Achieving Forecasts in the Thermosphere and Ionosphere with Lead Times of a Few Days

Abstract:

Forecasting space weather requires the development of first-principles-based models for the coupled Sun-Earth System. To achieve lead times of a few days for disturbances in planetary thermospheres and ionospheres, models of the solar wind propagating through the heliosphere are required. An active research community has achieved a suite of models that describe conditions within, and coupling between, the solar wind, the Earth’s magnetosphere and the ionosphere. At least one version of each model in the suite is available for broad community use and investigation. While these models represent an important step towards the goal of achieving accurate space weather forecasts, it is recognized that certain physical processes are not represented in these models, with unknown impact on the forecasts. We suggest an approach towards improved space weather forecasts that emphasizes model evaluation techniques, providing detailed information on how the physical processes represented in the models affect forecasts based on those models. Such detailed information permits the models to be used for investigating science questions, and permits observations to be the basis for improving the models and increasing scientific understanding. In this talk, we present upper atmosphere forecasting results using community models of the coupled Sun-Earth system. We describe our approach to analyzing the physics of ionospheric storms as represented in the Global Ionosphere Thermosphere Model developed at the University of Michigan. Such analysis involves both an approach to model diagnostics and the use of a comprehensive set of observations. We compare forecast results for high-speed solar wind stream storms and storms initiated by solar coronal mass ejections. First-principle and empirically based approaches to coupling between the solar wind and ionosphere are compared. These comparisons provide insight into the strengths and limitations, and areas for future improvement, of first-principles models as they are used to forecast the Sun-Earth interaction.