateral project. On August 16, the US National various high-level publications, including the ESO Annual Report. Science Board authorized the US share of the ALMA construction. Depending on qualification, expertise, and personal interest, you may utilise up to 20% of your time to conduct scientific research. The aim of this one-day meeting is to provide an overview and update to the European astronomy community of the ALMA project Education: University degree in astronomy, physics, general sci- as it enters the construction phase, and to solicit input from the ence, scientific journalism or equivalent. community on science operations and user support for ALMA. Knowledge and Experience: The successful candidate should Topics will include: combine a strong interest in science communication with a good • Overview of project and project status; knowledge of modern astronomy. The ideal candidate should have desktop publishing experience, Web related abilities, and be con- • Major science drivers; ALMA as a complement to facilities at versant with a range of text and image processing software, and at other wavelengths; least one major data reduction package such as MIDAS, IRAF or • Science operations plan; IDL. Excellent communication skills, a very good command of the • Regional Support Centres: core and additional functions; English language, and the ability to fit into a small and active team •Toward a European Regional Support Centre; are imperative, and knowledge of other European languages would • Open discussion be an asset. More details on ALMA can be found in the March 2002 issue of Duty station: Garching near Munich, Germany The Messenger, see http://www.eso.org/projects/alma/info/ Starting date: as soon as possible brochure.pdf, and the ALMA web site http://www.eso.org/projects/ Contract: This appointment is based on a fixed-term contract of alma/ three years with the possibility of extension or permanence. To register for the meeting and obtain further information, please Remuneration: We offer an attractive remuneration package in- send a message to Samantha Milligan ([email protected]) by cluding a competitive salary (tax-free), comprehensive social ben- October 5. There will be no registration fee. A second announce- efits, and provide financial support in relocating families. ment will be distributed to participants in October. Applications consisting of your CV (in English language) and the Organizing Committee: ESO Application Form (to be obtained from the ESO Home Page at http://www.eso.org) should be submitted by 12 October 2002. R. Bachiller, A. Benz, R. Booth, P. Cox, E.F. van Dishoeck, S. For further information, please consult the ESO Home Page or contact Ms. Guilloteau, R. Kurz, J. Richer, P. Schilke, P. Shaver, M. Walmsley. Angelika Beller.

GRAZIAN, Andrea (I), Student PERSONNEL MOVEMENTS KIM, Tae-Sun (ROK), Fellow PANCINO, Elena (I), Student International Staff RIVINIUS, Thomas (D), Fellow (1 July 2002 – 30 September 2002) STOLTE, Andrea (D), Associate CHILE ARRIVALS ATHREYA, Ramana (IND), Fellow EUROPE BÖHNHARDT, Hermann (D), Data Flow Operations Astronomer BLONDIN, Stéphane (F), Student HUTSEMEKERS, Damien (B). Operations Staff Astronomer ESCHBAUMER, Siegfried (D), Infrared Laboratory Technician JONES, David Heath (AUS), Fellow IVANESCU, Liviu (R), Assembly Integration and Testing RABELING, David (NL), Associate Engineer RATHBORNE, Jill (AUS) Associate SEST LEONI, Marco (I), Astrophysical Virtual Obs. Archive Software Engineer MENGEL, Sabine (D), User Support Astronomer Local Staff MOTTINI, Marta (I), Student SCALES, Kevin (USA), Optical/Electrical Engineer (1 June 2002 – 30 June 2002) THILLERUP, Jesper (DK), Electronics Technician VAN DEN ANCKER, Mario (NL), User Support Astronomer ARRIVALS WEGERER, Stefan (D), Mechanics Technician CERDA HERNANDEZ, Susana, Telescope Instrument WERNER, Daniela (D), Associate Operator PREMINGER HEYUM, Daisy, Data Handling Administrator CHILE DALL, Thomas (DK), Fellow NÜRNBERGER, Dieter (D), Fellow CORRIGENDUM DEPARTURES The Horsehead Nebula on page 34 of the June 2002 issue of The Messenger (No. 108) was erroneously attributed to the ESO/ EUROPE MPG 2.2-m Telescope. BRANDNER, Wolfgang (D), AO Instrument Scientist The photo was produced from three images, obtained with the CHADID, Merieme (MA), Fellow FORS2 multi-mode instrument at the 8.2-m KUEYEN telescope at DEVILLARD, Nicolas (F), Astronomical Data Reduction Paranal. The images were prepared by Cyril Cavadore (ESO- Specialist ODT), by means of Prism software.

59 ESO, the European Southern Observa- tory, was created in 1962 to “... establish Contents and operate an astronomical observatory in the southern hemisphere, equipped Catherine Cesarsky: ESO Turns 40 ...... 1 with powerful instruments, with the aim of furthering and organising collaboration ESO COUNCIL MEETING IN LONDON, 8–9 JULY 2002 in astronomy...” It is supported by ten countries: Belgium, Denmark, France, Speeches to Mark the Accession of the UK to ESO Germany, Italy, the Netherlands, Portu- Lord Sainsbury, UK Science Minister ...... 2 gal, Sweden, Switzerland and the United Dr. Arno Freytag, President of the ESO Council ...... 3 Kingdom. ESO operates at two sites in Dr. Catherine Cesarsky, ESO Director General ...... 3 the Atacama desert region of Chile. The Gerry Gilmore: ESO and the UK. Why Does the UK Need More Astronomy? . 4 new Very Large Telescope (VLT), the largest in the world, is located on ESO TURNS 40 Paranal, a 2,600 m high mountain approximately 130 km south of Antofa- Perspectives from the Directors General, Past and Present gasta, in the driest part of the Atacama Adriaan Blaauw: Reflections on ESO, 1957–2002 ...... 7 desert where the conditions are excellent Lodewijk Woltjer ...... 9 for astronomical observations. The VLT Harry van der Laan: From Sest to ALMA, from NTT to OWL: consists of four 8.2-metre diameter tele- Of Vision, Dreams and Realities ...... 10 scopes. These telescopes can be used separately, or in combination as a giant Riccardo Giacconi ...... 11 interferometer (VLTI). At La Silla, 600 km Catherine Cesarsky ...... 11 north of Santiago de Chile at 2,400 m Some Snippets of History ...... 12 altitude, ESO operates several optical telescopes with diameters up to 3.6 m TELESCOPES AND INSTRUMENTATION and a submillimetre radio telescope (SEST). Over 1300 proposals are made L.-Å. Nyman, P. Schilke and R.S. Booth: The Atacama Pathfinder Experiment 18 each year for the use of the ESO tele- O. Le Fèvre, D. Mancini, M. Saïsse, S. Brau-Nogué, O. Caputi, L. Castinel, scopes. The ESO headquarters are lo- S. D’Odorico, B. Garilli, M. Kissler, C. Lucuix, G. Mancini, A. Pauget, cated in Garching, near Munich, Ger- G. Sciarretta, M. Scodeggio, L. Tresse, D. Maccagni, J.-P. Picat, many. This is the scientific, technical and G. Vettolani: VIMOS Commissioning on VLT-Melipal ...... 21 administrative centre of ESO where tech- L. Germany: 2.2-m Team ...... 27 nical development programmes are carried out to provide the Paranal and La Silla observatories with the most ad- REPORTS FROM OBSERVERS vanced instruments. There are also ex- M. McCaughrean, H. Zinnecker, M. Andersen, G. Meeus, N. Lodieu: Standing tensive astronomical data facilities. ESO employs about 320 international staff on the Shoulders of a Giant: ISAAC, Antu, and Star Formation ...... 28 members, Fellows and Associates in L. Kaper, A. Castro-Tirado, A. Fruchter, J. Greiner, J. Hjorth, E. Pian, Europe and Chile, and about 160 local M. Andersen, K. Beuermann, M. Boer, I. Burud, A. Jaunsen, B. Jensen, staff members in Chile. J.M. Castro-Cerón, S. Ellison, F. Frontera, J. Fynbo, N. Gehrels, J. Gorosabel, J. Heise, F. Hessman, K. Hurley, S. Klose, The ESO MESSENGER is published C. Kouveliotou, N. Masetti, P. Møller, E. Palazzi, H. Pedersen, four times a year: normally in March, L. Piro, K. Reinsch, J. Rhoads, E. Rol, I. Salamanca, N. Tanvir, June, September and December. ESO P.M. Vreeswijk, R.A.M.J. Wijers, T. Wiklind, A. Zeh, E.P.J. van den Heuvel: also publishes Conference Proceedings, Gamma-Ray Bursts: the Most Powerful Cosmic Explosions ...... 37 Preprints, Technical Notes and other ma- R.E. Mennickent, C. Tappert, M. Diaz: Cataclysmic Variables: Gladiators terial connected to its activities. Press Releases inform the media about partic- in the Arena ...... 41 ular events. For further information, con- L. Wang, D. Baade, P. Höflich, J.C. Wheeler: Supernova Polarimetry with the tact the ESO Education and Public VLT: Lessons from Asymmetry ...... 47 Relations Department at the following address: OTHER ASTRONOMICAL NEWS EUROPEAN E. Mason and S. Howell: An Exciting Working Session on Cataclysmic SOUTHERN OBSERVATORY Variables at ESO/Santiago ...... 51 Karl-Schwarzschild-Str. 2 W. Hillebrandt and B. Leibundgut: Conference Summary: From Twilight to D-85748 Garching bei München Highlight: the Physics of Supernovae. ESO/MPA/MPE Summer Workshop Germany 2002 ...... 52 Tel. (089) 320 06-0 J.R. Walsh and M.M. Roth: Developing 3D Spectroscopy in Europe ...... 54 Telefax (089) 3202362 [email protected] (internet) D. Hofstadt, L.-Å. Nyman: Obituary Guillermo Delgado (1961–2002) ...... 55 URL: http://www.eso.org C Madsen: Forty Years ESO – Public Anniversary Activities ...... 56 http://www.eso.org/gen-fac/pubs/ A. Bacher and R.M. West: First Teachers Training Course at ESO HQ was messenger/ a Great Success ...... 57

ANNOUNCEMENTS The ESO Messenger: Editor: Peter Shaver Workshop on “Structure Evolution and Cosmology: New synergy between Technical editor: Kurt Kjär ground-based observations, space observations and theory” to be held at ESO/Santiago ...... 58 Printed by Workshop on “Stellar Candles for the Extragalactic Distance Scale” to be held Universitätsdruckerei WOLF & SOHN at the Universidad de Concepción ...... 58 Heidemannstr. 166 Meeting on “Science Operations with the Atacama Large Millimeter Array” D-80939 München to be held at ESO, Garching ...... 59 Germany ESO Vacancy: Editor (EDG604) ...... 59 Personnel Movements ...... 59 ISSN 0722-6691 Corrigendum ...... 59