Moonlit sky polarization patterns from Cerro Paranal
B. Pereira, S. Gonzalez-Gaitan, A. M. Mourao, J. Rino-Silvestre, A. Paulino-Afonso, J. P. Anderson, A. Cikota, A. Morales-Garoffolo
TL;DR
This study addresses moonlit sky polarization as a contaminant in ground-based polarimetry by comparing full-Moon observations at Cerro Paranal with atmospheric scattering models. It finds that pure Rayleigh scattering cannot explain the data; anisotropic Rayleigh with a depolarization factor that grows with wavelength, plus a Mie component from larger aerosols and an unpolarized multiple-scattering contribution, best reproduce both polarization and intensity across BVRI. The results show Mie and MS become more important at longer wavelengths, and multi-wavelength fits suggest larger aerosol fractions and non-negligible imaginary refractive indices may be needed for simultaneous description. Practically, a simple anisotropic Rayleigh model with wavelength-dependent depolarization provides a reliable proxy for correcting moonlit sky polarization in astronomical observations, while more complex models are required to capture intensity and cross-band coherence.
Abstract
We investigate the polarization patterns from the moonlit sky as observed from the European Southern Observatory at Cerro Paranal. The moonlit sky background can be significant in astronomical observations and thus be a source of contamination in polarimetric studies. Based on sky observations during full Moon with FORS2 in imaging polarimetric mode, we measure the polarization degree and intensity at different wavelengths and scattering angles from the Moon, and compare them to theoretical and phenomenological single and multiple scattering models. Single scattering Rayleigh models are able to reproduce the wavelength dependence of the polarization as long as strong depolarization factors that increase with wavelength are introduced. Intensity data, however, require the inclusion of single Mie scattering from larger aerosol particles. The best models that simultaneously fit polarization and intensity data, are a combination of both single scattering processes, Rayleigh and Mie, plus an unpolarized multiple scattering component. Both Mie and multiple scattering become more dominant at longer wavelengths. Other factors like cloud depolarization and the sunlight contribution during the twilight were also investigated. The present study underscores the importance of accounting for moonlight scattering to enhance the accuracy of polarimetric observations of astronomical targets.
