Is there a magnetospheric ground state? Do the position and size of the auroral oval depend on the magnetic dipole tilt angle at quiet magnetospheric conditions? In order to address these questions, northern hemisphere images of the aurora at 1356 ¿, obtained by Polar BEAR at solar minimum (beginning of 1987), were related to high temporal resolution IMP 8 measurements of the interplanetary magnetic field, to solar wind velocity, and to the ground-based activity index Kp. The first problem was addressed by a two-dimensional correlation study of the repeatability of auroral emissions in corrected geomagnetic space at conditions of minimum energy transfer from the magnetosphere. The correlation measure of auroral images was 0.6--0.85. Error simulations indicate that given the uncertainties in pixel position and intensity, the maximum expected value of the correlation measure is 0.65--0.9. The notion of a ground state magnetosphere is therefore supported by our data. Repeatability was found at the same level regardless of time or reconfigurations of the magnetosphere between images and independent of magnetic time sector. The second problem was addressed by relating latitudinal shifts of the aurora with dipole tilt angle without resorting to auroral boundary specification.

Our data indicate that the latitude of the continuous aurora is related to the dipole tilt angle at quiet magnetospheric conditions. In the winter hemisphere a 10¿ increase in the dipole tilt angle causes a 1¿ decrease (increase) in the latitude of auroral emissions at noon (midnight). The magnetic local time distribution of the latitudinal shifts with dipole tilt angle support a simple model in which the dipole tilt angle determines the position of the center of the auroral circle along the magnetic meridian 1320--0120 MLT (for IMF By positive) and does not affect its radius. ¿ American Geophysical Union 1993