Geomagnetic storms over the last solar cycle: A superposed epoch analysis

Abstract

[1] Presented here is a discussion of the results of a superposed epoch analysis of geomagnetic storms over the last solar cycle. Storms, identified by means of their characteristic SYM-H evolution, are separated by size into weak (−150 < SYM-H ≤ −80) nT, moderate (−300 < SYM-H ≤ −150) nT, and intense (SYM-H ≤ −300) nT categories. Where possible, the corresponding solar wind (SW) onset mechanisms were located by means of 1 min ACE OMNI data. Intense storms were observed to be driven solely by coronal mass ejections (CMEs); moderate storms were dominated by CME onset, while only weak storms were driven by both CMEs and corotating interaction regions (CIRs) at a ratio of ∼2:1, respectively. As might be expected, more intense storms resulted from the largest SW enhancements. Individual storm phase durations for different storm sizes were investigated, revealing that the duration of the main phase increases with storm size to a critical point, then decreases for more intense storms, contrary to the findings of a previous study by Yokoyama and Kamide (1997). Various SW-magnetosphere coupling functions were investigated for this data set in an attempt to estimate storm size from SW conditions.

Influence of the interplanetary driver type on the durations of the main and recovery phases of magnetic storms

“…For example, Wang et al [2003] found that a southward field component Bz ≤ −3 nT with a duration of Δt ≥ 1 h results in moderate magnetic storms (Dst min ≤ −50 nT) and threshold values of Bz ≤ −6 nT with a duration of Δt ≥ 2 h result in strong magnetic storms (Dst min ≤ −100 nT), and these results differ somewhat from the results of previous papers by Russell et al [1974], Gonzalez and Tsurutani [1987], and Gonzalez et al [1994]. The papers mentioned above analyzed the duration of the southward IMF Bz component which is enough to generate magnetic storms, but the duration of the main phase was studied less intensively and it was shown that the duration of the main phase may be from 2 h to 1 day [see, e.g., Yokoyama and Kamide, 1997;Vieira et al, 2004;Vichare et al, 2005;Gonzalez and Echer, 2005;Yermolaev et al, 2007aYermolaev et al, , 2007bHutchinson et al, 2011;Nikolaeva et al, 2012, and Though it is well known that the dynamics of magnetic storms depends on the type of interplanetary drivers (see, e.g., papers by Borovsky and Denton [2006], Yermolaev et al [2010Yermolaev et al [ , 2012a, Guo et al [2011], Liemohn and Katus [2012], Nikolaeva et al [2013], Cramer et al [2013], and references therein), the majority of previous works have not made a selection by types of solar wind streams which generated the storms. In other works, the selection was either performed only for a limited type of solar wind or for a complex of types.…”

“…In the case of selecting by type of driver, our results show a small (within the statistical dispersion) peak in Figure 2 at Dst min in the range from −100 to −125 nT only for MC-induced storms (red line), while for the combined "ICME" = MC + Ejecta + Sheath type (similar to the type used in the work by Hutchinson et al [2011] and shown in Figure 2 by a black line), no maximum is observed.…”

“…Figure 2 shows the dependencies of the main phase durations!ΔT on storm intensity Dst min for magnetic storms induced by various types of solar wind streams when the data are selected in 25 nT bins of Dst min . Different color symbols show various types of drivers in accordance with the legend under the figure, and "ICME" events, which include intervals of MC and Ejecta, and Sheath before them (the sum of Sheath, MC, and Ejecta), and is similar to "CME" used in the paper by Hutchinson et al [2011]. Taking the large spread of the data into account, we can suggest that there is a slight increase of the duration ΔT in the region of (−125 ... −100 nT) only for MC.…”

“…For instance, the authors of several investigations (see, e.g., papers by Turner et al [2009], Richardson and Cane [2012], Keesee et al [2013], Potapov [2013], Yuan and Zong [2013], and Cramer et al [2013]) compared CIR-and CME-driven impacts on the magnetosphere, but they did not take into account that a CME on the Sun results in two different geoeffective types of solar wind streams at 1 AU: an interplanetary CME (ICME-magnetic cloud MC or Ejecta) and a compressed region Sheath before it and did not make a selection between Sheath-and ICME-driven storms. As in this paper, we plan to compare our investigations with results of recent paper by Hutchinson et al [2011], we indicate a definition of "CME" type in the caption of Figure 2 of this paper: "A typical coronal mass ejection (CME) trace seen in ACE OMNI data, with simultaneous increases in all components: interplanetary magnetic field (IMF) magnitude, solar wind (SW) speed, pressure, density, and temperature. " This is the definition of compression region Sheath before an ICME.…”

“…The dependencies of the durations of the main and recovery phases of storms on the storm intensity (Dst min ) have been studied in several papers [Yokoyama and Kamide, 1997;Hutchinson et al, 2011]. Yokoyama and Kamide [1997] showed that there is a correlation between the duration of both phases and Dst min .…”