Organizers: Rupert Klein, Sebastian Reich, Daniel Baum, Stephan Pfahl

Part B: Computation and Data Analysis

Abstract:

Significant atmospheric phenomena, such as fronts, cyclones, and heat waves, stand out amongst the daily ups and downs of the weather. They are well recognizeable owing to their specific spatial structure and their longevity relative to the ubiquitous much faster transients associated, e.g., with the convection and turbulent processes that constitute them. They are "persistent" in this sense, and it is this persistence that makes these structures identifiable in observational or simulation data and particularly accessible to theoretical analysis, efficient numerical approximation, and data-based tracking or modelling. This minisymposium collects reports from such diverse approaches towards improved scientific understanding of these often severe weather patterns.

Speakers

14:30h: Ann Almgren (LBNL)

Balancing Theory, Numerics and Observations in the Development of a new Regional Atmospheric Model

The development of a new code for regional atmospheric modeling requires a large number of design choices, from the choice of variables and fundamental equations to the choice of temporal and spatial discretizations to the question of which physical processes to include and in what form. In this talk I describe the path we have followed in the development of the Energy Research and Forecasting Model (ERF). I will present some of the successful validation results as well as some of the unexpected challenges we have faced.

15:00h: Amir Maghoul (FU Berlin)

Numerical and asymptotic analysis of atmospheric flows on the quasi-geostrophic and Obukhov scales

This talk presents two applications of asymptotic scale analysis to geophysical fluid dynamics, focusing on the development of reduced models and their numerical implementation. In the first part, motivated by the lack of a unified theoretical framework describing the full life cycle of atmospheric blocking, we introduce the Obukhov scaling regime, in which the characteristic length scale of the flow is comparable to the external Rossby radius. Through a multiscale analysis of the Obukhov and quasi-geostrophic (QG) regimes, we show that the resulting governing equations provide additional insight into the dynamics and persistence of atmospheric blocking beyond what can be obtained from the single-scale QG equations. In the second part, we introduce the approach for applying asymptotic scaling analysis directly to a numerical discretization. Using this method, we derive the QG regime asymptotic equations within the BK19 finite volume solver and introduce corresponding blending switches. This enables seamless access to the QG regime without modifying the underlying discretization. The approach provides a general framework for including reduced models into numerical schemes while preserving their original structure.

15:30h: Ingrid Hotz (Linköping University)

Topological Feature Tracking in Visualization

Feature tracking is a fundamental problem in the analysis of time-dependent data, arising in a wide range of applications. In practice, it involves two closely connected challenges: defining an appropriate feature and establishing a stable method for tracking its evolution over time. For both, the specific application context plays an essential role. Topological methods provide a powerful basis for feature definition. They offer mathematically well-understood descriptors that are robust to perturbations, naturally support multi-scale analysis, and allow for a flexible range of feature definitions. Over the past decades, a variety of methods have been developed to track such topological features, reflecting the diverse requirements of different applications. In this talk, I will first give a short overview of a unifying framework for classifying existing feature tracking methods. The framework highlights common principles underlying different approaches and provides a basis for understanding their relationships and differences. I will then discuss an example motivated by cyclone tracking, illustrating how the application context influences both the definition of relevant topological features and the design of a suitable tracking method.