[time-nuts] Water on Enceladus - What does this imply about NASA'a ability to measure frequency?

[Jim Lux]

On 4/4/14 5:01 AM, Dr. David Kirkby wrote:
> On 4 Apr 2014 08:55, "Tom Knox" <actast at hotmail.com> wrote:
>>
>>   90 microns  is approx a freq res of about 1 x 3.66 -12
>>
>> Thomas Knox
>
> Since the Doppler shift is prortional to the frequency,  I can't see how
> one can determine the absolute frequency.
>
> But given light travels at 3e8 m/s and they can resolve 9e-5 m/s, I would
> have thought that the frequency resolution needed was  9e-5/3e8=3e-13. We
> are differing by more than a factor of 10.
>
>
It's actually even more tricky, if you think about it, because what you 
are really doing is making the measurement over some time period, and 
the path length of signal is continuously varying during that time.

Not only is Cassini doing it's flyby of Enceladus (and you're looking 
for small deviations in trajectory from those due to an idealized point 
source masses), but you've also got your ground stations on Earth moving 
due to planetary motion, daily rotation, as well as things like solid 
earth tides moving the DSN station up and down by tens of cm during the 
measurements.

Gravity science in deep space is a very time-nutty activity.. it's 
basically finding all the various sources of change, modeling them, and 
driving the uncertainties as low as possible.

They use a collocated radiometer to compensate for the extra delay of 
the atmosphere of earth.  JPL has all those folks computing earth 
rotation models, and that figures in (hey, you need to know the 
rotational velocity of earth pretty accurately, to take that out of the 
equation).


The folks who do this spend a lot of time looking at "residuals" plots 
and trying to make them look like a flat line of zero width.