This paper presents a detailed model-data comparative study of the 17 March 2015 geomagnetic storm using the high-resolution version of the thermosphere-ionosphere-electrodynamic general circulation model and the total electron content observations from a dense global navigation satellite system network. Driven by time-dependent high-latitude ionospheric convection and auroral precipitation inputs, together with an empirically defined subauroral plasma stream (SAPS) field, our simulation reproduce many observed storm-related ionospheric phenomena, including large-scale traveling ionospheric disturbances over Europe, the effects of prompt penetration electric field over South and Central America, and the formation of a storm-enhanced density (SED) plume across the continental United States. Our simulation results reaffirm a number of important characteristics concerning the SED plume: (1) enhanced background ionospheric density is a necessary but not sufficient condition, and enhanced ion drift is required to form the SED plume; (2) the SAPS flow channel does not directly transport the plasma from midnight to postnoon via dusk to form the SED plume, instead, the SED plume is formed at the equatorward and westward edge of the SAPS channel; and (3) the SED plume appears to subcorotate with respect to the Earth.Plain Language Summary Storm-induced ionospheric density variations are a major concern of near-Earth space environment as they could drastically disrupt satellite navigation and telecommunication systems. Although ionospheric disturbances such as traveling ionospheric disturbances (TIDs) and storm-enhanced density (SED) are commonly observed during geomagnetic storms, accurate specification of these phenomena remains a great challenge for geospace models. This paper presents a detailed model-data comparative study of the well-known March 2015 St. Patrick's Day storm. The simulated TIDs and SED structures from the thermosphere-ionosphere-electrodynamic general circulation model (TIEGCM) are compared with the total electron content observations for a dense global navigation satellite system (GNSS) network over Europe, South and Central America, as well as North America. While the model reproduces many observed storm-related ionospheric features, quantitative differences between the simulation results and the GNSS data indicate that further improvements to the TIEGCM (such as a more realistic, self-consistent electrodynamic coupling of the ionosphere and magnetosphere) are required in order to meet the challenges of space weather specification and eventually forecast. This study not only serves a meaningful validation of our numerical model but also sheds some new lights on an apparent paradox concerning the formation of the SED plume.
Key Points:• We have carried out a detailed model-data comparison as a way to validate the model outputs • Our simulation results reproduce many observed features, including traveling ionospheric disturbances and the storm-enhanced density plume • Our results shed new lights on th...