Geophysical Research Letters

Post-equinox periodicities in Saturn's energetic electrons

Authors


Abstract

[1] Since Saturn's vernal equinox in August 2009 (day 223), energetic electrons (110–365 keV) have exhibited a variety of periodic and aperiodic behavior within a spectral window of 5–15 hours. From late 2009 through the end of 2010, when the observed at dusk, a single period near 10.7 hours dominated the Lomb spectra of these particles. Near the end of 2010, however, the energetic electrons displayed multiple periods, with the strongest at 10.65 hours. The periodicity observed after equinox has a mean value of 10.69 ± 0.06 hours and agreed closely with that of Saturn kilometric radio (south) emissions. By early 2011, when the observer had moved to the dayside, the periodicities abruptly disappeared and the Lomb spectra show no periodicity. This behavior may suggest changes in Saturn's ionosphere as a result of seasonal change, or may alternately imply a local time dependence of periodicity caused by magnetodisk thickness asymmetry.

1. Introduction

[2] The periodicity of Saturn's magnetosphere presents one of the most interesting problems in planetary physics. The Pioneer and Voyager flybys in the 1970's and 1980's discovered strong periodicities near 10.65 hours in Saturn's kilometric radio (SKR) emission of the planet as well as in energetic charged particles and magnetic fields [Desch and Kaiser, 1981; Carbary and Krimigis, 1982; Espinosa and Dougherty, 2000]. Later observations from Ulysses demonstrated Saturn's radio period changes slowly by ∼1% over several years in a manner incompatible with the planet's internal rotation [Galopeau and Lecacheux, 2000], which should remain stable to much better than 1% over periods of several years [e.g., Anderson and Schubert, 2007]. Observations made by the Cassini spacecraft in 2004 confirmed that the SKR period had changed to ∼10.7 hours [e.g., Gurnett et al., 2005; Kurth et al., 2007]. By 2009, curious dual periods of ∼10.6 and ∼10.8 hours had been recognized in radio emissions, charged particles, and magnetic fields in Saturn's magnetosphere [Gurnett et al., 2009; Carbary et al., 2009a; Southwood, 2011]. The dual periods have been associated with radio sources, one in the southern hemisphere and one in the northern hemisphere. The two radio periods evolve in such a way that they may merge and cross shortly after equinox [Gurnett et al., 2010]. The source of these north/south periodicities and their variability remains mysterious and unknown, although many investigators consider the dual periods as seasonal effects in Saturn's ionospheric conductivity [e.g., Gurnett et al., 2007, 2009, 2010; Andrews et al., 2010; Southwood, 2011].

[3] Periodicities of energetic particles (E > 10 keV) in Saturn's outer magnetosphere (>20 RS, 1 RS = 60238 km) are usually considered as oscillations of the planet's plasma sheet where such particles are concentrated [e.g., Carbary et al., 2007; Khurana et al., 2009], or in terms of a rotating anomaly [Carbary and Krimigis, 1982], or in terms of periodic releases via plasmoid motion down the magnetotail, possibly in connection with magnetospheric injection events driven by centrifugal interchange instabilities [Brandt et al., 2008, 2010]. The periodicities of energetic particles apparently have counter-parts in the magnetic field oscillations [Espinosa and Dougherty, 2000; Giampieri et al., 2006; Andrews et al., 2010] and thermal plasma periodicities [Arridge et al., 2008; Burch et al., 2008, 2009; Jackman et al., 2009]. The precise connection between the magnetic field periodicities and particle periodicities is not fully understood [see Southwood, 2011, and references therein].

[4] This paper describes the periodicity in fluxes of energetic electrons (110–365 keV) in Saturn's outer magnetosphere (>15 RS) after the planet's vernal equinox, which occurred on day 223 in 2009. A number of previous investigations have demonstrated the efficacy of using these electron fluxes to examine periodicities in the Saturn's outer magnetosphere [e.g., Carbary et al., 2007, 2009a, 2009b]. The flux oscillations may represent motions, possibly “flapping,” of the plasma sheet [e.g., Arridge et al., 2008; Carbary et al., 2008; Jackman et al., 2009]. As shown here, the electron periodicity becomes singular immediately after equinox, then abruptly reverts to multiple periodicity, and finally becomes aperiodic during 2011.

2. Instrumentation and Method of Analysis

[5] The energetic electron data are from the Low Energy Magnetospheric Measurement System (LEMMS) on the Cassini spacecraft. LEMMS is a subsystem of the Magnetospheric IMaging Instrument (MIMI) [Krimigis et al., 2004]. This study uses data from the lowest energy channel of the “E” detector, or E0, which measures electrons from 110 to 365 keV. This LEMMS channel offers low noise, nearly continuous coverage, and strong foreground signal throughout the magnetosphere. The use of E0 for studying magnetospheric periodicity and morphology has been discussed previously, and several recent investigations have employed this particular channel [e.g., Carbary et al., 2009a, 2009b].

[6] The E detectors have an ultimate time resolution of ∼6 seconds, but to improve the signal-to-noise ratio and reduce the data processing burden, the differential electron fluxes were averaged into 15-minute time intervals. Then E0 fluxes inside radial distances of 15 RS were filtered out, as were fluxes outside the model magnetopause of Arridge et al. [2006]. Fluxes in the inner magnetosphere were removed to eliminate effects of satellite absorption as well as possible Doppler effects of the spacecraft motion; outside the magnetosphere the E0 fluxes drop to essentially background, rendering magnetopause filtering somewhat superfluous. Similar data conditioning has been performed to examine periodicities in previous epochs [e.g., Carbary et al., 2009a, 2009b]. Measurements were used from day 280 2009 to day 270 201, the latest available to date. These amounted to over 34000 15-minute samples from the outer magnetosphere. This interval is after equinox when the spacecraft moved exactly in the equatorial plane and both north and south periods could be sampled.

[7] The E0 fluxes were subjected to Lomb periodogram analyses, which can reveal both short-duration (∼10 hr) and long-duration (∼10 day) periodicities [e.g.,Carbary et al., 2009a, 2009b]. The periodogram window was restricted between five hours and 15 hours to examine short-duration periods comparable to SKR periods. The signal-to-noise ratio (SNR), which is the ratio of the peak maximum to the mean of the secondary peaks, reveals the strength and accuracy of the periodicity. For the 5-10-hour window used here, an SNR of ∼10 or more indicates a strong signal whose peak period is known to within 2%, while an SNR above 20 indicates an extremely strong signal with a peak accuracy better than 1% (1% ≈ 6 minutes for a 10.8 hour period). These uncertainties can be determined from Monte Carlo simulations of noisy oscillations (seeauxiliary material).

3. Periodograms

[8] Figure 1 (top) summarizes the principal results of this investigation, while Figures 1 (middle) and 1 (bottom) show the local time and distance of the spacecraft at apoapsis. Figure 1(top) shows periodograms “stacked” in time (x axis), with the color indicating the Lomb power at a particular period (y axis). Reds and yellows indicate high power; blues and greens indicate low. In sliding increments of 10 days, periodograms were constructed in 100-day intervals from day 280 2009 to day 270 in 2011. To regularize the display grid, each individual periodogram was also re-binned into regular increments of 0.05 hours. The resulting composite shows how the periodicity varies over two years, although the rebinning does smooth the periodogram peaks. Similar displays of radio and electron periodicities have been made for earlier epochs [e.g.,Gurnett et al., 2009, 2010; Carbary et al., 2009a].

Figure 1.

Lomb spectrograms of energetic electrons, (top) stacked by time compared to the (middle) local time and (bottom) periapsis of the Cassini spacecraft. In Figure 1 (top), red and yellow colors indicate enhancements in the spectral power, while greens and blues represent low power. Crosses show peaks in the spectral power; three separate intervals for detailed analysis are indicated. The SKR north and south periods derived by Lamy [2011] are overplotted for comparison.

[9] After equinox, the energetic electrons displayed three types of periodicity. During most of 2010, when the spacecraft had regularized its orbit at the equator near dusk, the electrons exhibited a strong, single period near 10.70 hours. Interval 1 typifies this single peridocity. The single period trended slightly downward (toward shorter periods) from the equinox through late 2010. This period is identical to that of the southern SKR [Lamy, 2011]. By late 2010, however, the single period splits into several new filaments. The strongest of these periods (the one with the most Lomb power) was near 10.65 hours. The spacecraft trajectories at this time were very close to dusk. Interval 2 indicates when the multiple periods occurred.

[10] By early 2011, the electron periodicities became noisy and essentially disappeared. This occurred in interval 3, which lasted essentially to the last available data. By this time, the spacecraft trajectory had migrated in local time to the dayside. By the middle of 2011, the Cassini apoapsis was near 15 hours.

[11] Figure 2 details the electron periodicities in each of the three intervals. Figure 2(top) is from the first interval when the spacecraft was on the nightside. The single period is very strong and has a signal-to-noise ratio larger than 20, which indicates an uncertainty in the period of less than 1% (see auxiliary material).Figure 2 (middle) shows the interval when multiple periods were observed. The main periodicity seemed to be breaking down at this time, when the spacecraft was moving from night to day. Finally, Figure 2(bottom) indicates the interval when Cassini was on the dayside in the mid-afternoon. Here, essentially no periodicity is apparent. The maximum in the Lomb periodogram is near 14.5 hours, but the signal is weak compared to the noise (SNR = 7.1) and essentially no periodicity is evident.

Figure 2.

Detailed Lomb spectra of energetic electrons in the three intervals delineated in Figure 1. (top) The first interval contains a strong monoperiodic signal. (middle) In the second interval, signals at four different periods appear. (bottom) In the third interval, no distinct periodicity can be recognized.

4. Discussion

[12] A general consensus is emerging in the Saturn magnetospheric community that the planet's bizarre periodicities have some connection with the planet's ionosphere [Gurnett et al., 2009, 2010; Carbary et al., 2009a; Andrews et al., 2010; Southwood, 2011]. As the planet's seasons change, the high-latitude ionospheric conductivities also change because of changing solar illumination. Seasonal variation in the ionospheric conductivities in turn alters the field-aligned currents that generate the SKR and, presumably, the other periodic magnetospheric phenomena. However, the exact mechanism of generating an m = 1 wave period, a different one for each hemisphere, remains elusive [e.g.,Southwood, 2011], and differentially-rotating hemispheres present obvious theoretical difficulties where their field lines meet at the equator. Furthermore, very recent investigations have suggested that both “south” and “north” periods can be observed at all latitudes, irrespective of hemisphere, in SKR and energetic particles [e.g.,Lamy, 2011].

[13] According to one conjecture, the SKR periodicities will become equal and reverse shortly after equinox when north and south ionospheric conductivities become equal under the influence of equal solar illumination [Gurnett et al., 2010]. At a time sufficiently removed from equinox, the north and south conductivities would again become different and prompt a separation of the periodicities, one for the north and one for the south.

[14] The post-equinox observations of the energetic electrons suggest their periodicities may evolve according to observer local time. On the nightside, periodicity was well organized, but as the observer moved from night to dusk to day, the periodicity fragmented into multiple and then non-existent threads. This behavior may represent a day-night asymmetry in the magnetodisk. Saturn's magnetodisk is much thinner on the nightside than on the dayside [Krimigis et al., 2007; Sergis et al., 2007, 2009]. Being confined to the equatorial plane, the satellite would be much more likely to observe oscillating motions of a thin magnetodisk on the nightside than a thick magnetodisk on the dayside. Another possibility might be that the disk itself may only oscillate on the nightside, and that such oscillations are severely damped on the dayside due to solar wind effects.

5. Conclusions

[15] Revealed in the fluxes of the energetic electrons, the periodicity in Saturn's outer magnetosphere (>15 RS) in the two-year period after equinox displays a single period at 10.69 hours immediately after equinox in 2010, but then separates into multiple periods by late 2010. By early 2011, evidence of periodicity has vanished. The usual north-south ionospheric conductivities that equalize near equinox might explain the behavior of the energetic electron periodicity. Alternately, the loss of periodicity might be attributed to spacecraft motion from a thin, oscillating magnetodisk on the nightside to a thick, possibly non-oscillating magnetodisk on the dayside.

Acknowledgments

[16] This research was supported in part by the NASA Office of Space Science under Task Order 003 of contract NAS5-97271 between NASA Goddard Space flight Center and the Johns Hopkins University.

[17] The Editor thanks an anonymous reviewer for his or her assistance in evaluating this paper.