Analysis of "fratricide effect" observed with GeMS and its relevance for large aperture astronomical telescopes

Large aperture ground-based telescopes require Adaptive Optics (AO) to correct for the distortions induced by atmospheric turbulence and achieve diffraction limited imaging quality. These AO systems rely on Natural and Laser Guide Stars (NGS and LGS) to provide the information required to measure the wavefront from the astronomical sources under observation. In particular one such LGS method consists in creating an artificial star by means of fluorescence of the sodium atoms at the altitude of the Earth's mesosphere. This is achieved by propagating one or more lasers, at the wavelength of the Na D₂a resonance, from the telescope up to the mesosphere. Lasers can be launched from either behind the secondary mirror or from the perimeter of the main aperture, the so-called central and side-launch systems, respectively. The central-launch system, while helpful to reduce the LGS spot elongation, introduces the so-called "fratricide effect". This consists of an increase in the photon-noise in the AO Wave Front Sensors (WFS) sub-apertures, with photons that are the result of laser photons back-scattering from atmospheric molecules (Rayleigh scattering) and atmospheric aerosols (dust and/or cirrus clouds ice particles). This affects the performance of the algorithms intended to compute the LGS centroids and subsequently compute and correct the turbulence-induced wavefront distortions. In the frame of the Thirty Meter Telescope (TMT) project and using actual LGS WFS data obtained with the Gemini Multi-Conjugate Adaptive Optics System (Gemini MCAO a.k.a. GeMS), we show results from an analysis of the temporal variability of the observed fratricide effect, as well as comparison of the absolute magnitude of fratricide photon flux level with simulations using models that account for molecular (Rayleigh) scattering and photons backscattered from cirrus clouds.