Atmospheric dynamical cores numerically represent the resolved fluid motions in weather and climate models and strongly influence their accuracy, stability, conservation, computational efficiency, and physical fidelity. Using the 2026 FIFA World Cup as an organizing analogy, I will survey the international landscape of dynamical-core development and the strategic choices underlying atmospheric general circulation models. These choices include governing equation sets, computational grids and staggering, vertical coordinates, numerical methods, physics–dynamics coupling, dissipation, filtering, and parameter tuning. Model intercomparison projects and hierarchical test suites serve as referees, exposing numerical artifacts, grid imprinting, excessive or insufficient diffusion, and divergent representations of precipitation and stratospheric circulation. Results demonstrate that seemingly technical design and tuning decisions can substantially alter simulated weather systems and large-scale circulation.

I will also examine emerging frontiers: global convection-allowing simulations at 1–3 km resolution, adaptive mesh refinement, nonhydrostatic deep-atmosphere equation sets, and potentially non-spherical Earth formulations. These advances require scalable, hardware-agnostic software that exploits graphics processing units and emerging programming frameworks such as C• /Kokkos, Python, and JAX. Machine-learning emulators, artificial-intelligence-assisted development, efficient data strategies, and models carrying hundreds of tracers offer additional opportunities. Together, these developments point toward more accurate, physically complete, computationally portable, and systematically evaluated dynamical cores for next-generation weather, climate, and whole-atmosphere prediction.

Bio:
Christiane Jablonowski is a Professor in the Department of Climate and Space Sciences and Engineering at the University of Michigan. She also worked at the National Center for Atmospheric Research (NCAR), NOAA's Geophysical Fluid Dynamics Laboratory, and the European Centre for Medium-Range Weather Forecasts (ECMWF). Dr. Jablonowski’s research lies at the interface between weather and climate modeling, scientific computing, and data science.

Her research portfolio includes atmospheric fluid dynamics, weather and climate modeling including high-resolution modeling, climate model hierarchies, scientific computing, and machine learning techniques for the climate sciences. She works with the weather and climate models from NCAR, NOAA, and the Department of Energy (DoE). Dr. Jablonowski is the recipient of a DoE Early Career Award and the Presidential Early Career Award for Scientists and Engineers (PECASE). She is a co-lead of NOAA's Convection-Allowing Model for short-range weather forecasts with the Unified Forecast System (UFS) and serves on the Steering Committee for NCAR’s Community Earth System Model (CESM).