4.1 Zone-I

Zone-I occurs at intermediate latitudes just poleward of the diffuse aurora. The exact latitudes depend on hemisphere and local time. This region is by far the narrowest in latitude among the auroral zones as shown in Figure 1, but its clearly defined equatorward and poleward boundaries, as well as the high repeatability among the various data sets, make it the most straightforward to identify. Mauk et al. (2020) characterized this region with dominant downward energetic electrons within the loss cone.

Kotsiaros et al. (2019) and Mauk et al. (2020) noted an agreement between the upward field-aligned current and predominantly downward energetic electrons for the PJ6S auroral pass (shown here in Figure 3), suggesting that Zone-I is associated with upward electric currents. Figures 2-5 corroborate this correspondence between the 50 and 1,000 keV downward electrons and the well-structured upward field-aligned current from δB_φ and confirm that most of the upward current is indeed carried by downward energetic electrons. It should be highlighted that although upward currents in Zone-I are well-ordered, the predominantly downward electron acceleration supporting these currents is via both inverted V and broadband distributions, often the latter attaining higher energies (Mauk et al., 2017b). These distributions have been observed serially within the same Zone-I pass and are occasionally overlaid onto one another (see figures 8 and 12 in Mauk et al. (2020)). While the domination of the downward energetic electron is a reliable predictor of Zone-I, there exists large variability in the size of the asymmetry between the downward and upward energy fluxes among the different events. This can be as large as 100× (e.g., PJ4S) and as relatively modest as 3–5× (e.g., PJ6S and PJ9S). The size of the asymmetry is likely related to both the nature of the acceleration region and Juno's proximity to it.

Kurth et al. (2018) showed for PJ7N that an interval of downward broadband electron distribution (in what was later identified as Zone-I) is coincident with brief but very intense broadband plasma waves in both the electric and magnetic spectra (∼01:15:51 in Figure 4). It appears that this correspondence is repeatable across events whenever broadband electron distributions are present, for example, at 13:39:07 during PJ4S in (Figure 2). There are, however, no plasma wave signatures that uniquely correspond to downward inverted V electron distributions. Kurth et al. (2018) proposed the importance of these intense broadband electromagnetic waves in intervals of broadband electron acceleration and determined the direction of their Poynting vector with respect to the Jovian magnetic field to show that they were propagating in the same direction as the predominantly downward energetic electrons. These waves were interpreted as being in the whistler mode as the frequency extended well above f_cH+ and assumed to cut off at the electron plasma frequency, f_pe, at ∼10 kHz (or n_e ≈ 1.2 cm⁻³), which represents the theoretical upper-frequency cutoff for whistler-mode waves in the presence of a strong magnetic field (e.g., Persoon et al., 2019). We will show in the next section, however, that Zone-I is a region where the electron densities are dramatically depleted to as low as <0.01 cm⁻³, or f_pe < 900 Hz. Densities could not be inferred within these brief intervals of broadband acceleration; therefore, the presence of the whistler mode would imply that the densities are anomalously greater during these intervals. Broadband electromagnetic waves are routinely observed over Earth's auroral regions, although typically confined to the downward current regions (Ergun et al., 1998b) and have also been reported in Jupiter's polar cap region (Elliot et al., 2020). We will revisit these features and show their correspondences against energy- and pitch-angle-time spectra when discussing Zone-II as they appear to be much more prevalent there.

Another important observation in Zone-I is the lack of, or significant reduction in, Alfvénic fluctuations compared to just equatorward in the diffuse aurora. Alfvén waves are known to develop parallel electric fields when finite electron mass is considered and their role has therefore been posited to explain the broadband nature of Jupiter's auroral electrons (Lysak et al., 2021; Saur et al., 2018). It is therefore likely that these waves have dissipated at higher Zone-I altitudes, lending most of their energy to electron acceleration. It is important to note that Jupiter's low-altitude region is characterized by very strong magnetic fields meaning any Alfvénic fluctuations present may just be too small to be picked up by the magnetometer. The Poynting flux is estimated as δB²v_A/μ₀, where v_A is the Alfvén speed which considerably rises in the presence of significant density depletions. Therefore, for a given Poynting flux, it follows that δB would decrease correspondingly.

It is worth emphasizing that the Alfvénic fluctuations are repeatable signatures of the diffuse aurora, but not Zone-I or Zone-II. The waves are clearly supported over a wide range of M-shells. Allegrini et al. (2020) presented a survey showing that the lower-energy 3–30 keV electrons typically peak just equatorward of the main oval (or what is now called Zone-I). It appears from Figures 2-5 that the poleward edge of the Alfvénic fluctuations is when the 3–30 keV electrons peak and precedes the higher 50–1,000 keV that dominate Zone-I. Interestingly, during PJ4S and PJ7N (Figures 2 and 4), the Alfvénic fluctuations diminish in the diffuse aurora as the 3–30 keV electron energy fluxes peak at ∼13:37:30 and ∼1:18:30, respectively, before recovering again. Li et al. (2021) applied a data-model comparison to show that whistler-mode waves are the driver of Jupiter's diffuse auroral precipitation above several keV via pitch-angle scattering, although this mechanism did not account for the observed precipitation of lower energies (< several keV) and was limited to lower latitudes (M-shells 8–18). Based on our observed correspondences, we postulate that Alfvén waves may indeed be responsible for precipitating lower energy electrons in the diffuse aurora at the higher latitudes.

The most prominent plasma wave signature in Zone-I are intense emissions below f_cH+ and f_cH3+. The electric and magnetic field spectral densities are enhanced over a broad range of low frequencies (few kHz bandwidths) and undergo a distinct drop in intensity at f_cH+ and/or at f_cH3+. This is usually an indication of strong damping via cyclotron resonance where the wave energy is transferred to the corresponding ions. This characteristic is consistent with ion cyclotron waves and their observation in the presence of upward energetic ions and downward energetic electron beams draw a strong analogy to both Earth's and Saturn's upward current regions where the correlation has been observed (e.g., Bader et al., 2020; Cattell et al., 1988; McFadden et al., 1998; Mitchell et al., 2009). Ion cyclotron waves in the auroral regions have been observed as both electrostatic (EIC) and electromagnetic (EMIC) modes. The strong magnetic component here is evidence that EMIC waves are present, though not necessarily in the absence of EIC, and the significance is that they carry Poynting fluxes.

Figures 6a–6d show an analysis of the Poynting vector direction for these waves during PJ4S. These are the emissions present below 1 kHz and the series of peaks and nulls in the electric field spectrum is due to spin modulations. The electric and magnetic field fluctuations have high coherency, C_Ey-Bz ≈ 1, and the combination of a phase φ_Ey-Bz ≈ −180° (or 180°) and a positive B_x/|B_x| in the southern hemisphere indicates an upward-propagating wave. Figure 6e shows that the power of these waves primarily resides perpendicular to the magnetic field. Here we compare the spin-modulations in the electric field spectral densities to the angle between the antenna dipole and background magnetic field and show that spectral densities peak (depress) when the antenna is perpendicular (parallel) to the magnetic field. At the measured frequency of ¼ f_cH+, the ratio of the components is E_⊥/E_|| = 200. Despite a strong magnetic component, the E/cB ratio (not shown here) is greater than one but of order unity. This can occur in the presence of an admixture of EIC and EMIC waves.

Although we cannot directly verify that they are intrinsically left-hand-polarized, we can indirectly infer this from the fact that their electric and magnetic fields are highly coherent, fluctuate perpendicular to the background magnetic field, and do not propagate above f_cH + or f_cH3+. Altogether, these are consistent with resonant absorption of left-hand-polarized ion cyclotron waves, a well-recognized mechanism for ion heating (e.g., André et al., 1998; Chang et al., 1986; Lysak, 1986). The observed (mostly) upward-propagation of these waves is somewhat in contrast to what is typically observed during low-altitude passes of Earth's aurora, where waves below f_cH + are more commonly observed to be downward propagating (Chaston et al., 1998; Gurnett et al., 1984). The difference at Jupiter may be either due to their sources originating at an altitude lower than Juno, that is, ⪅1 R_J above the one-bar level, or a different generation mechanism altogether. Electron drifts as the source of free energy driving ion cyclotron instability have been invoked to explain their correlation with auroral field-aligned currents (Cattell et al., 1998). Testing whether this hypothesis holds at Jupiter requires solving dispersion relations with modeled particle distributions which is beyond the scope of this study. It has been further demonstrated that broadband EMIC waves can also accelerate cold secondary electrons to form counterstreaming field-aligned electrons (McFadden et al., 1998). Since bidirectional electrons are a key feature of Jupiter's auroral zones, the role of EMIC waves should not be neglected.

The coincident field-aligned H⁺ fluxes suggest that any perpendicular heating by the ion cyclotron waves is not sufficient to deviate the pitch angle from the field-aligned direction and generate conics. The measured electric field spectral density of 10⁻⁵ V m⁻¹ Hz⁻¹ near f_cH+ (Figure 6e) yields a maximum cyclotron resonant heating rate of ∼500 eV/s (Chang et al., 1986) and is comparable to that measured in Io's Main Alfvén Wing where, by contrast, H⁺ conics were detected (Clark et al., 2020; Sulaiman et al., 2020). The difference is likely due to the interaction time, proximity to, nature of the acceleration region or a combination thereof. Szalay et al. (2021) concluded, based on the presence of H⁺ inverted V distributions, that quasi-static parallel potential structures drove the acceleration of H⁺ away from Jupiter's high-latitude ionosphere. This is further supported by the disappearance of upward H⁺ during intervals of broadband acceleration within Zone-I shown by Mauk et al. (2018). The observation of both downward electron and upward H⁺ beams at these altitudes would suggest that Juno was in or close to a unidirectional acceleration region, that is, an upward parallel potential. Therefore, it is possible that the perpendicular heating supplied by ion cyclotron waves is overcome by the action of more powerful parallel potentials that deposit much larger amounts of energy along the field line. The ion cyclotron waves (shown to be upward propagating) may have their source in the ionosphere where the density is high and enough ions exist to significantly dampen the waves. Cold ionospheric ions are bound by Jupiter's large gravitational potential (the gravitation binding energy of H⁺ is ∼20 eV and H₃⁺ is ∼60 eV) and to be admitted into the electrostatic potential at higher altitudes, a means of energization is required to escape the gravitational potential. When ions are heated perpendicular to the magnetic field in the presence of a diverging magnetic field, they experience a mirror force that transports them to a region of weaker magnetic field, that is, higher altitudes, as a parallel velocity component develops to conserve kinetic energy and the first adiabatic invariant.

In summary, a multi-instrument in-situ analysis shows that the following criteria identify Zone-I in Jupiter's low-altitude auroral region: (a) presence of a gradient in the B_φ perturbation that is indicative of an upward field-aligned current, as measured by MAG; (b) greater downward electron energy fluxes than upward, as well as greater than outside the loss cone, accompanied by broadband and/or inverted V distributions as measured by JEDI; (c) the low-frequency portion of intense, apparently dispersive, coherent, mostly upward-propagating ion (H⁺ and/or H₃⁺) cyclotron waves, as measured by Waves; and (d) presence of field-aligned upward flowing H⁺, as measured by JADE and JEDI. These observations are unique to Zone-I and highly repeatable, such that any one of them is highly predictive of Zone-I. Furthermore, they exhibit distinct and unambiguous equatorward and poleward edges that are consistent with the main oval emission shown in Figure 1. The boundary at which Alfvénic fluctuations significantly decrease reliably marks the entry into Zone-I from the diffuse aurora. The deficiency in observed Alfvénic fluctuations, however, is not a unique marker of Zone-I as this is continuous into Zone-II.

4.2 Electron Density Depletions in Zone-I

Electron density depletions occur within Zone-I, exhibiting large variability and with an equatorward edge. The scatter plot in Figure 7a shows the electron number density variation with increasing M-shell. This is color-coded in altitude over a range of 0.6–1.7 R_J above the one-bar level. The direction of increasing M-shell translates into Juno sampling the auroral regions in the poleward sense, beginning with the equatorward edge of the broad diffuse aurora through to the poleward edge of Zone-I. The M-shells here are likely overestimated since the auroral regions are believed to be mapped to ∼30 R_J in the equatorial plane. The purpose of this figure is to examine how the electron densities vary on different auroral field lines. It should be noted that different internal field and/or current sheet models will yield different M-shell values. We therefore identify the auroral crossings based on in-situ observations and do not rely on the values provided by M-shell mapping.

We digitize the densities by identifying Ordinary (O) mode waves that are sometimes present during the auroral passes. The O-mode is a transverse electromagnetic wave. Its dispersion relation is one of two derived from the cold-plasma approximation for propagation perpendicular to the B—the other being the extraordinary (X) mode more familiarly recognized as radio waves. The main difference is the relatively simpler dispersion relation of the O-mode which is the same as that for an unmagnetized plasma. The waves are evanescent below f_pe and therefore exhibit a low-frequency cutoff, as shown in the first panels of Figures 2-5 (see Sulaiman et al. (2021) for the theoretical background as well as early and more recent implementations of this technique by Gurnett and Shaw (1973) and Elliott et al. (2021)). Strictly speaking, this cutoff is an upper limit to the local electron plasma frequency due to the possibility of a higher-density region existing between the source and the spacecraft. When this occurs, the measured cutoff corresponds to the maximum density between the source and the spacecraft; however, the cutoffs observed here are usually well-defined and continuous which suggest the densities are local. Since f_pe ∝ √n_e, the total electron number density is straightforwardly obtained and this is in excellent agreement with the electron partial density derived by JADE for overlapping intervals (see Figure S2 in Supporting Information S1—the electron partial densities derived by JADE were mostly limited to the diffuse aurora; Allegrini et al., 2021). Despite the limited coverage in altitude shown here, the expected anti-correlation between density and altitude is present, giving confidence in our method. We obtain density measurements whenever the O-mode waves are present and discernible. The circled points highlight measurements taken when Juno was magnetically connected to Zone-I using all the criteria whenever the O-mode was present during the first 10 perijoves. When the pitch angle coverage was suboptimal, only criterion (c) was used. Recall that any one of the criteria alone is a sufficient marker of Zone-I.

Figure 7a exposes a boundary between the diffuse aurora and Zone-I. Within Zone-I, the electron densities are, on average, significantly depleted, by up to 2 orders of magnitude down to below 0.01 cm⁻³. In Zone-II, the sub-f_cH + band of the O-mode waves becomes “washed out” in the spectrogram due to the presence of intense broadband low-frequency electromagnetic emissions; therefore, it is not possible to determine, based on this technique, how far they remain depleted and whether/where they recover. All Zone-I verified densities are below 0.1 cm⁻³ with a subset below 0.01 cm⁻³.

Density depletions, or auroral cavities, are known to be intimately related to auroral acceleration processes (e.g., Paschmann et al., 2003; Persoon et al., 1988). Their association is well supported by theoretical modeling (Block & Fälthammar, 1968; Knight, 1973) and repeatedly corroborated by experimental evidence (Ergun et al., 2002; Hull et al., 2003) although much of the focus has been on the development of parallel potentials in the context of inverted V distributions. The basic principle is that density depletions reduce the number of charge carriers thereby limiting the ability of plasmas to carry strong field-aligned currents. This “current choke” results in the development of parallel electric fields as the displacement current term of Ampère's law builds up to ensure ∇ × B is balanced (Ray et al., 2009; Song & Lysak, 2006).

Although turbulence-induced broadband processes are typically associated with weaker Alfvénic aurora at Earth, they are believed to be of greater importance than electrostatic acceleration processes in generating Jupiter's most intense aurora (Clark et al., 2018; Saur et al., 2018). Parallel electric fields from Alfvén waves become important when the k_⊥²λ_e² term is large, where λ_e is the electron inertial length given by c/2πf_pe and k_⊥ is the wave vector component perpendicular to the background magnetic field. A large k_⊥ can be satisfied by a converging flux tube as the area is inversely proportional to B. A low-density region, or greater λ_e, means Alfvén waves undergoing a turbulent cascade are dissipated “earlier” in k-space. The measured densities in Zone-I equate to λ_e as large as 50 km, larger than 20–30 km modeled by Saur et al. (2018), thereby further lowering the threshold for Alfvénic dissipation to be achieved in the high-latitude region. Dispersive Alfvén waves have been observed within deep density cavities over Earth's auroral oval together with upgoing transversely heated ionospheric ions and downgoing field-aligned electrons. This has been interpreted as evidence for a positive feedback mechanism, whereby small-scale Alfvén waves erode the auroral ionosphere by facilitating ion outflow, which in turn leads to deeper density cavities that maintain the production of small-scale Alfvén waves via refraction and phase mixing of incoming large-scale Alfvén waves (Chaston et al., 2008; Rankin et al., 1999). More recently, Lysak et al. (2021) proposed that an ionospheric Alfvénic resonator (IAR) operating at Jupiter can account for the observed broadband electron distributions. This is a widely accepted model used to explain similar distributions in the case of Earth, whereby the propagation of Alfvén waves is facilitated by a rapid decrease in density (Lysak, 1991). The corresponding increase in Alfvén speed gives rise to partial reflection of Alfvén waves which become trapped. At large enough k_⊥, the parallel electric field fluctuating at some resonant frequency can result in electron acceleration over a broad range of energies.

Figure 7b combines the electron densities with measured magnetic field strengths to express f_pe/f_ce variations. This ratio is especially important for the generation of radio emissions via the Cyclotron Maser Instability (Wu & Lee, 1979). This mechanism requires f_pe/f_ce ≪ 1 in the presence of a positive gradient in the perpendicular velocity distribution of weakly relativistic electrons. It is clear that the necessary low f_pe/f_ce is well satisfied, particularly in Zone-I, thus will provide further constraints on Jupiter's radio sources (e.g., Imai et al., 2019; Louis et al., 2019).

4.3 Zone-II

Among the three zones, Zone-II occurs at the highest latitudes just poleward of Zone-I. This region has a clearly defined equatorward boundary, but its poleward boundary with the polar cap is often ambiguous. Mauk et al. (2020) characterized this region with upward energetic electrons with energy fluxes greater than or equal to the downward component within the loss cone. Another key difference is the bidirectional electrons are almost always broadband in energy. On the other hand, downward H⁺ inverted Vs have been observed intermittently and, by contrast to Zone-I's highly field-aligned H⁺ beams, exhibit a nearly isotropic pitch angle distribution with an empty upward loss cone (Mauk et al., 2020). Whereas Zone-I features are typically (but not always) continuous within its boundaries, Zone-II features are spatially or temporally sporadic.

Kotsiaros et al. (2019) and Mauk et al. (2020) noted agreements between the downward field-aligned currents and Zone-II during the PJ6S auroral pass (Figure 3), although this is usually limited to the most intense portion of the energetic particles and not as simple as the more ordered Zone-I. Again, Figures 2-5 corroborate this correspondence. Observed Alfvénic fluctuations in Zone-II remain relatively low/absent and comparable to Zone-I. This could also be evidence of dissipation, especially in a region supported predominantly by broadband, bidirectional energetic electrons (Lysak et al., 2021; Saur et al., 2018) and in the absence of strong evidence for inverted Vs and thus local parallel potentials. The plasma wave emissions, on the other hand, are the most intense of all zones with the largest average amplitudes in both the electric and magnetic fields. These are present throughout Zone-II and majority of the power is confined to frequencies below f_cH+ (Figures 2-5), and are often accompanied by brief, intense emissions that extend well above f_cH + that resemble those sometimes observed in Zone-I. The difference is that these brief and intense emissions occur intermittently in Zone-I, whereas they appear to be a key feature of Zone-II and are correlated with the intervals of most intense energetic electrons which are in turn correlated with downward currents.

The downward current region is fundamentally different from its upward counterpart. The charge carriers are abundantly sourced from the cold, dense ionosphere as electrons and are accelerated by many orders of magnitude above their thermal energy. What is peculiar about Jupiter's Zone-II is that although the downward electron energy fluxes are generally no greater than the upward energy fluxes, they can be as intense or greater than the downward energy fluxes in Zone-I and sufficient to produce observable auroras (Mauk et al., 2020; and see Figure 1 here), in contrast to the “black aurora” at Earth and Saturn that are connected to flux tubes carrying downward currents. It is clear based on the difference in fields and particles characteristics that the acceleration mechanism in Zone-II is distinct and more observationally complicated than that supporting Zone-I. While Juno does not carry a DC electric field instrument, the various characteristics highlighted in the previous section support the sporadic presence (although not exclusively) of parallel potential structures in Zone-I. Other than the downward H⁺ inverted Vs that are sometimes observed in Zone-II and not least that they are quasi-isotropic, the evidence for a stable parallel potential is inconclusive. The bidirectional electrons might be interpreted as originating from potential structures above and below the spacecraft, however, this is not consistent with their broadband energy.

We emphasized in the previous section that EMIC waves should not be neglected in the context of electrons since their link has been established (McFadden et al., 1998), whereby cold secondary electrons are trapped and accelerated to form counterstreaming populations. It is therefore probably not a coincidence that the most intense waves below f_cH + occur in Zone-II, where bidirectional electrons are present.

An important piece of the puzzle for broadband electrons may be in the contemporaneous broadband emissions in the frequency-time spectrograms shown in Figures 8 and 9. In the frequency domain, large-amplitude solitary (or “spiky”) structures in the waveform typically manifest as broadband noise. In other words, their steepness results in a broad range of frequencies displayed in the frequency domain; however, the underlying physics is in the waveforms. It is clear in Figures 8 and 9 that where broadband plasma wave emissions are seen, there are large-amplitude, spiky, and nonlinear electric field structures. Electrostatic solitary waves (ESWs) have been proposed to play a key role in accelerating electrons by carrying substantial potentials and are most often observed in Earth's downward current regions and in the presence of density depletions (Ergun et al., 1998b; Temerin et al., 1982). The ubiquity of these broadband emissions in Zone-II might be explained by the highly nonlinear evolution of two-stream electron beam instabilities, set up by bidirectional populations, that give rise to sharp pulses in the electric field (Matsumoto et al., 1994), as shown in Figures 8 and 9. Field-aligned electrons are then accelerated to a broad range of energies by the sum of individual micro-potential drops as they travel through ESWs. Despite their electrostatic nature, it is possible to measure an associated magnetic component (not shown here) which would result from the Lorentz field of a traveling charge.

Although the electron densities cannot be inferred within Zone-II, we can say with reasonable confidence that they remain low. The O-mode emissions above f_cH + appear continuous well into Zone-II with its low-frequency edge in the region below f_cH + that is dominated by intense electromagnetic turbulence. We therefore set f_cH + to be the approximate upper limit of f_pe and conclude that the electron densities within Zone-II are <0.1–0.01 cm⁻³. Therefore, the correspondingly large electron inertial lengths in Zone-II would similarly lower the threshold for Alfvénic dissipation, which remains the leading mechanism to account for the observed electron spectra (Lysak et al., 2021; Saur et al., 2018). Whether the densities are comparable to Zone-I, of similar variability and/or spatial scales are important questions that are beyond the reach of our present density digitization methods.

Perhaps the most recognizable and commonly observed plasma wave feature above auroral regions is the whistler-mode auroral hiss. In a frequency-time spectrogram, they are easily identified by their characteristic funnel or V-shape (Gurnett, 1966; James, 1976) which arises when the wave normal angle approaches the whistler-mode resonance cone (Santolík & Gurnett, 2002). The favored generation mechanism is a coherent beam-plasma instability at the Landau velocity (Farrell et al., 1989; Maggs, 1989), that is, ω/k_|| ≈ v_||. Since the auroral regions, including satellite auroral flux tubes, are a site for electron beams, whistler-mode auroral hiss are often observed and are often a reliable diagnostic for field-aligned currents (Gurnett et al., 1983, 2011; Sulaiman et al., 2018, 2020). That said, these plasma wave features are not as clearly identifiable in Jupiter's low-altitude auroral zones, contrary to expectation.

Figure 10 shows a rare example when the funnel-shaped auroral hiss was observed in the southern auroral zone during PJ12S. Although it appears like there are two similar emissions above and below f_cH+, they are fundamentally different and not connected since, above f_cH+, the timescales fall below the ion gyroperiod and the ions are effectively unmagnetized. The emission below f_cH+ is the H⁺ cyclotron mode commonly observed in Zone-I, likely propagating along its resonance cone. On the other hand, the emission above f_cH+ is the whistler-mode auroral hiss, propagating along its own resonance cone. Auroral hiss is typically not seen to propagate down to as low as f_cH+. Along its resonance cone, the lower hybrid frequency, f_LH, represents a lower limit through which they cannot propagate but instead reflect off. In this highly magnetized regime, that is, f_ce ≫ f_pe, we find f_LH ≃ f_cH+ (Sulaiman et al., 2021) and therefore conclude the waves are reflecting at the f_cH + boundary. While the whistler mode is typically observed as electromagnetic, its propagation along the resonance cone is quasi-electrostatic and this is supported by the relatively weaker magnetic component and an E/cB ratio of ∼10. This mode is characterized by an index of refraction that is much greater than unity, that is, a phase velocity that is low. Therefore, the Landau resonance condition requires low-energy electrons for the beam-plasma instability. Higher-energy electrons that interact with higher phase velocities can generate electromagnetic waves that cease to exhibit the characteristic funnel shape. And even higher energies that exceed the maximum phase speed allowed by the dispersion relation will result in no Landau resonance altogether. This likely explains why quasi-electrostatic auroral hiss is not as common a feature at Jupiter's low-altitude region as at Earth owing to the much higher electron energies at play.

Finally, what has not been covered in this study are the properties of heavy ions. The clear cutoff of plasma waves in Zone-I at f_cH3+ is indicative of H₃⁺ cyclotron waves and is strong (indirect) evidence for presence of upward H₃⁺. However, H₃⁺ ions in the auroral zones have not been reported by the particle instruments at the time of writing. The presence of multiple heavy ions would have a significant impact since each additional ion introduces five characteristic frequencies: the standard cyclotron and plasma frequencies plus the more complex ion hybrid, multi-ion cutoff, and crossover frequencies, which require numerical solving. The latter three are highly sensitive to the fractional abundance of ions, let alone any individual density. This also means that composition can be constrained by modeling and correctly diagnosing wave modes and their characteristic frequencies. The significance of an ion hybrid frequency in a multicomponent plasma is that it modifies the wave mode's dispersion relation and therefore how it propagates through the medium. For example, a resonance cone can develop above each hybrid frequency (Santolík et al., 2016). The crossover frequency is that which the waves reverse their intrinsic polarization (left to right or vice versa) and can therefore affect the nature of wave-particle interactions.