INTERPLANETARY ORIGIN OF GEOMAGNETIC-ACTIVITY IN THE DECLINING PHASE OF THE SOLAR-CYCLE

Interplanetary magnetic field (IMF) and plasma data are compared with ground-based geomagnetic Dst and AE indices to determine the causes of magnetic storms, substorms, and quiet during the descending phase of the solar cycle. In this paper we focus primarily on 1974 when the AE index is anomalously high (<(AE)over bar> = 283 nT). This year is char acterized by the presence of two long-lasting corotating streams assoc iated with coronal holes. The corotating streams interact with the ups tream low-velocity (300-350 km s(-1)), high-density heliospheric curre nt sheet (HCS) plasma sheet, which leads to field compression and simi lar to 15- to 25-nT hourly average values. Although the B-z component in this corotating interaction region (CIR) is often < -10 nT, typical ly the field directionality is highly variable, and large southward co mponents have durations less than 3 hours. Thus the corotating stream/ HCS plasma sheet interaction region can cause recurring moderate (-100 nT less than or equal to Dst less than or equal to -50 nT) to weak (- 50 nT less than or equal to Dst less than or equal to -25 nT) storms, and sometimes no significant ring current activity at all (Dst > -25 n T). Storms of major (Dst less than or equal to -100 nT) intensities we re not associated with CIRs. Solar wind energy is transferred to the m agnetosphere via magnetic reconnection during the weak and moderate st orms. Because the B-z component in the interaction region is typically highly fluctuating, the corresponding storm main phase profile is hig hly irregular. Reverse shocks are sometimes present at the sunward edg e of the CIR. Because these events cause sharp decreases in field magn itude, they can be responsible for storm recovery phase onsets. The in itial phases of these corotating stream-related storms are caused by t he increased ram pressure associated with the HCS plasma sheet and the further density enhancement from the stream-stream compression. Altho ugh the solar wind speed is generally low in this region of space, the densities can be well over an order of magnitude higher than the aver age value, leading to significant positive Dst values. Since there are typically no forward shocks at 1 AU associated with the stream-stream interactions, the initial phases have gradual onsets. The most dramat ic geomagnetic response to the corotating streams are chains of consec utive substorms caused by the southward components of large-amplitude Alfven waves within the body of the corotating streams. This auroral a ctivity has been previously named high-intensity long-duration continu ous AE activity (HILDCAAs). The substorm activity is generally most in tense near the peak speed of the stream where the Alfven wave amplitud es are greatest, and it decreases with decreasing wave amplitudes and stream speed. Each of the 27-day recurring HILDCAA events can last 10 days or more, and the presence of two events per solar rotation is the cause of the exceptionally high AE average for 1974 (higher than 1979 ). HILDCAAs often occur during the recovery phase of magnetic storms, and the fresh (and sporadic) injection of substorm energy leads to unu sually long storm recovery phases as noted in Dsr. In the far trailing edge of the corotating stream, the IMF amplitudes become low, <3 nT, and there is an absence of large-amplitude fluctuations (Alfven waves) . This is related to and causes geomagnetic quiet. There were three ma jor (Dst less than or equal to -100 nT) storms that occurred in 1974. Each was caused by a nonrecurring moderate speed stream led by a fast forward shock. The mechanisms for generating the intense interplanetar y B-s which were responsible for the subsequent intense magnetic storm s was shock compression of preexisting southwardly directed B-z (B-s) for the two largest events and a magnetic cloud for the third (weakest ) event. Each of the three streams occurred near a HCS crossing with n o obvious solar optical or X ray signatures. It is speculated that the se events may be associated with flux openings associated with coronal hole expansions. In conclusion, we present a model of geomagnetic act ivity during the descending phase of the solar cycle. It incorporates storm initial phases, main phases, HILDCAAs, and geomagnetic quiet. It uses some of the recent Ulysses results. We feel that this model is s ufficiently developed that it may be used for predictions, and we enco urage testing during the current phase of the solar cycle.