This task calculates and removes the instrumental polarisation signal from SCUBA polarimeter data.

For a fractional instrumental polarisation $P_I\left(e\right)$ at angle $\theta \left(e\right)$ and elevation $e$ (the instrumental polarisation is elevation dependent) the correction to the measured flux at waveplate angle $W$ is:

$$S_{Actual}=S_{Measured}-S_{Mean}{P}_I\left(e\right)\left(1+cos\left(4W-2\theta \left(e\right)\right)\right)$$

(1)

For SCUBA, we can not trivially measure the mean flux level (since each bolometer is jiggling to different areas on the sky) so this task uses an approximate form of the above equation to correct for instrumental polarisation that is valid for IPs of a few percent:

$$S_{Actual}\approx S_{Measured}\ast \left(1-P_I\left(e\right)\left(1+cos\left(4W-2\theta \left(e\right)\right)\right)\right)$$

(2)

where we have made the approximation that $S_{Mean}\approx S_{Measured}$.

IN = NDF (Read)

IPFILE = FILE (Read)

MSG_FILTER = CHAR (Read)

OUT = NDF (Write)

remip file1 ipfile.dat $\setminus$$\setminus$

Correct file1.sdf using ipfile.dat and write IP corrected data to the default output file (eg file1_ip).

• Variance is propagated correctly.

• This task writes out the waveplate angles and rotation angles. The waveplate angle per integration is written to .MORE.REDS.WPLATE. The rotation angle (waveplate 0 to X pixel axis) is written to .MORE.REDS.ANGROT (angle per integration). The angle between nasmyth and the ra/dec frame (ie ANGROT - 90 degrees) is stored in .MORE.REDS.NASMYTH_ANG (angle per sample). These are written as NDFs and so can be displayed in the normal way. The angles are in degrees.

• An array containing the fast axis angle is also written to the REDS extension (FAST_AXIS). The size of this array matches the number of sub-instruments in the file.