The main aim of this proposal is to implement a synergistic approach leveraging innovative High-Performance Computing techniques and observations to advance fundamental knowledge on the dynamics underlying the emergence of large scale extreme events and local instabilities in stratified and rotating turbulent fluids and their feedback on mixing and transport properties of such flows. This project has the ambition to achieve an unprecedented statistical and phenomenological characterization of large scale extreme events and their feedback on the small scales in a novel paradigm in fluid turbulence: that of three dimensional rotating and stratified flows where the energy goes to both large and small scales with a dual constant flux cascade. The fundamental study proposed is therefore a synthesis of major research themes of the Axe 7.1 The capability to design state of the art high-resolution DNS of rotating and stratified turbulence able to capture all these phenomena in a parameter space compatible with the real flows, together with the expertise and means present within the scientific team of the EVENTFUL project to design experiments and field campaigns in the stratosphere and Mesosphere-Lower Thermosphere (MLT), will lead to a comprehensive assessment of the emergence and dynamics of large-scale powerful events in the active flow fields, by means of an – innovative and synergistic – combined use of observations, DNS and machine learning techniques, as detailed in the following. Objectives will be accomplished taking advantage of the extensive experience gained by the scientific team in implementing the proposed methodologies and in the investigation of anisotropic turbulent flows.