[daip] Re: forwarded message from Daniel Lebach

[Daniel Lebach]

Hi Joan,

I do indeed estimate rates in my FRING solutions.  However, your
e-mail raises an excellent point: if the rate of change of
parallactic angle effects are sufficiently large, then they could
well contribute systematic differences between the RCP and LCP phase
residuals that are estimated.  I think the issue is not so much
whether I estimate rates in my solutions, but rather whether the
rates from parallactic angle effects are included in CL tables from
CLCOR w/ OPCOD 'PANG'.  It seems they aren't.  Thus one might expect
that I will estimate different residual rates for RCP vs. LCP in
FRING, because the rate contribution from parallactic angle effects
have the opposite sign (I think) for RCP vs. LCP.

I investigated the residual rates I obtained for SC in BR071D, the
same experiment from which I provided plots in my original e-mail.
Unfortunately, the results are inconclusive.  Far away from source
transit, the magnitude of the differences between RCP and LCP
phase-delay rates are < 0.01 ps/s (i.e., phase rates < ~0.084 mHz)
with only few exceptions.  Near source transit, the differences in
magnitude in phase delay rates grow up to ~0.045 ps/s (phase rates
~0.38 mHz).  However, the scan lengths of usable data near source
transit are short, so the uncertainties of rate estimates are
correspondingly large, and the phase rate differences are only
marginally statistically significant from zero.  The bottom line is
that I can't say for sure whether the rate differences I see between
RCP and LCP are just another symptom of the problem or in fact are
the cause.  Perhaps with further investigations at your end, someone
will figure that out.  In any case, I don't consider this RCP vs. LCP
phase-difference problem for sources near zenith to be of serious
significance.

Best regards,
Dan

In response to:
>
> Date: Thu, 26 Sep 2002 08:23:57 -0600 (MDT)
> From: Joan Wrobel <jwrobel at aoc.nrao.edu>
> To: Dan Lebach <del at cfa.harvard.edu>
> cc: daip at primate.aoc.nrao.edu, Eric Greisen <egreisen at cv3.cv.nrao.edu>
> Subject: Re: forwarded message from Daniel Lebach
> 
> Dan:  When you phase reference, do you make use of the rates?  (Some
> people just do phases these days.)  Fitted rates can differ slightly 
> for RCP and LCP, so I am thinking that could be a way to get mildly
> wrong phases during interpolation when working so close to the zenith
> at SC and MK.  Cheers.  - Joan
>