N. U. Crooker

Relationship of halo coronal mass ejections, magnetic clouds, and magnetic storms

Halo coronal mass ejections (CMEs) had been rarely reported in coronagraph observations of the Sun before the Solar and Heliospheric Observatory (SOHO) mission. Since mid-1996, however, the SOHO Large Angle Spectrometric Coronagraph (LASCO) instruments have observed many halo or partial-halo CMEs. A halo CME, especially when associated with solar activity near sun center, is important for space weather concerns because it suggests the launch of a potentially geoeffective disturbance toward Earth. During the post-solar minimum period from December 1996 to June 1997, we found that all six halo CMEs that were likely Earthward-directed were associated with shocks, magnetic clouds, and moderate geomagnetic storms at Earth 3–5 days later. The results imply that magnetic cloud-like structures are a general characteristic of CMEs. Most of the storms were driven by strong, sustained southward fields either in the magnetic clouds, in the post-shock region, or both. We discuss the characteristics of the halo events observed during this period, their associated signatures near the solar surface, and their usefulness as predictors of space weather at Earth.

Multiple heliospheric current sheets and coronal streamer belt dynamics

The occurrence of multiple directional discontinuities in the coronal streamer belt at sector boundary crossings in the heliosphere, often ascribed to waves or kinks in the heliospheric current sheet, may alternatively be attributed to a network of extended current sheets from multiple helmet streamers with a hierarchy of sizes at the base of the corona. Frequent transient outflows from these helmets can account for a variety of signatures observed at sector boundaries, including ordered field rotations, planar magnetic structure, and sandwichlike plasma structure.

Lobe cell convection as a summer phenomenon

Patterns of average potential over the high-latitude ionosphere in winter show that the dusk convection cell dominates the dawn cell, consistent with the presence of a day-night conductivity gradient, as predicted by a number of models. However, in the summer hemisphere, when IMF B{sub y} is strongly positive, the dusk cell so dominates the dawn cell that the latter nearly disappears; and when IMF B{sub y} is strongly negative, the cells are most nearly equal. The difference between winter and summer can be explained by the addition in summer of a single lobe cell, that is, a cell confined to open field lines, circulating within the dusk cell of the two-cell pattern when B{sub y} is positive and within the dawn cell when B{sub y} is negative. The result is consistent with predictions of the overdraped lobe model, that lobe cells occur in only one hemisphere at a time, and that their occurrence is controlled by dipole tilt. 35 refs., 3 figs.

The semiannual variation of great geomagnetic storms and the postshock Russell‐McPherron effect preceding coronal mass ejecta

The occurrence rate of great geomagnetic storms displays a pronounced semiannual variation. Of the forty-two great storms during the period 1940–1990, none occurred during the solstitial months of June and December, and 40% (17) occurred during the equinoctial months of March and September. This suggests that the semiannual variation found by averaging indices is not the result of some statistical effect superposed on the effects of random storm occurrence but rather is dominated by the storms themselves. Recent results indicate that the intense southward interplanetary magnetic fields (IMFs) responsible for great storms can reside in the postshock plasma preceding the driver gas of coronal mass ejections (CMEs) as well as in the driver gas itself. Here we propose that strong southward fields in the postshock flow result from a major increase in the Russell-McPherron polarity effect through a systematic pattern of compression and draping within the ecliptic plane. Differential compression at the shock increases the Parker spiral angle and, consequently, the azimuthal field component that projects as a southward component onto Earths dipole axis. The resulting prediction is that southward fields in the postshock plasma maximize at the spring (fall) equinox in CMEs emerging from toward (away) sectors. This pattern produces a strong semiannual variation in postshock IMF orientation and may account at least in part for the observed semiannual variation of the occurrence of great geomagnetic storms.

Opening the cusp

Defining the equatorward boundary of the cusp region in the ionosphere as the projection of the merging line from the magnetopause, we use a quantitative, geometrically realistic model to show how the local time span of the cusp increases with increasing merging rate for southward interplanetary magnetic field. Since the merging rate is fixed by the magnitude of the magnetic field component normal to the model magnetopause and then normalized to the cross-polar-cap potential, the result gives the variation of cusp local time span as a function of potential. From 0 kV up to 60 kV, the cusp expands from a point at noon to 2 hours on either side. Nearly tripling the voltage to 170 kV adds one more hour to each side, yielding a span from 0900 to 1500 LT. As an example of broad local time span during magnetically active periods, we present spacecraft observations of cusp particles during the great storm of March 1989 that cover more than 8 hours of local time. An important aspect of the result is the demonstration that a merging line of fixed length on the magnetopause, as assumed in the model, maps to a projected length in the ionosphere that increases as the funnel-shaped cusp opens. This behavior contrasts with earlier models that have cleft rather than cusp geometry, where the projected merging line length is proportional only to its length on the magnetopause. The model results are used to construct the footprint of a flux transfer event caused by time variations of the merging rate, uniform along the length of the merging line. The cusp geometry distorts the field lines mapped from the magnetopause to yield footprints with dawn and dusk protrusions into the region of closed magnetic flux.

On the low correlation between long-term averages of solar wind speed and geomagnetic activity after 1976

During solar cycle 20, the first full cycle with measurements of solar wind parameters, geomagnetic activity measured by Ap was found to correlate with the square of solar wind speed V, and activity measured by Dst was found to correlate with the product of V and the southward component of the interplanetary magnetic field, B[sub s]. Both of these correlations break down during cycle 21. In the case of Ap, the much stronger variation of B[sub s] in cycle 21 compared to cycle 20 makes clear that the B[sub s] contribution to activity is important on yearly as well as shorter time scales. The product B[sub s]V[sup 2] gives an excellent correlation with Ap over both cycles. In the case of Dst, the stronger variation of B[sub s] in cycle 21 causes a stronger variation in B[sub s]V, which is not reflected in Dst, perhaps because Dst also depends upon solar wind dynamic pressure in a nonlinear way. 12 refs., 4 figs.

A test of source-surface model predictions of heliospheric current sheet inclination

The orientation of the heliospheric current sheet predicted from a source surface model is compared with the orientation determined from minimum-variance analysis of ISEE 3 magnetic field data at 1 AU near solar maximum. Of the 37 cases analyzed, 28 have minimum variance normals that lie orthogonal to the predicted Parker spiral direction. For these cases, the correlation coefficient between the predicted and measured inclinations is 0.6. However, for the subset of 14 cases for which transient signatures (either interplanetary shocks or bidirectional electrons) are absent, the agreement in inclinations improves dramatically, with a correlation coefficient of 0.96. These results validate not only the use of the source surface model as a predictor but also the previously questioned usefulness of minimum variance analysis across complex sector boundaries. In addition, the results imply that interplanetary dynamics have little effect on current sheet inclination at 1 AU. The dependence of the correlation on transient occurrence suggests that the leading edge of a coronal mass ejection (CME), where transient signatures are detected, disrupts the heliospheric current sheet but that the sheet re-forms between the trailing legs of the CME. In this way the global structure of the heliosphere, reflected both in the source surface maps and in the interplanetary sector structure, can be maintained even when the CME occurrence rate is high.

Properties of interplanetary magnetic sector boundaries based on electron heat‐flux flow directions

We have used the solar wind electron heat flux flow directions to determine the interplanetary magnetic field polarities for the ISEE 3 period from 1978 to 1982. This technique assumes that the heat flux electrons flow away from the Sun along magnetic field lines. It provides the field polarities independently of the field directions. The resulting distribution of sector durations and the changes in that distribution with solar activity cycle are presented for four 1-year periods. The large-scale sectors expected from extrapolation of the Stanford source surface maps are present along with a population of small-scale sectors with a peak in the time range of 9 hour to 1 day. About half the small-scale sectors contain bidirectional electron (BDE) flows, suggesting their origins in coronal mass ejections. We also examine cases of false polarities, in which the directions of the fields imply polarities opposite to those determined from the heat-flux directions. These constitute only 6 to 8% of all the hourly averages of the data. The majority (78%) of these false polarity regions were not associated with BDEs, and 75% were of only 1 or 2 hour durations. False polarity regions tended to lie nearly orthogonal to the spiral field angles and at relatively high latitudinal angles. While multiple (≥ 3 in 24 hours) current sheet crossings are common, we find no cases consistent with a wavy current sheet.

The polarities and locations of interplanetary coronal mass ejections in large interplanetary magnetic sectors

We have surveyed the ISEE 3 electron heat flux data to determine the polarities of the interplanetary magnetic field for the period 1978–1982. Many intervals of bidirectional electron (BDE) heat fluxes, signatures of the closed fields of interplanetary coronal mass ejections (ICMEs), were distributed among the magnetic sectors, often carrying sector boundaries. Here we examine the distribution of BDE intervals within the longest lived sectors of the ISEE 3 period to determine where ICMEs occur relative to the heliospheric current sheet (HCS). The intrasector BDE intervals, with polarities matching those of the sectors, appear to be uniformly distributed within the sectors, indicating that those ICMEs are not closely confined to the HCS. We estimate the ratio of intrasector matching polarity BDE intervals to opposite polarity BDE intervals to be about 10. The small number and short duration of opposite polarity BDE intervals suggest that injections of opposite polarity coronal mass ejections into sectors are rare. We found a number of mixed polarity BDE intervals, containing both matching and opposite polarities. These mixed polarity BDE intervals occur a factor of 5 less often than the single polarity BDE intervals alone do but are statistically more than twice as long in duration. We discuss possible origins of mixed polarity BDE intervals.

Heliospheric current sheet inclinations predicted from source surface maps

The inclinations of the neutral line at the ecliptic plane derived from source surface model maps of coronal fields are measured for the interval from June 1976 to March 1992. The mean and median values of 53° and 57° are close to the average inclinations determined earlier from minimum variance analyses of solar wind measurements at sector boundaries, but the mode falls in the 80°-90° bin. This result, which is based on the model assumptions implicit in deriving the source surface maps, predicts that the heliospheric current sheet typically intersects the ecliptic plane nearly at right angles, even without steepening by stream interaction regions. High inclinations dominate the solar cycle for about 7 years around solar maximum. Dips to lower inclinations occur near solar minimum, but high variance admits a wide range of inclinations throughout the cycle. Compared to the smooth solar cycle variation of the maximum latitudinal excursion of the neutral line, often treated as the tilt angle of a flat heliospheric current sheet, the noisy variation of the inclinations reflects the degree to which the neutral line deviates from a sine wave, implying warps and corrugations in the current sheet. About a third of the time the neutral line so deviates that it doubles back in longitude.