This is an animation of electron density during the St. Patrick's Day solar storm in 2015. The model assimilates GNSS data from dual frequency receivers both on the ground and in orbit (COSMIC) in a Kalman filter to optimally estimate the state of the ionospheric electron density.

Questions:
Are the poll-ward extending features (e.g. 3/17 18:30) real?
Are they enough to cause increases in GPS errors?
Are they possible evidence of the penetration electric fields?

Answer:
They look physically realistic. They are consistent with expected dayside build up of plasma due to a combination of penetration electric fields and winds. The penetration E fields are themselves a combination of (magnetic) zonal electric fields, which tend to move the plasma vertically and poleward on the dayside during the driven pat of the storm, and the meridional (perpendicular to B) that can cause the plasma to stagnate and sit in sunlight and build up. What exactly is the combination of drivers is tricky without a physical model simulation. This plasma structure would negatively impact GPS receiver accuracy.