REGIONS OF ABNORMALLY LOW PROTON TEMPERATURE IN THE SOLAR-WIND (1965-1991) AND THEIR ASSOCIATION WITH EJECTA

Occurrences of solar wind plasma with abnormally low proton temperatures (T-p) have long been associated with the interplanetary manifestations (which we term ''ejecta'') of coronal mass ejections (CMEs). We survey the National Space Science Data Center Omni solar wind database for 1965-1991, and data from the Hellos 1 and 2 spacecraft for more limited periods, to identify plasma in which T-p is less than the temperature expected (T-ex) from the well-established correlation between the solar wind speed and T-p for normal solar wind expansion. The occurrence rate of low-temperature plasma (specifically with T-p/T-ex less than or equal to 0.5) is shown to be correlated with solar activity levels. Individual low-temperature regions have durations from 1 to similar to 80 hours with a mean of similar to 10 hours. Around one third are encounters with the heliospheric plasma sheet (HPS). These events are observed most frequently during the increasing phase of solar activity when the HPS lies closer to the ecliptic. The abnormally low temperatures may be intrinsic to the HPS or may provide support for the proposal that the coronal streamer belt underlying the HPS is a frequent source of ejecta. The remaining events have an occurrence rate which shows a particularly clear correlation with solar activity levels and with the CME rate at the Sun (when CME observations are available), consistent with an association with ejecta. These events also show greater than chance associations with other ejecta signatures. We suggest that the Omni plasma data can provide information on the rate of ejecta passing the Earth, and hence give an indication of the CME rate, for a period commencing before spacecraft coronagraph CME observations became available in the early 1970s. Our findings suggest that T-p depressions may provide a more comprehensive indication of the presence of ejecta than other ejecta signatures, such as bidirectional solar wind electron heat fluxes and energetic ion flows, which alone do not identify all ejecta.