next up previous 600
Next: Sky noise removal (b)- calcsky
Up: A Fuller Reduction: Removing bad bolometers, sky-noise and spikes
Previous: Initial Despiking


Sky noise removal (a)- remsky.

In Section [*] we already used remsky to remove sky noise by choosing a ring of bolometers around the center and we could even have used all bolometers in the array. However, for short chop throws, or where we have an extended source or where the source is not centered in the array one should be more careful. A good way to identify suitable sky bolometers (i.e. bolometers without any source emission) is to convert the map into Az/El (assuming we chopped in Az) so that we can identify bolometers at the edge of the array, which are not affected by the chop nor by any source emission. We regrid the extinction corrected data with the SURF task rebin, setting the parameter $\rm OUT\_COORDS$ to az. In the example below we use the same data set on IRC$+$10216 as we used before. We plot the map immediately with display using the option faint and overlay the bolometer array using scuover (use noname if you want the bolometers as numbers).

% display axes clear
IN - NDF to be displayed /@i86_lon_reb/ > 
DEVICE - Name of display device /@epsfcol_p/ > xwindows
MODE - Method to define the scaling limits /'fa'/ > 
Data will be scaled from -0.0012709096772596 to 0.013754481449723.
% scuover
DEVICE - Name of graphics device /@xwindows/ > 
Current picture has name: DATA, comment: KAPPA_DISPLAY.
Using /home/sandell/dec8/i86_lon_reb as the input NDF.
SURF: file contains data for 4 exposure(s) in 3 integration(s) in 1
measurement(s)

IRC$+$10216 is extended at 850$\mu$m, because of its large CO envelope, and we therefore want to use edge pixels away from the chop direction (usually az) to avoid removing any real signal from the data. In our map of IRC+10216 (Fig. [*]) one would therefore take bolometers 4,9,15 and 22 in the south and 16, 29, and 34 in the north. If we now run remsky on the extinction corrected data, we notice that the background level added back to the map was lower than when we used ring 2, which therefore picks up some of the faint extended emission surrounding IRC$+$10216.

 % remsky
IN - Name of input file containing demodulated map data 
/@o86_lon_sky/ > o86_lon
_ext
SURF: run 86 was a MAP observation with JIGGLE sampling of object 
IRC+10216
OUT - Name of output file /'i86_lon_sky'/ > 
BOLOMETERS - The Sky bolometers, [a1,a2] for an array /'r2'/ > 
[4,9,15,16,29,34]

SURF: Using 6 sky bolometers
MODE - Sky removal mode /'median'/ > 
Sky noise: 0.0005996103 (192 points)
Adding mean background level back onto data (value=6.3598651E-5)

Figure: This image of IRC+10216 has been regridded in Az/el and overlaid with the bolometer array in order to allow us to identify bolometers for sky noise removal. For any extended source we want to avoid bolometers in the chop direction, and we therefore choose bolometers at the bottom and top of the map, i.e. for example 4,9 and 22 as well as 16, 29, and 34.
\includegraphics[width=\textwidth]{sc11_fig4.eps}



next up previous 600
Next: Sky noise removal (b)- calcsky
Up: A Fuller Reduction: Removing bad bolometers, sky-noise and spikes
Previous: Initial Despiking

The SCUBA map reduction cookbook
Starlink Cookbook 11
G. Sandell, N. Jessop, T. Jenness
Joint Astronomy Centre, Hilo, Hawaii
29th October 2001
E-mail:ussc@star.rl.ac.uk

Copyright © 2008 Science and Technology Facilities Council