Title Tracing the photoprocesses shaping the Horsehead nebula Pi J. Pety Time 103 hrs ************************************************************************* 1. Name of program and authors Program: Tracing the photoprocesses shaping the Horsehead nebula Authors: J.Pety, M.Gerin and J.R.Goicoechea 2. One short paragraph with science goal(s) The Horsehead nebula is a fantastic laboratory of physics and chemistry. It is a typical pillar shaped by the photoevaporation of low density material which was protected by the shadow of denser material (Pound et al. 2003). This is the first pillar where rigid rotation of gas around its axis has been demonstrated (Hily-Blant et al. 2005). The horsehead contains two dense cores surrounded by a lower density halo (Ward-Thompson et al. 2006; Philipp et al. 2006). But its stunning particularity is the relatively simple geometry of the illuminated cloud edge: Abergel et al. 2003 and Habart et al. 2005 demonstrated that the PDR at the edge of the west condensation can be accurately modeled as a 1D PDR-structure, seen almost edge-on. For that reason, several observational studies at medium resolution (5") have been triggered with the Plateau de Bure Interferometer (PdBI) (e.g. Pety et al. 2005 and Goicoechea et al. 2006), the idea being to use this source as a reference for PDR chemical models. It is proposed here to map the top of the Horsehead (whose shape gave birth to its name) in the different 13CO lines observable with the ALMA receivers as well as the CI lines. This program would enable 1) to study the dynamics of photoevaporation of the dense material at the pillar top, 2) to precisely determine the thermal profile (by having access to diagnostics of the warm gas), 3) to observe the transition between atomic carbon and carbon monoxide. Note that complementary C+ observations towards the Horsehead will be done within the framework of the HIFI/Herschel guaranteed-time key-programs. Current state-of-the-art PDR models show that the typical spatial scale of the physical and chemical gradients range from 1" to 50", making this source an excellent target for ALMA. The choice of the 13CO isotopologue enable to minimize optical depth effects (as probed by PdBI 12CO observations) while still observing bright lines. C18O and continuum would also be observed in the same tunings. The Horsehead nebula is one of the most famous object of the sky. We thus propose to incorporate the nose and the mane in the maps as the obtained maps would thus be an excellent advertisement of the ALMA capabilities in the wide-field mapping area. 3. Number of sources The top of the Horsehead, i.e. a region of 150"x300". 4. Coordinates: 4.1. Rough RA and DEC Ra 2000: 05:40:54 Dec 2000: -02:28:00 4.2. Moving target: No 4.3. Time critical: No 4.4. Scheduling constraints: Avoid windy periods to ensure high precision mosaicing. 5. Spatial scales: 5.1. Angular resolution (arcsec): 1" 5.2. Range of spatial scales/FOV (arcsec): 150"x300" We propose to make Nyquist sampled mosaics following an hexagonal compact pattern. This implies: 49 pointings at 110 GHz (Band 3) 195 pointings at 220 GHz (Band 6) 507 pointings at 330 GHz (Band 7) 869 pointings at 440 GHz (Band 8) 1193 pointings at 492 GHz (Band 8) 2165 pointings at 661 GHz (Band 9) In view of the total number of fields, it is clear that such a program would benefit from the On-The-Fly interferometric mode that IRAM will try to prototype for ALMA in the coming three years. 5.3. Required pointing accuracy: (arcsec) 0.6" rms to ensure high precission mosaicing. 6. Observational setup 6.1. Single dish total power data: Required Observing modes for single dish total power: * On-The-Fly is required by the field of view. * The narrow zero-power linewidth (3 km/s) makes frequency switch suitable to this project, which will improve single dish signal-to-noise ratio by sqrt(2). 6.2. Stand-alone ACA: Required 6.3. Cross-correlation of 7m ACA and 12m baseline-ALMA antennas: This question is difficult to answer. We do not know of any public studies that shows that cross-correlation between ACA and ALMA would be beneficial. Several points need to be clarified: 1) Does the collecting surface of ACA is enough to ensure alone (i.e. without a correction for the integrating time) good sensitivity at spatial frequencies around 7m relatively to the sensitivity at frequencies measured by ALMA alone? 2) Cross correlating 2 different interferometers implies a multiplication of the respective primary beams in the measurement equation: Does this implies in practice a limitation of the field of view of the small antennas to the field of view of the large antennas? 6.4. Subarrays of 12m baseline-ALMA antennas: No 7. Frequencies: 7.1. Receiver band: Band 3, 6, 7, 8, 9 7.2. Lines and Frequencies (GHz): Band 3: 110 GHz. Band 6: 220 GHz. Band 7: 330 GHz. Band 8: 440 and 492 GHz. Band 9: 661 GHz. 7.3. Spectral resolution (km/s): 0.2 km/s 7.4. Bandwidth or spectral coverage (km/s or GHz): 3 km/s for lines but this implies a much larger bandwidth for the single dish frequency switched observations. 8. Continuum flux density: 9. Line intensity: 9.1. Typical value (K or Jy): 0.5 K in all bands giving a peak signal-to-noise ratio of at least 20. 9.2. Required rms per channel (K or Jy): 9.3. Spectral dynamic range: Between 20 and 100 (Equal to the peak signal-to-noise ratio as we are trying to resolve the PDR edge). 9.4. Calibration requirements: Absolute: 5% The main goal of this project is a quantitative comparison of observations with PDR model predictions. We thus need as high as possible absolute precision. But making a large mosaic that associates ALMA, ACA and Single-Dish measurements is increasing the complexity of the calibration. We will thus pragmatically take what we will get but having a 5% absolute precision would be a tremenduous progress compared to today. We assume that a given absolute precision implies the same level of repeatability and relative precision. 10. Polarization: No 11. Integration time for each observing mode/receiver setting (hr): Due to Nyquist sampling of the mosaics, the pointings are not independent. The standard sensitivity formula to use in such a case implies to divide the number of pointings by 1.7. Using the ALMA time estimator, this gives: 7.5 hrs at 110 GHz (band 3). 3.8 hrs at 220 GHz (band 6). 10.0 hrs at 330 GHz (band 7). 27.0 hrs at 440 GHz (band 8). 33.7 hrs at 492 GHz (band 8). 21.2 hrs at 661 GHz (Band 9). 12. Total integration time for program (hr): 103 hrs. 13. Comments on observing strategy: We need an homogeneous data set to make precise comparisons with models. We thus propose to observe exactly the same field of view at the same resolution in all lines. The proposed resolution (1") ensures to use only the compact configuration at the highest frequency while it needs moderately extended configuration at the lowest frequency (largest baseline: 550 m). *************************************************************************** -------------------------------------------------- Review v2.0: DRSP (2.3.9): "Tracing the photoprocesses shaping the Horsehead nebula", and I have a following minor comment. This is an interesting program as a whole, but I cannot understand whether the resolution of 1" is necessary for all the species. The abundance gradient is different from molecule to molecule, and hence, only the total power map may be enough for some species. R.: ne of the goal of this project is to test the transition between CI and CO in the PDR. As shown in the figures of the contribution we wrote for the ALMA conference (attached at end of this email), the data from single-dish instruments and from Plateau de Bure of several of the species proposed in this DRSP is already available. It is clear that the single-dish resolution is not large enough to reach the above goal. Indeed, we know from emission map of 2.12 micron H2 line that the PDR profile is structured at scales up to at least 1". And PDR models predict a transition from C+/CI/12CO in a few arcsec. This is why our goal is a resolution of 1". We propose this resolution in all the observed species because this is the best way to ensure that we will not have to make any assumptions of the beam filling factor when modelling the different intensity brightness. Also it is important to accurately locate the position of the peak emission of the various lines in order to precisely constrain the steep temperature gradient (from 300~K to 30~K in about 10 arcsec).