A three-dimensional model of the generalized polar wind

The dynamic behavior of the generalized'' polar wind is investigated using a three-dimensional (3-D) dynamic model. The model is composed of two components. The high-altitude component is based on a macroscopic particle-in-cell (mac-PIC) approach that extends from an altitude of 1200 km to several Earth radii. The lower boundary conditions of the mac-PIC model are provided by a 3-D fluid-like model (low-altitude component) that extends down to 100 km in altitude. With the coupled model, the relevant equations are solved along magnetic flux tubes that convect across the high-latitude region. A large number (similar to 1000) of plasma flux tubes are followed. The total number of simulation particles in the mac-PIC component is 10(8)-10(9). The generalized polar wind is simulated for an idealized geomagnetic storm, with a time step of 2.5 s. The model properly accounts for many physical mechanisms such as ion-ion collisions, wave-particle interactions, magnetospheric energetic electrons, and low-altitude ion energization. The computing-intensive nature of the model requires utilization of supercomputers with thousands of processors. A 3-D picture is assembled from the temporal evolution of the individual flux tubes by keeping track of their locations. The resulting 3-D dynamic picture is investigated with special emphasis on the difference between the behaviors of the O+ and H+ ions. The main conclusions are as follows: (1) during the storm maximum phase, O+ may remain dominant for altitudes up to several Earth radii; (2) the O+-to-H+ density (n(O+)/n(H+)), velocity (UO+/UH+), and flux (FO+/FH+) ratios tend to be greatly enhanced during the storm with time delays of 0.5 to 1 hour; (3) the O+ downward flow tends to occur in the subauroral region and the dawnside of the polar cap; and (4) the O+ downward flux in the polar cap tends to occur below 1 RE altitude and during the storm decay phase.