Title Molecular line studies of submm galaxies --constraining dust obscured galaxy formation Pi C. Carilli Time 170 hrs 1.1.5: Name -- Molecular line studies of submm galaxies -- constraining dust obscured galaxy formation Authors: C. Carilli 2. Science goal: The discovery of the IR background, and the SCUBA/MAMBO population of dusty, star forming galaxies at high redshift, has transformed our understanding of galaxy formation. It is now clear that a significant fraction (of order 50%) of star formation in the cosmos occurs in galaxies that are heavily obscured by dust, and that this fraction may rise with redshift, possibly corresponding to the formation of spheroidal galaxies in active starbursts. One highly uncertain aspect of the study of submm galaxies is their redshift distribution. Optical redshifts remain problematic for the majority of such sources, and can be misleading due to possible mis-identifications. We propose a three part program -- (1) a 'blind' search for CO emission from a representative sample of submm galaxies to constrain their redshift distribution, (2) high resolution imaging of a sub-sample to determine the gas distribution and dynamics on sub-kpc scales and (ii) a search for HCN emission to search for dense gas directly associated with star formation. A representative sample of sources will be chosen from standard (sub)mm continuum surveys with large single dishes or by ALMA itself. Part 1: Redshift search -- I assume redshifts will be 'narrowed-down' via photometric techniques (optical and/or radio) to +/- 0.5 in dz, and that the typical source redshift is between 2 and 3. Using band 3 then requires 3 settings between 90 and 116 GHz to get the CO(3-2) line. This will take about 1hr per source (see below). A second search will then be required to look for higher/lower order transitions to confirm the redshift. This will take another hour. This second search will also give some indication of CO excitation conditions. I assume the characteristic line flux density is of order 1 mJy. Part 2: High resolution CO imaging -- I assume the characteristic source size is >= 1 kpc (0.2"), and intrinsic brightness temperature >= few K. Part 3: HCN emission is typically 10x fainter than CO. To get a 4 sigma detection will then take of order 10hrs/source. 3. Number of sources: 50 4. Coordinates: 4.1. Any 4.2. Moving target: no 4.3. Time critical: no 5. Spatial scales: 5.1. Angular resolution: small configuration preferred - (1) and (3) A configuration for (2) 5.2. Range of spatial scales/FOV: 5.3. Single dish: no 5.4. ACA: no 5.5. Subarrays: no 6. Frequencies: 6.1. Receiver band: Band 3 -- initial search, high res imaging, HCN search Band 6 -- 220 GHz follow-up for redshift verification 6.2. Lines and Frequencies CO, HCN at z=1 to 5, various transitions 6.3. Spectral Resolution (km/s) 100 km/s 6.4. Bandwidth or spectral coverage: 8 GHz (for search) 7. Continuum flux density: 7.1. Typical value: 1 mJy 7.2. Continuum peak value: 7.3. Required continuum rms: 7.4. Dynamic range in image: 8. Line intensity: 8.1. Typical value: 1 mJy -- CO(3-2), 0.1mJy -- HCN 8.2. Required rms per channel: 100 GHz -- 0.14 mJy (20min), 0.026mJy (10hrs) 220 GHz -- 0.2 mJy Imaging -- T_B = 0.25 K at 0.2" res in 1hr note: this corresponds to intrinsic T_B = 0.9 K (1sigma) at z=2.5 8.3. Spectral dynamic range: 9. Polarization: no 10. Integration time per setting: (1a) Search -- 1/3hr per setting x 3 settings x 50 srcs with band 3 = 50hr (1b) 1 hr/src for verification x 50 sources with band 3 or 6 = 50 hr (2) CO imaging -- 1hr/src x 20 srcs with band 3 = 20 hr (3) 10 hr/src for HCN search for 5 sources with band 3 = 50 hr 11. Total integration time for program: 170 ============================================================ Review Pierre Cox: This programme has essentially the same scientific goals has the previous programme, except that there is an explicit mention of HCN. So a slightly redundant. Not clear on what the selection of the 50 sources will be based, sources detected from bolometer surveys such as planned with APEX? The integration time is correct. Comment EvD: Keep this program in DRSP; no need to merge -------------------------------------------------- Review v2.0: 1.1.5 Redshift determination of optically dark SMGs using CO/CI/CII lines is one of the most important science objectives of the ALMA. Large samples of luminous southern SMGs will be provided by continuum camera mounted on APEX, ASTE, CCAT, and so on. Just a minor comment on the integration time; the required rms level (per channel) of 0.14 mJy and 0.026 mJy at 100 GHz will need 32 min and 16 hr, respectively, according to the current specification. It seems OK for 220 GHz estimation. The resultant total integration time is about 230 hours or so. Note that the 8 GHz band width of ALMA will enable us to observe both HCN and HCO+ lines simultaneously. They have comparable intensities and are indeed useful as tracers/diagnostics of dense ISM.