A statistical study of the ionospheric convection response to changing interplanetary magnetic field conditions using the assimilative mapping of ionospheric electrodynamics technique

We examine 65 ionospheric convection changes associated with changes in the Y and Z components of the interplanetary magnetic field (IMF). We measure the IMF reorientations (for all but six of the events) at the Wind satellite. For 22 of the events the IMF reorientation is clearly observed by both Wind and IMP 8. Various methods are used to estimate the propagation time of the IMF between the two satellites. We find that using the magnetic field before the IMF orientation change gives the smallest error in the expected propagation time. The IMF is then propagated to the magnetopause. The communication time between when the IMF encounters the magnetopause and the start of the convection change is estimated to be 8.4 (±8.2) min. The resulting change in the ionospheric potential is examined by subtracting a base potential pattern from the changing potential patterns. From these residual patterns, a number of conclusions are made: (1) the location of the change in convection is stationary, implying that the change in convection is broadcast from the cusp region to the rest of the ionosphere in a matter of seconds and that the elctric field mapped down the cusp controls the entire dayside ionospheric convection pattern; (2) the shape of the change in the ionospheric convection is dependent on the IMF component that changes, which is indicative of the change in the merging rate on the dayside magnetopause; (3) 62% of the events change linearly form one state to another, while 11% of the events change asymptotically; (4) the change in the ionospheric potential is linearly related to the magnitude of the IMF orientation, with B _z changes having a larger proportionality constant than B _y changes; (5) the ionospheric convection takes, on average, 13 min to completely reconfigure; and (6) some of the ionospheric convection changes occur on a timescale shorter than that of the corresponding IMF reorientation, possibly as a result of thresholding in the dayside merging region.