Is Our Understanding of Solar-Wind/Magnetosphere Coupling Satisfactory?

An assessment of our physics-based understanding of solar-wind/magnetosphere coupling finds that the understanding is not complete. Solar-wind/magnetosphere coupling is foundational to magnetospheric physics and it is a key to comprehending and predicting space weather. We are modestly successful at correlating solar-wind variables with geomagnetic indices, but we lack the full knowledge to describe in detail how the shocked solar-wind plasma transports mass, momentum, and energy into the magnetosphere-ionosphere system and how the shocked solar wind drives geomagnetic activity and magnetospheric evolution. The controlling solar-wind factors that govern the driving of the magnetosphere-ionosphere system are not accurately known. Without this knowledge accurate predictions of the magnetospheric behavior cannot be made and no magnetosphere-ionosphere model will work correctly if it is driven incorrectly. Further, without a fundamental understanding, the prediction of the system reaction to some as-yet-unseen extreme solar-wind conditions will not be possible. In this perspective article several gaps in our knowledge are cataloged. The deficiencies in our physical understanding of solar-wind/magnetosphere coupling constitute a major unsolved problem for space physics (and for astrophysics), a problem that demands enhanced, coordinated research.


INTRODUCTION
It is known that the solar wind drives geomagnetic activity in the Earth's magnetosphere-ionosphere system: as the properties of the solar wind change, activity in the magnetosphere changes. Solarwind-magnetosphere coupling is the study of how the solar wind produces that activity along with other changes in the Earth's magnetosphere. Solar-wind-magnetosphere coupling is a scientific topic that reaches into all aspects of magnetospheric physics and it is the core physics underlying space weather.
The basis for most of our thinking about solar-wind/magnetosphere coupling originates from the 1961 papers of Jim Dungey and Ian Axford (Axford and Hines, 1961;Dungey, 1961). The quantitative data-analysis study of solar-wind/magnetosphere coupling started in earnest with the works of Heppner et al. (1963) and Snyder et al. (1963), published shortly after the first spacecraft measurements of the solar wind were made, (e.g. Neugebauer and Snyder, 1962). Via data analysis and computer simulations, more than 50 years of work by the international space-physics community has delivered a lot of progress on this topic (cf. the reviews by McPherron (1995), Russell (2000Russell ( , 2007, Cowley et al. (2003), Otto (2005), Koskinen (2011), and Walsh et al. (2014a), and Borovsky (2016)). However glaring gaps in our knowledge prevent us from declaring this problem solved (cf. Denton, 2020).
Nine quantitative deficiencies in our physical and quantitative understanding of how the solar wind couples to the magnetosphere-ionosphere system are collected into Table 1 and are elaborated upon in the following section. The paragraphs are numbered according to the items in Table 1. An assessment and recommendations are given in the final section.

THE GAPS IN OUR KNOWLEDGE
1) It is generally accepted that the rate of magnetic-field-line reconnection between the magnetosheath plasma and the magnetospheric plasma across the dayside magnetopause controls the majority of the coupling (Dungey, 1961). On this issue there is an unsettled debate about what physically controls the dayside reconnection rate and what factors in the solar wind control the coupling. For decades it has been assumed that the solar-wind motional electric field v sw × B sw directly influences the electric field v inflow B inflow in the reconnection inflow on the dayside magnetosphere, (e.g. Gonzalez and Mozer, 1974;Goertz et al., 1993;Milan et al., 2012;McPherron et al., 2013). However, numerical tests of that assumption (Birn and Hesse, 2007) indicate that, in general, the reconnection electric field is not equal to the external driving electric field. On the contrary it has been argued that the dayside reconnection electric field (and hence the reconnection rate) is controlled by local plasma parameters at the magnetopause (Borovsky and Birn, 2014;Wang et al., 2015): specifically, the Alfven speed and the magnetic-field strength in the magnetosheath and in the magnetosphere. The electric-field argument is that the reconnection rate is determined by the amount of magnetic flux delivered to the magnetopause by the solar-wind flow (Newell et al., 2007;Lopez et al., 2010) while the local-plasma argument is that the reconnection rate determines the flow boundary condition into the magnetopause (Borovsky and Birn, 2014). This unsettled dispute about what controls dayside reconnection (and consequently what controls the amount of solar-wind/ magnetosphere coupling) prevents us from elucidating the root cause of space weather. Statistical studies of solar-wind/ magnetosphere coupling have not settled this what-controlsreconnection issue. Related controversies involve whether the nature of the solar-wind coupling varies with the phase of the solar cycle and with the type of solar wind (McPherron et al., 2013;Dremulkhina et al., 2018) or whether it does not (Borovsky and Denton, 2018). Further, kinetic processes in the bow-shock and magnetosheath (often triggered by microstructure in the solar-wind plasma) also play roles in controlling the coupling of the solar wind to the magnetosphere, (e.g. Omidi et al., 2016;Hoilijoki et al., 2017;Newman et al., 2020), and see also Item (7).
2) Theoretical (Borovsky and Steinberg, 2006a;Cassak and Shay, 2007), simulation Ouellette et al., 2014), and experimental (Walsh et al., 2014b) evidence has been gathered to support the idea that the mass density of magnetospheric plasmas can reduce the rate of dayside reconnection. When the magnetospheric mass density is comparable to the magnetosheath mass density, the magnetosphere can mass load the reconnection process and accordingly reduce the coupling of the solar wind to the magnetosphere. This reduction is most important during geomagnetic storms (Borovsky et al., 2013). Unfortunately, our ability to assess the magnitude of this "mass-loading" effect is hindered by a lack of surveys of the mass density of the dayside magnetosphere  and by a lack of understanding of the origin and spatial-temporal evolution of the oxygen-rich warm plasma cloak (Chappell et al., 2008). The lack of knowledge of the time-dependent mass density of the dayside magnetosphere prevents us from accurately implementing straightforward formulas for the quantitative reduction of the dayside reconnection rate in studies of solarwind/magnetosphere coupling and in predictions of geomagnetic activity.
3) As noted in Item (2), the mass density of the magnetospheric plasma plays a role in determining the reconnection rate at the dayside magnetopause. A controversy about this has arisen, and not been settled, as to the global reaction of the solar-wind driving in response to spatially localized reductions of the reconnection rate, (e.g. by the localized dense plasmaspheric drainage plume). The specific controversy is whether or not the total integrated reconnection TABLE 1 | Outstanding questions associated with the nine deficiencies discussed in this perspective.

Item
Outstanding question 1 What physically controls the dayside reconnection rate and what are the solar-wind controlling factors? 2 What are the mass densities of the dayside-magnetosphere plasmas, how do they evolve, and what controls them? 3 What is the reaction of the solar wind to localized magnetospheric mass loading and how is the total reconnection rate affected? 4 How does the viscous interaction work and what are the controlling factors? 5 How does polar-cap potential saturation work? 6 Why does geomagnetic activity increase with increasing levels of upstream solar-wind turbulence? 7 What are the entry mechanisms of solar-wind plasma into the magnetosphere and what solar-wind parameters govern the amount of entry? 8 What are the roles of magnetospheric preconditioning, the ionosphere, and feedback in solar-wind/magnetosphere coupling? 9 In solar-wind/magnetosphere correlations, how do we isolate physics from mathematics?
Frontiers in Astronomy and Space Sciences | www.frontiersin.org February 2021 | Volume 8 | Article 634073 rate is affected by localized magnetospheric mass loading (Lopez, 2016;Ouellette et al., 2016;Zhang et al., 2016Zhang et al., , 2017. 4) It has been widely argued that there is a viscous-like interaction between the solar wind and the magnetosphere 1) that drives geomagnetic activity (Axford and Hines, 1961;Farrugia et al., 2001) and 2) that might be very important for solar-wind-plasma entry into the magnetosphere. There are magnetospheric convection features (Lundin et al., 1995;Farrugia et al., 2001) and ionospheric convection features (Drake et al., 2009;Sundberg et al., 2009) that are interpreted as signatures of this viscous interaction, and there is residual geomagnetic activity under strongly northward IMF that is attributable in part to this viscous interaction (Lockwood and Moen, 1999). *Two viscous mechanisms that have been considered for the transport of momentum into the magnetosphere are 1) plasma diffusion driven by waves (Eviatar and Wolf, 1968) and 2) Kelvin-Helmholtz waves on the magnetopause (Nykyri and Otto, 2001); like dayside reconnection, Kelvin-Helmholtz waves are affected by the mass density of the magnetospheric plasma (Walsh et al., 2015;Delzanno et al., 2020). Clear quantitative evidence for particular physical mechanisms underlying the viscous interaction is absent and the solar-wind controlling factors for the strength of this viscous interaction are unknown (cf. Sect. 2.1 of Borovsky et al., 2020). 5) Dayside reconnection magnetically connects the moving solar-wind (magnetosheath) plasma to the magnetosphereionosphere system, and the rate of dayside reconnection governs the amount of solar-wind magnetic flux connected (Lockwood et al., 1990). With the exception of the build-up of magnetic flux in the magnetotail, (e.g. Birn and Hesse, 2014), the physics of the coupling after reconnection has been little explored and so "post-reconnection coupling" is insufficiently understood. This lack of understanding of post-reconnection coupling is glaring in the problem of stormtime polar-cap potential saturation: at least nine different saturation mechanisms have been suggested (cf. Table A1 of Borovsky et al., 2009) but no community consensus has been reached on the basics of how polar-cap saturation works, (e.g. Gao et al., 2013;Myllys et al., 2017;Lakka et al., 2018). Another unsolved post-reconnection issue is the relative importance of the solar-wind forcing of antisunward ionospheric flow in the polar caps (Siscoe and Siebert, 2006) for driving the magnetospheric and ionospheric convection patterns. 6) Statistical analysis finds that the level of geomagnetic activity increases as the amplitude of magnetic-field fluctuations in the upstream solar-wind plasma increases (Borovsky and Funsten, 2003;D'Amicis et al., 2007D'Amicis et al., , 2010D'Amicis et al., , 2020Osmane et al., 2015). This statistical increase is seen even when controls are added to hold the dayside reconnection rate constant and when the fluctuating solarwind magnetic field remains purely northward (Borovsky, and Steinberg, 2006b) (This latter restriction rules out the driving of geomagnetic activity by on-and-off dayside reconnection enabled by large solar-wind magnetic-field fluctuations, (e.g. Tsurutani and Gonzalez, 1987;Tsurutani et al., 1995).) Physical mechanisms underlying this "upstream-turbulence effect" have been suggested, (e.g. Borovsky, 2006;Nykyri et al., 2017;Nakamura et al., 2020), but no evidence has been found to support one suggestion over the others.
7) It is known that solar-wind ions (Lennartsson and Shelley, 1986) and solar-wind electrons ) populate the magnetosphere and there are estimates of their transport timescales from the solar wind (Denton and Borovsky, 2009). Unknowns about solar-wind plasma entry are 1) the mechanisms that are conveying plasma from the magnetosheath into the magnetosphere, 2) the relative importance of various pathways into the nightside plasma sheet (low-latitude boundary layer vs. cusp-mantle), 3) estimates of how much mass enters, and 4) the solar-wind factors that control the pathways and the transport rates. Recent observations find that solar-wind directional discontinuities cause a jetting of plasma in the magnetosheath leading to renewed interest in the impulsive penetration of solarwind plasma through the dayside magnetopause into the magnetosphere, (e.g. Heikkila, 1992;Dmitriev and Suvorova, 2015;Plaschke et al., 2018;Escoubet et al., 2020).
8) The state of the magnetosphere-ionosphere system is known at a rudimentary level to impact the driving of the system by the solar wind. Typically, the state of the magnetosphere-ionosphere system depends on the time history of the solar wind. Investigations have examined the preconditioning of solar-wind/magnetosphere coupling via the roles of 1) the state of the plasmasphere and the state of the plasma cloak (Borovsky and Steinberg, 2006a;Borovsky et al., 2013), 2) the presence of a cool dense plasma sheet (Thomsen et al., 2003;Lavraud et al., 2006), 3) ionospheric conductivity (Kivelson and Ridley, 2008;Wang et al., 2013), 4) the state of the ring current (Dmitriev et al., 2011), and 5) the overall morphology of the magnetosphere and in particular the shape of the dayside magnetopause (Merkin et al., 2005;Ober et al., 2006). Feedback on the strength of solar-wind coupling via activity-driven ionospheric outflows and changes to the system (Kan et al., 2010;Borovsky, 2014a) needs to be understood and quantified, particularly when the magnetosphere is strongly driven by the solar wind. 9) Ideas about solar-wind/magnetosphere coupling are tested by cross correlating temporal variations in solar-wind properties with temporal variations in geomagnetic activity, (e.g. Newell et al., 2007;Badruddin and Aslam, 2013;Tenfjord and Østgaard, 2013;McPherron et al., 2015) (See also Wang et al. (2014) for cross correlations in global simulations.) The correlation strengths measure how well solar-wind variances describe geomagnetic variances. It is generally assumed that if a solarwind formula has a better correlation with geomagnetic indices, then that formula better describes the physics of the coupling. An alternate possibility is that a solar-wind formula that better describes important solar-wind variances has a better chance of describing geomagnetic variances. One such "unphysical" solar-wind formula that has better correlations with geomagnetic activity is v sw + 56B south (Borovsky, 2014b), which beats all of the standard solar-wind functions, (e.g. Newell et al., 2007;Borovsky, 2008). Multivariate fits between a single geomagnetic index and multiple solar-wind variables (cf. for yielding high correlation coefficients. This math-versusphysics dilemma of correlations casts some doubt on our ability to test the accuracy of our physics understanding by using solar-wind/magnetospheric data correlations. Three other problems that hinder the ability to test physics concepts of coupling are 1) imperfect solar-wind values owing to the upstream monitors being off the aberrated and time-variable solar-wind flow streamline that hits the Earth (Borovsky, 2018a;Walsh et al., 2019;Burkholder et al., 2020), 2) the strong intercorrelations of all pertinent solar-wind variables (Borovsky, 2018b(Borovsky, , 2020, and 3) geomagnetic indices being imperfect measures of solar-wind coupling.

ASSESSMENT AND THE FUTURE
How well do we understand solar-wind/magnetosphere coupling? Certainly it is not understood sufficiently well for a full explanation of how the magnetosphere-ionosphere system is driven: not understood are the viscous interaction, the upstream-turbulence effect, plasma entry, and polar-cap potential saturation. It is not understood sufficiently well that we know all of the physical processes acting to ensure that modeling codes capture them. Further, we do not know the solar-wind controlling factors for the dayside reconnection rate, for the strength of the viscous interaction, or for the rate of plasma entry. And we don't know the geomagnetic-activity controlling factors that govern the mass density of the dayside magnetosphere, which enters into the dayside reconnection rate. A potential issue for the future is the role of the Earth's magnetotail in solar-wind/magnetosphere coupling: specifically whether the rate of nightside reconnection impacts the rate of dayside reconnection (Lockwood et al., 1990) and the role of magnetotail processes in mass transfer from the solar wind into the magnetosphere (Otto and Fairfield, 2000). There are surprisingly many unsolved issues about the physics of the coupling. The extent of the deficiencies is underappreciated by the broad spaceresearch and space-weather communities.
To obtain a full grasp of the physics underlying the solar wind driving of the Earth's magnetosphere, 1) more open discussion of our deficiencies is needed in the broad space-physics community, 2) funding agencies need to become aware of the gaps in our knowledge and the adverse impacts of those gaps, and 3) multiple coordinated research efforts are needed to solve the questions of Table 1.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
The author performed all research on this perspective and wrote the manuscript.

FUNDING
Work was supported at the Space Science Institute by the NSF GEM Program via grant AGS-2027569, by the NASA Heliophysics LWS program via award NNX16AB75G, by the NASA Heliophysics Guest Investigator Program via award NNX17AB71G, by the NSF SHINE program via grant AGS-1723416.