ALMA Polarization

[iancze]

Is your feature request related to a problem or opportunity? Please describe.
This is a conceptual stub summarizing what would be necessary to expand MPoL to work for ALMA polarization observations.

ALMA's polarization capabilities are summarized in Section 8.7 of the Technical Handbook.

Describe the solution you'd like
This is an incomplete list of the things that would need to happen to accommodate polarization observations.

• Expand the base cube representation with an additional axis for XX, XY, YY, and YX visibility cross correlations.
• Alternatively, should the base representation be I, Q, U, and V Stokes cubes? We would also need methods providing easy conversions between the representations.
• Adjust FFT and sampling routines to accommodate the extra cube dimension. It would also be nice (perhaps necessary) to include convenience routines to seamlessly get rid of the extra polarization dimension for non-polarization observations. The same could also be done to get rid of the image cube dimension for continuum observations.
• Given V_XX, V_XY, V_YY, and V_YX visibility measurements, how does the likelihood function change? What is the noise covariance between polarizations? Is this contingent on the instrumental polarization? Currently we assume the user provides V = average(V_XX, V_YY), where V_XY and V_YX are assumed to be negligible.
• Explore priors on polarization degree.

Please feel free to comment with additional suggestions.