Plasma wave observations from Juno spacecraft at the Jovian bow shock Authors: J Joseph, W S Kurth, L B Wilson, J E P Connerney, F Allegrini, R J Wilson, A H Sulaiman, R W Ebert, J B Faden, C W Piker, A N Jaynes, B H Mauk, S J Bolton Published: 2026 ABSTRACT AbstractRecent observations of Jupiter’s bow shock have provided new insights into the energy dissipation mechanisms associated with bow shocks of outer planets. At greater distances from the Sun, the typically increasing super-Alfvénic nature of the solar wind leads to higher-energy bow shocks with enhanced dissipation requirements. Research on Earth’s bow shock suggests that plasma waves (e.g., ion-acoustic waves, waves driven by the electron cyclotron drift instability, electrostatic solitary waves, and whistler mode waves) play a critical role in converting solar wind bulk kinetic energy into thermal energy. However, the lack of high-resolution plasma wave data, which is essential for detecting these waves, has previously hindered their identification at the outer planets. Here we demonstrate, based on two case studies, that similar waves also exist at Jupiter’s bow shock. But, unlike their terrestrial counterparts, Ion-acoustic waves at Jupiter may display harmonic structures, indicating possible particle trapping, and intense electron cyclotron drift instability may occur to accommodate the shock’s intensity. Magnetic field measurements suggest possible shock reformation.Subject terms: Giant planets, Magnetospheric physics Introduction On the sun-facing side of a solar system planet, a bow shock forms because the solar wind must flow around a nearly impenetrable boundary created by the interaction of the solar wind and the planetary magnetic field or ionosphere. At the bow shock, the solar wind is abruptly slowed down, and the excess kinetic energy is primarily converted into heat. In collisional media, this energy dissipation is facilitated by particle-particle collisions. In ionized gases, this role is played by Coulomb collisions between electrons and ions1, which act as an effective frictional mechanism. However, in the collisionless environment of the solar wind and planetary bow shocks, Coulomb collisions are too rare to account for the observed dissipation. Instead, energy conversion must proceed through alternative mechanisms. These include dispersive radiation2, quasi-static field structures3, particle reflection at the shock front4, and wave-particle interactions5. Shock reformation further complicates the dissipation process and may result from nonlinear wave dynamics6–8, accumulation of reflected ions9,10, or compressive magnetic pulsations driven by reflected ion populations11,12. Despite these processes, all viable dissipation pathways converge at the microscale through wave-particle interactions. Theoretical studies5,13–16 predict that high-frequency electrostatic waves may provide the key dissipation channel in collisionless shocks. Observations17 indicate that in supercritical terrestrial bow shocks, ion entropy increases primarily in the shock foot and ramp regions. This enhancement is linked to the presence of both reflected ions and upstream solar wind ions, leading to a growing anisotropy and the generation of ion-scale electrostatic wave bursts. In contrast, electron entropy shows little change across the shock, implying mostly adiabatic heating, though non-adiabatic effects emerge at high-Mach-number shocks18,19. Wilson et al. demonstrated20 that wave-particle interactions govern the energy dissipation process of collisionless shocks at Earth, and high-frequency, broadband electrostatic waves, such as ion acoustic waves (IAWs), electron cyclotron drift instabilities (ECDIs), and electrostatic solitary waves (ESWs) are responsible. From a broad comparative examination of the planetary bow shocks, it is expected that the electrostatic wave-particle interactions at the outer planets may be significantly more effective in producing shock dissipation than their analogs at Earth21. However, no detailed study of energy dissipation at the bow shock of an outer planet, where the Mach numbers are typically higher than those of Earth22,23, has been reported. Until recently, the detailed structures of magnetic perturbations and the presence of various high-frequency plasma waves—key candidates for energy dissipation—at Jupiter’s bow shock remained poorly understood due to instrumental and telemetry limitations. Prior to the Juno mission, eight spacecraft had encountered Jupiter’s bow shock: Pioneers 10 and 11, Voyagers 1 and 2, Ulysses, Galileo, Cassini, and New Horizons. Bow shock crossings are typically evident in magnetic field data, and numerous studies have reported their locations based on such measurements. Averaged plasma wave spectra from these past missions are also generally available. Although Juno was equipped with burst-mode waveform capture from the start of its mission, this capability was not directed at the bow shock region until November 2024, in order to conserve telemetry resources. In the past, a few snapshots of high-resolution (28 kHz) electric field measurements at Jupiter’s bow shock were available from the Voyager spacecraft. However, these data suffered from a limited instantaneous dynamic range due to their insufficient (4-bit) amplitude resolution, which obscured many fine-scale plasma wave features. Here, we show high-resolution (50 kHz) and high dynamic range (16-bit) plasma wave observations from Juno at the Jovian bow shock. These observations, supplemented by magnetometer and charged particle data, are analyzed to offer a comprehensive view of Jupiter’s shock environment. Our findings suggest that, while some energy dissipation mechanisms at Jupiter’s bow shock resemble those at Earth, distinct features—such as strong nonlinear plasma wave behavior and discrete periodic shock reformation structures—appear to be different in the Jovian context. Results Bow shock crossings On 10 December 2024 at approximately 17:25UT, Juno traversed from Jupiter’s magnetosheath into the solar wind, before re-entering the magnetosheath on 11 December 2024 at around 03:05UT. During this interval, the spacecraft crossed the bow shock twice, exhibiting distinctly different characteristics on each occasion. These crossings occurred at a radial distance of about 81.44RJ (RJ= radius of Jupiter = 71,492 km) from Jupiter, at (JRM33 model24) approximate southern magnetic latitudes of 61.5° (outbound) and 59.7° (inbound), both on the dusk flank of the magnetosphere, corresponding to approximate magnetic local times of 16.02 and 16.4, respectively (Fig. 1). Observations during these intervals include magnetic field data from the MAG instrument25, electric fields from the Waves instrument26, and plasma measurements from the JADE instrument27,28. In addition to the one-second-averaged plasma wave electric and magnetic field power spectral densities, burst-mode electric field waveforms sampled at 50 kHz were available intermittently (duty cycle 12.5%). Magnetic field measurements were recorded at 8 Hz (typically 1 Hz) near the shock, while ion data were acquired at a two-minute cadence. Jovian bow shock observations on December 10 and 11, 2024. A Spectrogram showing the electric field power spectral density; B Omnidirectional ion count rates; C Magnitude of the background magnetic field. Regions of interest corresponding to bow shock crossings are indicated with magenta (Case I) and red (Case II) dashed lines. In (A), Langmuir waves detected by Juno during its traversal through the solar wind are marked with arrows. In (B), both incoming solar wind ions and reflected ions are labeled. Arrows indicate the identified bow shock crossings by step-like changes in the magnetic field (C) and shifts in solar wind populations (B). In (C), the thickness of the plot varies as a result of changes in the magnetic field sampling rate. A–C share the same x-axis, and the detailed x-axis label is shown only in (C). Bow shock crossings are identified by distinct step-like changes in the magnetic field (Fig. 1C) and by the appearance or disappearance of the solar wind proton population (Fig. 1B). The behavior of a shock is primarily governed by two parameters: the Mach number and the angle (θBn) between the upstream magnetic field and the shock normal. When θBn is less than approximately 30–45°—meaning the magnetic field29 is nearly parallel to the shock normal—protons and heavier ions reflected upstream do not reencounter the shock and instead escape upstream. These reflected ions interact with the incoming solar wind, driving instabilities and generating wave activity in an extended region known as the ion foreshock30. Conversely, for larger θBn, indicating a more perpendicular shock geometry, ion reflection becomes more localized, as specularly reflected ions execute partial gyro-orbits upstream before reencountering the shock. In this case, the shock transition occurs over a shorter region known as the “foot,” located just ahead of the shock ramp. The foot spans a distance on the order of the reflected ions’ gyroradius and arises from the magnetic field orientation constraining ion motion along the shock surface. We infer that both shock types were observed during our study, despite instrumental limitations. At the shock front, solar wind electrons are accelerated and reflected back into the upstream region, generating electron beams that travel along magnetic field lines. These backstreaming electrons generate plasma waves, particularly Langmuir waves31. During the inbound crossing on December 11, 2024 at about 3:05 UT, intense Langmuir wave activity was observed immediately adjacent to the shock transition (Fig. 1A). Similar wave signatures were present near the outbound shock on December 10, 2024 at about 17:25 UT, but these were spatially offset from the shock itself (Fig. 1A). Instead, a secondary ion population—interpreted as reflected solar wind ions (Figs. 1B, 3C)—was detected near the outbound shock. A schematic overview of the bow shock crossings is provided in Fig. 2. We refer to the outbound and inbound encounters as Case I and Case II, respectively, throughout this study. December 10, 2024 (Case I) bow shock features. A Magnitude of the background magnetic field. B Spectrogram of electric field power spectral density. C Omnidirectional ion count rates. Panels D–F show different high-frequency waves detected at the shock crossings. Time-domain variations of the electric field associated with electrostatic solitary waves (ESWs) and ion acoustic waves (IAWs) are shown in (D) and (E), respectively. F shows the time-domain waveform of the electric field (top) and the power spectral density (bottom) associated with the electron cyclotron drift instability (ECDI). Red vertical dashed lines across A–C indicate distinct regions the spacecraft traversed: the downstream region, ion foreshock-like, and electron foreshock-like regions. An ultra-low-frequency (ULF) magnetic oscillation in the ion foreshock-like region is marked in (A). A Langmuir wave signature within the electron foreshock-like region is noted in (B). In (C), the solar wind extending across the electron and ion foreshock-like regions, as well as reflected solar wind in the ion foreshock-like region, is identified. Artist’s impression (not to scale) of Juno’s trajectory through various regions near the bow shock. Blue lines represent the nominal (Parker spiral) background magnetic field, and black arrows indicate the direction of the solar wind. The electron foreshock-like (yellow) and ion foreshock-like (red) regions are shown along with unshocked region (light blue), magnetosheath (dark blue), and magnetosphere (gray). Juno’s outbound path (Case I) and inbound path (Case II) are marked by magenta and red arrowed lines, respectively. This illustration is a conceptual rendering based on the interpreted spacecraft trajectory and plasma wave regions. Approximate times and magnetic latitudes, where Juno crosses different regions, are marked by white crosses. Parts of the background of the illustration is adapted from Tsurutani and Rodriguez70. Both shock crossings had an Alfvénic Mach number of approximately 9. For Case I, the θBn was 17° ± 14⁰, while in Case II, it was approximately 54° ± 8⁰. Figure 3A–C provide a closer examination of Case I, highlighting distinct regions: the downstream region and different parts of the foreshock, including reflected/back streaming ion population region and electron beam–dominated regions. In this study, we use the term “electron foreshock-like region” to refer to the electron beam-dominated region and ion foreshock-like region to refer to the “reflected/back streaming ion population region”. The ion foreshock-like region exhibits minimal high-frequency plasma wave activity (Fig. 3B). Within this region, ultra-low-frequency32 (ULF) magnetic oscillations (about 7 mHz) are detected (Fig. 3A), accompanied by solar wind–reflected ions (Fig. 3C). The reflected ion count rate in this region shows significant variability by a factor of approximately 2. Beyond the ion foreshock-like region lies the electron foreshock-like region, characterized by strong Langmuir waves near 9 kHz (Fig. 3B). Figure 4A–C provide a closer examination of Case II, highlighting distinct regions identified as the downstream region, the foot, and an electron foreshock-like region. Unlike Case I, which featured an extended ion foreshock-like region, Case II exhibits only a compact foot region located just ahead of the shock front (Fig. 4A, B). Adjacent to the foot, Langmuir waves, likely generated by electron beams, are observed. Additionally, the magnetic field data (Fig. 4A, C) reveal periodic structures recurring approximately every 8.3 seconds. December 11, 2024 (Case II) bow shock features. A Magnitude of the background magnetic field. B Spectrogram of electric field power spectral density. C A zoomed-in view of the background magnetic field variation near the shock transition. D Time domain variation in electric field showing electrostatic solitary waves (ESW). E Time domain variation in the electric field showing ion acoustic waves (IAW). Red vertical dashed lines across A–B indicate distinct regions the spacecraft traversed: the electron foreshock-like region, foot, and downstream. Periodic magnetic oscillations in the foot and downstream region are marked in (A). A Langmuir wave signature within the electron foreshock-like region is noted in (B). The Relation between the periodicity of the oscillation (8.3 s) and proton gyro period is highlighted in (C). High-frequency plasma waves All the usual broadband electrostatic waves33 —namely, IAWs, ECDIs, ESWs, and Langmuir waves—were observed in and around the shock crossings. These wave modes were identified through distinct waveform signatures (Figs. 3D–F, 4D, E, 5B, E) and/or power spectral density (PSD) characteristics (Fig. 5C, F). ESWs were visually identified based on well-established criteria34–38. In this case, ESWs exhibited lifetimes of approximately 0.2 ms and amplitudes of several tens of mV/m. Nonlinear plasma wave observations on December 11, 2024 (Case II). A Magnitude of the background magnetic field, highlighting two key regions around the shock: the Ion-Acoustic Wave (IAW) section 1 (marked in blue) and the Electron Cyclotron Drift Instability (ECDI) section 2 (marked in red). B Time-domain waveform of the electric field in Section 1 (IAW region). C Power spectral density (PSD) of the electric field in Section 1, with five 20.8 ms segments color-coded to match their corresponding time intervals in (B). The colored dotted horizontal lines show the 0 dB reference x-axis for each interval. D Zoomed-in view of a subsection from Section 1, showing a distorted sine wave with a flat top during the red segment interval. E Time-domain waveform of the electric field in Section 2 (ECDI region). F PSD of a highlighted subsection of Section 2 (indicated in red in E), showing enhanced power at frequency intervals corresponding to the electron cyclotron frequency. In (C), frequency-domain structures for the five color-coded segments of the IAW region (from B) are shown, with a noticeable enhancement in power at harmonic frequencies (1st, 2nd, and 3rd harmonics) for the red trace. D provides a closer look at the nonlinear waveform structure observed during the same time interval. While it is possible to mistake a dust impact for an ESW39, the probability of such a misidentification is minimal—though not entirely negligible40—due to the frequent occurrence of ESWs at planetary bow shocks. Electron Bernstein waves were confirmed via the periodic enhancements in the PSD at an interval of the electron cyclotron frequency (Fig. 5F), consistent with earlier findings41. IAWs were identified both visually—as localized wave packets (Figs. 3E, 4E)—and via their broadband PSD signature, which appears Doppler-shifted and sometimes shows drift in the IAW peaks, likely due to variations in plasma parameters42,43, in the electric field spectrum (Fig. 5C). These IAWs typically occurred in discrete packets with amplitudes rarely exceeding ±0.01 V/m, and packet durations ranging from a few milliseconds to several tens of milliseconds. The Electron Bernstein mode waves observed here were superimposed on ion-acoustic waves, confirming the presence of ECDI. A narrow spectral line around 9.6 kHz (Figs. 3B, 4B), observed upstream of the shock, corresponds to Langmuir waves excited near the local solar wind plasma frequency by upstream-propagating electron beams along foreshock magnetic field lines. Although lower hybrid waves were expected, their presence could not be confirmed. This is due to instrumental limitations: the relevant lower hybrid frequency range was not captured by either the fluxgate magnetometer or the Waves instrument. Around planetary bow shocks, beam-plasma instabilities are common, and wave–particle interactions can produce harmonic wave features through both intrinsically multi-harmonic instabilities (e.g., ECDI) and nonlinear processes, such as wave steepening or particle trapping. Some interesting wave behavior with distinct harmonic structures was detected in regions highlighted by the blue and red lines in Fig. 5A. In the first region, at the shock foot (blue lines, Fig. 5A), a burst of 6144 electric field samples—covering 123 ms at a 50 kHz sampling rate—was analyzed (Fig. 5B). Strong IAW activity was observed between 3:05:55.78 and 3:05:55.82 UT (Fig. 5B, C). For spectral analysis, the dataset was divided into six segments of 1024 samples each, and a Fast Fourier Transform (FFT) was applied. The resulting spectra from five relevant segments are shown in Fig. 5C. Here, the center frequency of Doppler-shifted IAWs generally peaked around 5 kHz, with some variation. The maximum amplitude ( ± 0.015 V/m) was recorded near 3:05:55.8 UT, along with a harmonic structure at 5610, 11130, and 16680 Hz, with corresponding PSD values of 3.5×10⁻⁸, 3.0×10⁻¹⁰, and 7.0×10⁻¹² V²m⁻²Hz⁻¹. The fundamental frequency (5610 Hz) of IAW is well above the electron gyrofrequency of 196 Hz under the background magnetic field of 7 nT. The second instance of strong harmonic behavior occurred at the shock ramp (red lines, Fig. 5A). A similar analysis was conducted using a burst of 6144 electric field samples. This dataset revealed strong ECDI activity (Fig. 5F), with approximately 18 peaks at intervals of about 300 Hz (electron cyclotron frequency) around 3:06:06.780 UT (Fig. 5F). The time-domain signal (Fig. 5E) exhibited a “peaky” structure, with maximum amplitudes reaching approximately 0.06 V/m. Several additional ECDI activities were observed in Case II, whereas only a few were detected in Case I. A schematic overview of the plasma wave observations across different regions is presented in Fig. 6. Illustration of types of plasma waves observed around the Jovian bow shocks. The sketch (right panel) shows various plasma waves detected in different regions near the bow shock, with colored dots indicating specific wave types. Example electric field waveforms corresponding to the colored dots are presented in the left panel. While these time-domain structures on the left panel are informative, identifying wave types typically requires frequency-domain analysis (not shown here). In the quasi-perpendicular shock case (bottom right), the observed waves include linear and nonlinear ion-acoustic waves (IAWs), electron cyclotron drift instabilities (ECDIs), and electrostatic solitary waves (ESWs). In contrast, the quasi-parallel shock configuration (top right) primarily features linear IAWs and ESWs with few instances of ECDI. Additionally, ultra-low-frequency (ULF) waves appear in the ion foreshock-like region, while Langmuir waves are detected in the electron foreshock-like region. Shock reformation In Case II (Fig. 4A, C), regular magnetic field structures with a period of about 8.3 s were observed at the bow shock. For an upstream magnetic field of 2 nT, the proton gyroperiod is 32 s, indicating that the observed periodicity corresponds to approximately 0.26 of the proton gyroperiod. The strongest magnetic field peak, identified as the shock ramp, reached around 10.5 nT at 3:06:04.1 UT. Two earlier peaks in the foot region occurred at 3:05:55.75 UT and 3:05:47.5 UT, with approximate amplitudes of 7.12 nT and 5.56 nT, respectively. Downstream of the ramp, four additional prominent peaks exceeding 8 nT were recorded at around 3:06:13.8, 3:06:23.1, 3:06:31.5, and 3:06:39.3 UT. In single-spacecraft observations, various magnetic structures can arise from repeated shock crossings caused by shock surface motion driven by foreshock ULF waves or solar-wind fluctuations. Such signatures, however, typically occur at irregular intervals and on timescales much longer than a fraction of the proton gyroperiod. For Case II, no evidence of foreshock driving is observed, and plasma-wave measurements show no consistent signatures of solar-wind–driven variability. Among the mechanisms capable of producing shock-front nonstationarity, the observed steady timescales, comparable to a fraction of the gyroperiod, indicate that shock self-reformation is the most likely explanation44–46. The underlying cause of the self-reformation44,47 may involve reflection-controlled processes and/or the modified two-stream instability (MTSI) and ion-scale whistler precursor48. Direct confirmation of ion reflection at this timescale (8 s, corresponding to 0.26 τcp) is not possible due to Juno’s 120 s ion measurement cadence. Discussion High-frequency plasma waves at Jupiter’s bow shock, as at Earth, are the main contributors to plasma heating, but quantifying their power dissipation is challenging because Juno’s instruments were not designed for this purpose. Ideally, such analysis20 requires full three-axis E and B field measurements across low (<10 Hz) and high ( > 2fpe) frequencies, along with high-frequency electron and ion spectra. Juno instead provides limited data: three-axis magnetic fields at 8 (1) Hz and a 50 kHz burst-mode (1/8 duty cycle) single-axis E-field in the spin plane. Consequently, estimates of wave power dissipation in this study are necessarily qualitative rather than quantitative. Stronger wave nonlinearity at Jupiter is expected due to the substantially larger temperature increase across its bow shock. Studies49–51 show ion temperature ratios of approximately 80 and electron temperature ratios of approximately 10 across the shock, exceeding typical Earth values of about 18 and 5, respectively. Some shock intervals in our study resemble Earth-like behavior, whereas others exhibit stronger high-frequency waves (IAWs, ECDIs) and fore shock structures that can produce much greater plasma heating at Jupiter’s bow shock. These processes are discussed below. It should be noted that comparisons of wave-field magnitudes between Jupiter (5 AU) and Earth (1 AU) must account for the weaker solar-wind magnetic field at Jupiter. The solar-wind magnetic field strength scales approximately as r−1.4, where r is the heliocentric distance from the Sun52. As a result, the field at 5 AU is lower by a factor of 5−1.4 relative to 1 AU. Therefore, comparisons with Earth should be based on wave amplitudes normalized by a factor of about 10 rather than raw observed amplitudes53. Linear, Doppler-shifted IAWs, generated by ion–electron relative drift or counter-streaming ion beams, are commonly observed at planetary bow shocks. In the nonlinear regime, these waves can steepen, distorting the electric field waveform and producing second- and higher-order harmonics of the fundamental frequency54–56. IAWs can interact with low-energy electrons through Landau resonance, leading to electron heating. The presence of multiple harmonics of IAWs broadens the range of electrons that can be energized. While the harmonic structure of IAWs has previously been observed in Earth’s auroral zone55 and near 35 solar radii by the Parker Solar Probe57, this appears to be the first direct observation of such occurrences at a planetary bow shock. Evidence of this phenomena is found in the flat-topped, sinusoidal (5610 Hz) electric field waveforms recorded between 03:05:55.786 and 03:05:55.806 (Fig. 5D), and is further supported by the detection of second and third harmonics at 11,130 Hz and 16,680 Hz, respectively, in the corresponding frequency spectrum (red trace, Fig. 5C). The waveform shown in Fig. 5D matches the expected profile of moderately steepened IAWs—displaying an asymmetric shape with flattened peaks, but without sharp electric field spikes58. The electric field amplitudes remain well below the analog-to-digital converter (ADC) saturation threshold, ruling out instrumental clipping as the cause of the observed harmonics. The observed fundamental frequency (5.61 kHz) falls above the local electron gyrofrequency (196 Hz) and below the electron plasma frequency (9.6 kHz) measured in the upstream foreshock. Due to Juno’s single-probe electric field configuration, wave polarization and phase velocity could not be determined. This kind of waveform distortion can be attributed to electrons trapped in the electrostatic potential of wave59. Consistent with auroral-zone observations, these IAWs do not necessarily exhibit higher amplitudes than their linear counterparts as particle trapping phenomenon is mostly independent of the wave amplitude55. Any repeated trapping and untrapping of electrons is substantially more effective for shock dissipation than the nonlinear steepening of electrostatic waves, as observed in cases of ECDI58 and oblique whistler waves59. The electron cyclotron drift instability generates waves (i.e., ECDI) consistent with a combination of Doppler-shifted IAWs and electron Bernstein modes, typically appearing at harmonic intervals of the electron cyclotron frequency (fce). In our observations (e.g., Fig. 5F), the ECDI manifests as distinct spectral peaks at multiple fce harmonics, superimposed on a broadband IAW spectrum extending up to the plasma frequency (fpe around 9.6 kHz, estimated from upstream Langmuir oscillations). This instability arises from the relative drift between incoming electrons and shock-reflected ions60–62. ECDI-driven waves play a key role in collisionless shock physics by resonantly interacting with reflected ions, thereby extracting bulk kinetic energy and preferentially heating the electron population61–63. The growth of the ECDI is favored when the electron-ion drift speed (vd) exceeds the perpendicular electron thermal speed (ve), with significant wave excitation occurring64 when the ratio vd/ve increases. Since ve is proportional to the ambient magnetic field strength, which diminishes with increasing heliocentric distance, ECDI activity is expected to be more prevalent at the bow shocks of the outer planets under comparable shock conditions. This is evident from Table 1, which shows that even under unfavorable antenna orientations ( ≪ 90°) in our observations, all ECDI peaks remain detectable until the receiver signal approaches the background noise level. In contrast, at Earth typically only a small fraction of the maximum possible (fpe/fce) ECDI peaks are observed. A richer set of resonance points generally enhances total electron heating, as it broadens the range of velocities and harmonics participating in the energy exchange process. So, ECDI-related power dissipations at Earth’s bow shock are more moderate than those detected at Jupiter20,65. Table 1Electron Cyclotron Drift Instability (ECDI) observations at the bow shocks of Earth and JupiterPlanetObservation (UT)Spacecraftfpe(kHz)/fce(kHz)Observed harmonics (N)Antenna orientation (⁰) w.r.t BNfce/fpeEarth652001-11-14T01:14:36WIND27/0.510 (y-antenna)5 (x-antenna)90⁰42⁰0.180.089Earth652006-11-17T20:13:58STEREO-B18/0.4613 (x-antenna)3 (y-antenna)90⁰90⁰0.330.077Earth202009-07-13T08:59:50THEMIS-B28/15 (x- antenna)3 (y-antenna)90⁰90⁰0.1780.107Earth692015-11-27T06:02:49MMS-440/0.2753 (x-antenna)3 (y-antenna)90⁰90⁰0.0210.021Jupiter2024-12-11T03:06:06.8Juno9.6/0.3018 (y-antenna)62⁰0.56Jupiter2024-12-11T03:06:00.8Juno9.6/0.3120 (y-antenna)63⁰0.645Jupiter2024-12-10T17:25:01.8Juno9.6/0.3518 (y-antenna)80⁰0.66fpe/fce is the electron plasma-to-cyclotron frequency ratio, and N is the number of observed ECDI harmonics. Column 6 gives the antenna orientation relative to the ambient magnetic field. The quantity Nfce/fpe listed in the final column provides a measure of ECDI severity.Columns 1–3 show the planet, observation time, and spacecraft, respectively. Electron Cyclotron Drift Instability (ECDI) observations at the bow shocks of Earth and Jupiter 10 (y-antenna) 5 (x-antenna) 90⁰ 42⁰ 0.18 0.089 13 (x-antenna) 3 (y-antenna) 90⁰ 90⁰ 0.33 0.077 5 (x- antenna) 3 (y-antenna) 90⁰ 90⁰ 0.178 0.107 3 (x-antenna) 3 (y-antenna) 90⁰ 90⁰ 0.021 0.021 fpe/fce is the electron plasma-to-cyclotron frequency ratio, and N is the number of observed ECDI harmonics. Column 6 gives the antenna orientation relative to the ambient magnetic field. The quantity Nfce/fpe listed in the final column provides a measure of ECDI severity. Columns 1–3 show the planet, observation time, and spacecraft, respectively. Shock self-reformation9,66,67 is expected to be more prevalent, stronger, and easier to observe at Jupiter than at Earth, owing to higher Mach numbers, lower plasma β, and larger ion kinetic scales in the outer heliosphere. Figure 7 highlights the importance of this phenomenon at Jupiter, as observed in Case II, by comparing it with typical cases at Earth and Saturn. In spacecraft observations, shock self-reformation appears on timescales of tens to hundreds of seconds as large-amplitude, quasi-periodic magnetic enhancements at Jupiter (Case II) and Saturn45 —much stronger and longer-lived than the often intermittent signatures at Earth, which last less than a couple of seconds. Each microstructure within the shock front is expected to have a characteristic width comparable to the electron inertial length, which is roughly an order of magnitude larger at Jupiter than at Earth. Wave–particle interactions within these microstructures (Fig. 4A, 5A) contribute to plasma heating. The associated energy-conversion processes (e.g., wave activity) exhibit the same cyclic timing (Fig. 4B). A prolonged sequence of shock self-reformation structures—commonly observed at the outer planets45 —would broaden the effective shock-dissipation region and could thereby enhance overall plasma heating. Comparison of foreshock reformation phenomena at different planets. A–C show fluxgate magnetometer observations at Earth, Jupiter, and Saturn, respectively. The red dashed line marks the bow shock location, while the blue dashed lines indicate successive reformation cycles in each case. In each panel, the time interval between successive reformation cycles and its correspondence to the proton cyclotron period (Tcp) are indicated. The shock normal angle (θBn), Alfvén Mach number (MA), and ion plasma beta (βi), where available, are also shown. In summary, our findings offer valuable insights into the mechanisms of energy dissipation at Jupiter’s bow shock. They reveal that the wave-particle interactions at moderately high Mach numbers and low plasma beta shocks in the outer planets share notable similarities with those observed at Earth, though nonlinear behaviors may occur more frequently. Although additional observational and theoretical studies are required to establish the broader applicability of these behaviors, our results extend current understanding of collisionless shock physics in a previously unexplored parameter regime of the solar system. Methods Calculation of shock normal angle Coplanarity theorem The events analyzed here exhibited significant magnetic field perturbations, resulting in local variations of the shock normal. To lessen these effects, we employed a 10-minute averaging window upstream and downstream of each shock crossing, excluding regions close to the shock that are influenced by foot structures, upstream wave activity, or downstream overshoot. For the outbound shock observed on 10 December 2024 at approximately 17:25 UT, the downstream and upstream intervals were selected as 17:10–17:20 UT and 17:35–17:45 UT, respectively. For the inbound shock on 11 December 2024 at approximately 03:05 UT, the upstream and downstream intervals were 02:30–02:40 UT and 03:20–03:30 UT, respectively. The shock normal vector where Bd and Bu are the mean magnetic fields in the downstream and upstream windows, respectively, and ΔB = Bd - Bu. The angle between the upstream magnetic field and the shock normal, θBn (degrees), was calculated as: with uncertainty where ΔBu and ΔBd are the standard deviations. Using this method, we find that θBn is 16.5° ± 14.3⁰ for the outbound shock and θBn is 53.5° ± 7.9⁰ for the inbound shock. Magnetic field data used are available from the Juno Fluxgate Magnetometer Calibrated Data V1.0 dataset at the NASA Planetary Data System: https://doi.org/10.17189/1519711. Calculation of Mach 006Eumber Upstream parameter window used for 2024−12-10 is18:00 – 18:40 and 2024-12-11 is 2:00-2:40. Magnetic field strength (upstream) about 2 nT Plasma frequency from upstream Langmuir wave about 9 kHz Using where ne is the electron number density, e and m are the electron charge and mass, respectively and ε0 is the permittivity of the free space, results in ne about 1 cm−3. In solar wind, the Alfven velocity where np (approximately same asne) is the number density of proton and mp is the mass of a proton. Solar wind speed Vs estimated from JADE is about 400 km/s (MMESH model68 is about 410 km/s), Alfven Mach number where θVn is the angle between the solar wind velocity and the shock normal. In the absence of direct measurements, θVn is assumed to be 0⁰, implying that this estimate applies at the nose of the shock. Electron number density estimates are derived from the Juno Waves Survey Data V2.0 dataset (JNO-E/J/SS-WAV-3-CDR-SRVFULL-V2.0) available through the NASA Planetary Data System. https://doi.org/10.17189/1520498. Supplementary information Transparent Peer Review file Source data Source Data