WAVEGUISE - Wavefield Analysis and Segmentation Unraveling an Interpretable Set of Wave Packets

Kurzfassung

Atmospheric gravity waves (GW) arise from a wide range of sources and span a broad spectrum of temporal/spatial scales and phase speeds. Depending on the source, we find e.g. GWs with short scales being stationary above a tiny island or with large scales propagating fast away from regions of geostrophic adjustment. GWs from various sources can superimpose and interact with each other such that an observer has a hard time to interpret his measurements. WAVEGUISE is a tool designed to disentangle this wave soup in atmospheric observations. It can be applied to data sets of up to three dimensions and is based on the continuous wavelet transform, enabling a decomposition of the observed wave field into localized spectral components while preserving spatial and temporal information. By exploring this spectral space, coherent wave packets emerge as multi-dimensional peaks in spectral power that share common scale and localization properties. WAVEGUISE segments these peaks and returns a set of individual wave packets that can be analyzed separately, providing a more interpretable representation of complex wave fields compared to existing approaches. In this presentation, I demonstrate three applications of WAVEGUISE to atmospheric observations: (i) a pressure time series from a long-duration balloon campaign, (ii) a satellite swath observation, and (iii) an airglow imager video sequence.