the variations in the solar wind and IMF, and thus high‐latitude convection electric field, or ion‐neutral interactions in the region of large electric fields (Anderson et al., 1988; Lockwood & Carlson, 1992; Rodger et al., 1994; Valladares et al., 1996, 1998; Zhang et al., 2013a). One mechanism, transient reconnection, is accompanied by soft electron precipitation from the magnetosheath. The precipitating electrons can cause ionization and heat the cusp plasma. Patches created by this mechanism are found to have lower density and lower optical luminosity (Hosokawa et al., 2016; Oksavik et al., 2006; Zhang et al., 2013b) than those originated from the dayside solar EUV‐produced plasma. Periodic poleward‐moving auroral forms (PMAFs) produced by those precipitating soft electrons shown in Hosokawa et al. (2016) only reached ~150 R, which is lower than the typical patch luminosity on the nightside (Hosokawa et al., 2006). In the future, it may be more sensible to treat the relatively low‐density patches and higher‐density patches separately, since they might be produced by different mechanisms.
Carlson et al. (2006) and Carlson (2004) suggest that dayside reconnection is the dominant mechanism responsible for the patch production in the European sector. Whether the dayside reconnection can also account for a majority of the patches observed in the American sector needs further study, since the northern magnetic pole is located within the Canadian sector and the inclination/declination angles of magnetic field lines in the European and North American sectors are very different. In addition, high‐resolution coupled ionosphere‐thermosphere models will be needed to quantitatively evaluate the different formation mechanisms of the polar cap patches.
On the nightside, patches exit the polar cap and become a boundary blob. A statistical study using 8 years of data from the meridian scanning photometer data from Ny‐Aalesund reveals that the patches exiting the nightside polar cap mainly and nearly symmetrically around the magnetic midnight (Moen et al., 2007). A subsequent study further identified that a clear preference for earlier premidnight/later postmidnight MLTs under positive/negative IMF By (Moen et al., 2015).
4.5 ION UPFLOW ASSOCIATED WITH POLAR CAP HIGH‐DENSITY STRUCTURES
The F‐region and topside ionospheric density is enhanced within the high‐density ionospheric structures. Convective transport of these high‐density structures into regions with enhanced precipitating particle fluxes has been suggested to be an important mechanism of generating large ion upflow fluxes (Lotko, 2007; Yau et al., 2011). Without particle precipitation, the field‐aligned plasma flows within these high‐density structures are usually downward (Ren et al., 2018; Sojka et al., 1997). When these high‐density structures convect antisunward following the convection flows to regions, such as the dayside cusp and the nightside auroral zone, intense type‐2 ion upflow fluxes (Wahlund et al., 1992), and even divergent fluxes, can form.
Observationally, a couple of fortuitous measurements showed large ion upflow fluxes on the nightside that can be related to the polar cap patches and SED plumes (Semeter et al., 2003; Tu et al., 2007; Yuan et al., 2008; Zhang et al., 2016; Zou et al., 2017b) . Using the Sondrestrom ISR, Semeter et al. (2003) reported an observation of strong ion upflow event due to drifting polar cap patch into particle precipitation at the nightside auroral poleward boundary. Combining GPS TEC and DMSP satellite observations, Yuan et al. (2008) reported large ion vertical fluxes of ~1.2 x 1014 m‐2s‐1 measured by DMSP satellite within the SED, when it reached the nightside polar cap boundary during the 20 November 2003 superstorm. Recently, during a strong polar cap expansion event on 1 June 2013, soft electron precipitations in the cusp region moved equatorward and crossed a preexisting SED plume, resulting in strong heating and divergent ion flows and fluxes (Zou et al., 2017b). Figure 4.7 shows the 2‐D TEC map at 0100 UT on 1 June 2013 with an evident SED plume, and the PFISR field‐aligned beam observations of the plasma density, temperature, flow, and flux within the plume. The peak upflow fluxes reached ~2 x 1014 m‐2s‐1 at ~600 km in this event. Using the radio plasma imager (RPI) on the IMAGE satellite, Tu et al. (2007) found that the plasma density at ~7 RE can increase and decrease subsequently, when the SED plume extends to the polar cap and then disappears, and thus suggested that the high‐altitude density increase is due to the enhanced cleft ion fountain effect.
Figure 4.7 (a) TEC map at 0100 UT on 1 June 2013 shows the extension of SED plume into the Alaska sector. The black segments highlight the field of view of the PFISR radar. (b)–(f) PFISR field‐aligned beam observation of the density, ion temperature, electron temperature, field‐aligned velocity, and flux. Divergent plasma fluxes are seen within the SED plume when the open‐closed boundary moved across the beam
(from Zou et al., 2017b; Reproduced with permission of John Wiley and Sons).
Besides the effect of soft electron precipitation, Zhang et al. (2016b) also reported ion upflow events in the polar cap patch due to enhanced frictional heating, that is, the type‐1 ion upflow defined in Wahlund et al. (1992), using DMSP satellite. In a subsequent study based on an extended DMSP database, Ma et al. (2018) found that the highest upflow occurrence rate was associated with hot patches, which are accompanied with particle precipitation, strong convection speed, and localized FACs.
Cold plasma of ionosphere origin has indeed been observed at very high altitude. For instance, Foster et al. (2014) described in situ observations of locally enhanced cold plasma density at the ~5 Re altitude of the Van Allen Probes RBSP‐A spacecraft on magnetic field lines mapping to the point where the TOI intersected the midnight auroral oval as seen in GPS TEC imagery. Similarly, Walsh et al. (2014) reported THEMIS observations at ~12 Re altitude of enhanced cold plasma density on reconnecting dayside field lines mapping to the point where the SED plume entered the polar cap at the noontime cusp. These observations suggest that the density enhancements seen as the SED plumes and TOI at ionospheric heights could extend to very high altitudes along magnetospheric and polar cap field lines. In addition to the direct contribution of ion upflows/outflows from the cusp to the dayside reconnection site, cold plasma of the plasmaspheric plume origin has also been observed in the reconnection region (Lee et al., 2016).
The subsequent impact of the large but intermittent ion upflow/outflow fluxes associated with polar cap patches and TOI on magnetospheric dynamics is of great interest but outside the scope of this chapter and thus will not be elaborated here. Future quantitative studies using numerical models are needed to further distinguish the classical cold and the newly identified hot patches, such as whether the classical patch evolves into the hot patch under the influence of particle precipitation and FACs, or these hot patches can be produced solely by particle precipitation.
4.6 OPTICAL EMISSION MECHANISMS AND VARIABILITY OF POLAR CAP PATCHES
Polar cap patches are traditionally defined by plasma density enhancement and thus typically observed by LEO satellite or remotely sensed by ground‐based radars, but they can also be observed using optical instruments, such as 630 nm red line all‐sky imagers. The classical nighttime source for generating O(1D) is through dissociative recombination:
This process is negligible in the sunlit atmosphere. The volume emission rate due to dissociative recombination in the absence of precipitating particles based on Link and Cogger (1988) is