For information:

This is a quick note about Cloudsat and potential implications
for ALMA.

    Cloudsat is a downwards-looking radar, operating in the 94-94.1 GHz
band, due to be launched next April (2005). It's likely to be in
operation for some years. It is very high power: 1.8 kW of peak
transmitter power, into a 63-dB gain antenna, giving an EIRP
of 4.3*10^9 watts.  The satellite will be in a polar orbit, with
an altitude of about 700 km and a period of about 90 minutes.
Its antenna beam is always directed directly down.
The radar pulses have a variable repetition rate around 4 kHz, with
a pulse width of 3.3 microseconds.  There is an article about the system,
from the American Meterological Society, Dec 2002 available at:
  ftp://ftp.cv.nrao.edu/NRAO-staff/hliszt/CloudSat/CloudSat.pdf .

Some of the implications for ALMA:

   If ALMA antennas are ever looking at the radar when it is looking
down directly at us, there is a possibility of the ALMA frontends
being destroyed.  Since the satellite radar beam always points
directly downwards, this means we should avoid letting our
antennas look at the zenith.
    In normal operation this main beam - main beam coupling is
statistically very unlikely, but it does imply that we should
never leave our antennas parked looking at the zenith. We should
take some additional precautions to protect the receivers if the
antennas are ever unavoidably pointing at our zenith - for example,
while being transported.  Putting our solar observing filters
(which I think will have about 12 dB attenuation?) in front
of the receiver feeds may be sufficient to protect the receivers
from destruction.  This should probably be the default for any
antennas not actually making observations.

    At the ALMA site there will be a few satellite passes a day, with
each pass taking up to about 15 minutes.  Provided ALMA antennas
are not actually looking at the satellite, and that the satellite
beam isn't directly beamed in our direction, the level of interference
into ALMA data is not likely to be significant.

    If an ALMA antenna happens to look at the satellite, even if the
satellite is not being aimed at ALMA, there is likely to be a
significant level of interference even well away from the radar
frequency, and probably saturation (but not destruction) of at
least our Band 3 receiver.  The same may be true when the ALMA
antenna is looking well away from the satellite, but the satellite
beam is directed at ALMA.  However, because of the satellite
motion neither occurrence will last more than a couple of
seconds per satellite pass.

   We may want to set some procedure in place to flag data
taken under these conditions.  You could also imagine some
special-purpose hardware or firmware to gate out radar pulses.
At the very least, our control software should warn when an antenna
is pointing near the zenith, and preferably prevent that from
happening unnecessarily.

     This is just for information.  Harvey Liszt, NRAO's
spectrum manager, is coordinating contact with the
Cloudsat folks.

                     Cheers,
                           Darrel.