Observations of the Fe and O charge states in the solar wind and interplanetary coronal mass ejections (ICMEs) generally exhibit a positive correlation between the average charge states of Fe and O ($ avQ Fe $ and $ avQ O $). Because Fe and O charge states freeze at different heights in the corona, this positive correlation indicates that conditions at different heights in the corona vary as a whole. We identify short time periods in the solar wind that exhibit anticorrelations between the average Fe and O charge states and investigate their properties. We aim to distinguish whether these anticorrelations are due to the related solar sources or to transport effects (e.g., differential streaming). We study kinetic properties of the solar wind related to these anticorrelated structures as well as heavy ion differential streaming in order to infer a possible relationship between conditions in coronal source regions and the reported in situ measurements. We employed a recently developed sliding-window cross-correlation method to locate anticorrelated structures in the solar wind composition measurements between 2001 and 2010 from the Advanced Composition Explorer (ACE). To account for fluctuations and measurement uncertainties, we varied the timescales and temporal lags. We determined the onset and end times of the gradual increases or decreases in the average charge states of O and Fe and analyzed the kinetic and plasma properties of the anticorrelated structures. We identified 103 anticorrelated structures both in the solar wind and in ICMEs. The behavior of $ avQ Fe $ is strongly related to solar wind kinetic properties, including proton speed, proton temperature, and the proton-proton collisional age. We find that the anticorrelation of $ avQ Fe $ and $ avQ O $ during these time periods cannot be explained by differential streaming nor by unrecorded hot plasma ejections. Thus, the measured anticorrelated variations in $ avQ Fe $ and $ avQ O $ probably indicate that changes in coronal conditions at different freeze-in heights may follow opposite monotonic trends.
Read full abstract