[daip] Re: your phone message

[Leonia Kogan]

Mark,

I have checked the new version of DELZN which has to work with the data 
it had not worked
with your data before.
Can you check this new DELZN's version with that data?

Now about last discussion about one entry outfile of DELZN.

I have carried out tests.
I ran DELZN, created OUTFILE, then edited it to get only one entry for 
each antenna.
Then I ran CLCOR('atmo') using this file as input file.
I found the expected correction for the column 'DELAY'and column 'RATE':
The first term of the fitting polinomial corrects column 'DELAY'; the 
second term of the polinomial corrects column 'RATE'.(with correction 
for sin(ELEV))
So DELZN=>CLCOR('atmo') works exactely as I expected (as I anderstand)

Now let me repeat your concern as I understand:
To exclude a confusion lets fis the terminology:
DELAY and RATE are the column of  CL table; DEL and DERIVATIVE are the 
value at the
relevant column of the the DELZN output file (which is inputfile for CLCOR).
In the case of the two term fitting polynomial (your case) DEL is the 
first term and DERIVATIVE is the second term of the fitting polinomial. 
Agree?
If there is only one entry for each antenna then interpolation in CLCOR 
gives a constant
(for each antenna) value for DELAY and RATE (with small change because 
of the elevation change).   Agree?

Now you apply CLCOR at FRING (???) and get another SN table which you 
use as input
for  DELZN. You probably use OPTYPE='MDEL'. Therefore column DELAY of 
the SN
table is used for fitting. The value of this collumn should be corrected 
by the first term of the initial fitting and therefore shuldbe equal 
zero. This is your result!
The value of the DERIVATIVE has to stay the same as in the initial 
fitting beacause
the correction of the CL table DELAYwas done by DEL and DERIVATIVE 
corrected RATE
and does not effect DELAY at all. So the value of the fitting DERIVATIVE 
should stay without change!!!! This is your result which is expected!!!!

May I say again the main point:
 DEL directs to correct DELAY and
 DERIVATIVE directs to correct RATE.
DEL does not effect RATE; DERIVATIVE does not effect DELAY.

So your test result coincides with DELZN=>CLCOR('atmo') as the define

Leonia
Mark Reid wrote:

>Leonia
>
>    I have to run today, but I wanted to give you
>one piece of information to clarify things.  The
>CLCOR ATMO input files were generated by my fitting
>program.  I've checked my program against DELZN and
>they agree within about +/-10%.   (Interestingly,
>this difference may be related to the mapping function
>used to translate zenith delays to low elevations.)
>
>Mark
>
>
>-------------------------------------------------------------
>Mark J. Reid                 Phone: 617-495-7470
>Harvard-Smithsonian CfA      Fax  : 617-495-7345
>60 Garden Street             Email: reid at cfa.harvard.edu
>Cambridge, MA 02138, USA     Web  : cfa-www.harvard.edu/~reid
>-------------------------------------------------------------
>
>  
>