P1.5
Operational Real-Time Run of the UAF Eulerian Parallel Polar Ionosphere Model (UAF EPPIM) and its Forecasting Capabilities.
Sergei Maurits, University of Alaska, Fairbanks, AK; and J. McAllister and B. Watkins
UAF EPPIM is the first principles model of the polar ionosphere, based on the flexible modular structure. It supports incorporation of various types of case-specific data for ionospheric simulations in the post-analysis mode, provided the data is available. Alternatively, when specific data is absent, EPPIM also has a capability to invoke statistical modules, which are controlled by the standard set of geophysical indexes such as Kp/Ap, F 10.7, and the upstream solar wind characteristics. This second mode of the model operation describes key ionospheric parameters as solar UV intensity, auroral precipitation, and electromagnetic drifts through respective statistical (or "climatological") modules, driven by the period-specific time series of the geophysical indexes. Their real-time availability at the WWW-site of NOAA Space Environment Center facilitates continuous forecasting run of the UAF EPPIM. This real-rime high-resolution capability is based on the Arctic Region Supercomputing Center computational and networking resources, combined with thorough computational optimization of the model code (see the UAF EPPIM WWW-site at http://www.arsc.edu/SpaceWeather for details and the list of available forecasts.)
This forecasting approach employs available "climatological" approximations, which are driven by the period-specific or "weather-scale" drivers, as a continuous feed for the ionospheric model. The validity and limitations of such approach are investigated in this research. Ongoing year-long massive scale (100,000+ of all cases) comparisons of the EPPIM-predicted critical plasma frequencies with the ionosondes' network measurements demonstrated that the approach forecasting accuracy is in the range of 30-70% for the day-side solar maximum winter ionosphere. Summer comparisons generally improve the statistical agreement up to 15% or even better for some locations. Night-side comparisons reveal systematic statistical bios of the model, which provided information for the empirical model corrections.
An experience of transfer of the research model to operational capability is described. Organization of the continuous WWW-based run is discussed, particularly a dynamic shift of the model time to accommodate the solar wind delay from the L1 sampling point while propagating toward the Earth. Research program for the next tier of comparisons and validations includes elements of the adaptive schemes and has an emphasis on the real-time data utilization when available. Such massive statistical validations contributes to a formulated goal of the National Space Weather Program of establishing the currently achieved capabilities in the polar ionospheric modeling and sets a target for possible improvements.
Poster Session 1, Space Weather Posters
Wednesday, 14 January 2004, 2:30 PM-4:00 PM, Hall AB
Previous paper