IMF BX and Dipole Tilt Dependence on Transpolar Arcs

Abstract

<p>Transpolar arcs (TPAs) are predicted by many models to appear in both hemispheres, as so-called conjugate TPAs. However, some observations have suggested that this is not always the case, and that there is an IMF B<sub>x</sub> dependence on whether TPAs appear on both hemispheres or not. Specifically, it has been suggested that TPAs only appear on the northern hemisphere for negative IMF B<sub>X</sub> and vice versa for positive IMF B<sub>X</sub>. Furthermore, a positive Earth dipole tilt is predicted to have a similar effect on TPA occurrences as a negative IMF B<sub>X</sub> and vice versa. It is also known that TPAs appear on different locations on the auroral oval, i.e., dawn-, dusk- or both sides of the oval, depending on IMF B<sub>Y</sub>. However, the role of IMF B<sub>X</sub> and IMF B<sub>Z</sub> for the TPA location remains unclear, with some previous observations suggesting a correlation with IMF B<sub>X</sub>.</p><p>In this study, we investigate the influence of IMF B<sub>X</sub> and dipole tilt on TPAs by statistically analyzing observational data. We analyze TPA datasets from four previous studies, as well as our own TPA dataset, created from DMSP satellite measurements. At first glance, the data suggests that there is a strong correlation between both IMF B<sub>X</sub> and dipole tilt, and TPA observations in a specific hemisphere. However, when normalizing the data to the solar wind distribution and when taking observational bias into account, this correlation disappears. We therefore conclude that there is no clear correlation between neither IMF B<sub>X</sub> nor dipole tilt and in which hemisphere a TPA appears. We further analyze four of the five datasets with respect to dawn-dusk appearances of TPAs and its correlation to IMF B<sub>X</sub>, B<sub>Y</sub> and B<sub>Z</sub>. Here, the results for the datasets mostly agree with previous observations. Finally, we discuss the potential causes for the few non-conjugate TPAs, by studying our own TPA dataset in further detail.</p>