The Observational Procedure

At least two exposures are required to estimate the degree and orientation of the polarization, taken with half-wave plate positions of 0 degrees and 22.5 degrees (see here). For ease of reference, these exposures are referred to here as $T_{0}$ and $T_{22.5}$. The $O$ and $E$ ray images in $T_{0}$ measure the intensities parallel and perpendicular to the reference direction. $T_{22.5}$ has an effective analyser angle of 45 degrees (twice the half-wave plate angle) and so measures the intensities at angles of 45 degrees and 135 degrees to the reference direction

Usually, a further two exposures are taken at half-wave plate positions of 45 degrees and 67.5 degrees (referred to here as exposures $T_{45}$ and $T_{67.5}$). These provide some redundancy in the data and enable internal consistency checks to be made during the data reduction stage.

These exposures are denoted by the letter $T$ to indicate that they are target exposures. In addition to these target exposures, some flat-field exposures are also required. These are used to correct for any spatial variation in the sensitivity of the system, and consist of exposures of a photometrically flat surface. Ideally, the flat-field source should be unpolarized. However, if the additional target exposures $T_{45}$ and $T_{67.5}$ are taken, then a spatially constant polarization across the flat-field source can be corrected for during data reduction. Since the polarization of the flat-field surface is rarely known to be zero, these additional target exposures should always be taken.

It is important that the flat-field has a good signal-to-noise ratio. For this reason, it is common practice to take one or more flat-field exposures at each half-wave plate position, and stack them together into a single master flat-field (although it is not strictly necessary to use different half-wave plate positions). Further discussion of the flat-fielding procedure is given here.