[daip] Re: CLCOR sign conventions

Leonia Kogan lkogan at aoc.nrao.edu
Tue Jan 28 13:56:43 EST 2003


Mark,

You wrote:

The fitting program provided the following results:
                                                            Vertical 
                                                            delay
 PARAMETER   LEAST SQUARE VALUE  1-SIGMA ERRORS    ST Elev  theory
                                                       (m)   (cm)
  A1 T(cm)    -215.238782          0.997365        BR  255    218
  A2 T(cm)    -187.219800          0.933735        FD 1615    186
  A3 T(cm)    -220.477216          1.964724        HN  309    217
  A4 T(cm)    -178.049764          0.629284        KP 1916    180
  A5 T(cm)    -177.211686          0.954120        LA 1967    179
  A6 T(cm)       0.000000          0.000000        MK 3720
  A7 T(cm)    -219.651005          1.576241        NL  241    219
  A8 T(cm)    -198.311136          0.731832        OV 1207    195
  A9 T(cm)    -168.768906          0.843504        PT 2371    170
  A10T(cm)    -228.808112          1.787768        SC   16    225

Using positive (+) vertical delays, I found the residual
delays got worse (nearly doubled).
 
Using negative (-) vertical delays, I found the residual
delays got much better (nearly zero) and were comparable to
those obtained from the WITH_ATM data.

The CLCOR help file reports:

                                                 The values given at the 
                       zenith delay correspond to the desired correction.
                       Therefore the given zenith delay should be 
                       subtructed from the full zenith delay.

If we take "the desired correction" from the last column of your table
then the help file recomends wrong sign.

If we take "the desired correction" from the last column of your table
then the help file recomends wrong sign

If we take "the desired correction" from the second column of your table
then the help file recomends right sign.

The desired correction is added to the relevant column of the CL table.
So it should be megative!!!!

I'll think more about the sign later

Thanks

Leonia



More information about the Daip mailing list