Estimation of the vertical wavelength of atmospheric gravity waves from airglow imagery

Atmospheric Gravity Waves;Airglow;Imaging;Parameter estimation;Gabor filter;Vertical wavelength;Greenline emission;New Mexico;Socorro;National Solar Observatory

Kanwar, Uday ; Makela ; Jonathan J.

关键词: Atmospheric Gravity Waves;    Airglow;    Imaging;    Parameter estimation;    Gabor filter;    Vertical wavelength;    Greenline emission;    New Mexico;    Socorro;    National Solar Observatory;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/29699/Kanwar_Uday.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Abstract In the summer of 2010, two imagers were installed in New Mexico with the objective of making stereoscopic observations of atmospheric gravity waves (AGWs). As AGWs propagate vertically, they spatially perturb the airglow emission layers in all three dimensions. Estimates of the vertical wavelength, horizontal wavelength, and the intrinsic frequency are needed to characterize an AGW and quantify its effects on upper atmospheric dynamics. The dispersion relation describes the relationship between vertical and horizontal wavelengths as a function of the intrinsic frequency. Thus, any two of the three aforementioned parameters can be used to determine the third. Mesospheric winds are hard to measure and consequently the intrinsic frequency is difficult to estimate. However, the horizontal wavelength can be directly measured from airglow imagery once the three-dimensional imager field of view is projected onto the two-dimensional image plane. This thesis presents a method to estimate the vertical wavelength using an airglow perturbation model proposed by Anderson et al. (2009). The model is subsequently validated using the observations from ground-based imagers installed in New Mexico. Abstract The perturbed airglow is modeled as a quasi-monochromatic wave and thus, it can be characterized using only a few parameters, one of which is the vertical wavelength. Because the vertical wavelengthis embedded in both the phase and the magnitude of this model, two values of the vertical wavelengthare estimated by applying two different parameter estimation techniques on the phase and magnitude. The estimation of the vertical wavelengthfrom the phase of the model entails solving an overdetermined system of linear equations by minimizing the sum of the squared residuals. This estimate is then compared to that obtained by iteratively finding the best approximation to the roots of a function, representing the magnitude of the perturbation model. These two techniques are applied on three nights in 2010, and the estimates for the vertical wavelength match to within a few kilometers. Thus, the perturbation model is validated using real data.