SCUCLIP

Simple sigma clipping for each bolometer

Description:

Each bolometer is analysed independently, the mean and standard deviation are calculated, any points greater than NSIGMA sigma from the mean are treated as spikes and removed. Note that for mapping this despiking algorithm is only useful for very weak sources; bright sources will be removed (since a bolometer jiggles on and off bright sources). Photometry observations do not suffer from this problem as the bolometers are always on source.

Usage:

scuclip in out nsigma

Parameters:

IN = NDF (Read)
This is the name of the input demodulated data file
MSG_FILTER = CHAR (Read)
Message output level. Allowed values are QUIET, NORM and VERB. Default is NORM. If MSG_FILTER is set to VERBOSE the number of spikes removed from each bolometer is reported.
NSIGMA = DOUBLE (Read)
Number of sigma beyond which data are thought to be spikes. If this number is positive an iterative clip will be performed (ie the standard deviation is recalculated after clipping and points lying outside the new range are removed). If it is negative then iterative clipping is switched off.
OUT = NDF (Write)
Output data file.

Examples:

scuclip infile outfile 5
Clip any data points that are further than 5 sigma from the mean. The clipping is done on a per bolometer basis and is iterative.
scuclip infile outfile -5
Clip any data points that are further than 5 sigma from the mean. The clipping is done on a per bolometer basis and only one pass of clipping is performed.

Notes:

Related Applications

SURF: scuquick, rebin, scuphot, sigclip, despike;
KAPPA: setbb

Implementation Status:

The despiking routine sets QUALITY bit 5 to bad. It does not affect the data. The effects of despiking can be removed by using the Kappa task setbb to unset quality bit 5.

Copyright © 1997-2000,2003 Particle Physics and Astronomy Research Council