Title:Three-Polarizer Treatment of Linear Polarization in Coronagraphs and Heliospheric Imagers

Abstract:Linear polarized light has been used to view the solar corona for over 150 years. While the familiar Stokes representation for polarimetry is complete, it is best matched to a laboratory setting and therefore is not the most convenient representation either for coronal instrument design or for coronal data analysis. Over the last 100 years of development of coronagraphs and heliospheric imagers, various representations have been used both for direct measurement and analysis. These systems include famous representations such as the (B, pB) system that is analogous to the Stokes system in solar observing coordinates, and also internal representations such as in-instrument Stokes parameters with fixed or variable "vertical" direction, and brightness values through a particular polarizing optic or set thereof. Many polarimetric instruments currently use a symmetric three-polarizer measurement and representation system, which we refer to as "(M, Z, P)", to derive the (B, pB) or Stokes parameters. We present a symmetric derivation of (B, pB) and Stokes parameters from (M, Z, P), analyze the noise properties of (M, Z, P) in the context of instrument design, develop (M, Z, P) as a useful intermediate system for data analysis including background subtraction, and draw a helpful analogy between linear polarimetric systems and the large existing body of work on photometric colorimetry.