JCMT SPECTRUM
N E W S L E T T E R O F T H E J A M E S C L E R K M A X W E L L T E L E S C O P E AUTUMN 2007 • #27
Legacy of SCUBA cosmology, maps, and catalogs
SCUBA Legacy Science Results JCMT Spectrum, Newsletter of the In This Issue James Clerk Maxwell Telescope, is published biannually and is edited by Gerald Schieven and Jonathan Kemp. From the Desk of the Director (Davis & Chrysostomou) ...... 3 ISSN 1351–5497. Copyright © 2007 by Science and Technology Facilities JCMT Science Council, Joint Astronomy Centre, Hilo, except where copyright re- SHADES: A Cosmological SCUBA Legacy (Dunlop) ...... 6 tained by named individual article or The SCUBA Legacy Maps and Catalogues image authors. SSUE (Di Francesco, Johnstone, Kirk, MacKenzie, & Ledwosinska ) ...... 8 I postal address
HIS Extreme Activity in a Biased Environment: James Clerk Maxwell Telescope An AzTEC 1.1 mm Survey (Geach) ...... 10 Joint Astronomy Centre T University Park N I The Detection of a Molecular Bipolar Flow 660 North A`ohoku Place in the Multipolar Planetary Nebula NGC 2440 (Hasegawa, Wang, & Kwok) ...... 12 Hilo, Hawai`i 96720–2700 USA Mapping Magnetic Fields in Planetary Nebulae and Post–AGB Stars with SCUBA (Sabin, Zijlstra, & Greaves) ...... 13 telephone +1 (808) 961 – 3756 CN Emission in the Sgr A* Environment — First Results (Stanković, Seaquist, Mühle, & Fuente) ...... 14 facsimile Mid–Infrared Spectral Diagnosis of Submillimetre Galaxies +1 (808) 961 – 6516
(Pope, Chary, Alexander, Armus, Dickinson, Elbaz, Frayer, Scott, & Teplitz) ...... 15 world wide web http://www.jach.hawaii.edu/JCMT/
JCMT Operations The Joint Astronomy Centre provides services and support to enable visit- 2 Starlink Software Collection: Puana Release (Cavanagh) ...... 4 ing and staff astronomers to under- First Light with the eSMA at 345 GHz (Bottinelli) ...... 5 take top-quality, front-line interna- tional-class research using the James The Night Shift (Warrington) ...... 11 Clerk Maxwell Telescope (JCMT) and the United Kingdom Infrared Tele- scope (UKIRT); to develop these fa- cilities in order to maintain their position as the most advanced of their kind in the world; to operate them in the most cost effective and efficient manner on behalf of the funding agencies; and to be respon- sive to the changing needs of the contributing organizations.
The JCMT is supported by the United Kingdom’s Science and Technology Facilities Council (STFC), the Na- tional Research Council Canada (NRC), and the Netherlands Organi- 2007 • #27 zation for Scientific Research (NWO); it is overseen by the JCMT Board.
UTUMN The JCMT is a member of the On the front cover: 850 μm SCUBA images of the supernova remnant Cas A (upper right), the Whirlpool RadioNet consortium. Galaxy M 51 (lower left), and the massive star forming On the rear cover: (top) eSMA: JCMT, CSO, & SMA region NGC 6334 (center). All of these images (and interferometric array. Courtesy Inge Heyer. (bottom) • A many more) are available for public download from the First fringes with the eSMA on baselines between one CADC web site. SHADES SCUBA 850 μm images of the SMA antenna (“6”), the JCMT (“10”), and the CSO (“9”). SXDF (upper left) and Lockman Hole (lower right) fields, The SMA was in its extended configuration yielding indicating the positions of the 120 most secure eSMA baselines up to ~450 m (baseline 6–9, lower sources. (Also see Figure 1 and article on page 6 of panel), about twice as long as the longest SMA baseline this issue and Figures 1, 2, and 3, and article on page (~225 m). (Also see Figure 1 and article on page 5 of PECTRUM 8 of this issue.) this issue.)
JCMT S JCMT JCMT S
From the Desk of the Director PECTRUM
Professor Gary Davis (Director JCMT) & Antonio Chrysostomou (Associate Director JCMT)
• A
Welcome to the summit pipeline. Indications thus far strument is
autumn 2007 are that the observing modes work scheduled to be UTUMN edition of Spec- well, although bad baselines are cur- shipped to Ha- trum. This edi- rently holding back the teams from waii in late tion contains a starting in earnest. This issue has 2007, with inte- 2007 • #27 number of excit- evolved to top priority for the Instru- gration and ing science arti- ment Science group but it is not a commissioning cles, confirming trivial problem to solve; it is in fact a through the first that the JCMT is combination of problems between half of 2008. a busy place. ACSIS and HARP whose many layers Modifications to Certainly all of mean that they can only be resolved the telescope Gary Davis, Director JCMT us who work one at a time. infrastructure to Antonio Chrysostomou, here know that! accommodate Associate Director JCMT In the meantime, there has been the instrument Since the last newsletter, there has significant progress in the JCMT Sci- are essentially complete and we are been a concerted effort to improve ence Archive (JSA). We are regularly waiting anxiously for this ambitious the operational performance of the shipping data to the CADC and you instrument, with its immense scien- facility, paying particular attention should all now be accessing your tific promise, to arrive. to the new instruments HARP and data via their portal. The future de- ACSIS, the telescope pointing model velopment of the project went Whilst HARP/ACSIS and SCUBA–2 are and the introduction of new (or im- through a review process this sum- expected to dominate the usage of proved) heterodyne observing mer. A review panel comprising Dr the telescope for several years, three modes. Progress on all these aspects Sean Carey (IPAC, Chair), Prof. Ian ancillary instruments are also in vari- has been sure. In parallel the Soft- Smail (Durham), Prof. Christine Wil- ous stages of development: ROVER, ware and Computing Services have son (McMaster) and Dr Michiel Ho- a polarimeter for use with HARP and 3 been working hard to improve the gerheijde (Leiden) took a careful RxA; POL–2, a polarimeter being computational infrastructure, in par- look at the Advanced Data Products developed at the Université de Mont- ticular the communications between (ADPs) that the JSA will produce as real for use with SCUBA–2; and FTS– the 25 computer systems at the well as the JSA infrastructure. The 2, a Fourier transform spectrometer JCMT. This effort has seen our fault Carey Report endorsed the project being developed at the University of rate gradually but surely decrease. and its aims, and a finalised docu- Lethbridge, also for use with SCUBA– At the time of writing, the 6 weeks ment describing the requirements 2. The ROVER hardware has been since the start of the new semester for these ADPs has been made avail- delivered and the required software have seen the fault rate remain able via the JLS web site (http:// development is underway as time steady at about 3% of the clear time www.jach.hawaii.edu/JCMT/surveys/). It allows amongst our higher-priority lost, a significant improvement over is now down to the survey teams to commitments. The latter two instru- where we were 6 months ago. consider the algorithms to be used ments can only operate in conjunc- D ESK which will calculate these ADPs. tion with SCUBA–2, and are sched-
Science Verification (SV) for the JCMT uled for delivery in 2008. OF Legacy Surveys (see the June 2006 Whilst the transformation of the issue of Spectrum) continues and JCMT’s spectroscopic capabilities is A major milestone was recently THE there have thus far been four now virtually complete, the transfor- achieved in our ongoing project to phases, each providing valuable mation of the continuum side awaits develop an interferometric capability D feedback to us. The Gould’s Belt the arrival of the SCUBA–2 instru- in collaboration with the SMA and IRECTOR Survey, Nearby Galaxies Survey and ment. The project team at the UK the CSO: strong fringes were ob- the Spectral Legacy Survey — as the ATC have been working very hard tained on all baselines at 345 GHz at only ones with an interest in HARP throughout the year to bring the the first attempt (see article on page data for their surveys — have each instrument to a state of readiness 5). We anticipate issuing a Call for sent observers who have really put for delivery, and there have been a Proposals in late 2007 for a pilot the telescope and instruments number of significant technical programme consisting of 42 nights through their paces, giving the achievements. At the same time, spread over several months starting whole system a thorough shake- however, there have been some in 2008. down. These teams should now have holdups, as is only to be expected in sufficient data to begin establishing a project of this complexity. At the Looking to the future, we reported quality assurance criteria for the time of writing this column, the in- (Director’s Desk, continued on page 4)
Starlink Software Collection: Puana Release
Brad Cavanagh (JAC Software Group)
The Joint Astronomy Centre an- and a 10 km/s bandpass centered plication plots astronomical objects nounces the Puana (Hawaiian for on 9.313 km/s. A channel map in azimuth and elevation for any Procyon) release of the Starlink Soft- made from this extracted central point on Earth. ware Collection. This release in- region is shown in Figure 1. cludes bug-fixes and feature en- Download links, release notes, patch hancements throughout the soft- A new application (but familiar to information, and mailing list sub- ware suite. Improvements that affect observers at the JCMT) known as scription details can be found at JCMT users include improved hetero- Sourceplot is now bundled. This ap- http://starlink.jach.hawaii.edu/ . PERATIONS dyne spectrum handling in SPLAT (the spectral line analysis tool), sig- nificant data handling improvements in MAKECUBE, and improved cube JCMT O JCMT handling in GAIA.
Conversions between NDF and FITS formats have been improved, includ- ing better support for CLASS-style WCS FITS headers.
The ORAC–DR data reduction pipe- line now reduces raw ACSIS time- series data. It creates a cube, base- lines it, then creates various display files, including integrated intensity 4 maps and channel maps. Develop- ment on ORAC–DR for ACSIS is on- going, and future releases will build upon this support.
NDF sectioning can now be done using World Coordinate System coor- dinates. Using an ACSIS cube as an example, to extract a central portion one would use NDFCOPY:
% ndfcopy orion'(5h35m14s~0.5m,- 5d22m32s~0.5m,9.312637~10.0)' orion_centre
This example extracts a cube with half-arcminute sides centered on the peak emission in the spatial axes Figure 1. — Channel map of Orion.
(Diretcor’s Desk, continued from page 3) Newsletter. Three vacancies have picking up some of the duties previ- in our previous column that all three been filled in Engineering and Tech- ously carried out by Kynan Delorey. 2007 • #27 agencies had indicated their wish to nical Services with the appointments Finally, Michele Mulkey joined us in continue operating the JCMT until at of Marcel Tognetti as an electronics July as Human Resources Manager, least 2012. This remains the case, technician, Ian Campbell as an elec- filling the post vacated by Ian Mid-
UTUMN and a formal agreement to this ef- tronics engineer, and Alan Ryan as a son. Ian returned to the UK this past fect is in preparation. It is hoped mechanical design engineer. Alan is summer to resume his employment that a final announcement on this employed jointly by the JAC and the with the Science and Technology
• A issue can be made within the next UH Institute for Astronomy, and his Facilities Council (formerly PPARC) few months. time will be shared between the two after 15 years at the JAC, and we organisations. Anubhav Agarwal re- wish him well for the future. Finally, there are some staffing cently began work at the JAC as a
PECTRUM changes to report since the previous software engineer, and he will be
JCMT S JCMT JCMT S
First Light with the eSMA at 345 GHz PECTRUM
Sandrine Bottinelli (Leiden) & The eSMA Commissioning Team
• A
Following the successful commis- present on August 30th were high! be obtained by going to the “tools”
sioning of RxWB on the JCMT (cf., They were not disappointed and af- page of the “SMA observer cen- UTUMN JCMT Spectrum, Spring 2007) and of ter carrying out the setup procedure ter” (http://sma1.sma.hawaii.edu/). Barney on the CSO earlier this year, I (including recabling the receiver, am pleased to report that these two pointing and focusing), fringes were Congratulations and many thanks to 2007 • #27 receivers were successfully con- rapidly acquired on all baselines (as everyone involved in reaching this nected to the SMA LO box and that partially shown in Figure 1), despite achievement! the first fringes at 345 GHz were average, but fortunately stable, obtained with the eSMA on the night weather. eSMA developments at the JCMT are of August 30th 2007. financially supported by a Nether- The excitement caused by the lands NWO–M grant awarded to The eSMA (extended Sub-mm Array) strings of dots of Figure 1 is justi- Ewine van Dishoeck (Leiden Observa- is a collaboration between the SMA, fied by the fact that operating the tory/MPE Garching). In the past, the the JCMT, and the CSO, to join all eSMA at the target frequency of 345 grant provided for the technical ex- three observatories into a single in- GHz will yield unprecedented spatial pertise and support from Leo de terferometric array (see photo on the resolution and sensitivities at these Jong (SRON) and Rob Millenaar rear cover). It will operate part-time frequencies, thanks to the added (ASTRON), who both came to the JAC in the 345 GHz atmospheric window baseline length and collecting areas for extended visits to assist with the to study Solar System bodies, proto- of the JCMT and the CSO. Hence the conversion and commissioning of planetary disks, star forming re- eSMA will naturally allow one to ob- RxWB. Currently, this grant supports gions, evolved star envelopes, tain unique science observations and S. Bottinelli (Leiden Observatory) to nearby galaxies, and ultraluminous achieve novel results in the afore- travel to the JAC to participate in the galaxies at cosmological distances. mentioned areas. The interested commissioning of the eSMA and cali- To be able to carry-out 345 GHz op- reader will find more information on brate the instrument. We also ac- erations, JCMT and CSO receivers the potential early science program knowledge the thesis work of Emily 5 must be compatible with SMA receiv- that was discussed at a workshop Curtis whose dedicated laboratory ers, in particular regarding the IF organized in Leiden on February 1-2 work on building and testing parts and bandwidth. Hence new 345 GHz 2007 (see http:// of RxWB contributed significantly to receivers have been built for the CSO www.strw.leidenuniv.nl/esma-workshop/). the successful completion of this and the JCMT, namely Barney and Moreover, information on the resolu- receiver and hence of the advance- RxWB respectively, to meet the re- tion and sensitivity of the eSMA can ment of the eSMA. quired technical specifications set by the SMA receivers.
Daytime testing carried out in early August enabled us to link the new 345 GHz receivers of the single-dish facilities to the LO box of the SMA JCMT O and a lock was obtained for all re- ceivers. The positive outcome of these tests presaged for a bright future. The critical night was set to PERATIONS August 30th, 2007. The previous eSMA run in May had already set a cornerstone in eSMA commissioning as strong fringes were observed on all baselines at 230 GHz and a high- quality baseline track was obtained. Hence the spirits of the commission- ing team [Ken Young (alias “Taco”), Rob Christensen, and Zack Gazak for the SMA, Richard Chamberlin and Hiroshige Yoshida for the CSO, Jim Hoge, Remo Tilanus, and Sandrine Figure 1. — First fringes with the eSMA on baselines between one SMA antenna (“6”), the JCMT (“10”), and the CSO (“9”). The SMA was in its extended configuration yielding eSMA baselines up to ~450 m (baseline 6-9, lower Bottinelli for the JCMT] who were panel), about twice as long as the longest SMA baseline (~225 m). (Also see rear cover of this issue.)
SHADES: A Cosmological SCUBA Legacy
James Dunlop (Edinburgh) & The SHADES Consortium
SHADES, the SCUBA Half Degree Eales et al. 2000). SHADES was moti- scheduled nights on Mauna Kea, Extragalactic Survey, commenced at vated by many science drivers, in often with an ailing SCUBA camera. the JCMT in early 2003 (http:// particular the desire to clarify the In the end, by the time SCUBA finally www.roe.ac.uk/ifa/shades/). The origi- number density, redshift distribu- died in 2005, we had succeeded in nal plan was to map a total area of tion, and clustering properties of the mapping approximately 0.25 deg2, 0.5 deg2, to an rms noise level of ~2 bright sub-mm selected galaxy split between two survey fields. mJy at 850 μm. The choice of survey population (see van Kampen et al. While short of our original goal, the CIENCE depth was designed to play to the 2005, Mortier et al. 2005). However, resulting 850 μm maps of the Lock- strengths of SCUBA on the JCMT — at a basic level, all involved accepted man Hole and SXDF/UDS fields significant numbers of extragalactic that, to make further progress in the shown in Figure 1 form the largest sources were known to be detect- field, what was now required was a extragalactic sub-mm imaging sur- JCMT S JCMT able at the ~8 mJy level (Scott et al. large and complete sample of 850 vey of meaningful depth ever under- 2002), and yet by staying reasonably μm sources (analogous to the 3CR taken, and provide a uniquely pow- bright we aimed to avoid the more radio source sample, which ulti- erful resource for the study of the severe effects of source confusion mately revolutionized extragalactic bright sub-mm galaxy population. resulting from the 14” beam. The radio astronomy). The problem was planned survey area was then cho- that the creation of such a sample As befits such a scarce resource, the sen to yield a statistically useful with SCUBA required one third of the SHADES 850 μm data have been sub- sample of a few hundred sub-mm available time on the JCMT over 3 jected to an extremely thorough sources. years. analysis, with 4 separate sub-groups within the consortium undertaking SHADES has been undertaken by a Since 2003 a large number of as- independent data reduction and large consortium of UK and Cana- tronomers within the SHADES con- source extraction. The final merged dian astronomers, many of whom sortium (ably supported by the staff results of this exhaustive process, 6 had previously worked in competi- of the JCMT) have worked extremely reported in Coppin et al. (2006), tion, undertaking the pioneering hard to complete this unique sub- include a new sample of 120 sub- sub-mm surveys which followed mm survey. The project required mm sources with statistically de- first-light with SCUBA in 1997 (Smail that consortium members staff a boosted fluxes, and a definitive et al. 1997; Hughes et al. 1998; very large number of flexibly- (SHADES, continued on page 7) 2007 • #27 UTUMN • A
Figure 1. — The SHADES SCUBA 850 μm images of the SXDF (left) and Lockman Hole (right) fields, indicating the positions of the 120 most secure sources. (Also see front
PECTRUM cover of this issue.)
JCMT S JCMT JCMT S
(SHADES, continued from page 6) required to extract astrophysical zona. These mid-IR data, coupled PECTRUM measurement of the 850 μm source results from 4σ peaks in SCUBA with optical and near infrared imag- counts in the range 1–10 mJy (Figure maps. In particular, the VLA has ing obtained from Subaru (Furasawa 2). These results, and the lessons been used to assemble extremely et al. 2007) and UKIRT (as part of learned in getting to grips with deep 1.4 GHz radio images in both UKIDSS; http://www.ukidss.org/) have • A these data, have played a major role the SHADES survey fields. As a re- enabled additional source identifica- in informing the design of the sult, ~70% of the SHADES sources tion, more sophisticated (and inde- SCUBA–2 Cosmology Legacy Survey have now been identified with radio pendent) redshift estimation, and a UTUMN (http://www.jach.hawaii.edu/JCMT/ sources, yielding positions accurate detailed exploration of the stellar surveys/Cosmology.html). to ~1 arcsec (Ivison et al. 2007), masses and star-formation histories
and a first estimate of the redshift of the SHADES galaxies (Figure 4; 2007 • #27 While some SHADES consortium distribution of the SHADES sample Dye et al. 2007; Clements et al. members worked to build the SCUBA has been derived (Figure 3; Aretxaga 2007). These studies indicate that maps, others have made extensive et al. 2007). These results, derived bright sub-mm sources are generally efforts to assembling the extensive by fitting the radio-submm-farIR housed in already massive galaxies (>1011 M ), with the “current” star- supporting multi-frequency dataset fluxes of SHADES galaxies to a range of observed low-redshift template burst having been preceded by pre- SEDs, will be further refined as addi- vious star-formation events which tional sub-mm and mm data become formed at least half of the final stel- available. This process has already lar mass. commenced, through the SHARC2 350 μm observations reported by Most recently we have been explor- Coppin et al. (2007), and new 1.1 ing the results of pushing the effec- mm JCMT AzTEC mapping of the tive depth of the SHADES imaging SHADES fields (see below). via a variety of stacking analyses (Serjeant et al. 2007) and investigat- SHADES has also benefited from the ing what constraints SHADES can advent of the Spitzer Space Tele- place on the clustering properties of scope. The SXDF field is part of the the sub-mm galaxy population (van SWIRE survey (http:// Kampen et al. 2008). Figure 3. — The first estimate of the redshift distribu- swire.ipac.caltech.edu/swire/), while 7 tion of the SHADES galaxies, derived by fitting their deep IRAC and MIPS imaging of the A total of ten SHADES papers have observed long-wavelength colours to a range of low- redshift template spectral energy distributions. The Lockman Hole has been obtained in now been published or submitted results broadly confirm previous evidence that the collaboration with the Spitzer GTO for publication. Nevertheless, and redshift distribution peaks in the range 2 Figure 2. — The 850 μm number counts, as derived from the images shown in Figure 4. — Examples of SED fitting to the full range of optical+infrared+submm Figure 1. The result is the best estimate to date of the number counts in the 1–10 data now available for the SHADES sources, from which redshifts, star-formation mJy range, and resolution of 20–30% of the background. histories, and galaxy masses can be estimated. The SCUBA Legacy Maps and Catalogues James Di Francesco, Doug Johnstone (NRC/HIA), Helen Kirk (Victoria), Todd MacKenzie (Prince Edward Island), & Elizabeth Ledwosinska (McGill) SCUBA was the flagship instrument lic download from Canadian Astro- To make useful maps of the cali- for the JCMT, and for nine years nomical Data Centre (CADC) from brated data, we divided the sky into (1996–2005) its combination of high the following URL: http:// 1.2 deg × 1.2 deg (“square-degree”) sensitivity and wide area coverage www.cadc.hia.nrc.gc.ca/community/ regions (overlapping with their allowed for some of the most stun- scubalegacy/ . In this brief report, we (Legacy, continued on page 9) ning results in modern astronomy. describe the results of this effort. CIENCE SCUBA data have been long available Extensive details of the project can to the public but utilization of these be found in a paper located at the data could be difficult for those not website that will be soon published well versed in the SURF reduction in the Astrophysical Journal Supple- JCMT S JCMT package or those who simply didn't ment Series (Di Francesco et al. have the time to reduce the data 2008, in press). themselves. We downloaded all 35,455 SCUBA To allow SCUBA data to be dissemi- data files in the archive that corre- nated as widely as possible to the sponded to 850 or 450 μm data community, we have uniformly re- taken in jiggle or scan mode, a total reduced all viable SCUBA jiggle and of 78.7 GB. We removed all Solar scan data at 850 and 450 μm. System objects from the ensemble, Where possible, data obtained at given the variations in their sizes, several epochs were combined to brightnesses and positions, as well Figure 1. — 850 μm SCUBA image of the supernova provide superior map quality. All as all obvious engineering data, remnant Cas A. All of these images are available for public download from the CADC web site. (Also see resulting maps are available for pub- leaving 28,534 files (69.9 GB). front cover of this issue.) 8 (SHADES, continued from page 7) of their SEDs and bolometric lumi- complete the analysis of the SHADES despite the premature death of nosities. SCUBA+AzTEC data in the context of SCUBA, there is still much to be the still-growing multi-frequency done. First, a lot has still to be Excitingly, this continually improving supporting dataset. The results will learned about the 850 μm sources multi-frequency follow-up will take include the clearest picture to date uncovered in the SCUBA map. In par- place within the context of a new of the number density, redshift dis- ticular, less than 10% of these JCMT SHADES dataset, which has tribution, and clustering properties sources have robust spectroscopic been provided by the AzTEC camera of the bright sub-mm/mm galaxy redshifts, despite many having been (PI Grant Wilson — see JCMT Spec- population. These measurements targeted during multi-object spectro- trum, Spring 2007). This was used are already providing some of the scopic campaigns on 8-m class tele- on the JCMT in 2005/06 to map the most demanding observational scopes. This is in part because many SHADES fields at 1.1 mm. These benchmarks for theoretical models of the SHADES galaxies are very red new SHADES–AzTEC maps have of galaxy formation, as well as help- galaxies, and we expect to improve more than covered the original ing us to refine the key outstanding the spectroscopic completeness of planned total area of 0.5 deg2. These questions to be answered by SCUBA– the sample via a major new ap- maps are thus yielding new bright 2. (Also see article on page 10.) proved programme of deep, red op- sources, as well as allowing us to tical FORS2 spectroscopy with the obtain 1.1mm flux measurements References VLT in the SXDF/UDS field (PI Al- for many of the 850 μm sources ex- Aretxaga, I., et al., 2007, MNRAS, 379, 1571 2007 • #27 maini), supplemented in the future tracted from the original SCUBA Austermann, J., et al., 2007, in prep. Clements, D., et a., 2007, MNRAS, submitted by near-infrared spectroscopy with maps. A new SHADES-AzTEC consor- Coppin, K., et al., 2006, MNRAS, 372, 1621 FMOS on Subaru. Improved Spitzer tium has now been formed, and the Coppin, K., et al., 2007, MNRAS, in press Dye, S., et al., 2007, MNRAS, submitted UTUMN data will also be supplied in early SHADES AzTEC maps and source Eales, S.A., et al., 2000, AJ, 120, 2244 2008 via a new Spitzer Legacy pro- lists will be presented in Auster- Furusawa, H., et al., 2007, in prep Hughes, D.H., et al., 1998, Nature, 394, 241 gram in the SXDF/UDS field (PI mann et al. (2007). New stacking Ivison, R.J., et al., 2007, MNRAS, 380, 199 van Kampen, E., et a., 2005, MNRAS, 359, 469 • A Dunlop), SMA observations are being analyses, and techniques for joint van Kampen, E., et al., 2008, MNRAS, in prep. used to clarify ambiguous radio source extraction are currently being Mortier, A., et al., 2005, MNRAS, 363, 563 identifications, and data anticipated explored. Scott, S., et al., 2002, MNRAS, 331, 817 Serjeant, S., et al., 2007, MNRAS, submitted from Herschel in the SHADES fields Smail, I., et al., 1997, ApJ, 490, L5 PECTRUM should revolutionize our knowledge Over the next 2 years we expect to JCMT S JCMT JCMT S (Legacy, continued from page 8) respectively. The effective resolu- logue (EMOC, with 6118 objects). PECTRUM neighbours by 0.1 deg). Maps were tions of the maps are 22.9” FWHM at The Catalogues also list for each created from all 850 or 450 μm jig- 850 μm and 17.3” at 450 μm. (An object the maximum 850 μm inten- gle or scan data within each respec- Extended Dataset containing all sity, sizes of each object defined by tive region using the “matrix inver- SCUBA map data regardless of inher- Clumpfind, and 850 μm fluxes, in • A sion” technique of Johnstone et al. ent calibration quality is also avail- addition to the associated 850 μm (2000). Full details of the data re- able for those looking for widest noise level. Where possible, the duction, calibration, map making areal coverage.) These resolutions Catalogues also list the maximum UTUMN and image improvement can be are larger than the diffraction limit 450 μm flux and peak flux density found in Di Francesco et al. (2008). of the JCMT because of the convolu- and the local 450 μm noise level. After calibrating, smoothing, and tions with narrow Gaussians to re- Finally, the Catalogues provide 2007 • #27 flattening the images, the absolute duce pixel-to-pixel noise and be- where possible a tentative associa- flux uncertainties of the maps are cause of the low-amplitude error tion with previously known objects ~20% at 850 μm and ~50% at 450 beam of ~40” FWHM at both wave- in the SIMBAD astronomical data- μm. lengths. base. All maps were visually inspected and Figures 1–3 show excellent exam- The Catalogues have been converted those not meeting a qualitative as- ples of 850 μm SCUBA maps pro- to the standard machine-readible sessment were removed from the duced from this effort, including M format of the Astrophysical Journal Datasets. Unfortunately, maps of 51, NGC 6334, and Cassiopeia A. and are also available for download the deep “cosmological” fields (e.g., All emission maps are available for from the aforementioned CADC web- the SHADES field, see article on page download from the CADC web site. site. Until SCUBA–2 is able to pro- 6) were removed since those made duce data that are orders of magni- with matrix inversion did not have To quantify the emission located in tude in scale larger than that pro- good quality. Those data were ob- the SCUBA Datasets, we constructed duced by SCUBA, we hope the SCUBA tained with a very specific chopping catalogues of objects within the 850 Legacy Maps and Catalogues will throw and angle, and the matrix in- μm maps. Object identification was allow a wider use of submillimetre version method works best when performed using the two- continuum data throughout the as- multiple scales are sampled. dimensional “clumpfind” algorithm tronomical community. (Williams, de Geus & Blitz 1994) The Fundamental Dataset down to a small multiple of the mini- References 9 (containing the best calibration) con- mum noise in each map. The result- Di Francesco, J., Johnstone, D., Kirk, H. M., MacKenzie, tains 1423 and 1357 deg2 maps at ing identifications form the bases of T. & Ledwosinska, E. 2008, ApJS, in press Johnstone, D., Wilson, C. D., Moriarty-Schieven, G., 850 and 450 μm respectively with the Fundamental Map Object Cata- Giannakopoulou-Creighton, J. & Gregersen, E. 2000, total sky coverages of 19.6 square logue (FMOC, with 5061 objects) ApJS, 131, 505 Williams, J. P., de Geus, E. J. & Blitz, L. 1994, ApJ, 428, degrees and 16.4 square degrees and the Extended Map Object Cata- 693 JCMT S CIENCE Figure 2. — 850 μm SCUBA image of the Whirlpool Galaxy, M Figure 3. — 850 μm SCUBA image of the massive star forming re- 51. (Also see front cover of this issue.) gion NGC 6334. (Also see front cover of this issue.) Extreme Activity in a Biased Environment: An AzTEC 1.1 mm Survey James Geach (Durham) & The JCMT/AzTEC MS 0451–03 Collaboration Background Current work — Identifying Counter- Finding counterparts to SMGs at parts to SMGs other wavelengths is extremely chal- The strong decline in the star forma- lenging, but essential to confirm tion rate of the Universe over the Our Spitzer 24 μm imaging has re- their identity. The large beam size of past 8 Gyrs is thought to be in part vealed a population of moderate AzTEC (FWHM~20”) means that any CIENCE due to the suppression of star for- starburst galaxies residing within individual detection could contain mation within gas-rich late-types as this cluster (SFR~50 M /yr, see several potential optical counter- ~ they are accreted into dense envi- Geach et al. 2006), already hinting parts within a reasonable search ronments during the build up of that this cluster harbours significant radius. Therefore, to help in this JCMT S JCMT groups and clusters (Tresse et al. activity which is obscured from opti- search, we have made use of the 24 2007). The observed paucity of in- cal views (the SFRs derived from the μm imaging (e.g., Ivison et al. 2007), tensely star forming galaxies in rich mid-infrared are on average a factor employing a two-stage match: first environments out to z~1, and pas- 5 higher than those derived from matching 1.1 mm detections to MIPS sive populations residing in the optical indicators). What of the most 24 μm sources, and then finding the cores of clusters seem to support extreme tail of this population? We most likely optical counterparts in this view (e.g., Dressler et al. 1997; are now searching for submillimetre deep optical images. At the time of Treu et al. 2003). What about star- galaxies (SMGs) in the cluster. (AzTEC, continued on page 11) burst galaxies that are obscured by dust? Dusty starbursts could be con- tributing a significant fraction of the total cluster star formation rate, but are likely to be severely under- estimated or even missed in magni- 10 tude limited optical surveys. Survival of starbursts within clusters could thus be an important stage in the late-development of massive galax- ies, and an alternative to the ‘dry- merger’ scenario of red galaxy for- mation (e.g., van Dokkum 2005). To address this issue, we have been undertaking a panoramic, panchro- matic survey of the rich cluster MS 0451–03 (z=0.55). Our large dataset includes extensive ground based optical/near-IR imaging and spec- troscopy, Spitzer and Hubble imag- ing. Our goal is to build up an unbi- ased picture of activity in the cluster — a complete census of star forma- tion. Our survey will benefit signifi- 2007 • #27 cantly from the most recent addition to our dataset: a 1.1 mm map ob- tained with the AzTEC camera on JCMT (Austermann et al. 2006; and UTUMN see the Spring 2007 issue of Spec- trum). The millimeter map will be essential to unveil any extremely • A luminous obscured starbursts within the cluster. Figure 1. — A 10’×10’ true colour (Bri) image of the cluster MS 0451-03 showing potential AzTEC sources as contours. The insets show close-ups of two AzTEC detections: the images are from HST–ACS, with 24 μm SST–MIPS contours overlaid. In each case the AzTEC centroid is indicated with a cross. We have a spectroscopic follow-up program to confirm the identity of counterparts to the SMGs — with the aim of detecting galaxies residing within PECTRUM the cluster at z=0.55. JCMT S JCMT JCMT S The Night Shift PECTRUM Ben Warrington (JCMT Telescope System Specialist) • A I have been asked to write an article has been switched over to SCUBA–2 ject is now underway to straighten from the TSS point of view for this preparations. We will likely be com- and widen Saddle Road. A six and a UTUMN newsletter. I have the sneaking sus- missioning that beast during 2008, half mile section from the summit picion that if I acquiesce, they will and I imagine that too will be, at turn-off leading west towards come back and ask me again for the times, an arduous process. It will be Waimea has been completed. It is a 2007 • #27 next newsletter, and the next. Oh worth it because, as I am sure you pleasure to drive; it is the closest well, I suppose I can manage to type are all aware, the potential is enor- thing that I have seen to a real high- a piece every six months. mous. way in this state. Before deciding to take this route to Kona side, how- By now, it is no longer news that the What would at any other time be ever, you may want to take note that JCMT is going through an extraordi- major developments, but which sort past this section, no work as been nary period of transition. The proc- of snuck by against the backdrop of done, and the road is in the same ess started about two years ago with all the activity above, was the com- state it has been for years. Improv- the arrival of ACSIS and, shortly missioning and first light for AzTEC ing and rerouting that part is for a thereafter, HARP. Accompanying and the development of eSMA. Due later phase of the project. They are these arrivals, of course, was the to the somewhat premature demise currently working on the section arrival of a whole new suite of ob- of SCUBA, not to mention the delays from the 19 mile marker, roughly serving software. Together, these in construction of the Large Millime- the end of the forest, to the summit have made for a substantial increase tre Telescope, JCMT had the oppor- turn-off. So far, they have done a in JCMT’s capabilities, but a capable tunity to host AzTEC in late 2005. good job of the destructive part of system is often complex, and a com- Quite a bit was written about the the project, so be prepared to drive plex system has lots of places for results of that in the last newsletter, on gravel for several miles, and po- bugs to hide. Speaking as a TSS, the so I will not add much except that it tentially face construction delays. commissioning process was long was mostly a fun experience to work and sometimes frustrating (I suspect with the AzTEC team. I think every- To close on a personal note, it 11 the hardware and software people one involved was quite pleased with seems that I have now been here for would agree), but it was satisfying to the results. The eSMA project linking three years. It doesn’t seem that see that progress was almost con- JCMT and CSO to the SMA for inter- long, which I suppose says good tinually being made. HARP and AC- ferometry continues to progress qui- things about the Joint Astronomy SIS were made available to the scien- etly with a few tests every month or Centre and the JCMT. tific community in the fall of 2006, two. Speaking as mostly an outside and the system has continued to observer, the progress seems good. improve to the point where observ- (See article in this newsletter.) ing with HARP is now more-or-less routine. Just in time, too, as most of As if all the activity at JCMT weren’t the engineering and software effort enough, a massive construction pro- JCMT O (AzTEC, Continued from page 10) far more complexity in the evolution cluster redshift would be a signifi- writing, we are in the process of ob- of galaxies in biased environments cant discovery — dramatically boost- taining deep Keck–DEIMOS spectros- than has previously been thought. ing the rate of stellar assembly in an copy of potential optical counter- The majority of stellar mass assem- environment traditionally thought to PERATIONS parts to 40 SMGs detected over ~0.5 bly in spheroids and bulges is be hostile to star formation. deg2 of the cluster field (Figure 1). thought to occur at high redshift, Our multiwavelength observations but starbursts in clusters at interme- References will then be essential for determin- diate redshift could also be contrib- Austermann, J., et al., 2006, AAS, 209, 8301 Dressler, A., et al., 1997, ApJ, 490, 577 ing the properties of these 1.1 mm- uting to the build-up of the red- Geach, J. E., et al., 2006, ApJ, 649, 661 detected SMGs (regardless of sequence — i.e., the characterising Ivison, R. J., et al., 2007, MNRAS, 380, 199 whether they are members of the feature of the galaxy populations of Tresse, L., et al., 2007, A&A, 472, 403 Treu, T., et al., 2003, ApJ, 591, 53 cluster, or reside at much higher local clusters. Having uncovered a van Dokkum, P. G., 2005, AJ, 130, 2647 redshift). population of moderately star form- ing galaxies with Spitzer–MIPS, we Summary are now searching for the most ex- treme starbursts within the cluster Dust-insensitive observations of star with AzTEC 1.1 mm mapping. The formation in clusters are revealing detection of just one SMG at the The Detection of a Molecular Bipolar Flow in the Multipolar Planetary Nebula NGC 2440 Tatsuhiko Hasegawa, Mei-Yan Wang (Academia Sinica), & Sun Kwok (Hong Kong) Our understanding of the morpholo- rich in CO in the NGC 2440 system. intersects with (and blows away) an gies of planetary nebulae (PNe) has early torus structure. undergone a significant change in There is no detectable velocity gradi- the last decade. When the effects of ent in the CO(3-2) line along the CO The present CO observation is in line projections and ionization are taken bilobal structure. The lack of a ve- with this scenario. In many plane- CIENCE into account, the apparent mor- locity gradient has also been found tary nebulae, even in an evolved one phologies of many PNe can be ex- in the slit spectra of nebular lines such as the Ring Nebula, the central plained by an equatorial torus plus along the optical PA 35 deg bilobal torus contains a significant amount two bipolar lobes (i.e., a bipolar structure by López et al. (1998). of molecular material. If the suppos- JCMT S JCMT structure). The optical spectra show scattered edly recent PA 35 deg lobes contain (by dust in the bilobal structure) material that has been blown away However, optical observations have (red-shifted) line-emissions that from the original, molecular torus, revealed a number of PNe that can- strongly indicate dust movements the presence of CO in the lobes can not be explained by this simple away from the central star at a be explained. The lack of CO in the model. Optical images with the speed of 150 km/s (López et al. east-west lobes is explained by Hubble Space Telescope (HST) show 1998). The optical PA 35 deg photodissociation of CO in the old that a number of PNe have multiple bilobal structure is therefore a bipo- bilobal structure. pairs of lobes (Sahai 2000) and that lar flow oriented nearly perpendicu- some PNe with an apparently typical lar to the line of sight. These results are discussed in more bipolar structure actually have multi- detail in Wang et al. (2007) and ple outflow axes. A phenomenologi- Optical observers have presented Mariappan et al. (2007). cal interpretation of these PNe is various scenarios about the complex “multiple episodic flows” or “bipolar, morphology of NGC 2440. Accord- References 12 rotating, episodic jets” (BRETS; ing to one plausible scenario Kwok, S. 2004, ASP Conference Ser. Vol 313, p 580 López et al. 1998). (Vázquez et al. 1999), a collimated, López, J.A., Meaburn, J., Bryce, M., & Holloway, A.J. 1998, ApJ, 493, 803 high-velocity jet emanates from the Mariappan, M., Wang, M.-Y., & Kwok, S. 2007 in prepa- NGC 2440 (Figure 1) is one of the central star. The postulated jet is ration Sahai, R. 2000, ASP Conference Ser. Vol 199, p 209 most dramatic examples of such episodic and its direction changes Vázquez, R., et al. 1999, ApJ, 515, 633 (optical) multipolar PNe (Kwok with time. The east-west bilobal Wang, M.-Y., Hasegawa, T.I., & Kwok, S. 2007 submit- ted to ApJ 2004). López et al. (1998), based on (bubble-like) structure may have deep images and slit spectra in the been carved by the jet at an earlier [N II] and [O III] lines, demonstrated time when the jet axis was in the the quadrupolar morphology of NGC east-west direction. The jet pre- 2440. We observed NGC 2440 in sumably changed its direction re- the J=3–2 line with RxB3 on the cently in such a way that the jet now JCMT. The CO(3–2) emission (Figure 2) traces the cigar-shaped bilobal structure that runs from NE to SW (PA = 35 deg). Optical and IR images of NGC 2440 show that identifying multipolar 2007 • #27 lobes in NGC 2440 is not a trivial matter. Selection of a probing line and other cares (data processing, exposure time, etc.) significantly UTUMN affect the identification of a bilobal structure. The CO rotational line turns out to be an excellent probe to • A identify a single bilobal structure. At the same time, the present CO(3– Figure 2. — Greyscale image of [N II] taken at CFHT 2) observation clearly indicates that Figure 1. — Three-colour image of NGC 2440 (Kwok 2004). Contours are of CO(3-2). The lowest the PA 35 deg bilobe is exceptionally (courtesy of NASA). Red is Hα and [N II], green is [O contour and the contour interval are both 0.63 K km/s, PECTRUM III], blue is [He II]. which corresponds to about 2σ. JCMT S JCMT JCMT S Mapping Magnetic Fields in Planetary Nebulae PECTRUM and Post–AGB Stars with SCUBA Laurence Sabin, Albert Zijlstra (Manchester), & Jane Greaves (St. Andrews) • A UTUMN The Hubble Space Telescope made bipolar targets: three planetary evolutionary phase of the nebulae famous the round, elliptical, bipolar nebulae, NGC 6302, NGC 6537 and may play a role in the field structure and point-symmetric shapes of NGC 7027 and the proto planetary and distribution. Organisation of planetary nebulae (PNe) and post- nebula (or post-AGB) CRL 2688. the field can also be related to the 2007 • #27 AGB stars (PAGBs). Although there is age of the nebulae, as the magnetic strong evidence that the departure NGC 6537 (Figure 1) presents a field appears more organised in the from spherical symmetry occurs in magnetic field mainly localised older ones (NGC 6537 and NGC the AGB phase, one can wonder how along the equatorial plane (and 6302). After the poloidal field is car- from an original sphere we end up hence perpendicular to the outflow). ried away with the outflow, the more with aspherical objects. With the The vectors are consistent with a stable toroidal field would appear. wind-wind interaction (the slower well defined toroidal magnetic field The transition would occur when the from the AGB and the faster from localised in a circumstellar torus. equatorial field is wound up through the post-AGB) and the companion the interaction with a stellar com- interaction in a binary system, mag- The PAGB CRL 2688 (Figure 2) shows panion. The main shaping agent netic fields are among the hypothe- the most complicated field structure would therefore be attributed to a ses used to explain the shapes ob- of the four studied. It covers the binary companion while the mag- served mainly in PNe and PAGBs. whole nebula and presents two dis- netic field would take a more impor- tinct orientations: one along the tant part later on. Following Greaves et al. (1999), we pole and the other along the equa- assume that submillimeter polarisa- tor. Some perturbations are also More observations are needed to tion is due to grain alignment, i.e., present at the tip of the outflow di- have a clearer idea of the role played in the presence of a magnetic field rection. by magnetic fields in planetary (B) dust grains will be aligned with nebulae and other late type objects. their long axis perpendicular to the In conclusion, we can say that all our These results are discussed in more 13 field. Thus the measured angle of nebulae show polarisation indicative detail in Sabin et al. (2007). polarisation is 90° rotated with re- of grain alignment by magnetic spect to B. Using SCUBA at 850 and fields. The three PN present a tor- References 450 μm, we obtained images of the oidal field localised at the equatorial Sabin L. , Zijlstra A.A, and Greaves J.S, 2007, MNRAS, in plane and CRL 2688 shows a clear press dust polarisation and, by extension Greaves J.S. et al., 1999, A&A, 344, 668-674 of the B field distributions, in four poloidal field. The chemistry and the JCMT S CIENCE Figure 1. — 850 μm map of the magnetic field distribution in NGC 6537. The bar at the bottom represents the length of a vector of 40% polarization. Figure 2. — 850 μm map of the magnetic field distribution in CRL 2688. CN Emission in the Sgr A* Environment — First Results Marija Stanković, Ernie Seaquist, Stefanie Mühle (Toronto), Asuncion Fuente (Observatorio Astronómico Nacional) The radical CN, through the CN/HCN encompassing the CND and its envi- tion. ratio, is an exceptionally good tool rons using HARP in the raster mode. to study photodissociation regions ACSIS bandwidth of 1 GHz was suffi- The observations conducted with the (PDRs; e.g., Fuente et al. 2005). The cient to simultanously observe two JCMT telescope (M06BC06) comple- circumnuclear disk (CND) surround- fine structure components in the CN ment ongoing studies at the lower ing Sgr A* is exposed to the strong spectra. A preliminary CN inte- CN and HCN transitions using the FUV radiation field originating from grated-intensity map and sample NRO 45 m and the IRAM 30 m tele- CIENCE the central stellar cluster and con- spectra are shown in Figure 1. The scopes. These data sets at the three tains potentially star forming clouds strongest CN emission corresponds different frequencies have similar as elucidated by interferometric HCN to the most massive cores from the resolution of 12”–15”, thus permit- observations by Christopher et al. CND. We detect CN emission to- ting accurate line ratios to be ob- JCMT S JCMT (2005). PDR models (Boger & Stern- wards the central pc as well, tained. To investigate the effects of berg 2005) predict that external FUV whereas other molecular species PDRs on star formation in the CND heating of gas clumps generates appear to be photodissociated by we will analyse these data using new photoevaporative outflows and the strong UV radiation originating PDR and radiative transfer models drives strong shocks compressing from the central stellar cluster, thus that can constrain the PDR structure the gas to higher densities. implying that the central cavity is and properties in different physical not devoid of gas, but that it is heav- regimes. As a part of our study of the exis- ily photodissociated. Our CN spec- tence and role of such PDR zones in tra show resolvable fine structure References the global photodissociation of the components which provide a diag- Boger, G.T. & Sternberg, A. 2005, ApJ, 632, 302 CND and in promoting or inhibiting nostic of the CN optical depth by Christopher, M. et al. 2005, ApJ, 622,346 Fuente, A. et al. 2005, ApJL, 619, 155 star formation, we mapped the CN comparison of their relative intensi- (N=3–2) emission in a 5’×5’ area ties within a single rotational transi- 14 2007 • #27 UTUMN • A Figure 1. — Preliminary CN(N=3–2) map integrated over all fine and hyperfine transitions of the central 5’×5’ region of our Galaxy (left). The observations were conducted using HARP in raster mode and ACSIS with 1 GHz wide bandwidth centered at ν =340.136 GHz, in between two fine structure components. This frequency is used as the o rest frequency for the velocity scale on the spectra shown next to the map. The contour lines denoted integrated CN(N=3–2) intensities of (5,7,9...15,19,23...39)×5 K km/s. PECTRUM The position of Sgr A* is marked with a blue cross. The CN(N=3–2) spectra at selected positions indicated in the map are shown on the right. JCMT S JCMT JCMT S Mid–Infrared Spectral Diagnosis of Submillimetre Galaxies PECTRUM Alexandra Pope (UBC), Ranga-Ram Chary (Cal Tech), David Alexander (Durham), Lee Armus (Cal Tech), Mark Dickinson (NOAO), David Elbaz (Saclay), David Frayer (Cal Tech), • A Douglas Scott (UBC), & Harry Teplitz (Cal Tech) UTUMN Perhaps the most enigmatic mem- nent plus an AGN continuum compo- We also find that the mid-IR proper- bers of the high redshift galaxy nent, together with extinction ap- ties of SMGs differ from those of population are the submillimeter plied to both. We find that the mid- typical 24 μm-selected ULIRGs at z~2 2007 • #27 (submm) galaxies (SMGs). Initially IR luminosity of SMGs is clearly (Sajina et al. 2007); the former are discovered about a decade ago with dominated by SF, with a maximum mainly dominated by SF, while the SCUBA on the JCMT, they have mid-IR AGN contribution of 30%. latter are a more heterogeneous proven to be challenging to study, Overall the mid-IR spectra of SMGs population, including a significant due to the relatively large JCMT are similar to a scaled (i.e., much fraction showing strong AGN activ- beamsize and the faintness of the brighter) spectrum of the local star- ity. This difference might be ex- counterparts at most other wave- burst galaxy M82. plained through an evolutionary sce- lengths (e.g., Ivison et al. 2002). nario, in which SMGs are an earlier, Nevertheless, it has become clear We also fit the mid-IR spectra of the cooler phase in the massive merger that they are very massive systems SMGs together with the submm and process, when the AGN has not yet (e.g., Swinbank et al. 2004; Borys et radio photometry using template become strong enough to heat the al. 2005; Greve et al. 2005) at z~1–4 spectral energy distributions (Chary dust and dilute the PAH emission. (Chapman et al. 2005; Pope et al. & Elbaz 2001). The best-fit models 2005, Aretxaga et al. 2007) which have cool dust temperatures Full details of the work reported are thought to be connected with ( 1999). A crucial issue in under- between L and L which has IR PAH,6.2 Aretxaga A., et al., 2007, MNRAS, 379, 1571 standing these connections is deter- been established for local starburst Armus L., et al., 2007, ApJ, 656, 148 Borys C., et al., 2003, MNRAS, 344, 385 15 mining what powers the extreme galaxies (Figure 1b). This suggests Borys C., et al., 2005, ApJ, 635, 853 infrared (IR) luminosities of SMGs: that PAH luminosity can be used as a Brandl, B., et al., 2006, ApJ, 653, 1129; Chapman S., et al., 2005, ApJ, 622, 772 active galactic nuclei (AGN) or star proxy for SF rate in SMGs. Chary R., & Elbaz D., 2001, ApJ, 556, 56 formation (SF)? Draine, B., 2003, ARA&A, 41, 241 Greve T., et al., 2005, MNRAS, 359, 1165 Local ultra-luminous IR galaxies Ivison R., et al., 2002, MNRAS, 337, 1 In order to better understand the (ULIRGs), which are often considered Lilly S., et al., 1999, ApJ, 518, 641 Pope A., et al., 2005, MNRAS, 358, 149 SMG population, we have targetted as the low-z analogues of SMGs ap- Pope A., et al., 2006, MNRAS, 370, 1185 about a dozen SMGs in deep integra- pear to be deficient in PAH luminos- Pope A., et al., 2007, ApJ submitted Sajina, A., et al., 2007, ApJ, 664, 713 tions with the Spitzer Space Tele- ity for a given L as compared to IR Swinbank A.M., et al., 2004, ApJ, 617, 64 scope's Infra-Red Spectrograph (IRS). local starburst galaxies and SMGs. Our sample was selected from sources in the Great Observatories Origins Deep Survey (GOODS) North field, specifically from those SMGs JCMT S having secure multi-wavelength counterparts in our GOODS–N SCUBA ‘Super-map’ (Borys et al. 2003; Pope CIENCE et al. 2006). We find strong PAH emission in all of our targets, which allows us to determine mid-IR-based spectroscopic redshifts, as well as placing constraints on the SF vs AGN contributions to the mid-IR emission (and by inference the observed sub- mm). We study the ensemble prop- erties of the SMGs by averaging to- Figure 1. — (left) SMG composite mid-IR spectrum (black curve) smoothed by a Gaussian with a full width at half gether the individual mid-IR spectra maximum (FWHM) of 0.12 μm which corresponds to the average instrument resolution. The shaded region shows the 1σ uncertainty. The red curve is the best-fit model which is composed of a PAH template (green curve) and a in the rest-frame (Figure 1a). We can power-law continuum component (blue curve), each with extinction applied to them. (right) Correlation between L IR fit this average spectrum to a model and the 6.2 mm PAH luminosity. SMGs are denoted by the large red circles while the local starbursts (Brandl et al. 2006) and the Bright Galaxy Sample (BGS) ULIRGs (Armus et al. 2007) are shown by the blue squares and green containing a starburst PAH compo- crosses, respectively. The line shows the best-fit relation for the local starburst galaxies. New Starlink Software Collection Release eSMA first fringes at 345 GHz with JCMT, CSO, and SMA www.jach.hawaii.edu/JCMT jk