1. Name: Infall toward protostars
    Authors: A. Wootten, ....

 2. Science goal: Detect molecular line absorption against the
continuum of a disk surrounding a protostar.  The program is based on
the detection of formaldehyde at 1.3 mm in IRAS4A by Di Francesco et
al. 2001, ApJ 562, 770. Using IRAM, they detected H$_2$CO absorption
at 1.3 mm of $T_b = 10$ K against a continuum of 3000 mJy with a
velocity resolution of 0.16 km/s.  This provides the best evidence for
infall, but it is currently only possible for the few brightest
sources. To generalize the result and to study the infall velocity
field in detail, we would like to do similar experiments on 30 sources
with 10 times weaker disks with a velocity resolution of 0.05 km/s.


3. Number of sources: 30

4. Coordinates: 

  4.1. 10 sources in Oph      (RA=16:30, DEC=-24)
       10 sources in Perseus  (RA=03,  DEC=+30)
       10 sources distributed over sky (RA=any, DEC=any visible)

  4.2. Moving target: no

  4.3. Time critical: no


5.  Spatial scales:

  5.1. Angular resolution: 0.5"
       
  5.2. Range of spatial scales/FOV: 11"x8"

  5.3. Single dish: yes
 
  5.4. ACA: yes

  5.5. Subarrays: no
 

6. Frequencies:

  6.1. Receiver band: Band 6 

  6.2. Line: H2CO 3_12 - 2_11
       Frequency: 226 GHz

  6.3. Spectral resolution (km/s): 0.05 km/s
  
  6.4. Spectral coverage (km/s or GHz): 20 km/s    


7. Continuum flux density:

  7.1. Typical value: O.3 Jy

  7.2. Continuum peak value: 0.3 Jy

  7.3. Required continuum rms: 0.0002 K

  7.4. Dynamic range in image:


8. Line intensity:

  8.1. Typical value: 0.03 Jy

  8.2. Required rms per channel: 0.1 K
  
  8.3. Spectral dynamic range: 20


9. Polarization: no


10. Integration time per setting: 30 x 10 hrs 

11. Total integration time for program: 300 hr

*************************************************************************

Review Munetake Momose: Basically this plan seems O.K., but there is a
minor comment on the sensitivity requirement. The ALMA sensitivity
calculator shows that the planned observing setup (f=226GHz,
dV=0.05km/s, 0.5" beam and 10-hour integration per source) will result
in 0.18K rms in line brightness or 1.84 mJy rms in flux density, which
is slightly higher than the required value quoted in the plan (0.1K
per channel). I think this difference is not essential because the
typical line intensity is ~ 30 mJy. If it is needed that all the
parameters are consistent, the required rms per channel should be
0.18K and the integration time is kept as 300 hrs.


Response Wootten: I agree with this assessment.  Thinking about this
further, if running the actual experiment I would be limited by the
exact values of velocity resolution available in the backend.  This
would in face limit me to choose either 0.08 km/s or 0.04 km/s.
Choosing the former, I would achieve rms 1.455 mJy; Line brightness
0.13934 K.

In fact, I would choose dual polarization no cross products, 8 digitizers
of 125 MHz bandwidth giving 2048 channels across each placement of the
digitizers.  I would place these on the line specified in this DRSP but
with the expected 8 GHz bandwidth of the B6 receiver, I could also cover
other lines.  I would choose the 3(0,2)-2(0,2) line at 218.222 GHz,
the 3(2,2)-3(2,1) line at 218.475 GHz, the 3(1,2)-2(1,1) H213CO line 
at 219.908 GHz and the primary target the 3(1,2)-2(1,1) H2CO line
at 225.698 GHz, on each of which would be placed a digitizer pair.

Comment Ewine: keep total integration time at 300 hrs