Magnetic reconnection is an important process that occurs at the magnetopause boundary of Earth's magnetosphere because it leads to transport of solar wind energy into the system, driving magnetospheric dynamics. However, the nature of magnetopause reconnection in the case of Saturn's magnetosphere is unclear. Based on a combination of Cassini spacecraft observations and simulations we propose that plasma βconditions adjacent to Saturn's magnetopause largely restrict reconnection to regions of the boundary where the adjacent magnetic fields are close to anti-parallel, severely limiting the fraction of the magnetopause surface that can become open. Under relatively low magnetosheathβconditions we suggest that this restriction becomes less severe. Our results imply that the nature of solar wind-magnetosphere coupling via reconnection can vary between planets, and we should not assume that the nature of this coupling is always Earth-like. Studies of reconnection signatures at Saturn's magnetopause will test this hypothesis.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
 The interaction between the flow of solar wind plasma from the Sun and a magnetized planet leads to a cavity surrounding the planet known as a planetary magnetosphere. The solar wind is largely excluded from such cavities; however, processes that take place at the boundary of a magnetosphere (the magnetopause) can lead to the transport of solar wind energy into the system. One of these processes is magnetic reconnection, which changes the magnetic field topology and converts energy stored in the magnetic field into particle kinetic energy [Dungey, 1961; Vasyliunas, 1975; Russell, 1976].
 The occurrence of reconnection at Earth's magnetopause is the main driver of dynamics in the terrestrial magnetosphere [Dungey, 1961; Russell, 1972] (see the review by Paschmann [2008, and references therein]). Evidence in the form of accelerated plasma flows (reconnection jets) has been detected at approximately half the crossings of Earth's magnetopause made by the Double Star TC1 spacecraft [Trenchi et al., 2008], and, although high magnetic shear conditions are more favorable, reconnection can occur when the magnetic shear across the boundary is as low as 90° [Pu et al., 2007; Trattner et al., 2007; Trenchi et al., 2008]. Conditions are more favorable when the plasma βin the near-magnetopause solar wind (the magnetosheath) is less than approximately 2 (plasmaβ is the ratio of plasma to magnetic field pressure) [Paschmann et al., 1986; Trenchi et al., 2008].
 Although the Cassini spacecraft's orbital tour of Saturn allows us to study the Saturnian magnetosphere in great detail, the nature of magnetopause reconnection at Saturn remains unclear. Cassini magnetopause crossings with signatures suggestive of reconnection have been reported [McAndrews et al., 2008], and Saturn's main auroral emission has been proposed to lie at the boundary between open and closed field lines [e.g., Cowley et al., 2005]. However, spacecraft attitude and a limited field of view severely limits the ability of Cassini plasma analyzers to detect unambiguous evidence for magnetopause reconnection in the form of reconnection jets, and no such evidence has been reported yet. Furthermore, no examples of the reconnection phenomenon of Flux Transfer Events (FTEs) at Saturn have been identified to date (FTEs often form at Earth's magnetopause [Russell and Elphic, 1978]), and neither Saturn's low-latitude boundary layer nor Saturn's auroral power show an Earth-like response to the orientation of the Interplanetary Magnetic Field (IMF) [Crary et al., 2005; Clarke et al., 2009; Masters et al., 2011].
 To address the subject of magnetopause reconnection at Saturn we can assess what the magnetized plasma conditions adjacent to Saturn's magnetopause current layer imply for reconnection onset. Theory suggests that a low value of the plasma β on either side of a current layer promotes reconnection onset, as does a low value of the absolute difference in plasma β across the layer (|Δβ|) [Quest and Coroniti, 1981; Swisdak et al., 2003, 2010]. The relative importance of these effects for producing the β-dependence of magnetopause reconnection at Earth is unclear. Observations of reconnecting solar wind current sheets [Phan et al., 2010] provide strong evidence for the |Δβ| effect known as diamagnetic suppression, which has been introduced based on simulations and theory [Swisdak et al., 2003, 2010]. The high fast magnetosonic Mach number of Saturn's bow shock compared to Earth's should produce a higher plasma β in the Saturnian magnetosheath, leading to less favorable conditions for magnetopause reconnection [Scurry and Russell, 1991; Mauk et al., 2009].
 In this paper we use data taken by the Cassini spacecraft to determine the magnetized plasma conditions at Saturn's magnetopause. With the support of a Particle-In-Cell (PIC) simulation of magnetic reconnection at a current sheet where adjacent conditions are typical of those at Saturn's magnetopause (based on the Cassini observations), we propose that magnetopause reconnection at Saturn is largely restricted to anti-parallel magnetic field geometries under nominalβ conditions.
2. Measuring Magnetized Plasma Conditions at Saturn's Magnetopause
Figure 1a shows the positions of 520 crossings of Saturn's magnetopause made by the Cassini spacecraft between June 2004 and August 2007, in the xy plane of the Cartesian Kronocentric Solar Magnetospheric (KSM) coordinate system (approximately the equatorial plane). The unit of distance used is Saturn radii (RS; 1 RS = 60,268 km). These crossings predominantly took place between magnetic latitudes of ±20° [Masters et al., 2011].
Figures 1b and 1cshow data taken by two Cassini instruments during a magnetopause crossing on 15 May 2007: The dual-technique magnetometer (MAG) [Dougherty et al., 2004], and anode 5 of the electron spectrometer (ELS) [Young et al., 2004]. Based on the measured magnetic field and ambient electron distributions, the spacecraft made a transition from the magnetosphere (higher magnetic field strength and hotter, more tenuous electron population) to the magnetosheath (lower magnetic field strength and colder, denser electron population). A planetary magnetopause is a current layer; Saturn's magnetopause current layer (MPCL) is evident in Figure 1b as the clear change in magnetic field orientation at ∼01:58.
 To determine the plasma β we require the magnetic field pressure and the pressure exerted by each plasma population. MAG data provides the magnetic field pressure, and moments derived from ELS data provide the thermal electron pressure [Lewis et al., 2008]. The Cassini magnetospheric imaging instrument (MIMI) [Krimigis et al., 2004] provides the energetic charged particle pressure (>∼10 keV electrons and ions), and moments derived from Cassini ion mass spectrometer (IMS) data provide the thermal ion pressures (protons: H+, species with mass-per-charge 2: H2+/He++, water group ion species with mass-per-charge between 16 and 19: W+) [Young et al., 2004; Thomsen et al., 2010]. Magnetic field-parallel and magnetic field-perpendicular temperatures are not calculated for these populations due to restricted pitch angle coverage. The combination of all these pressures givesβ ∼ 1 in the magnetosphere and β ∼ 9 in the magnetosheath for this particular crossing (Figure 1d).
 Mean pressures from MAG, ELS, MIMI, and IMS were determined in intervals of 1 minute, 5 minutes, 10 minutes, and 15 minutes immediately on either side of each MPCL transition, respectively (data cadences used: 1 second – MAG; moments based on 32 second-averaged distributions – ELS; 5 minute-averaged pressures – MIMI; and irregular cadence moments – IMS). At 387 of the 520 crossings the MPCL is unambiguous, and MAG, ELS, and MIMI-derived pressures are available, defining a partial plasmaβ without thermal ion pressures. Pointing constraints generally prohibit the derivation of reliable thermal ion moments [Thomsen et al., 2010] (see auxiliary material), leading to a full plasma β that includes thermal H+ and H2+/He++ pressures on both sides of the MPCL for 70 of these 387 crossings.
 Reliable thermal W+ moments in the magnetosphere are only available at 15 of the 70 crossings, and at 10 of these also in the magnetosheath (likely due to finite gyroradius leakage through the MPCL). The paucity of reliable W+ moments may be due to W+ densities below the IMS detection threshold in the vicinity of the magnetopause, or the limited energy range of IMS [Thomsen et al., 2010]. When measured, the thermal W+ pressure was ∼10% of the total plasma pressure in the magnetosphere and ∼2% of the total plasma pressure in the magnetosheath.
 We note that the lack of simultaneous observations of conditions on either side of Saturn's MPCL may affect our results. Furthermore, mirror mode waves in Saturn's magnetosheath can produce large variations in the local plasma β [e.g. Violante et al., 1995]. 1-second cadence magnetic field data taken during the 15-minute magnetosheath intervals used in this study define a mean field strength perturbation (δB/B) of 0.36, confirming that mirror mode waves can strongly influence magnetosheath β conditions. However, we argue that the number of crossings used in this study account for these temporal variability issues, revealing the prevailing β conditions (see error analysis in auxiliary material).
3. Implications of Plasma β Conditions for Magnetopause Reconnection
 Histograms of plasma β measured in the magnetosheath and magnetosphere adjacent to Saturn's magnetopause are shown in Figures 2a and 2b. Figure 2a includes all 70 crossings, whereas Figure 2b only includes crossings with magnetospheric W+ pressures. These measurements reveal a typical plasma β in Saturn's magnetosheath of ∼10, with extreme values of order 1 and of order 100. In Saturn's magnetosphere the plasma β is typically ∼2, ranging between extreme values of ∼0.3 and of order 10. These ranges do not appear to be sensitive to the inclusion of thermal W+ pressures. For 93% of the crossings β was higher in the magnetosheath than in the magnetosphere.
 The plasma β in Earth's magnetosheath immediately adjacent to the terrestrial magnetopause is typically ∼1, with extreme values of order 0.1 and of order 10 [Trenchi et al., 2008]. Our results confirm that Saturn's magnetosheath is a higher plasma β environment than Earth's magnetosheath. Reconnection at Earth's magnetopause is more likely to occur when the magnetosheath plasma β is below ∼2 [Paschmann et al., 1986; Trenchi et al., 2008], suggesting that reconnection at Saturn's magnetopause is most likely to occur when the Saturnian magnetosheath plasma β is relatively low (see Figure 2a).
 The theory of diamagnetic suppression of reconnection suggests that a higher |Δβ| across the current layer is less favorable for reconnection [Swisdak et al., 2003, 2010]. The principle underlying diamagnetic suppression is that the drift of charged particles within a current sheet can disrupt the reconnection jets, suppressing reconnection when this disruption is sufficiently large. When the reconnecting fields are perfectly anti-parallel the drift with respect to the X-line is perpendicular to the reconnection jets (outflows); however, when the fields are not anti-parallel the drift has a non-zero component along the outflow direction, promoting outflow on one side of the X-line and opposing it on the other (Figures 3b and 3c). Reconnection is suppressed when this component of the drift is greater than the speed of the outflows, and the following condition is satisfied:
where L is the width of the density gradient layer across the current layer, di is the ion inertial length, and θ is the magnetic shear across the current layer. Note that this is the general diamagnetic suppression condition, introduced by Swisdak et al.  and tested by Phan et al. .
Figure 2c shows the measured conditions at Saturn's magnetopause in |Δβ|-magnetic shear parameter space. Magnetic shears are based on average fields in 1-minute intervals either side of the MPCL. This parameter space is roughly separated into a region where the diamagnetic suppression condition (given byequation (1)) is satisfied (reconnection suppressed) and a region where it is not satisfied (reconnection possible). Saturn's low-latitude magnetopause generally lies in the region where reconnection is suppressed. The issue of W+ pressure inclusion does not appear to strongly affect the range of |Δβ| covered by the data points, and neither do estimates of the measurement uncertainties (see auxiliary material). Comparing to Earth's magnetopause, if we assume a magnetospheric plasma β equal to 0, the terrestrial boundary lies in the |Δβ| regime of ∼0.1 to ∼10, where reconnection is possible for a larger range of magnetic shears.
 Low plasma β and |Δβ| conditions appear to be a necessary, but not sufficient, requirement for reconnection to occur [Phan et al., 2011]. The Cassini magnetopause crossing with evidence for reconnection reported by McAndrews et al.  was included in this study. Although this crossing is not associated with a full plasma β (and so is not shown in Figure 2c) it is associated with a partial plasma β of 0.2 ± 0.1. Since the typical partial magnetosheath plasma β is ∼5, and full and partial β are well correlated (see auxiliary material), it is very likely that this magnetopause crossing corresponded to low-β conditions in Saturn's magnetosheath.
 To support these findings we simulated magnetopause reconnection in two dimensions using a PIC code [Swisdak et al., 2003] (see auxiliary material). Two runs were carried out: One where β either side of the current layer was equal to 1 (“Case A”, Figure 4a), and one where the β conditions were typical of Saturn's magnetopause (10 and 1, “Case B”, Figure 4b). In both cases an out-of-plane magnetic field produced a magnetic shear across the layer of 120°. Note that inFigures 4a and 4bthe out-of-plane current density is shown rather than the flow field. This is because current density reveals the magnetic structure of the X-line in more detail. We refer the reader toSwisdak et al.  for a detailed discussion and presentation of the flow field in such simulations.
Figure 4ashows that in Case A the structure of the current sheet on either side of the X-line is similar, whereasFigure 4b shows that in Case B the structure is more asymmetric. The rate of increase of total reconnected magnetic flux is higher in Case A than in Case B (see Figure 4c; the reconnection rate is the slope of the plotted curve). The non-zero total reconnected flux at a simulation time of zero is an artifact of the perturbation used to initialize reconnection. The structural asymmetry in Case B suggests that |Δβ|-related diamagnetic suppression plays a role. We have not simulated higherβ conditions due to computational constraints, and we note that Case B lies at the approximate boundary of strong suppression (see Figure 2c). Under conditions clearly in the suppressed regime we expect the reconnection rate to fall to zero.
 We have examined the magnetized plasma conditions at Saturn's low-latitude magnetopause and found that plasmaβconditions should largely restrict reconnection to regions where the adjacent magnetic fields are close to anti-parallel, severely limiting the fraction of the magnetopause surface that can become open. The implications of these results are that conditions are less favorable for reconnection at Saturn's magnetopause than at Earth's, and the magnetosheath plasmaβshould play a greater role in controlling the suitability of near-magnetopause conditions for reconnection onset at Saturn. This study suggests that we should not assume that the interaction between the solar wind and a planetary magnetosphere via magnetopause reconnection is always Earth-like. Comprehensive studies of reconnection signatures at Saturn's magnetopause are required to test this hypothesis.
 We acknowledge the support of the CAPS and MAG data processing/distribution staff, and L. K. Gilbert and G. R. Lewis for Cassini ELS data processing. This work was supported by UK STFC through rolling grants to MSSL/UCL and Imperial College London, and an STFC Advanced Fellowship awarded to JPE. Work at Los Alamos was conducted under the auspices of the U.S. Department of Energy, with support from NASA's Cassini program.
 The Editor thanks Lorenzo Trenchi and an anonymous reviewer for assisting with the evaluation of this paper.