Abstract. In February 1996, the POLAR spacecraft was placed in an elliptical orbit with a 9 RE geocentric distance apogee in the northern hemisphere and 1.8 RE perigee in the southern hemisphere. The Thermal Ion Dynamics Experiment (TIDE) on POLAR has allowed sampling of the three-dimensional ion distribution functions with excellent energy, angular, and mass resolution. The Plasma Source Instrument (PSI), when operated, allows sufficient diminution of the electric potential to observe the polar wind at very high altitudes. In this paper, we describe the results of a survey of the polar wind characteristics for H + , He + , and O + as observed by TIDE at -5000 km and -8 RE altitudes over the polar cap during April-

Low‐energy (below approximately 50 eV) ionospheric ions, injected into the magnetosphere at the dayside cleft, are studied using data from the retarding ion mass spectrometer (RIMS) experiment on the Dynamics Explorer 1 satellite. It is concluded that upwelling ions at the cleft form an ion fountain and are blown into the polar cap by antisunward convection. At high Kp (>4), convection is generally strong enough to fill the entire polar magnetosphere with low‐energy O+ ions, whereas at low kp (<2) they are largely restricted to the dayside half of the cap. Using a two‐dimensional kinetic ion trajectory model, the locations where RIMS detected O+ within the cap are shown to be consistent with the spatial distributions of O+ density, predicted for an upwelling ion source at the cleft and various dawn‐dusk convection electric fields. A detailed study is made of one polar pass of DE 1, during which RIMS detected He+, N+, O+, and O++ ions, the ion trajectory model being used to trace all these ions back to a common source at an observed upwelling ion event near the cleft. All observed species are deduced to be falling earthward in the nightside of the cap, as predicted from the model, indicating the dominance of gravity over upward field‐aligned acceleration (such as by the ambipolar electric field). Comparison of field‐aligned velocities observed for O+ and O++ ions defines a maximum limit to the upward electrostatic acceleration present within the cap which was only sufficient to eject ionospheric H+ ions, all heavier ions being supplied from the dayside by the cleft ion fountain.

A region of density enhancements of thermal heavy ions (O+, O++, and N+) has been observed on numerous occasions by the retarding ion mass spectrometer on the Dynamics Explorer 1 satellite. Outer plasmasphere densities of heavy ions are often observed to be up to 2 orders of magnitude higher than equatorward densities within an orbital pass. This phenomenon is almost always observed in the region of the plasmasphere just inside the plasmapause and has been seen at all local times. A statistical study of these heavy ion density enhancements, covering almost 600 passes of DE 1 through the plasmasphere, shows that O+ and O++ enhancements occur over about 64% of the observed passes, with the highest frequency of occurrence being found in the late evening and morning regions. O+ enhancements tend to be seen more frequently in the morning, while O++ enhancements are more likely to be observed in the evening. O+ appears to have a higher outer plasmasphere density enhancement than O++ in all local time regions. Evidence showing that the L shell of the heavy ion density enhancement peaks is dependent upon Dst is presented. Finally, mechanisms for the creation of the heavy ion density enhancements are briefly discussed, including the “geomagnetic mass spectrometer” and plasmapause‐associated electron heating of the topside ionosphere.

No abstract

[1] We review observations and theories of the solar ablation of planetary atmospheres, focusing on the terrestrial case where a large magnetosphere holds off the solar wind, so that there is little direct atmospheric impact, but also couples the solar wind electromagnetically to the auroral zones. We consider the photothermal escape flows known as the polar wind or refilling flows, the enhanced mass flux escape flows that result from localized solar wind energy dissipation in the auroral zones, and the resultant enhanced neutral atom escape flows. We term these latter two escape flows the ''auroral wind.'' We review observations and theories of the heating and acceleration of auroral winds, including energy inputs from precipitating particles, electromagnetic energy flux at magnetohydrodynamic and plasma wave frequencies, and acceleration by parallel electric fields and by convection pickup processes also known as ''centrifugal acceleration.'' We consider also the global circulation of ionospheric plasmas within the magnetosphere, their participation in magnetospheric disturbances as absorbers of momentum and energy, and their ultimate loss from the magnetosphere into the downstream solar wind, loading reconnection processes that occur at high altitudes near the magnetospheric boundaries. We consider the role of planetary magnetization and the accumulating evidence of stellar ablation of extrasolar planetary atmospheres. Finally, we suggest and discuss future needs for both the theory and observation of the planetary ionospheres and their role in solar wind interactions, to achieve the generality required for a predictive science of the coupling of stellar and planetary atmospheres over the full range of possible conditions.

Thermal ion composition measurements from several successive dusk sector passes by the DE‐1 satellite show the formation of the new outer plasmasphere and double plasmapause following a sharp decrease in geomagnetic activity. In less than one day after the magnetic activity decrease, the outer plasmasphere formed and consisted of cold, essentially Maxwellian plasma with ion composition and thermal energy characteristics generally similar to those of the inner plasmasphere, albeit at significantly lower densities. There is also evidence that at times the thermal O+ density is comparable to the H+ density within the plasmasphere.

A statistical study of plasmaspheric density profiles and their boundaries is performed, using measurements of core (<50 eV) ions by the retarding ion mass spectrometer (RIMS) on Dynamics Explorer 1. The plasmasphere density profiles are classified into essentially six different categories, indicative of density structure out to the outer boundary of observable cold, isotropic plasma. The most common profiles observed are those which are relatively featureless out to this outer boundary and those which exhibit multiple plateaus and plasmapauses. The multiple plateau profiles occur predominantly in the afternoon and evening sectors, while the “featureless” profiles are most common in the midnight through morning sectors. In the multiple plateau profiles, the average MLT (magnetic local time)‐L locus of the inner “plasmapause” gradients is nearly circular at L = 3–4. Profiles with significant density troughs occur most often in the evening sector, and the troughs are widest in this sector as well. The MLT‐L local time shape of the outer boundary of RIMS‐observed cold ions (termed here low‐energy ion transition, or LEFT) is similar to previously reported average plasmapause shapes, though its bulge region occurs in the afternoon sector. This MLT‐L shape also tends to become more circular with increasing geomagnetic activity. Using simultaneous plasma density measurements from the plasma wave instrument, it is found that the LEIT typically is located between densities of 10 and 100 cm−3, with an average plasma density there of about 60 cm−3.

[1] The structure of the density discontinuity across the plasmapause is often based on electron and H + density profiles with the contribution of heavy ions (He + , O + etc) neglected. Electron and ion density measurements in this region may differ significantly due to the presence of heavy ions and it is important for the intercomparison of different datasets to understand these differences. Dynamics Explorer (DE-1) magnetic field and plasma composition data have been used to compare heavy ion responses across the plasmapause and to calculate the mass loaded ion density (r) profiles. To illustrate this we investigate mass loading through radial profile variations in the Alfven velocity (V A ). Results show that the gradient in r and V A across the plasmapause is modified when mass loading due to multiple heavy ion species is included, particularly in the presence of the O + torus. Application to ultra-low frequency (ULF) field line resonance is used as an example where the contribution from heavy ions smoothes out the expected ULF wave resonant frequency discontinuity at the plasmapause.