Figure 17.3 schematically illustrates the local time and geomagnetic latitude dependences of the observed rate of precipitation of auroral particles into the upper atmosphere and the different associated classes of auroral emissions [Hartz, 1971; Carlson and Egeland, 1995].
Figure 17.3: Schematic illustration of the patterns of particle precipitation into Earth's
upper atmosphere and the different classes of auroral emissions [Hartz, 1971; Carlson and
Egeland, 1995], as functions of local time and geomagnetic latitude. The direction to the Sun is at local
noon (12 hours) while the center of the tail lies at local midnight (0 hours). The
radial direction corresponds to geomagnetic latitude; the geomagnetic pole is at the
cross symbol. The figure is described more in the text.
The average flux is represented schematically by the density of the dot, triangle and star symbols, measured during multiple spacecraft transits through these regions.
The star symbols represent low energy
polar cusp plasma that originates in the magnetosheath and solar wind and has characteristic
energies of order 100 eV. As expected from Lecture 14, the polar cusp precipitation region is
on the dayside and at relatively high geomagnetic latitudes 
degrees. These precipitating
particles form ``dayside cusp auroras'' [Carlson and Egeland, 1995]. These are typically sub-visual and
relatively weak, with
maximum intensities in atomic rather than molecular lines due to their generation at larger
heights than other aurorae, typically in the F layer of the ionosphere. Dayside cusp aurorae
apparently tend to be relatively uniform and diffuse, without arcs or localized fine structure but are
relatively tightly confined in local time to near noon. Cusp aurorae usually change substantially in location and
intensity in response to variations in the solar wind 
component. For 
(and
so greater efficiencies for dayside magnetic reconnection and better connection of the cusp to the solar wind),
cusp aurorae broaden, extend equatorward and brighten considerably. Sometimes this
dayside motion equatorward coincides with the nightside aurora moving poleward, as expected for
increased loading of the tail, but not always.
The dot symbols in Figure 17.3 represent where high energy (> 20 keV) auroral particles precipitate.
These high energy particles precipitate on a circle
with constant geomagnetic latitude 
degrees, as expected for trapped particles leaking out
of the loss cone as they drift around Earth in the ring current. It turns out that these high energy particles
are relatively ineffective in producing auroral emissions.
The triangle symbols in Figure 17.3 show where medium energy (0.2 - 20 keV) particles precipitate and produce the majority of visual aurorae at Earth. The footprint made by these particles is magnetically connected to the plasmasheet and the plasmasheet boundary layers, as discussed in Lectures 14 and 15. Note that the particles precipitate onto an oval shape, not a circular shape, due to the influence of magnetic reconnection on the magnetic field lines connecting to the plasma source regions. This oval shape is the so-called auroral oval where aurorae are observed, as verified in Figure 17.4.
Figure 17.4: Variation in the size and location of the auroral oval with geomagnetic activity
[Feldstein and Starkov, 1967; Carlson and Egeland, 1995]. The
shaded areas show the spatial regions with maximum auroral activity, averaged over many episodes,
in the northern hemisphere. The coordinate system is similar to that in Figure 17.3.
At least two classes of auroral activity are associated with the medium energy plasmasheet particles in Figure 17.3 and 17.4. First, there is the so-called ``diffuse'' aurora, which is fairly weak and smooth and is present relatively continuously around most of the auroral oval (except near noon). This emission is just due to particles precipitating from the plasmasheet and its boundary layer during their bounce motion, either due to their being in the loss cone or being scattered into the loss cone (see Lecture 2).
The second class is due to ``discrete auroral arcs'' which are typically bright, localized structures that are primarily (but not always) produced during the auroral brightenings that accompany magnetospheric substorms. These structures are usually elongated in the east-west direction (along the auroral oval) but are narrow in the north-south direction (normal to the oval). These occur primarily in the nightside auroral oval and typically vary rapidly with time. Spacecraft observations show that the electrons responsible for discrete arcs originate in the plasmasheet but are energised by passing through a potential drop of several to a few tens of keV, showing up as very field-aligned beams with energies of 2 - 20 keV.
Auroral displays associated with electrons and with protons are observed at Earth, having different spectral properties and typically different locations too. Typically, however, electron-driven auroras dominate at Earth. This need not be the case at other planets, or in stellar magnetospheres.