New Lidar Technique to Eliminate Noise Floor from Spectra: A Demonstration using Antarctic Observations and Modeling

Noise in lidar data greatly complicates the derivation of crucial second-order parameters (such as atmospheric wave energies, fluxes, and spectra), especially when the signal-to-noise ratio is low, leaving a bias in the derived parameters. Following studies exploring the use of covariance to eliminate the noise bias in the temporospatial domain, this study explores using the cross power spectral density to eliminate the noise floor in spectral data. It is found that the method is effective in most cases, and the study also explores the accuracy and precision of the approach.