Professor Manuel Leote Eshkivel from the University of Lisbon NOVA, Portugal delivered a lecture for undergraduates and postgraduates of SFedU as part of the "Student Mathematical Seminar" of the SFedU Mathematical Center.

At the lecture, Professor Eshkivel showed a multiscale stochastic model of atmospheric phenomena capable of generating extreme precipitation leading to flooding. The model is based on a deterministic continuum-mechanical model based on an extended system of Navier-Stokes equations, with a probabilistic multiplicative cascade representing unresolved subcellular discontinuity. The cascade is dynamically conditioned by optimal transfer constraints and, as shown, approximates the stochastic continuity equation in the measure space. The statistics of extreme precipitation are derived using the theory of Gaussian multiplicative chaos, which leads to a power law in the tail of the distribution, determined by a physically parameterized discontinuity coefficient. This model provides a fundamental link between fluid mechanics, stochastic geometry, and statistical risk assessment.