Mediterranean Tropical-Like Cyclones (MTLCs), commonly referred to as Medicanes, are tropical-like storms increasingly affecting the Mediterranean basin. Their dynamics result from the interaction of convection, boundary-layer processes, and mesoscale circulation, leading to a multiscale organization that is still only partially understood. This study examines the internal structure of Medicane Ianos by combining a 1 km Weather Research and Forecasting (WRF) simulation with two complementary data-driven approaches: Proper Orthogonal Decomposition (POD) for the spatial organization of the flow, and Empirical Mode Decomposition with Hilbert Spectral Analysis (EMD-HSA) for its temporal scaling properties. The POD results reveal a vertically stratified system dominated near the surface by boundary-layer forcing, with energy concentrated in a small number of coherent modes. Higher in the troposphere, the flow becomes more uniform and isotropic, while small-scale features persist as embedded structures shaped by the evolving circulation. Temporal fluctuations inside the eyewall display clear changes with height: temperature variability shows strong persistence in the lower troposphere, while correlations weaken progressively at higher levels, a pattern confirmed by the vertical distribution of Hurst exponents. Overall, the analysis depicts Ianos as a layered multiscale system and demonstrates how data-driven decomposition can effectively complement dynamical modeling in the study of MTLCs. • Ianos displays a vertically stratified structure, with planetary boundary layer forcing dominating the lowest levels and a transition toward more uniform flow in the free troposphere. • Energy distribution across scales reveals that large vortical structures govern the system, while smaller features persist as embedded patterns shaped by the evolving circulation. • Temporal fluctuations inside the eyewall show stronger persistence near the surface and a gradual weakening aloft, revealing a layered nature of the cyclone’s internal variability.