Event Details

Mathew Owens -University of Reading, UK

Abstract

As scientists, we strive to construct models and simulations that describe natural systems in greater and greater fidelity. But this comes with a cost; increased code complexity and computational demand. The solar wind is typically simulated using three-dimensional magnetohydrodynamics (3DMHD), providing both a valuable research tool to understand the dynamic heliosphere, and a means to forecast space weather a few days ahead. But beyond the upper corona, the magnetic field is largely passive and the flow is highly radial. Thus one-dimensional, hydrodynamic solutions to the same solar wind initial conditions provide a good “surrogate” for 3DMHD using a much simpler code base and at a fraction computational cost. For space-weather forecasting, this is a valuable complimentary approach to traditional 3DMHD. It enables rapid forecasts and large ensembles of model solutions to comprehensively investigate parameter space, and hence quantify forecast uncertainty. It also permits the use of powerful data assimilation techniques to exploit the available solar wind observations and provide much needed additional constraints on the simulations, limiting how far they can diverge from reality.