Title Dynamical parameters of extrasolar planets by ALMA astrometry Pi J-F Lestrade Time 100 hrs 1. Name: Dynamical parameters of extrasolar planets by ALMA astrometry Authors: J-F Lestrade, K.M. Menten (Jean-François.LESTRADE@obspm.fr) 2. Science goal: ALMA Astrometry with a theoretical precision of 0.1 milliarcsecond can yield important dynamical parameters of extrasolar planets - masses, inclinations and nodes of orbits - that radial velocity measurements cannot provide. ALMA has the capability to detect directly the thermal emission of the photospheres of 446 nearby stars in the Gleise and Jarheiss (1991) catalogue with reasonable integration times. Wobbles of these stars could be searched to discover new planets with masses as low as 0.1 Jupiter and orbital periods of ~ 10 years. Differently, ALMA could complement the determination of the system orbital parameters for the 46 stars that have known planets from radial velicity surveys and are also detectable by ALMA. We anticipate that the first series of observations will have to demonstrate that the phase-referencing technique combined with the fast switching mode of observation can provide the required high sensitivity (submJy) and proper calibration of the rapid phase fluctuations of the atmosphere. This astrometric programm requires that the most extended configuration of baselines be used and, possibly, militate for a larger array than currently designed. 3. Number of sources: 446 4. Coordinates: 4.1. All over the ALMA sky 4.2. Moving target: no 4.3. Time critical: no 5. Spatial scales: 5.1. Angular resolution: < 0.020" 5.2. Range of spatial scales/FOV: 0.005'' - 0.001'' 5.3 Positional accuracy 6. Observational Setup 6.1. Single dish total power data: no 6.2. Stand-alone ACA: no 6.3. Cross-correlation of 7m ACA and 12m baseline-ALMA antennas: no 6.4. Subarrays of 12m baseline-ALMA antennas: no 7. Frequencies: 7.1. Receiver band: 345 GHz 7.2. Line: N/A 7.3. Spectral resolution (km/s): N/A 7.4. Spectral coverage (km/s or GHz): N/A 8. Continuum flux density: 8.1. Typical value: 0.2 mJy 8.2. Required continuum rms (Jy or K): 0.01 mJy 8.3. Dynamic range within image: low 8.3. Required continuum rms: 0.01 mJy 8.4. Calibration requirements: absolute N/A repeatability N/A relative N/A 9. Line intensity: 9.1. Typical value : N/A 9.2. Required rms per channel: N/A 9.3. Spectral dynamic range: N/A 10. Polarization: no 10.1. Required Stokes parameters: 10.2. Total polarized flux density (Jy): 10.3. Required polarization rms and/or dynamic range: 10.4. Polarization fidelity: 10.5. Required calibration accuracy: 11. Integration time per setting: 5 hrs 12. Total integration time for program: 100 hr 13. Comments on observing strategy : The tricky part of this project will be instead PHASE CALIBRATION, both instrumental and atmospheric. We expect that phase referencing relative to an angularly nearby calibrator will do it as succesfully demonstrated at lower frequencies. Position accuracy required for the calibrators is 10 milliarcsecond or better. *************************************************************************** Review Leonardo Testi: These two are essentially the same project and should be combined. The requirements set by the two authors are, however very different both in terms of the number of sources and the integration time per source and the expected return. It is hard for me to judge who is right: 4.4.2 asserts that you need to reach 0.01 mJy/beam sensitivity and that you can only access ~450 sources; 4.4.3 requires 0.1 mJy/beam and claim that one can observe up to ~1500 stars. Note that the time estimate in 4.3.2 is wrong: 5h are needed to reach the S/N per star and per single observation (one needs many to detect the wobble), the total time of 100h is for 1 observation of 20 objects ?? A key point which is missing is the required positional accuracy and its relation with the observing parameters (flux of targets, signal to noise required etc)... especially in 4.4.3 which quotes flux densities down to 0.1mJy and requires rms~0.1mJy. It is clearly a project that will be tried with ALMA. My feeling is that we should have one such proposals in the DRSP with something like 100-150 hours, which will correspond to observations of a sample of a few tens of the best candidates. Reply Butler: Merging 4.4.2 and 4.4.3 makes perfect sense to me. I point to my recent memo (with al and bob brown) for justification of the numbers I used at: http://www.alma.nrao.edu/memos/html-memos/abstracts/abs475.html . There is also a section in that memo on direct detection. One difference between the two "proposals" is that 4.4.2 suggests to also observe in detail the stars which already have confirmed planets (from radial velocity studies). i didn't include that part in 4.4.3. Beyond that difference, yes, i think they've overestimated the number of stars we can do this on. But in the end, the two "proposals" ask for roughly the same amount of time, so i think it makes sense to just set aside 100-150 hours for this kind of work, and not worry to much about the details. Comment Ewine: keep two separate programs; 250 hrs in total for this important science seems very reasonable. -------------------------------------------------- Review v2.0: Review of 4.4.1-4.4.3 No updated required w.r.t. DRSP 1.1. Integration times still formally hold for 64 antennas, so either increase the total times by 30% or reduce the sensitivity by 14%. Other, unknown factors in the sensitivity and performance will be larger than this correction. R.: This program is not jeopardized by the reduction of the number of antennae. Target list can be adjusted. The key point of this astrometric programm is the performance of ALMA in the extended configuration to benefit of the highest angular resolution of the instrument.