Calibration and validation of Airborne Multiangle SpectroPolarimetric Imager (AirMSPI) polarization measurements

Appl Opt. 2018 Jun 1;57(16):4499-4513. doi: 10.1364/AO.57.004499.


The Airborne Multiangle SpectroPolarimetric Imager (AirMSPI), a precursor to the future Multi-Angle Imager for Aerosols satellite instrument, is a remote-sensing instrument for the characterization of atmospheric aerosols and clouds. To help discriminate between different aerosol particle types, which is crucial to improve our understanding of their impact on climate and air quality, AirMSPI acquires imagery over multiple view angles in the ultraviolet, visible, and near-infrared, and it employs dual photoelastic modulators (PEMs) to target an uncertainty requirement of ±0.005 in the degree of linear polarization (DoLP) at selected wavelengths. Laboratory polarimetric calibrations using a second-generation Polarization State Generator-2 (PSG-2) and validation measurements at 0<DoLP<1 demonstrate a systematic uncertainty of <0.002. In-flight calibrations of the temperature sensitivity of the PEMs, which could otherwise introduce DoLP errors up to 0.02, are extracted from onboard "validator" measurements as well as from the AirMSPI imagery of the Earth. After this in-flight calibration, the stability of measurements of the validator's DoLP throughout the POlarimeter Definition EXperiment and Studies of Emissions and Atmospheric Composition, Clouds and Climate Coupling by Regional Surveys field campaigns is ±0.001. Comparisons of DoLP data from these campaigns with the Research Scanning Polarimeter show root-mean-square differences ranging between 0.003 and 0.006, while the regression slopes deviate from unity by currently unexplained values up to 0.024. The observed differences are the result of measurement errors in both instruments, as well as imperfections in the intercomparison. A complete polarimetric uncertainty model for AirMSPI is presented, including the effects of absolute calibration uncertainty, in-flight modulation uncertainty, and random noise.