CASSINI–HUYGENS: MAGNETOMETER Cassini’s magnetometer at Saturn

Michele Dougherty and the Cassini magnetometer team pick some highlights of this successful discovery mission.

xactly one month shy of 20 years after its 15 October 1997 launch from Cape ECanaveral, the Cassini–Huygens (hereafter referred to as Cassini) NASA– ESA spacecraft will end its life by burning up in the atmosphere of Saturn. This con- 1 Plumes of water ice and vapour erupting from locations along the “tiger stripes” near the south pole of clusion has been designed to protect any of Saturn’s moon Enceladus. (NASA/JPL/Space Science Institute) the potentially habitable moons of Saturn (in particular Enceladus and Titan) from science return from other instruments, but et al. 2017). On this first Cassini fly-by, MAG possible contamination by the spacecraft. It were accepted as necessary for the MAG observations revealed a clear perturbation ends a mission that has been a resounding team to fulfil their science objectives. near the moon, which was interpreted as and demonstrable success: many scien- To cover all of the science return from a signature of the nearly corotating Saturn tific discoveries, thousands of published the MAG team is beyond the scope of this plasma, and the magnetic field that was research papers, hundreds of graduated article. Instead, we focus on some of the “frozen in” to this plasma, being deflected PhD students, and widespread excitement highlights, including the MAG-led discov- and slowed around the moon; Enceladus and inspiration among the general public ery of a water vapour plume at Enceladus; seemed to be acting as an unexpectedly and schoolchildren alike. The Cassini mis- planetary-period oscillations large obstacle. In addition, sion has been a truly inter­national endeav- which fill the magnetosphere “Enceladus is one of there was an increase in our in which thousands of scientists and and potentially mask the sig- the prime potentially ion cyclotron wave activity engineers from around the world, and from nals of the internal dynamo habitable locations in produced by water group many different cultures, worked together planetary magnetic field; our solar system” ions near Enceladus, imply- towards a common goal. field-aligned currents (FACs) ing that the moon itself was The UK-led magnetometer (MAG) team, and the resulting aurora. We will also adding water group ions to the flowing, has Imperial College as the principal highlight results related to the moon Titan ambient magnetospheric plasma. investigator institute, UK co-investigators and the magnetosphere of Saturn. We end The second planned Enceladus fly-by based at the University of Leicester and the article with a description of the end-of- on 8 March 2005 reached a closer alti- University College London, and inter­ mission science orbits – the “Grand Finale” tude of 500 km. MAG data revealed very national co-investigators from Germany, (figure 9) – which were designed with MAG similar signatures, both the “draping” of Hungary and the United States. The instru- and gravity observations in mind. the magnetic field around the moon and ment is a dual-sensor suite, with a fluxgate an increase in the power of water group magnetometer (FGM) designed and built Discovering the plume at Enceladus ion cyclotron waves. This confirmed the at Imperial College, and a vector helium/ On 17 February 2005, the first targeted instinct of the team that there was some scalar sensor (V/SHM) designed and built fly-by of the moon Enceladus took place atmospheric interaction – with unknown at the Jet Propulsion Laboratory, Califor- at a distance of 1265 km (the diameter of source – at Enceladus. Because the gravita- nia. These sensors are located halfway Enceladus is 500 km). Before this, ground- tional field of Enceladus is relatively small, along and at the end of the spacecraft’s based observations and data from the such a source would need to be strong 11 m magnetometer boom. A year after Pioneer and Voyager spacecraft (in the late in order to maintain the presence of an Saturn orbit insertion, which occurred on 1970s and early 1980s) had indicated that “atmosphere” for both fly-bys. The team 1 July 2004, the V/SHM stopped operating, the surface of Enceladus had relatively produced a schematic of the potential dif- resulting in a much more complicated data few craters and was mainly smooth with fuse, extended atmosphere (figure 2a). calibration procedure, involving regular some extensive linear cracks. The surface Based on the observations from these rolls of the entire spacecraft in a quiet back- was dominated by water ice and seemed two fly-bys, the MAG team made the case ground magnetic field. These rolls must be to have been resurfaced. It had also been to the Cassini Project that there was poten- executed about two distinct axes to enable postulated that Enceladus could be the tially an atmosphere of water group ions at calibration of FGM data. This illustrates the source of the material in Saturn’s extensive, Enceladus, which was holding off the Sat- collaborative approach within the Cassini diffuse E ring (for a summary of the history urn field lines from the surface of the moon. team: these calibration rolls impact on the of Enceladus observations, see Dougherty The team requested that the third fly-by,

Downloaded4.36 from https://academic.oup.com/astrogeo/article-abstract/58/4/4.36/3988906 A&G • August 2017 • Vol. 58 • aandg.org by Institute of Child Health/University College London user on 10 January 2018 CASSINI–HUYGENS: MAGNETOMETER

on 14 July 2005, approach much closer to 2 (a) A schematic (with (a) the surface in order to investigate. This was Saturn and Enceladus agreed by the project team and Cassini’s not shown to scale) third fly-by reached an altitude of 173 km. showing the corotat- This time, multiple Cassini instruments ing Saturn magnetic obtained definitive evidence for active ejec- field and plasma being tion of water vapour and ice particles from draped ahead of the south pole of Enceladus. Enceladus by a diffuse, The resulting magnetic field observa- extended atmosphere. tions confirmed the atmospheric signature (Dougherty et al. 2006) but indicated that the “atmosphere” was (b) Revised schematic focused at the south pole, as revealed in showing the corotat- figure 2b. The various instrument data ing Saturn magnetic sets from this third fly-by revealed a moon field and plasma being with internal heat leaking out of cracks at perturbed by the polar the south pole, and a water-vapour plume plume of water vapour filled with dust and organic material rising generated at the south hundreds of kilometres above the surface pole of Enceladus. (b) (figure 1, see Dougherty et al. 2017). (Dougherty et al. 2006) Based on this plume discovery, the Cas- sini extended missions were designed to 3 Twelve days of Cas- further investigate Enceladus, which is now sini magnetospheric regarded as one of the prime potentially data during Rev. 17 in habitable locations within our solar system. 2005, with multiple magnetopause bound- Planetary period oscillations ary crossings inbound The phenomenon of planetary period at mid-morning on oscillations (PPOs) appears to be unique days 298 and 299, to Saturn’s magnetosphere. In the PPOs, periapsis near dusk at all the magnetospheric field and plasma the end of day 302, parameters oscillate at near the planetary and an outbound rotation period, despite the planetary magnetopause­ cross- magnetic field being, as far as we know, ing pre-dawn on day perfectly symmetrical about the planet’s 307. The top panel spin axis. The plasma parameters oscillate shows a radio wave (3) at the planetary rotation period in Jupi- power spectrogram ter’s magnetosphere, but that is because from 5 kHz to 2 MHz, Jupiter’s magnetic dipole is tilted by ~10° to the second panel a its rotation axis, so that both the field and thermal electron flux the embedded plasma “rock” up and down spectrogram from at the rotation period as the planet spins. ~0.5 eV to ~30 keV, and Cassini MAG data have shown, however, the four lower panels that the tilt of Saturn’s planetary magnetic show the three compo- dipole is less than ~0.1° (Burton 2010). nents and magnitude The existence of PPOs at Saturn was first of the magnetic field. detected in power modulations of plan- The components are etary radio emissions observed by the two spherical polar, refer- Voyager spacecraft in 1980. These radio enced to the northern modulations were interpreted as revealing spin/magnetic axis of the deep rotation period of the planet via the planet, and have some rotating magnetic anomaly – similar, the internal field of the in principle, to what happens at Jupiter. The planet subtracted. The period thus determined, ~10.656 h, remains data at the bottom as the IAU System III period (Desch & give the day number Kaiser 1981), in the absence of a more defini- and spacecraft position tive value. Oscillations near a ~10 h period – local time (hours), were also observed in the energetic particle co-latitude and radial and magnetic data from the Pioneer 11 distance (in Saturn

and Voyager fly-bys (Carbary & Krimigis radii, RS = 60 268 km). 1982, Espinosa & Dougherty 2000), but they PPO-related ~10 h proved not to be consistent with a rotat- modulations are evi- ing magnetic anomaly within the planet dent in all parameters, (Espinosa 2003). Remote radio observations and inbound magneto­ by the Ulysses spacecraft, over ~10 years pause location. (Modi- starting in 1993, showed that the radio fied from Gérard et al. modulation period varies slowly with time, 2006)

DownloadedA& Gfrom • August https://academic.oup.com/astrogeo/article-abstract/58/4/4.36/3988906 2017 • Vol. 58 • aandg.org 4.37 by Institute of Child Health/University College London user on 10 January 2018 CASSINI–HUYGENS: MAGNETOMETER

by up to ~1% per (Earth) year (Galopeau & Lecacheux 2000), much too fast to be associ- ated directly with planetary rotation. Data acquired after Cassini’s Saturn orbit insertion in 2004 showed that PPOs are ubiquitous in Saturn’s magnetosphere, and are associated with modulations of the magnetic field, plasma particle proper- ties, and waves that rotate around the planet with a ~10 h period throughout the magneto­sphere, as illustrated in figure 3. Radio observations subsequently revealed two PPO systems, simultaneously, at Sat- urn: one associated with the northern polar region and the other with the southern, rotating with slightly different (and drift- ing) periods (Gurnett et al. 2009). During the early Cassini mission (~2004–2007, post- 4 The upper and lower rows of the figure illustrate the northern and southern PPO-related current solstice southern summer, figure 3), the systems (green lines and symbols) and perturbation magnetic fields (blue lines and symbols), southern magnetic oscillation was found to respectively. Circled dots and crosses denote vectors pointing out of and into the plane of the diagrams. be dominant by a factor of ~3 in amplitude, The sketches on the left show views of the polar ionospheres looking down from the north, “through” and to have a longer period (~10.8 h) than the planet in the case of the southern system. The red dashed lines show the sense of the driving the northern oscillation (~10.6 h). twin-cell flows in the thermosphere/ionosphere. Position with respect to these systems is defined by

Further analyses have revealed that the the northern and southern phases, ΨN and ΨS, as shown. The central sketches show the principal PPO-

PPOs are not driven from Saturn’s interior, related currents in the ΨN,S 90°–270° meridian planes, where the black lines show the near-axisymmetric but arise from large-scale, rotating systems background magnetic field. The sketches on the right show the PPO-related magnetic perturbations in

of electric current, which extend out from the orthogonal ΨN,S 0°–180° meridian planes. (Adapted from Hunt et al. 2015) the two polar ionospheric regions, as illus- trated in figure 4 (Southwood & Kivelson 5 Plot showing PPO 2007, Andrews 2010a, 2010b, Southwood properties over the & Cowley 2014, Hunt et al. 2015). These 11 years from 2005 currents appear to be generated by rotat- to 2015. Time at the ing twin-vortex flows in the northern and bottom is given in days southern polar thermosphere–ionosphere since the start of 2004, regions (shown as red dashed lines on with year boundaries at the left side of figure 4). The effect of these the top, together with vortical flows, whose physical origin is Cassini Rev. numbers uncertain, then spreads outwards along at periapses. The top the polar field lines (arrowed black lines) panel shows the solar into the magneto­sphere, propagating in latitude at Saturn, wave-like fashion through the subcorotat- spanning nearly half ing magnetospheric plasma (Jiaet al. 2012, a Saturn year. Vernal Hunt et al. 2014). The associated current equinox (August 2009) systems are shown in the centre of fig- is marked by the red ure 4, where the FACs generated in one vertical dotted line. hemisphere close partly across the field The centre panel within the magnetosphere, and partly in shows the northern (blue) and southern (red) PPO periods derived from Cassini MAG data. The blue the opposite ionosphere. The currents act dotted line shows the Voyager northern PPO period (see text). The bottom panel shows the north/south to transmit force from the atmospheric flow PPO amplitude ratio k, plotted directly for 0 < k < 1 (southern dominant), and as 1/k in the upper half for to the magnetosphere, causing “cam-like” 1 < k < ∞ (northern dominant). Dotted lines indicate lower limits on k for intervals when the southern PPO rotating displacements of plasma. The could not be discerned (indicating k > 5). effects of these plasma displacements are seen in figure 3 (Burchet al. 2009), and give perturbations produced by the rotating on equatorial orbits correspond to super­ rise to PPO coupling between the northern currents is shown by the blue lines and position of the two perturbation field and southern hemispheres (Hunt et al. 2015, symbols in figure 4, both systems produc- patterns, leading to beat effects (with time- Provan et al. 2016). Radio enhancements ing a region of quasi-uniform perturbation scale typically tens of days). The northern occur whenever the main rotating region of field in the planet’s equatorial region clos- and southern oscillations are discernible (upward-directed) FAC in a given hemi- ing over the corresponding planetary pole because they produce different polariza- sphere passes over the auroral emission (Provan et al. 2009) (right-hand diagrams). tions in the field components. source regions being observed – when this Observed field oscillations can yield a Such analyses have allowed the deriva- happens, electrons are further accelerated model of the orientation of the current tion of the rotation phase and period of the downward, leading to stronger emission systems over time, and of the two rotation two PPO systems over essentially the whole of both radio waves (Saturn kilometric periods (Andrews et al. 2010b, 2012). Field of the Cassini mission to date (Andrews radiation or SKR, Lamy et al. 2010, 2013) and oscillations observed over either planetary et al. 2012, Provan et al. 2013, 2016). These ultraviolet auroras (Nichols 2010a,b). pole correspond solely to the relevant ephemerides also help organize other The rotating pattern of magnetic hemisphere, while oscillations observed observations, such as auroral emissions

Downloaded4.38 from https://academic.oup.com/astrogeo/article-abstract/58/4/4.36/3988906 A&G • August 2017 • Vol. 58 • aandg.org by Institute of Child Health/University College London user on 10 January 2018 CASSINI–HUYGENS: MAGNETOMETER

determination one Saturn year later, indi- cating some consistency with season.

FACs and the aurora Before the Cassini mission, Cowley et al. (2004a,b) modelled the main plasma flows in Saturn’s magnetosphere and the associated electric current systems in the 6 Figures illustrating the magnetic field perturbations and currents associated with plasma subcorotation magnetically conjugate regions of Sat- in Saturn’s magnetosphere. (a) shows the “lagging” field that results from the magnetosphere– urn’s polar ionosphere. With increasing ionosphere coupling currents viewed looking down on Saturn’s north pole. Arrowed black lines equatorial radial distance from Saturn, the represent the magnetic field lines. (b) and (c) show the currents (green arrowed lines and symbols) first component is a region dominated by and perturbation fields (blue arrowed lines and symbols) in both the northern ionosphere and in a planetary rotation, where the plasma sub- meridian plane, respectively. Circled crosses and dots indicate vectors pointing into and out of the corotates on closed magnetic field lines, as plane of the diagrams, respectively. The arrowed black lines indicate the near-axisymmetric background first described for Jupiter by Hill (1979). Sur- magnetospheric field. (Adapted from figure 1 of Hunt et al. 2014) rounding this region, the second compo- nent is where plasma is lost down the dusk 7 Joint observations tail via the stretching of magnetic field from two consecutive lines followed by plasmoid formation and HST images (A and B) pinch-off, as first described for Jupiter by taken in January 2007. Vasyliunas (1983). The third, outer region The in situ data panels is driven by the solar wind interaction at below indicate a CAPS the dayside and in the dawn tail, forming a energy–time electron modified Dungey cycle for Saturn (Dungey spectrogram (top 1961). These subcorotating plasma flows panel), the azimuthal produce a four-ring pattern of FACs, with magnetic field (MAG) distributed downward current across the component in nT polar cap, a narrow ring of upward cur- (middle panel), and rent at the boundary between open and the HST UV auroral closed magnetic field lines, and regions of brightness in kR distributed downward and upward FAC on (bottom panel). (From closed field lines at lower latitudes, associ- Bunce 2012) ated with angular momentum transfer from ionosphere to magnetosphere. Cowley et al. (2004a,b) proposed that the main auroral emission at Saturn is asso- ciated with the narrow ring of upward current located near the open–closed field line boundary. A simplified large-scale current system that is fundamental to this and boundary motions (e.g. Clarke et al. amplitudes varied strongly on ~100–200- subcorotation process is shown in green in 2010, Badman et al. 2012, Thomsen et al. day intervals, with fluctuating periods. The the right-hand sketch of figure 6. This cur- 2017). Rotation periods can also be esti- origin of this behaviour remains uncertain, rent system flows down field lines (black in mated from analysis of SKR modulations with debates regarding the possible roles figure 6) into Saturn’s ionosphere at higher when separated into northern and southern of the Great White Spot storm in December latitudes, equatorward in the ionosphere components (details in e.g. Gurnett et al. 2010 and new solar activity, following the and upward at lower latitudes, forming 2009). The SKR- and MAG-derived rotation recent prolonged solar cycle minimum an axisymmetric ring around the poles in periods generally show good agreement (Provan et al. 2015). Following the disturbed both hemispheres (shown in the centre of (Provan et al. 2014, 2016). interval, however, the periods coalesced figure 6 for the northern hemisphere). The PPO characteristics derived from Cassini for about ~1 year (mid-2013 to mid-2014), upward-directed currents are positioned magnetic data over the interval 2005–2015 implying interhemispheric coupling, before just equatorward of the boundary between are shown in figure 5, from just after south- separating again with the northern period open and closed field lines, and are directly ern solstice (October 2002) to just before enduring longer than the southern for the related and co-located with the remotely northern solstice (May 2017), via vernal first time in the mission. Seasonal depend- observed auroral and radio emissions. equinox in August 2009. Southern oscil- ence of the PPO periods is thus indicated, These FACs then close in the magneto- lations dominated (k < 1) during the early possibly via varying insolation and iono- sphere through the equatorial plasma as part of the Cassini mission, with southern spheric conductivity, although additional outward radial currents, thereby complet- period longer than northern. Across the factors are implied by the complex post- ing the circuit. The resulting azimuthal

equinox, however, the two amplitudes equinox behaviour. The blue dotted line perturbation field B( ϕ) is shown by the blue became near-equal (k ~ 1), while the two in figure 5 marks the Voyager “discovery” symbols. In addition, superimposed on this periods slowly converged towards the end PPO period, observed in radio emission system are the two PPO current systems, of 2010, leading to the expectation of north- during approach near vernal equinox in whose currents flow at similar latitudes ern system dominance and period reversal 1980, but shifted in time by one Saturn year. and with similar amplitude. The addition of as northern spring progressed. Instead, This value corresponds to the northern the subcorotating and PPO current systems there followed a ~2.5-year interval of system PPO period (right-hand polarized results in a combined FAC pattern which is disturbed behaviour in which the relative SKR), and agrees well with the Cassini strongly modulated according to the phase

DownloadedA& Gfrom • August https://academic.oup.com/astrogeo/article-abstract/58/4/4.36/3988906 2017 • Vol. 58 • aandg.org 4.39 by Institute of Child Health/University College London user on 10 January 2018 CASSINI–HUYGENS: MAGNETOMETER

of the PPO field perturbations (see above). and position were shown to be strongly location is controlled by total pressure Depending on these PPO phases, vastly dif- modulated by the southern PPO phase. The balance, and is typically ~25 Saturn radii

ferent FAC current signatures are observed northern hemisphere data set for this time (RS) sunward of the planet along the in the Cassini magnetometer data (Hunt et interval was more complex. Here, the FAC Saturn–Sun line (Arridge et al. 2006, Achil- al. 2014, 2015, 2016). morphology and strength were modulated leos et al. 2008, Pilkington et al. 2014, 2015a). Orbits in 2006 and 2007 provided the first not only by the northern PPO system, but Saturn has no clear “cushion region” in the opportunity for Cassini to directly sample also by the southern PPO system (Hunt et outermost magnetosphere, as observed the large-scale FAC current systems at al. 2015). This gave the first direct evidence at Jupiter (Went et al. 2011b); however, a Saturn associated with magneto­sphere– for the interhemispheric closure of the similarity with Jupiter’s magnetosphere ionosphere coupling in Saturn’s dayside southern PPO system into the northern lies in the radial distension of the planetary magnetosphere. Near-simultaneous Hub- ionosphere (Southwood & Kivelson 2007, magnetic field lines to form a “magneto- ble Space Telescope (HST) observations 2009). There is a direct link between disc” (Arridge et al. 2007, 2008a, Bunce et showed Saturn’s auroral oval and strong regions of upward FAC and downgoing al. 2007, Achilleos et al. 2010a,b). This arises upward FAC were co-located close to the electrons and PPO modulations have been from the appreciable amount of magneto- boundary between open and closed field observed in Saturn’s infrared and ultra­ spheric plasma that is forced to rotate with lines (Bunce et al. 2008), and were in good violet auroral intensity, power and location the planetary magnetic field, one result of agreement with theoretical discussions of (Nichols et al. 2008, 2010a,b, Badman et al. which is a radially outward centrifugal Cowley et al. (2004a,b). Figure 7 summa- 2012, Bunce et al. 2014). force that contributes to an equilibrium rizes these conjugate observations. The top During the high-latitude passes, we have magnetodisc field structure. two panels show UV images from HST of also seen unusual FAC events associated Saturn’s obliquity and near-aligned Saturn’s southern auroral oval (A and B). with solar-wind-driven global magneto­ rotation and magnetic dipole axes produce The white tracks show the magnetically spheric dynamics. For example, during a seasonal effect where the outer dayside mapped footprint of Cassini. The red dots revolution 89 in 2008, an magnetodisc is pushed show the position of Cassini at the time of extraordinarily strong “The influence of northward, because of the the exposure. The panels below show the upward-directed FAC Saturn’s intrinsic proximity of the subsolar in situ electron energy spectrogram, the signature was observed, magnetic field extends magnetopause, under south-

azimuthal field component Bϕ, and the UV accompanied by hot plasma far out into space” ern summer conditions (and intensity from the image A (red) and image signatures at very high lati- probably vice versa) (Arridge B (blue). The red vertical lines show the HST tude in the polar cap. It was suggested that et al. 2008b). On the nightside, the magnetic image-adjusted times. The green arrows this event was due to an episode of rapid field lines of the outer magnetosphere show the region of upward FAC as identi- tail reconnection, triggered by a sudden extend to form a long magnetotail (Arridge

fied from Bϕ, as also shown on the HST solar-wind shock compression of the mag- et al. 2008b, Jackman & Arridge 2011b). images. Bunce et al. (2008) showed that this netosphere (Cowley et al. 2005, Bunce et al. The structure of Saturn’s inner magneto- upward current region was adjacent to the 2010). A further example was seen in 2013, sphere shows less deviation from the pure open field region, identified from the lack additionally revealing unusual auroral planetary field. However, the magnetom- of electrons. It was also shown that these morphology (Badman et al. 2016). eter discovery of the moon Enceladus as a currents could account for the observed strong plasma source explains much of the UV intensity, given the acceleration of Saturn’s magnetosphere configuration of Saturn’s global magneto- downgoing electrons (Cowley et al. 2008). Saturn’s magnetosphere is one of the larg- sphere as a consequence of significant inter- This interval allowed examination of the est in the solar system. Cassini magnetic nal plasma mass loading (Dougherty et al. auroral region on the dayside of the planet. field observations have allowed charac- 2006, Kivelson 2006), as shown in figure 8. A second interval of high-latitude explora- terization of the highly variable solar wind Cassini magnetic field observations tion in 2008 consisted of 40 passes through at Saturn orbit (e.g. Jackman & Arridge have also shown how dynamic Saturn’s the auroral FACs near midnight. They 2011a). The typical solar magnetic field at magnetosphere is. The solar wind dramati- revealed two clear morphological states: Saturn is consistent with the “Parker spiral” cally compresses and expands the system type 1, a mainly “lagging” azimuthal field prediction (Jackman et al. 2008a), and the (Arridge et al. 2006, Achilleos et al. 2008), configuration; and type 2, with a strong solar wind generally comprises alternat- capable of producing an auroral response “leading” azimuthal field (leading relative ing regions of compression and rarefaction (Badman et al. 2005, 2008), and with addi- to the ionospheric footprint of the field line) (Jackman et al. 2004). The shock wave in tional control of the magnetopause position (Talboys et al. 2011), although the origin of the solar wind in front of the magneto- related to the internal state of the magneto- these states was not then understood. sphere and the downstream magneto­ sphere (Pilkington et al. 2015b, Sorba et al. Recently, with new knowledge of the sheath region have also been investigated 2017) and to the influence of the PPOs. Mag- PPOs, the 2008 data set was re-examined. (Bertucci et al. 2007, Achilleos et al. 2006, netic reconnection at the magnetopause The FAC signatures were organized Went et al. 2011a, Masters et al. 2011, 2013, drives a cycle equivalent to the terrestrial according to the phase of the rotating PPO Sulaiman et al. 2014, 2015, 2016), confirm- Dungey cycle (Cowley et al. 2004a, Jackman systems, as determined by Andrews et al. ing that the shock has one of the highest et al. 2005, Badman et al. 2005, Masters et al. (2012). By exploiting the anti-symmetry Mach numbers in the solar system. Much 2014, Masters 2015). Even the dayside mag- of the subcorotation (PPO-independent) work has focused on spacecraft crossings netodisc is sensitive to the solar wind, since current system shown in figure 5, and the of the magnetopause boundary of Saturn’s compression of the magnetopause toward PPO current systems shown in figure 3, the magnetosphere, revealing mass and energy the planet can effectively “dipolarize” the two systems were separated and shown to transport processes (Masters et al. 2009, field (Leisner et al. 2007, Arridge et al. 2008a). be comparable in amplitude and approxi- 2010, 2012, Cutler et al. 2011, Lai et al. 2012, Magnetic signatures of FACs have mately co-located just equatorward of the Badman et al. 2013). revealed how the magnetosphere and open–closed field line boundary (Hunt Thanks to Cassini, the typical internal planetary upper atmosphere (ionosphere) et al. 2014, Jinks et al. 2014). The southern configuration of Saturn’s magnetosphere is are coupled, central to a global understand- hemisphere FAC morphology, strength now well understood. The magnetopause ing of the physics of the system (Bunce et

Downloaded4.40 from https://academic.oup.com/astrogeo/article-abstract/58/4/4.36/3988906 A&G • August 2017 • Vol. 58 • aandg.org by Institute of Child Health/University College London user on 10 January 2018 CASSINI–HUYGENS: MAGNETOMETER

8 Schematic of Saturn’s boundary moved out over Cassini; Titan magnetosphere. (Taken too had been in the magnetosphere, for at from Kivelson 2006) most three hours. After another ~15 min- utes in the magnetosheath (the magneto­ pause having now moved back inwards past the spacecraft), Cassini reached its closest approach (CA) to Titan, which had again been exposed to magnetosheath plasma, for at least ~15 minutes. At CA, the magnetic field, remarkably, had a negative Z component (i. e. point- ing southward). By contrast, the ambient magnetosheath field surrounding the CA

interval had positive BZ. The “negative BZ” near CA is consistent with the draped field that would have been seen had Titan been continually immersed in Saturn’s al. 2008, Talboys et al. 2009a,b, Badman et al. close to what we believe is the true rotation magnetosphere. But at CA, both Titan 2012, Hunt et al. 2015, 2016). Recently, peri- period of the planet – around 10.7 hours. and Cassini were in the magnetosheath. odic magnetic field variability, not directly This flapping does not come from any tilt Titan had been there for at least 15 minutes related to planetary rotation nor to PPOs, in Saturn’s magnetic equator (Saturn’s after spending up to three hours in the has been reported (Yates et al. 2016), and is internal field is almost perfectly aligned magnetosphere. Hence, the field that was a first step in exploring Saturn’s magneto- with the planet’s rotational axis). Rather, imprinted on Titan’s ionosphere during its spheric pulsations. Magnetic signatures of it arises from a rotating wave-like pattern magnetospheric excursion survived there explosive energy release within the mag- imposed on the sheet by rotating systems for at least ~15 minutes. If we assume that netotail from magnetic reconnection and of current in the magnetosphere, flowing the long (10 hour) magnetosheath excursion related plasma transport have been widely on field lines extending to ~10–15 Saturn well before CA completely removed any reported and investigated, including their radii. As the plasma sheet moves, the imprint of Saturn’s magnetospheric field relationship to auroral emissions (Bunce upstream field changes, being dominantly from Titan, then the later magnetospheric et al. 2005, Jackman et al. 2007, 2008b, 2010, north–south when Titan is near the centre imprint must have then replaced this 2011, 2014, 2015, Lai et al. 2016, Smith et al. of the plasma sheet, and dominantly radial pure magnetosheath imprint during the 2016). Remaining open issues concerning (towards or away from Saturn) when Titan subsequent ~3 hours that the moon spent in Saturn’s magnetosphere include the mass is just outside the sheet. These changes the magnetosphere. Hence, this time range budget of the system, and how internal and were characterized by Bertucci et al. (2009), (~15 minutes – ~3 hours) constrains the external driving of the system combine to who surveyed Cassini magnetometer imprint or fossil field lifetime at Titan, rais- produce the magnetospheric dynamics data during spacecraft fly-bys of Titan. ing the intriguing prospect of a “magnetic revealed by Cassini. Later, Achilleos et al. (2014) used a model archaeology” where close fly-bys of Titan of the plasmasheet (magnetodisc) to study could potentially reveal details of ambient Titan and its environment one fly-by in detail. They found that the fields to which Titan has been exposed up Saturn’s largest moon, Titan, is unusual, magnetospheric flux tubes that flow closest to about three hours in the past. with a dense smoggy atmosphere and to Titan may carry with them the imprint hydrocarbon lakes described in this issue of a very different kind of upstream field The beginning of the end by Coates (p4.20) and Zarnecki (p4.31). compared to the imprint carried by plasma After 13 years in orbit around Saturn, the Titan’s orbit at ~20 Saturn radii places it, in the far-Titan space. This is because the Cassini mission has entered its final phase. most of the time, within Saturn’s magneto- upstream field is continually changing. The first stage of the Grand Finale involved sphere, immersed in rapidly rotating, mag- This “change in magnetic imprint” could a ring-grazing orbital phase in which the netized plasma that moves past the moon become more pronounced if the boundary periapse distance was just outside of the at ~100 km s–1, far faster than Titan’s orbital of Saturn’s magnetosphere moves inward edge of the visible rings, enabling a focus speed of ~6 km s–1. As the plasma flows past or outward past Titan. When this happens on ring science as well as measurement of and around Titan, the magnetic lines that (albeit relatively rarely), Titan transitions the FACs and the PPOs. Analysis of data are “frozen” to the flowing plasma drape between a magnetospheric and a magneto­ from this phase is critical to the success of around the moon, forming downstream sheath/solar-wind regime, a process first the Grand Finale MAG science, because lobes of strongly azimuthal field generally discovered by Bertucci et al. (2008) in the an understanding of both FAC and PPO pointing towards Titan in one lobe, and T32 Titan fly-by of Cassini. behavior is vital in order to resolve the away in the other. Bertucci et al. (2008) noted three phases. magnetic signatures from Saturn’s interior. The magnetic field configuration associ- Before the Titan encounter, Cassini The 22 Grand Finale orbits, with periapse ated with this Titan–Saturn interaction observed a magnetic field (and plasma distance inside the rings, began on 28 April depends on the conditions in Saturn’s rotat- parameters) consistent with a 10-hour 2017, each lasting 6.4 days. These orbits took ing magnetospheric plasma upstream of excursion into the magneto­sheath region three years of planning in order to ensure Titan. Before Cassini, the common percep- of Saturn, i.e. shocked solar wind plasma. the best possible science return from all the tion was that the upstream field orientation Cassini was inside Titan’s orbit during instruments, complicated by the fact that would be north–south, the equatorial direc- this interval, so it follows that Titan was neither the spacecraft nor the instruments tion of Saturn’s dipole field. Cassini has immersed in magnetosheath plasma for at were designed to carry out such manoeu- shown a different picture. The magneto- least 10 hours. Then the spacecraft encoun- vres. MAG is prime on two of the 22 orbits spheric, disc-like plasma sheet continually tered magnetospheric plasma for approxi- in which the spacecraft is rolling around flaps up and down past Titan with a period mately three hours as the magnetopause periapse, in order to enable us to calibrate

DownloadedA& Gfrom • August https://academic.oup.com/astrogeo/article-abstract/58/4/4.36/3988906 2017 • Vol. 58 • aandg.org 4.41 by Institute of Child Health/University College London user on 10 January 2018 CASSINI–HUYGENS: MAGNETOMETER

the instrument (around closest approach 9 Schematic of the we are in a measuring range not active end of mission phase since Earth swingby in 1999). On two more with the F ring phase in orbits, we are riding along with the gravity white, the grand finale experiment (Radio Sub System) and rolling phase in blue and the around the z-axis of the spacecraft, again final impacting orbit in for calibration. orange. Cassini MAG measurements to date have greatly enhanced our understand- ing of Saturn’s intrinsic magnetic field, but have raised many new questions as well. Saturn’s intrinsic magnetic field has always been regarded as somewhat anomalous since the very first measurements made by the Pioneer 11 magnetometer (Smith et al. 1980). The first surprise concerned the relative weakness of the field: Saturn’s surface magnetic field is only ~20 000 nT axisymmetric offset dipole, although the understand the intrinsic planetary field in near the equator, which is ~20 times weaker spacecraft did not spend long enough close its proper context. than Jupiter’s surface magnetic field, and enough to the planet to reveal any longitu- The Cassini end-of-mission orbits will even weaker than Earth’s surface magnetic dinal structure of that asymmetry. enable paradigm-shifting discoveries about field. This apparent weakness of the field is The primary MAG objective of the Grand Saturn’s intrinsic magnetic field and interior particularly surprising, given Finale orbits (figure 9) is to due to the close proximity to Saturn, as well that Saturn’s emitted power “One surprising feature determine any asymmetric as their highly inclined nature. With high- is the second highest in our is the close alignment component of the intrinsic quality MAG measurements along Cassini solar system after Jupiter. of Saturn’s magnetic planetary field. With an accu- proximal orbits, we will be able to address The second surprising axis with its spin axis” rate determination, Saturn’s several key questions concerning Saturn’s feature is the close align- rotation rate could finally be interior: the rotation rate of Saturn at depth, ment of Saturn’s magnetic axis with its spin inferred directly. This is of great impor- the degree of asymmetry of the intrinsic axis. Pioneer 11 measurements were able tance for construction of accurate interior field, and the depth to the dynamo region. to place an upper limit on a dipole tilt of models. It also sets the reference state With accurate field models based on meas- 1°, to describe the possible departure from against which the differential rotation of urements made during the proximal orbits, axisymmetry (Smith et al. 1980). Such a the atmosphere and its latitudinal depend- we will be able to compare characteristics close alignment seems to be in contradic- ence is defined, important for understand- of the large-scale structure of Saturn’s field tion to Cowling’s well-known anti-dynamo ing the angular momentum budget of with other planetary dynamos and dynamo theorem, which states that a purely the atmospheric circulation. Continuous simulations. This improved knowledge will axisymmetric magnetic field cannot be monitoring of the external magneto­spheric greatly enhance our understanding of the self-sustained. Only Saturn orbit insertion field and accurate tracking of the PPOs present state of Saturn’s interior, as well as

data have revealed any departure from an is required in order to determine and its formation and evolution. ●

AUTHORS Bunce E J et al. 2006 Adv. Space Res. 38 806 116 A05224 Pilkington N M et al. 2015a J. Geophys. Res. Space The Cassini magnetometer team are: Michele Bunce E et al. 2007 J. Geophys. Res. 112 A10202 Jackman C M et al. 2004 J. Geophys. Res. 109 A11203 Physics 120 7289 K Dougherty, Blackett Lab., Physics Dept, Imperial Bunce E J et al. 2008 J. Geophys. Res. 113 A09209 Jackman C M et al. 2005 J. Geophys. Res. 110 A10212 Pilkington N M et al. 2015b Geophys. Res. Lett. 42 6890 College London; Nicholas Achilleos, Physics and Bunce E J et al. 2010 J. Geophys. Res. 115 A20238 Jackman C M et al. 2007 Geophys. Res. Lett. 34 L11203 Provan G et al. 2009 J. Geophys. Res. 114 A02225 Astronomy, University College London; Emma Bunce, Bunce E J et al. 2014 J. Geophys. Res 119 3528 Jackman C M et al. 2008a J. Geophys. Res. 113 A08114 Provan G et al. 2013 J. Geophys. Res. 118 3243 Physics Dept, University of Leicester; Stanley Cowley, Burch J L et al. 2009 Geophys. Res. Lett. 37 L14203 Jackman C M et al. 2008b J. Geophys. Res. 113 A11213 Provan G et al. 2014 J. Geophys. Res. 119 7380 Leicester; Gregory Hunt, ICL; Adam Masters, ICL; and Burton M E et al. 2010 Geophys. Res. Lett. 37 L24105 Jackman C M et al. 2010 J. Geophys. Res. 115 A10240 Provan G et al. 2015 J. Geophys. Res. 120 9524 Gabby Provan, Leicester. Carbary J F & Krimigis S M 1982 Geophys. Res. Lett. Jackman C M et al. 2011 J. Geophys. Res. 116 A10212 Provan G et al. 2016 J. Geophys. Res. 121 9829 9 1073 Jackman C M et al. 2014 J. Geophys. Res. Space Smith E et al. 1980 Science 207 407 references Clarke K E et al. 2010 J. Geophys. Res. 115 A08209 Physics 119 Smith A W et al. 2016 J. Geophys. Res. Space Physics Achilleos N et al. 2006 J. Geophys. Res. 111 A03201 Cowley S W H et al. 2004a Ann. Geophys. 22 1379 Jackman C M et al. 2015 J. Geophys. Res. Space Physics 121 2984 Achilleos N et al. 2008 J. Geophys. Res. 113 A11209 Cowley S W H et al. 2004b J. Geophys. Res. 109 A05212 120 3603 Sorba A M et al. 2017 J. Geophys. Res. Space Physics Achilleos N et al. 2010a Mon. Not. R. Astron. Soc. Cowley S W H et al. 2005 J. Geophys. Res. 110 A02201 Jia X et al. 2012 J. Geophys. Res. 117 A04215 122 1572 4014 2349 Cowley S W H et al. 2008 Ann. Geophys. 26 2613 Jinks S L et al. 2014 J. Geophys. Res 119 8161 Southwood D J & Cowley S W H 2014 J. Geophys. Achilleos N et al. 2010b Geophys. Res. Lett. 37 L20201 Cutler J C et al. 2011 J. Geophys. Res. 116 A10220 Kivelson M G 2006 Science 311 1391 Res. 119 1563 Achilleos N et al. 2014 Geophys. Res. Lett. 41 8730 Desch M D & Kaiser M L 1981 Geophys. Res. Lett. 8 253 Lai H et al. 2012 J. Geophys. Res. 117 A05222 Southwood D J & Kivelson M G 2007 J. Geophys. Arridge C S et al. 2006 J. Geophys. Res. 111 A11227 Dougherty M K et al. 2006 Science 311 1406 Lai H R et al. 2016 J. Geophys. Res. Space Physics 121 Res. 112 A12222 Arridge C S et al. 2007 Geophys. Res. Lett. 34 L09108 Dougherty M K et al. 2017 Enceladus and the Icy 3050 Southwood D J & Kivelson M G 2009 J. Geophys. Arridge C S et al. 2008a J. Geophys. Res. 113 A04214 Moons of Saturn in press (University of Arizona) Lamy L et al. 2010 Geophys. Res. Lett. 37 L12104 Res. 114 A09201 Arridge C S et al. 2008b J. Geophys. Res. 113 A08217 Dungey J W 1961 Phys. Rev. Lett. 6 47 Lamy L et al. 2013 J. Geophys. Res. 118 4817 Sulaiman A H et al. 2014 J. Geophys. Res. Space Andrews D J et al. 2010a J. Geophys. Res. 113 A04212 Espinosa S A & Dougherty M K 2000 Geophys. Res. Leisner J S et al. 2007 Geophys. Res. Lett. 34 L12103 Physics 119 5651 Andrews D J et al. 2010b J. Geophys. Res. 115 A12252 Lett. 27 2785 Masters A et al. 2009 Planet. Space Sci. 57 1769 Sulaiman A H et al. 2015 Phys. Rev. Lett. 115 12 Andrews D J et al. 2012 J. Geophys. Res. 117 A04224 Espinosa S A et al. 2003 J. Geophys. Res. 108 A21085 Masters A et al. 2010 J. Geophys. Res. 115 A02725 Sulaiman A H et al. 2016 J. Geophys. Res. Space Badman S V et al. 2005 J. Geophys. Res. 110 A11216 Galopeau P H M & Lecacheux A 2000 J. Geophys. Masters A et al. 2011 J. Geophys. Res. 116 A10107 Physics 121 4425 Badman S V et al. 2008 Ann. Geophys. 2612 3641 Res. 105 13089 Masters A et al. 2012 Geophys. Res. Lett. 39 L08103 Talboys D L et al. 2009a J. Geophys. Res. 114 A06220 Badman S V et al. 2012 J. Geophys. Res. 117 A09228 Gérard J-C et al. 2006 J. Geophys. Res. 111 A12210 Masters A et al. 2013 Nat. Phys. 9 164 Talboys D L et al. 2009b Geophys. Res. Lett. 36 L19107 Badman S V et al. 2013 Geophys. Res. Lett. 40 1027 Gurnett D A et al. 2009 Geophys. Res. Lett. 36 L16102 Masters A et al. 2014 Geophys. Res. Lett. 41 1862 Talboys D L et al. 2011 J. Geophys. Res. 116 A04213 Badman S V et al. 2016 Icarus 263 83 Hill T W 1979 J. Geophys. Res. 84 6554 Masters A 2015 Geophys. Res. Lett. 42 2577 Thomsen M F et al. 2017 J. Geophys. Res. 122 280 Bertucci C et al. 2007 J. Geophys. Res. 112 A09219 Hunt G J et al. 2014 J. Geophys. Res. 119 9847 Nichols J D et al. 2008 J. Geophys. Res. 113 A11205 Vasyliunas V M 1983 Physics of the Jovian Magneto- Bertucci C et al. 2008 Science 321 1475 Hunt G J et al. 2015 J. Geophys. Res. 120 7552 Nichols J D et al. 2010a Geophys. Res. Lett. 37 L15102 sphere (CUP, Cambridge and New York) 395 Bertucci C et al. 2009 Planet. Space Sci. 57 1813 Hunt G J et al. 2016 J. Geophys. Res. 121 7785 Nichols J D et al. 2010b Geophys. Res. Lett. 37 L24102 Went D R et al. 2011a J. Geophys. Res. 116 A07202 Bunce E J 2012 Geophys. Monograph Series 195 397 Jackman C M & Arridge C S 2011a Sol. Phys. 274 481 Pilkington N M et al. 2014 J. Geophys. Res. Space Went D R et al. 2011b J. Geophys. Res. 116 A04224 Bunce E J et al. 2005 Geophys. Res. Lett. 32 L20S04 Jackman C M & Arridge C S 2011b J. Geophys. Res. Physics 119 2858 Yates J N et al. 2016 Geophys. Res. Lett. 43 11102

Downloaded4.42 from https://academic.oup.com/astrogeo/article-abstract/58/4/4.36/3988906 A&G • August 2017 • Vol. 58 • aandg.org by Institute of Child Health/University College London user on 10 January 2018