Effects of high‐latitude conductivity on observed convection electric fields and Birkeland currents

Relationships between convective electric fields, Birkeland currents, and ionospheric conductivity at high latitudes are investigated by using simultaneous measurements of electric and magnetic fields and electron fluxes from the S3‐2 satellite. Cases in which the polar cap is in sunlight and darkness are studied to assess the importance of ionospheric conductivity. We find that when the ionospheric conductivity is high, as it normally is in the auroral zones and the sunlit polar cap, Birkeland current sheets exist that deflect the magnetic field approximately in the direction of the convection, typically antisunward in the polar caps and sunward in the auroral zones. Under winter conditions (a dark polar ionosphere), no large‐scale antisunward deflection of magnetic field lines is found in the polar cap. Magnetic field deflections in the auroral zone are the result of the region 1/region 2 current systems discussed by Iijima and Potemra (1976). In the cases studied, region 1 currents are not observed to extend significantly poleward of the electric field reversal. We interpret this result as indicating the existence of strong conductivity gradients near the poleward boundary of the auroral oval, even in sunlight conditions. Deflections of magnetic field lines in the summer polar cap, first reported by McDiarmid et al. (1978), result from currents which serve to link the high‐latitude generator to the polar ionospheric load. Two types of linkage appear in the data; (1) Region 1 currents are greater than region 2 currents, with the excess region 1 current flowing across the polar cap from dawn to dusk. (2) An extra set of field‐aligned currents exist which are clearly detached from those of region 1 and are entirely poleward of the convective electric field reversal. We find that the evidence for antisunward convection on closed field lines threading the boundary layer, cited by McDiarmid et al. (1978), is inconclusive. With our more complete data set there is some evidence at ionospheric altitudes for a boundary layer, but in such cases there is only a modest (a few kilovolts) potential drop across the layer and there are insignificant Birkeland currents.