Skydips

From the listing in the previous section we can see that Skydip data was taken at scans 54 and 98. From the RO (either by using mapsum or photsum on the ro data or by using hdstrace) file we can see that the fitted taus were 1.140 (short) and 0.220 (long) for scan 54 and 1.042 (short) and 0.187 (long).

In most cases these numbers will be sufficient for use by extinction but it is possible to recalculate the tau by using the skydip task. As an example here is the result of skydip on scan 54:

% skydip 54
SURF: Opening apr8_dem_0054 in /scuba/observe/apr8/dem
SURF: run 54 was a SKYDIP observation
SURF: observation started at sidereal time 11 47 24 and ended at 11 54 07
SURF: file contains data for the following sub-instrument(s)
 - SHORT with filter 450
 - LONG with filter 850
SUB_INSTRUMENT - Name of sub-instrument to be analysed /'SHORT'/ > l
SURF: file contains data for 20 integration(s) in 10 measurement(s)
T_HOT - Temperature of hot load (K) /278/ > 
T_COLD - Temperature of cold load for LONG_DC /73.6/ > 
ETA_TEL - Telescope efficiency /0.91/ >
B_VAL - B parameter /-1/ > 
SCULIB: fit for filter 850 and sub-instrument LONG_DC
 eta =  0.91 +/-  0.00  b =  0.86 +/-  0.00  tau =   0.220 +/- 0.002
 Standard Deviation of fit residual =   0.74 K (X=     1.0 N=    9)

Figure: The Skydip result for scan 54. The crosses are the measured sky temperatures and the line is the fitted model.

The results of the fit are displayed in figure . Points worth noting are that the local sidereal time of the observation is printed (this is useful later when running extinction), a fixed $\eta_{tel}$ and a floating value of B (the default value for $\eta_{tel}$ is read from the file header) were used and the tau agrees with the on-line system (which is not surprising since the same code is used on-line as in SURF). The errors derived for the fit can sometimes be suspect since the parameters are not completely independent. The standard deviation of the fit residual gives a measure of the scatter in the points about the model fit. Note also that the `X' indicates the reduced $\chi^2$ of the fit (forced to be approximately 1.0 by the program when it determines the errors) and the `N' indicates the number of iterations required to converge on the fit.

Occasionally, it is necessary to remove some points from the fit. This can be achieved by using reduce_switch and change_quality before running skydip. An example of this can be found in §.

The sdip script can be used to automate the procedure of running skydip and displaying the results with KAPPA's linplot. More information on skydipping can be found in Appendix .