Search for: All records

Abstract. Medium-scale gravity waves (MSGWs) are atmospheric waves with horizontal scales ranging from 50 to 1000 km that can be observed through airglow all-sky images. This research introduces a novel algorithm that automatically identifies MSGWs using the keogram technique to study the waves over the Antarctic Peninsula. MSGWs were observed with an all-sky airglow imager located at the Brazilian Comandante Ferraz Antarctic Station (CF, 62° S), near the tip of the Antarctic Peninsula. Several preprocessing techniques are necessary to extract the parameters of MSGWs from the airglow images. These include projecting the images into geographical coordinates, applying a flat-field correction, performing consecutive image subtraction, and employing a Butterworth filter to enhance the visibility of the MSGWs. Additionally, a wavelet transform is used to identify the primary oscillations of the MSGWs in the keograms. Subsequently, a wavelet transform is also used to reconstruct the MSGWs and obtain the fitting coefficients of phase lines. The fitting coefficients are then used to calculate the MSGW parameters and assess the quality of the results. Simulations with synthetic images containing typical propagating gravity waves were conducted to evaluate the errors generated during the MSGW calculations and to determine the threshold for the fitting parameters. This methodology processed a year's worth of data in less than 1 h, successfully identifying most waves with errors lower than 5 %. The observed wave parameters are generally consistent with expected results; however, they show differences from other observation sites, exhibiting larger phase speeds and wavelengths.