The Aurora that brilliantly colors the night sky in the polar regions is generated by high-energy charged particles (i.e., ions and electrons) entering the earth’s atmosphere from the magnetotail, a region that exists at an altitude of about 100,000km on the night side of the earth. It is known that most of the charged particles originate from plasmas (high-temperature, thin ionized gas) blowing out from the sun at a supersonic speed, called the solar wind. But the earth’s magnetic field forms the magnetosphere and thus should play the role of a shield against the solar wind. So how does the solar wind enter the earth’s magnetosphere?
The solar wind is a high-temperature corona gas released from the sun together with the solar magnetic field and fills the entire solar system. The solar wind magnetic field is affected by explosive phenomena near the solar surface, such as flares, or is deformed in the outer space. As a result, its direction at times becomes the same as that of the earth’s magnetic field and, at other times, the opposite. We already know how gaps form in the shield of the earth’s magnetic field when the solar wind magnetic field is in the opposite direction to the earth’s, i.e., southward. By a mechanism called “magnetic reconnection,” the earth’s magnetic field lines interconnect with those of the solar wind and, consequently, the solar wind particles get free access to the magnetosphere. This reconnection process is also known to be associated with solar flares or other astronomical phenomena. On the other hand, it was accepted that, when the solar wind magnetic field is directed northward, magnetic reconnection does not take place and, accordingly, the solar wind is blocked by the shield of the earth’s magnetic field and veered off to the flanks. About twenty years ago, however, NASA’s ISEE 1 satellite discovered that, when the solar wind magnetic field is northward, the earth’s magnetosphere is filled with even more solar wind plasma.
When the earth’s magnetic field is acting as a stronger shield, how can solar wind enter the magnetosphere? This had been a big question for years. From recent studies, however, we found that vortices developed through “Kelvin-Helmholtz instability” may be a clue to answer the question.
Kelvin-Helmholtz instability in collisionless plasma
Kelvin-Helmholtz instability is a fluid instability that grows at the interface between two kinds of fluids streaming at a relative velocity to each other. Examples are surface waves generated when the wind is blowing over water, or vortices emerging in cloud layers. Already half a century ago, it was predicted that this instability could occur in the magnetospheric boundary layer that exists between the solar wind and the magnetosphere (Fig. 1). This is because plasmas in the magnetosphere are almost stagnant since they are trapped by the earth’s magnetic field, while the solar wind flows just outside the magnetosphere at a very high speed (several hundreds km/sec.).
When the Kelvin-Helmholtz instability grows sufficiently, a chain of vortices forms. We often see vortices facilitating the mixing of materials in our daily life, for instance, when we pour milk into coffee and stir. In this case, mixing is caused by collision of molecules with each other and resulting diffusion. In the space gas, however, because of its ultra-thin and high-temperature nature, collisions among charged particles hardly take place. In such a collisionless plasma, some cause-and-effect relationship between the existence of vortices and the mixing of materials is not a necessity.
Recent computer experiments, however, suggest that mixing of collisionless plasmas can take place efficiently when vortices exist. When vortices grow, magnetic-field lines in the vortices are twisted and, consequently, oppositely-directed magnetic-field components are produced, which boost magnetic reconnection. This is one explanation. If so, the next question is whether or not vortices can really roll up (i.e., the Kelvin-Helmholtz instability can grow vigorously) in the magnetosphere. The magnetic field, however, simultaneously plays a role like surface tension for the ordinary fluids and tends to hamper the instability growth. In addition, the magnetosphere has a complex 3D structure (Fig. 1). Considering these facts, we were not confident that vortices (i.e., the instability) grow sufficiently to roll up.
Detection of rolled-up vortices
As seen in the results of a computer experiment shown in the bottom of Fig. 1, rolled-up vortices have a very complicated shape. It is therefore almost impossible to identify the existence of such structures by conventional observation methods using only a single satellite. In this situation, CLUSTER satellites were launched in an attempt to unveil the temporal variation and spatial structure of the earth’s magnetosphere. The CLUSTER mission was initiated jointly by ESA and NASA and comprises four satellites. It is the first formation-flying satellites accomplished in the world. In the belief that the constellation of satellites must be able to dissect and pinpoint the structure of the magnetospheric boundary layer, we scrutinized the obtained data to find some signatures of vortices. Finally, in the data recorded on November 20, 2001, we found evidence of rolled-up vortices. On this day, according to solar-wind observation made by the ACE satellite, the solar wind magnetic field was pointing northward for a long period.
The simultaneous observation by the four satellites confirmed a unique plasma density structure (lower panel of Fig. 1) that emerges only when vortices roll up. More importantly, we found evidence that solar wind plasmas exist in the rolled-up vortices, in other words, evidence that solar wind entered into the magnetosphere. This agrees precisely with the prediction by computer experiments that vortices induce plasma mixing. It was also evident that each of the vortices detected by CLUSTER has a length of about 40,000km. From the computer experiments, we already know that the length-to-width ratio of a well-developed vortex is about 2:1. Accordingly, the thickness of the plasma-mixing layer is estimated to be about 20,000km, or three times the earth radius. It turned out that the mixing layer with a thickness non-negligible compared to the width of the magnetotail (about 20 times the earth radius) can be produced by vortices.