EOMF-11. The Application of New Methods for Removing Photon-Noise Bias in Second-Order Parameters Derived from Lidar Measurements of Atmospheric Waves

Abstract
Lidar systems provide unrivaled monitoring of the middle and upper atmosphere, allowing for continuous, high-temporal and spatial resolution measurements of atmospheric parameters and constituents, which in turn enable the observation of processes such as wave properties and constituent/heat/momentum fluxes (Chu and Papen, 2005. Gardner and Liu, 2014). Many of these measurements require the use of higher-order statistics like variance and covariance. When these are calculated from lidar data, they are strongly influenced by photon noise, which is an inherent part of the photon collection process. Here, two innovative methods of photon-noise bias removal are analyzed: the spectral proportion method and the interleaved method . These are tested on 10 years of lidar measurements taken at McMurdo Station, Antarctica and also on a forward model developed for this study. Efficacy under both winter and summer (full-night and full-day) conditions are compared here. The techniques utilized are referred to as the Spectral Proportion method (Chu et al. 2018) and the Interleaved method (Gardner and Chu 2020). Spectral proportion uses Monte-Carlo simulation to estimate the level of noise present in the signal to allow for removal, and the Interleaved method takes advantage of noncorrelation of the noise-components in adjacent atmospheric perturbation measurements to remove the noise-bias from the measurement. Both methods have differing advantages and disadvantages depending on the situation in which they are used. As would likely be expected, each technique performs better in high SNR conditions such as at low altitude or data collected with a low solar background. The Spectral Proportion method can be more easily used on a smaller sample size but is somewhat computationally expensive. Interleaved yields highly accurate results when large datasets can be averaged together, preventing its use in cases with small datasets and concealing run-to-run variations in the data. Overall, application of these techniques has provided new insights into the Antarctic gravity wave propagation behaviors, yielding wave energy measurements with a much higher level of confidence and accuracy than previously attainable and highlighting season trends in wave strength.