1. Name: Evolved stars and mass loss in the Magellanic Clouds Authors: Aalto, Lindqvist, Schoeier, Olofsson, Black et al 2. Science goal: The final stages of stellar evolution are associated with considerable mass loss in the form of an intense stellar wind. This mass loss has a profound effect on the evolution of a low- and intermediate-mass the star on the asymptotic giant branch (AGB), eventually terminating its existence as a star. The circumstellar envelope (CSE) formed by the wind contains the products of internal nuclear processes, as well as chemical reactions in the photosphere and envelope itself, and hence contributes to the chemical evolution of the cosmic gas and dust. Red supergiants will have similar cirumstellar envelopes. It is important to establish an understanding of the circumstellar emission of evolved stars. The mass return of such stars, and hence their contribution to the galactic chemical evolution, is determined by the rare high mass-loss rate objects. These are highly obscured and their photospheric abundances cannot be determined using classical methods such as visual and near-IR spectroscopy. One must rely upon estimates based on circumstellar line emission. Observations of molecular line and continuum (sub)millimeter-wave emission have proven to be one of the best tools for studying the structure, kinematics, and chemistry in CSEs. Recent systematic surveys and subsequent modellings of circumstellar CO and SiO radio line emission of evolved low- to intermediate-mass stars (AGB-stars) (Schoeier & Olofsson 2001, A&A 368, 969; Olofsson et al. 2002, A&A 391, 1053; Gonzalez Delgado et al. 2003, A&A in press) support the idea that these winds are generally driven by radiation pressure on dust grains which condense in the cool external layers of the photosphere, and that pulsation plays an important role in the grainformation process. Similar studies in the LMC/SMC would be of great importance in order to study the impact of a lower metallicity environment on the behaviour of the mass loss process and its evolution, and the chemistry of the envelope. The importance of obtaining an iso-distance sample should be emphasized. This will allow for reliable correlations between varios stellar and circumstellar parameters to be established, critically needed to test current dynamical wind models. We propose here to observe a sample of 60 evolved stars (mainly AGB, but also some red supergiants) that differ in mass loss rate characteristics and chemistry. We suggest using stellar samples from ISO/Herschel. Line surveys: ------------- A.) A detailed CO survey of a sample of 20 stars with high mass-loss rates (10E{-5} Mo per year) at a resolution of 0.05" (0.012 pc) and a velocity resolution of 1 km/s to compare structures and expansion velocities in low metallicity environments to those of stars in our Galaxy. Such spatial resolution makes it possible to also resolve the largest CSEs. B.) Furthermore, we suggest a lower resolution survey of 40 fainter, intermediate mass-loss rate (10E{-6} Mo per year) stars at a resolution of 0.1" (0.025 pc) and 2 km/s. It might be possible to push things down even further if the resolutions are reduced to 0.2" and 2 km/s - stars with mass-loss rates around 10E{-7} Mo per year could be detected. This means that a fair fraction of the AGB mass-loss rate distribution can be covered. Continuum survey of dusty CSEs: ------------------------------- A sample of 40 stars. According to Olofsson (2002; ALMA Science Day, Munich) the expected 340 GHz brightness for a mass-loss rate of 10E{-6} Mo per year, and a distance of 50 kpc is 0.025 mJy, which is the continuum sensitivity we expect after one hour of observing time at 345 GHz. It is therefore feasible to study the dusty envelopes of high to intermediate mass loss stars in LMC/SMC. At these wavelengths the CSEs are optically thin in dust emission and complementary information on mass-loss rates can be obtained. 3. Number of sources: 20+40+40 4. Coordinates: 4.1. 60 sources in 30 Doradus and N159, LMC (RA=3D05h40m, DEC=3D-69d) 40 sources in the SMC (RA=3D01h, DEC=3D-73d) 4.2. Moving target: no 4.3. Time critical: no 5. Spatial scales: 5.1. Angular resolution: 0.05" (for 20 sources); 0.1"=0.025 pc linear 0.1" (for 40 sources) 5.2. Range of spatial scales/FOV: At 345 GHz (Band 7) the field-of-view is about 18" 5.3. Single dish: no 5.4. ACA: no 5.5. Subarrays: no 6. Frequencies: 345 GHz 6.1. Receiver band: Band 7 6.2. Lines: CO 3-2 Frequency: 345 GHz 6.3. Spectral resolution (km/s): 1 km/s - 4 km/s 6.4. Spectral coverage (km/s or GHz): 50 km/s 7. Continuum flux density: 7.1. Typical value: 0.27 - 0.027 mJy 7.2. Continuum peak value: 7.3. Required continuum rms: 0.014 mJy 7.4. Dynamic range in image: 8. Line intensity: 8.1. Typical value: Stars with high mass loss: (CO 2-1: 30 K) CO 3-2: 30 K Intermediate mass loss: (CO 2-1: 3 K) CO 3-2: 3 K 8.2. Required rms per channel: 1 - 6 K 8.3. Spectral dynamic range: 9. Polarization: no 10. Integration time per setting: 2 - 4 hrs 11. Total integration time for program: 20x2 hrs=40 hrs 40x4 hrs=160 hrs 40x3 hrs=120 hrs ---------------- 320 hrs High mass loss: 2 hrs of CO 3-2 high resolution (0.05" and 1 km/s) observing gives a sensitivity of 6 K which we expect would be a 5 sigma detection. Intermediate mass loss: 4 hrs of CO 3-2 low resolution (0.1" and 2 km/s) observing gives a sensitivity of 0.77 K which would be a 4 sigma detection of a star with an order of magnitude lower mass loss than IRC10216. ********************************************************** Review Christine Wilson: interesting but I really worry about their S/N and time calculations. Can ALMA observe the CO 2-1 line with 0.05" and a resolution of 1 km/s and get enough sensitivity to detect a 30 K line? I didn't have time to check the integration time calculator but someone should. Reply Aalto: I went back and double (triple) checked with the time calculator, and the numbers I get can be found below - but the summary is that it takes one hour to get a 3sigma detection of a 30 K line. Looks like a piece of cake for ALMA - and this also agrees with calculations made by Hans Olofsson (see proceedings of last years ALMA meeting at ESO) plus models by Fredrik Schoier. Input parameters Central frequency (GHz): 230.00 Velocity Resolution (km/s): 1.00 Angular Resolution (arcsec): 0.050 Integration Time (seconds): 3600.000 Conversion Factors Wavelength (micron) 1303.4 Baseline (km) 5.38 Jy per K Ratio (Jy/K) 0.00010815 Output parameters Tsys (K): 167.38 Continuum flux density (mJy): 0.012937 Continuum brightness (K): 0.11962 Line flux density (mJy): 1.3211 Line brightness (K): 12.215 All sensitivities are calculated for 1 sigma. Note also that the proposal does not suggest CO 2-1 observations - but CO 3-2 observations - and only for the high mass stars do we suggest a resolution of 0.05". I will resend the proposal with some modifications just to make this clear. I have double checked all time estimates and they are still sound. Comment Ewine: new DRSP is now baseline