Direct Observations of Reconnection Fronts in Earth's Turbulent Magnetosheath

Magnetic reconnection, a fundamental plasma process during which magnetic field lines get cut and reconnected, has been well documented as the basic engine capable of efficiently converting magnetic field energy into thermal and kinetic energies of charged particles (e.g., Yamada et al. 2010). Reconnection has been thought to be initiated within a local diffusion region, in which particle frozen-in condition is broken and magnetic field lines are allowed to reconnect (e.g., Burch et al. 2016a). Its application to planetary magnetospheric physics leads to the well-known Dungey cycle (Dungey 1961), during which plasmas are accelerated and transported in a global scale. The path, from the local diffusion region to the global plasma convection, is primarily established by outward-propagating reconnection ejecta, particularly reconnection jets and fronts. Reconnection jets, which are usually dubbed as bursty bulk flows (e.g., Baumjohann et al. 1989; Angelopoulos et al. 1992; Cao et al. 2006, 2010), can propagate over a long distance more than tens of Earth Radii (e.g., Runov et al. 2009; Cao et al. 2013), carrying a larger amount of reconnected magnetic flux and accelerated particles toward the inner magnetosphere and thus driving geomagnetic disturbances during auroral substorms (e.g., Cao et al. 2019; Yu et al. 2017; Sitnov et al. 2019).

Reconnection fronts (RFs) are typically characterized by dramatic increase of reconnected magnetic field, and usually serve as leading boundaries of the reconnection jets (e.g., Sitnov et al. 2009). After being generated inside the diffusion region, RFs can propagate as coherent structures during their interactions with ambient plasma (e.g., Runov et al. 2009; Liu et al. 2022a), thus facilitating cross-scale energy transport. The interaction between the RFs and ambient plasma usually leads to intense energy transfer, which has been studied both numerically and experimentally, revealing that energy loads contributed mainly by ion currents typically dominate (e.g., Angelopoulos et al. 2013; Lapenta et al. 2014; Huang et al. 2015; Yao et al. 2017; Liu et al. 2018, 2022b, 2023; Lu et al. 2018; Zhong et al. 2019; Wang et al. 2020; Shu et al. 2021). The RF-driven energy transfer is usually manifested in the form of particle heating and acceleration (e.g., Fu et al. 2020; Birn et al. 2013; Duan et al. 2014; Runov et al. 2015; Gabrielse et al. 2016; Lu et al. 2016; Zhou et al. 2018; Liu et al. 2017) and wave/turbulence generation (e.g., Zhou et al. 2009; Khotyaintsev et al. 2011; Huang et al. 2012; Hwang et al. 2014; Divin et al. 2015; Deng et al. 2010; Liu et al. 2019). The released magnetic field energy is predominately manifested in particle enthalpy and heat fluxes (e.g., Liu et al. 2021).

Albeit the ubiquity of magnetic reconnection in space, RFs have hitherto been observed only in the planetary (e.g., Earth, Saturn, and Venus) magnetotails (e.g., Fu et al. 2020). In the magnetotails, RFs are also termed as dipolarization fronts (DFs) since the increased northward magnetic field at the RFs is reminiscent of dipolarized magnetic field in the inner magnetosphere (e.g., Nakamura et al. 2002; Runov et al. 2009; Schmid et al. 2011). RFs have been well documented to play a crucial role in the tail dynamics (e.g., Liu et al. 2013; Sitnov et al. 2019). Their absence in other magnetospheric regions, such as magnetopause and magnetosheath, raises the question whether RF is a unique structure that exists only in the planetary magnetotails where magnetic reconnection is typically symmetric with weak guide fields. In this study, with the aid of high-cadence measurements from NASA's Magnetospheric Multiscale (MMS) mission (Burch et al. 2016b), we present the first observation of successive RFs formed inside an ion diffusion region in the Earth's turbulent magnetosheath.

The event of interest was observed by MMS on 18 February 2018, from 12:55:00.0 to 12:56:30.0 UT, when MMS, located near [8.1, −12.6, 1.0] RE (Earth Radii), was inside the terrestrial magnetosheath. Separation of the spacecraft tetrahedron is close to 18 km (∼0.3 di, where di is local ion inertial length). High-cadence data from the fluxgate magnetometer (Russel et al. 2016), electric double probe (Ergun et al. 2016; Lindqvist et al. 2016), and fast plasma investigation (Pollock et al. 2016) instruments on board MMS, are used and presented in the geocentric solar ecliptic (GSE) coordinates unless specified otherwise.

We begin with presenting a long-time (from 12:55:00.0 to 12:56:30.0 UT) observation to provide an overview for the event, as shown in Figure 1. During the interval, MMS were cruising in the magnetosheath after several magnetopause crossings (not shown). Fluctuating magnetic fields, as manifested in the sharp changes in all components of B (Figure 1(a)), were observed. Clear field rotations indicated by bipolar changes of B components suggest presence of several current sheets (CSs) encountered by MMS. Plasma density of the compressed solar wind was very variable, ranging from ∼8 to 30 cm⁻³ (Figure 1(b)). In addition, the ion speed exhibits clear variations in all components (Figure 1(c)). Electron and ion energy spectrums also display fine structures in association with the field variations (Figures 1(d) and (e)). These features indicate that magnetic reconnection may develop in this turbulent region.

Zoom In Zoom Out Reset image size

We now present a zoom-in view for one CS crossing (between 12:55:26.0 and 12:55:34.0 UT), as shown in Figures 1(f)–(j). During this period, magnetic field is basically dominated by Bz and By components, which exhibit clear bipolar changes near 12:55:30.5 UT (Figure 1(f)), indicating presence of a CS. Current density, calculated by using the Ampere's law, indeed show a large increase associated with the field changes, reaching up to 600 nA/m2, with its parallel and perpendicular components (with respective to local magnetic field) being comparable in magnitude (Figure 1(g)). The propagating speed of the intense CS, determined by four-spacecraft timing analysis, is −251×[0.98, −0.21, 0.04] km s⁻¹. Considering its duration (∼0.5 s), the CS thickness is 126 km, or equivalently ∼2.5 di, given a local plasma density of 21 cm⁻³. The CS speed is basically close to the ambient plasma flow speed (∼−280 km s⁻¹; Figure 1(h)), indicating that it was basically convected along with the ambient flow. Nevertheless, both ion and electron flow speed show large local variations during the CS crossing, such as quasi-bipolar changes in Vi,y and Ve,y components (Figures 1(h) and (i)). In addition, electric fields were also locally enhanced near the CS. Hence the CS was potentially reconnecting.

To examine reconnection features inside the CS in detail, a local LMN coordinate system is established via minimum variance analysis of B, yielding: L = [−0.27, −0.92, −0.25 (along the antiparallel magnetic field direction), N = [0.95, −0.21, −0.22] (points along the CS normal), and M = N × L (in the out-of-plane direction). The CS normal direction obtained is close to the CS propagating direction estimated by the timing analysis, indicating that the normal direction is well resolved. Figure 2 shows the profiles of electromagnetic fields and particles in the LMN coordinate system. One can see that BL component shows a bipolar variation during the interval, from ∼−30 to ∼29 nT (Figure 2(a)). Along with the BL change (marked by red-shaded area), speeds of ion and electron flow in the L direction (the outflow direction) were clearly enhanced (approaching 220 km s⁻¹; Figures 2(c) and (d)), indicating crossing of a reconnection outflow region. The plasma outflow speed is close to 1.3 VA, where VA is Alfvénic speed estimated based on upstream magnetic field and density (BL = 30 nT and N = 17 cm⁻³). Super-Alfvénic electron flow in the M direction was also detected (Ve,M = −350 km s⁻¹; Figure 2(d)), carrying the main out-of-plane current JM which approaches 630 nA m⁻² (Figure 2(e)). The reconnection X-line also hosts other typical reconnection features, including density depletion (from 26 to 11 cm⁻³; Figure 2(g)), ion heating (Ti, from 340 to 390 eV; Figure 2(h)), and strong energy conversion in electron rest frame $J\cdot {E}^{{\prime} }$ (approaching 440 pW/m3; Figure 2(j)).

Zoom In Zoom Out Reset image size

Inside the reconnection outflow region (Vi,L ≈220 km s⁻¹), the BN component (reconnected magnetic field) exhibits clear variations. In particular, BN was locally enhanced near 12:55:28.2 UT, from 4.6 to 10.4 nT (dubbed as the first hump), and near 12:55:29.4 UT, from −1.0 to 8.9 nT (dubbed as the second hump; Figure 2(b)). Considering durations (∼0.35 s) of the two magnetic humps, their thicknesses are close to 70 km, or equivalently ∼1.3 di. The ion-scale, monotonical increase of these two structures is in line with the key features of RFs (e.g., Sitnov et al. 2009; Liu et al. 2019). Hence, they are pristine RFs (called RF1 and RF2 hereafter) developing inside the reconnection diffusion region. The two RFs, however, host some different characteristics, in terms of density (clearly increases at RF1 but weakly drops at RF2; Figure 2(g)), ion temperature (weakly decreases at RF1 but increase at RF2; Figure 2(h)), electron temperature (increases at RF1 but decreases at RF2), and energy conversion (negligible at RF1 but intense at RF2). Considering that the two RFs were in different locations with respect to the reconnection X-line (RF1 and RF2 are approximately 6 and 2 di away from the X-line, based on outflow speed and RF–X-line time span), the RF dynamics may evolve dramatically during their outward propagation. Interestingly, RF2, which is closer to the X-line, hosts much stronger energy conversion (approaching 1300 pW/m3; Figure 2(j)) than the X-line itself.

We further utilize a nonlinear magnetic field reconstruction method to determine the magnetic field topology in the X-line vicinity (Denton et al. 2020). We select several time lines corresponding to the RF2 interval to reconstruct magnetic field topology since RF2 is closer to the X-line. The reconstruction results are shown in Figure 3, where one can see that an X-line indeed stands near MMS (Figures 3(b)–(e)). During the interval of reconstruction, MMS encountered the RF inside the outflow region and were moving toward the X-line. At 12:55:29.7 UT, the RF is ∼1.4 di away from the X-line, in agreement with the RF–X-line distance (2 di) estimated above. Hence, the magnetic field reconstruction further verifies that MMS indeed detected RFs near the X-line.

Zoom In Zoom Out Reset image size

RFs have been widely observed inside the planetary (e.g., Earth, Saturn, and Venus) magnetotails and well documented to play a vital role in energy transport in the magnetosphere (e.g., Angelopoulos et al. 2013; Liu et al. 2022a). Nevertheless, they have never been clearly identified in other magnetospheric regions, such as magnetopause and magnetosheath, where reconnection has also been frequently observed. The underlying reason has been thought to be related to the fact that reconnection in different magnetospheric regions operate in different patterns: reconnection in the magnetotail is typically symmetric with weak guide field, while inside the magnetopause and magnetosheath, reconnection is usually asymmetric with large guide field. However, recent kinetic simulations have suggested that RFs can also develop during asymmetric reconnection with strong guide fields (e.g., Song et al. 2019; Shu et al. 2021). Hence from the simulation side, RFs are not exclusive structures hosted only by symmetric reconnection, and may be common in regions favorable for reconnection. Rather, the scarcity of observations of RFs in the magnetopause and magnetosheath should be ascribed to RFs' outward propagation, which may potentially alter the RF dynamics. The present observation of two ion-scale RFs in the turbulent magnetosheath partly verifies this posit.

The two RFs were detected inside an ion diffusion region and embedded within a steady ion outflow. They host different plasma activity, in terms of variations of particle density and temperature, as well as energy conversion rate. These differences indicate that RF dynamics may evolve dramatically over the distance relative to the central X-line, and the underlying physics may be related to their interactions with the surrounding plasma. RF1, which is relatively far away from the X-line, may have approached a steady state and thus hosts negligible energy conversion. On the contrary, RF2, which stands much closer to the X-line, drives intense energy conversion. Interestingly, the RF-driven energy conversion is much stronger (by a factor of ∼3) than that at the X-line. Hence, RFs cannot only play a crucial role in the global-scale energy transport driven by reconnection, but also in the local energy transfer inside the reconnection diffusion region. Note that across the fronts, plasma density exhibits a slight increase (at RF1) or weak variation (at RF2), akin to one state of dipolarization fronts observed in the flow braking region in the magnetotail (e.g., Schmid et al. 2015).

In summary, we present the first observation of successive RFs inside an ion diffusion region in the Earth's turbulent magnetosheath, via using MMS-cadence data and a nonlinear reconstruction method. The study not only sets new paths for understanding energy conversion driven by reconnection in the planetary magnetosheath, but also opens new possibility for searching for RFs in other magnetospheric regions, such as in bow shock and magnetopause of Earth, Saturn, and Venus.

The present study is supported by the National Natural Science Foundation of China (grants 41821003 and 42104164). We greatly thank the entire MMS team for making the high-cadence data available. The data used in the present study are collected by the NASA's MMS mission and are publicly available at https://lasp.colorado.edu/mms/sdc/public/about/browse-wrapper/.