Portal de Periódicos da CAPES

Coronal mass ejections: a driver of severe space weather

2015; Wiley; Volume: 70; Issue: 1 Linguagem: Inglês

10.1002/wea.2437

1477-8696

Lucie M. Green, Deb Baker,

Astro and Planetary Science

Observational studies of the solar atmosphere in the early 1970s revealed that the Sun sporadically ejects vast quantities of matter into the Solar System in addition to the constant outflow known as the solar wind (Tousey, 1973). These sporadic ejections are known today as coronal mass ejections (CMEs). They are seen in images that use an occulting disc to block the dazzling light emitted by the photosphere, creating an artificial solar eclipse and allowing the outer atmosphere to be viewed (Figure 1). CMEs are then revealed as outward moving structures that exhibit a range of morphologies and which carry around 1012 kg of magnetised plasma into the heliosphere with an average speed of ~490kms−1 (Webb and Howard, 2012). Upon eruption, these magnetic structures quickly expand to become larger than the Sun itself. CMEs are common events and occur with a frequency that follows the (on average) 11-year solar cycle with approximately one CME being formed per day at cycle minimum and as many as five per day at cycle maximum (Webb and Howard, 2012). The solar cycle is driven by an evolving global magnetic field that changes in strength and complexity and which manifests itself in the photosphere by the presence of sunspots – regions of concentrated magnetic field. As the solar cycle ebbs and flows, the magnetic field in the solar atmosphere is transformed between a magnetically simple state when there are correspondingly few sunspots at the surface, and a magnetically complex configuration when there are many sunspots. In this way, CMEs can be understood as a mechanism by which the Sun ejects atmospheric magnetic field that harbours both magnetic energy and a complex morphology (Low, 1996). CMEs play an important role in the magnetic cycle of the Sun. For recent review articles on CMEs see, for example, Webb and Howard (2012), Chen (2011) and Forbes et al. (2006). The discovery of CMEs goes beyond elucidating important aspects of the Sun's magnetic field evolution. The discovery of these ejections answered an outstanding question in solar-terrestrial physics when it was realised that they are the solar phenomenon that are the cause of major, but transient, disturbances to the geomagnetic field (e.g. Gosling, 1993). When CMEs collide with the Earth's magnetic field (the magnetosphere) they can compress its dayside and, if the magnetic field of the CME is of the correct orientation, magnetic reconnection between the CME and the magnetosphere can occur. Magnetic reconnection is a fundamental physical process in which plasma and magnetic field become momentarily decoupled, leading to a reconfiguration of the field and the conversion of magnetic energy into other forms of energy. The magnetic field at the sub-solar point of the magnetosphere (directly between the Sun and the Earth) is strongly northward, so a strong southward directed magnetic field within the CME could lead to magnetic reconnection and a joining of the magnetic field of the CME with the magnetosphere. If this occurs, the dayside magnetospheric flux gets dragged over to the nightside by the motion of the CME, and an accumulation of magnetic flux then occurs in this part of the magnetosphere, followed by magnetic reconnection. This reconfiguration of the nightside magnetosphere allows magnetic flux to move back to the dayside, where it can undergo reconnection with the CME once again. This process is known as the Dungey cycle (Dungey, 1961). During these times the magnetosphere is driven into a disturbed state, producing geomagnetic sub-storms and storms which then lead to a multitude of space weather effects (Pulkkinen, 2007). CMEs, which are the bulk ejection of magnetised plasma, are often temporally and spatially associated with another form of solar activity known as a solar flare – sudden bursts of electromagnetic radiation (decametre radio waves to gamma-rays) and high-energy particles (Benz, 2008). This CME-flare association has its origins in the physical processes that are common to both phenomena. Solar flares arise from the conversion of magnetic energy into other forms during magnetic reconnection, and this energy conversion leads to the production of electromagnetic radiation. However, magnetic reconnection also reconfigures the magnetic field during a CME, which accelerates the magnetised plasma away from the Sun. Despite these common aspects, CMEs and flares are very distinct in terms of their physical guise. Solar flares have a space weather impact through their high-energy radiation that ionises the Earth's upper atmosphere, and through the protons that are accelerated to near-relativistic energies, posing a threat to spacecraft and humans in space. The most extreme space weather event in our records is the geomagnetic storm that began on 2 September 1859 (Tsurutani et al., 2003). This storm has become known as the Carrington event, named after the Victorian amateur astronomer who observed the associated flare, which occurred in a large sunspot group he was drawing at his observatory in Redhill (Surrey). Less than a day later, an intense display of the aurora erupted, which was seen down to exceptionally low latitudes as an intense geomagnetic storm took hold. Measurements from a magnetometer in Bombay, India, revealed the exceptional changes to the Earth's magnetic field that occurred (Siscoe et al., 2006). The significant aspect of this event, which was unknown to Carrington at the time, was that the sunspot group had produced a CME at the same time as the flare. The CME reached the Earth after only 17.5 h, giving it a mean velocity along the Sun–Earth line of ≈ 2300kms−1. The most intense geomagnetic storms (Gonzalez et al., 1994) are associated with CMEs that have a large velocity as this can lead to a large inflow of magnetic flux. Multiple CMEs may also be important as earlier eruptions could create the solar wind conditions necessary for a subsequently fast CME. It looks likely that the Carrington CME occurred at a time of heightened solar activity, and the geomagnetic records suggest another CME had occurred just a few days before (Boteler, 2006). Intense geomagnetic storms are also related to CMEs that have a high dynamic pressure as this leads to a high momentum flux (Svalgaard, 1977). The great geomagnetic storm of 13 March 1989 (Allen et al., 1989) is an example of an extreme geomagnetic storm that was driven by a CME that was not exceptional in terms of its velocity; its mean Sun-Earth transit time was 770kms−1 (Feynman and Hundhausen, 1994). However, Feynman and Hundhausen (1994) remark that the CME was exceptionally bright in the coronagraph data, which strongly supports an interpretation that it was a very massive eruption (the coronagraph images detect photospheric light scattered by coronal electrons, and brightness is therefore a function of density) and therefore could have had a high dynamic pressure when it reached the magnetosphere. Indeed, the dayside magnetosphere is thought to have been compressed towards the Earth, by more than a factor of two, for an extended period of time (Allen et al., 1989). In addition to the above, it is possible that the pre-existing condition of the magnetosphere plays a role in severe geomagnetic storms. There are studies that indicate preconditioning may not play a significant role in strengthening the storm intensity, but it may lengthen the storm duration (e.g. Xie et al., 2008). One thing remains clear though: the most intense geomagnetic storms will not be triggered unless the CME has the correct magnetic field configuration. Intense geomagnetic storms require the presence of a strong southward magnetic field component more than any other plasma parameter, (Gonzalez et al., 2011) and this southward field must persist for several hours (Russell et al., 1974). This is because the strong southward component provides an effective coupling of the magnetic field of the CME with the magnetosphere, which controls the transfer of energy. NASA's ACE spacecraft at the first Lagrange point constantly monitors this magnetic field component, 1.6 million km upstream of the Earth. It is referred to as the BZ component as its axis is parallel to the ecliptic pole in the Geocentric Solar Ecliptic coordinate system. The role of CMEs in driving extreme geomagnetic activity has led to efforts in both modelling and observation to predict their velocity and arrival times at Earth. During their eruption, CMEs are rapidly accelerated at low altitudes in the solar atmosphere to velocities across a broad range from 100kms−1 to more than 3000kms−1. The travel time of CMEs from the Sun to 1 AU (1 AU is the astronomical unit for the Sun-Earth distance) varies from less than 24 h to more than 3 days; a surprisingly small range, given the distribution of initial velocities. This is explained by the fastest CMEs being decelerated in the solar wind (Woo et al., 1985; Watari and Detman, 1998) and the slower ones being accelerated by the solar wind flow (Lindsay et al., 1999) so that by the time CMEs arrive at 1 AU the majority are travelling at a speed close to that of the ambient solar wind flow (between 350 and 550kms−1, Gopalswamy et al., 2000). So, for an extreme event the CME needs to have a high velocity at 1 AU, which means it needs to have a high velocity when leaving the Sun and experience a low aerodynamic drag force in the solar wind. In addition, it needs to hit the magnetosphere head-on (i.e. not give us a glancing blow), have a high dynamic pressure and have a magnetic configuration with a strong and sustained southward component. It is this combination of factors that needs to be accurately identified in a CME from the plethora that leave the Sun. The propagation time of CMEs means that forecasting their occurrence may not necessarily be required. In the majority of cases, space weather users have upwards of a day to make their preparations once an eruption has been observed. Once a CME has been observed, models exist to forecast the direction of propagation and, if Earth-directed, the CME arrival time at 1 AU (Cargill and Schmidt, 2002; Owens and Cargill, 2004). However, for a Carrington-type event the CME transit timescale may well be insufficient for producing a timely forecast once the delay in receiving the observational data has been factored in. In light of this, attempts are being made to determine the likelihood of a CME occurring in the solar atmosphere based on observations. A brief summary of eruptive indicators and time scales is given in Table 1. Currently, the longest observational warning of an Earth-directed CME occurrence is provided by the presence of a so-called sigmoid – an S-shaped structure seen in soft X-ray or EUV images of the solar atmosphere (Canfield et al., 1999). Once this shape is seen, the region is likely to erupt within hours (i.e. not days) (Green and Kliem, 2014). Due to projection effects, these features are best seen when the magnetic configuration is viewed from above, that is, when they are near Sun centre. Eruptions from these features are therefore likely to be Earth-directed. This is in contrast to another observational indicator of an ensuing CME, namely streamer blowouts, which are seen on the solar limb and are not so likely to be Earth directed. Identifying and modelling the pre-eruption magnetic configuration of sigmoids using solar observations raises the opportunity of inserting this magnetic configuration into the propagation models mentioned above and eventually achieving data-driven simulations of CMEs from Sun to Earth in the coming years. It is also worth noting that, since there is a close association between CMEs and solar flares, the use of sigmoids to predict the occurrence of an eruption is also applicable to efforts in the area of flare forecasting. There is currently a major gap in our forecasting capability though. We still do not have a model that uses solar observations to build a realistic magnetic configuration of a CME as it leaves the Sun. Such a model would enable us to understand CME evolution through the inner heliosphere and therefore forecast the strength of the BZ field component at 1 AU. This is the current space weather challenge. However, the first steps towards achieving this have been taken in fundamental solar and solar-terrestrial research. When CMEs are observed in situ in the inner heliosphere they often exhibit a coherent structure in the solar wind that has an increased magnetic field strength as compared to the ambient solar wind flow, an anomalously low proton and electron temperature, and a large magnetic field rotation that is indicative of a particular configuration containing twisted, or helical, magnetic field known as a flux rope (Burlaga et al., 1981; Klein and Burlaga, 1982; Cane and Richardson, 2003). Figure 1 shows a CME whose plasma density distribution strongly supports such a twisted configuration. Indeed, it seems that most CMEs may be well modelled by a flux rope magnetic configuration regardless of their source region (Jian et al., 2006). This has motivated a search for flux ropes in the solar atmosphere, prior to their eruption as a CME. If found, the global aspects of their magnetic field can be studied to obtain an indication of the CME's geo-effectiveness from the orientation of the flux rope axis and the organisation of the twist, leading to a forecast of the presence of a southward-directed (BZ) magnetic field. We note that CMEs may develop sub-structure en route to the Earth (Steed et al., 2011), but this level of complexity is beyond the scope of current forecasting activity. There is increasing evidence that magnetic flux ropes are indeed present in the solar atmosphere before CME onset, perhaps unsurprisingly in the S-shaped (sigmoidal) regions that are highly eruptive (Green and Kliem, 2009; 2014; Liu et al., 2010; Green et al., 2011; Zharkov et al., 2011). The three-dimensional structure of the coronal field is not directly measurable, but the flux rope configuration can be inferred from observations of the solar atmosphere in soft X-ray and EUV emission. This identification allows us to make a direct link between the CME at the Sun and the flux rope measured in situ at 1 AU. The eruption of a flux rope in the solar atmosphere is another key focus of CME research. Their loss of equilibrium due to an ideal magnetohydrodynamic instability, catastrophe or force imbalance has so far been well studied and modelled in the idealised case (e.g. Hood and Priest, 1981; Forbes and Isenberg, 1991; Kliem and Török, 2006; Mackay and van Ballegooijen, 2006; Démoulin and Aulanier, 2010). Simulations of the formation of an idealised flux rope and its subsequent eruption have also been achieved (Amari et al., 2010; Aulanier et al., 2010). Observations of flux ropes prior to their eruption are crucial for future space weather forecasting as they can be used as the input to create more realistic data-driven CME simulations. Flux ropes are modified during their eruption as this involves magnetic reconnection within the overlying/surrounding magnetic field. This reconnection adds more magnetic flux to the rope, cuts the tethers of the overlying and constraining field, and aids the acceleration of the rope away from the Sun. It is this field that forms the external field of the rope (which is likely to be more twisted than the core) and which also needs to be known to forecast its coupling to the magnetosphere. Therefore, solar observations during the dynamic phase of the eruption must also be captured. The starting point for a data-driven simulation is to model the magnetic flux rope before its eruption as a CME. The solar observations can be used to determine the spatial extent of the flux rope, the pre-eruptive orientation of the flux rope axis and even an estimation of the axial flux content in special cases (Green et al., 2011). The S-shaped nature of the magnetic configuration is highly suggestive that the flux ropes are weakly twisted structures containing, in most cases, little more than one turn in the magnetic field vector. The solar data must then be used as the boundary condition for modelling the pre-eruption flux ropes. This has already been investigated using three approaches. First, by the flux rope insertion method (van Ballegooijen, 2004). This method inserts a flux rope into a magnetic field configuration that is extrapolated from solar observations. The location of the rope is guided by the observation of a filament – a dark and sinuous feature in the solar atmosphere. A match between the modelled magnetic structure and observed structure in the solar atmosphere is used to identify a realistic configuration. Many studies have been carried out using this technique, and they have found weakly twisted flux ropes in the pre-eruptive magnetic field (Bobra et al., 2008; Su et al., 2009; Savcheva and van Ballegooijen, 2009). Second is the approach that generates a time-series of 3D coronal magnetic field configurations using observations of the Sun's surface magnetic field (Mackay et al., 2011); as the magnetic field evolves, flux ropes can form (Gibb et al., 2014). Third, magnetic extrapolations are also used to reconstruct the magnetic field from observations of the three components of the vector magnetic field measured in the photosphere (e.g. Valori et al., 2012), but as yet the identification of flux ropes using this technique remains a challenge. As an example of the promise of the above flux rope models, we briefly discuss here a CME productive region that was on the Sun during December 2007, named NOAA active region 9077. The active region was studied from its birth on the Sun until it produced a CME, allowing the formation of the flux rope to be studied in detail. The flux rope was built in three phases. Phase one is a phase of increasing shear, driven by the dispersal of the surface magnetic field and magnetic reconnection that removes magnetic flux from the surface layer, following the model of van Ballegooijen and Martens (1989). In phase two, there is an accumulation of a significant amount of flux that runs almost parallel to the line that separates the main magnetic polarities of the region. In phase three, further magnetic reconnection produces field lines that are twisted around the axial flux (which look S-shaped) and which contribute poloidal flux and define the presence of a flux rope. Images in Figure 2 show these three phases for this solar active region and three others for emphasis. The observational approach to studying the magnetic flux content of the rope was brought together with the flux rope insertion method in Savcheva et al. (2012), who suggested the reliability of both modelling and observational approaches. Gibb et al. (2014) also studied this region and managed to reproduce the main features of the magnetic field evolution including the formation of the flux rope in the correct location. Figure 3 shows the results of these three studies and directly compares the observed and modelled magnetic configurations. LMG is grateful to the Royal Society for the award of a University Research Fellowship. DB acknowledges support by STFC Consolidated Grant ST/H00260/1. The authors thank Gordon Gibb for supplying a magnetic field extrapolation for Figure 3.