B Calculating the Polarization for Dual-beam Data

This section gives a mathematical description of the calculation of the degree and orientation of the polarization for dual-beam data, based on the observed intensities. It is assumed that any required corrections (such as flat-fielding, sky-subtraction, etc), have already been applied.

Each target exposure measures the components of the incoming light polarized in two different orthogonal directions (depending on the orientation of the half-wave plate). If the symbol Iα is used to represent the intensity of the component polarized at an angle α to the reference direction, then in each exposure the O ray image records Iα and the E ray image records Iα+90.

The first exposure (T0) is taken with the half-wave plate in its 0°  position. The O ray image will then record the intensity I0 and the E ray image will record the intensity I90. Malus’ law gives these intensities as:

I0 = Ip. cos 2θ + Iu 2 I90 = Ip. cos 2(90 θ) + Iu 2 = Ip. sin 2θ + Iu 2

Here, Ip and Iu are the polarized and unpolarized intensities in the incoming light, and θ is the angle between the plane of polarization and the reference direction (i.e. the 0°  position). The total intensity I is the sum of Ip and Iu, and can be found as follows:

I0 + I90 = Ip.(cos 2θ + sin 2θ) + Iu = Ip + Iu = I

Thus, summing the O and the E ray images gives the total intensity image.

The half-wave plate is now rotated by 22.5°  and another exposure (T22.5) is taken. Rotating the half-wave plate by 22.5°  is equivalent to rotating the analyser by 45° , and so the O and E ray images now record the intensities I45 and I135, where:

I45 = Ip. cos 2(45 θ) + Iu 2 I135 = Ip. cos 2(135 θ) + Iu 2 = Ip. sin 2(45 θ) + Iu 2

Again, the sum of the O and E ray intensities (I45 + I135) gives the total intensity I.

The mathematical description of polarization can be simplified by using the quantities Q and U defined as:

Q = Ip. cos 2θ U = Ip. sin 2θ

Together with the total intensity, I, these quantities are known as Stokes parameters11. Using these definitions, the polarized intensity, Ip, is:

Ip = Q2 + U2

and the orientation of the plane of polarization is:

θ = 0.5. arctan(U/Q)

The degree of polarization, p, is the ratio of polarized to total intensity, Ip/I. Using the expressions for I0 and I90 above, it can be seen that:

I0 I90 = Ip. cos 2θIp. sin 2θ = Ip. cos 2θ = Q


I45 I135 = Ip. cos 2(45 θ) Ip. sin 2(45 θ) = Ip. cos 2.(45 θ) = Ip. cos(90 2θ) = Ip. sin 2θ = U

Thus, using the four intensities I0, I45, I90 and I135 (obtained on two exposures with half-wave plate positions 0°  and 45° ), both Q and U can be found, together with two independent estimates of I. This allows the polarized intensity, the degree of polarization and the orientation of the plane of polarization to be found using only two exposures. However, it is usually advisable to obtain additional exposures at half-wave plate positions of 45°  and 67.5°  in order to correct for any difference in the sensitivity of the two channels of the polarimeter (such as may be produced for instance by a polarized flat-field).

11There is a fourth Stokes parameter, V, which is always zero for linearly polarized light, but is non-zero for circularly polarized light.