Figure 1: A whistler group, the discrete components
of which are due to energy from a single lightning flash
taking different paths through the magnetosphere.
Abstract We discuss the whistler ghost phenomenon and
suggest that
whistlers can trigger electric discharges in the atmosphere by
inducing electron precipitation from the magnetosphere.
Sommaire. Nous analysons le phénomène de ``sifflements'' et
suggérons que ces sifflements peuvent provoquer des décharges
électriques dans l'atmosphère en induisant une précipitation
d'électrons issus de la magnétosphère.
It is well known that particles from the earth's magnetosphere produce spectacular auroral displays in regions around the earth's magnetic poles. It is also known that particles from the magnetosphere can cause interference with radio communications and induce electric currents in power lines that can burn out transformers causing loss of electric power to whole regions. In March 1989, during a great magnetic storm, auroral currents induced surges in power lines that damaged transformers and left nine million people in Canada without power. Similar disturbances have disabled navigation satellites and have induced currents that have promoted rust in long oil pipe-lines. Evidence is now emerging of another magnetospheric effect in which lightning flashes may sometimes be triggered by particles entering the earth's atmosphere. The evidence for this is outlined here.
Energy is generated in a lightning flash over a wide range of frequencies. We of course see the visible flash and the `atmospherics' heard on radio receivers during a thunder storm are evidence of energy generated in the radio part of the spectrum. At audio frequencies a phenomenon known as a whistler occurs. This is produced when part of the radio energy, guided by the earth's magnetic field, travels through the ionosphere and magnetosphere and returns to earth in the opposite hemisphere. The frequencies which give rise to whistlers lie in the VLF frequency range 50 Hz to 20 kHz and can easily be observed in the opposite hemisphere with a simple antenna and audio amplifier. A whistler on its path may travel out to tens of thousands of kilometres from the earth. The energy starts as a sharp pulse lasting a few milliseconds which is stretched out in the magnetosphere because higher frequencies travel faster than lower frequencies. What is heard in the opposite hemisphere is a falling tone (the whistler) which lasts for more than a second. A spectrogram of such a whistler is shown in Figure 1. In fact the radio energy can take many different paths (Figure 2) through the magnetosphere, giving rise to a group of whistlers with slightly different travel times. It is also found that part of the energy can be reflected from the lower edge of the ionosphere, giving rise to whistler echoes which are heard in both hemispheres.
Whistlers have many interesting properties, one of which is that they can resonate and exchange energy with high energy charged particles in the magnetosphere. In the process the particle pitch angles are modified and some of them may precipitate into the atmosphere. These particles are normally trapped in the Van Allen belts around the earth and increase in number during magnetic storms. What is now being suggested is that these charged particles when induced by whistlers to enter the atmosphere can trigger further lightning flashes.
The evidence for the triggering of lightning rests on a phenomenon known as a whistler ghost. This was discovered by researchers from the University of Natal in data gathered during a 3-week scientific campaign at the South African base on Marion Island. A whistler ghost group is shown in Figure 3. The ghost whistler group follows the initial group after a time interval of about 600 ms.
One of the experiments deployed in the Marion Island campaign was a VLF radio direction finder to study whistlers and other VLF phenomena. In a two and a half hour period of recording it was noted that each strong whistler group was followed, after about 600 ms, by a fainter group which had components with the same dispersion as the original group. It was unlikely that the repetition of such a time delay would occur by chance. In other words, it is unlikely that two lightning flashes would occur repeatedly separated by the same time interval and so it was important to establish the cause of the second flash. It turns out that the time interval between the two flashes is just right for the second flash to be triggered by electrons precipitated into the atmosphere by the first. The process works like this. A lightning flash close to Marion Island's (46° 53' S, 37° 52' E) magnetic conjugate gives rise to a whistler which is received at Marion Island about a second later. On its path the whistler encounters high energy electrons with which it resonates, causing a reduction in their pitch angles. These particles are normally trapped in the earth's magnetic field and bounce back and forth, mirroring high above the atmosphere. After resonating with a whistler some of the electrons are no longer reflected but precipitate into the atmosphere in the northern hemisphere and trigger another lightning discharge. The second lightning flash produces the ghost whistler. The conditions for the wave-particle resonance are most favourable close to the equatorial plane. If this picture is correct then the time delay between the two groups would be determined by the time it takes the waves to travel up to the equatorial plane plus the time it takes the electrons to travel down to the atmosphere. It turns out that this time would be about 800 ms. To get the observed 600 ms delay it is necessary to shift the interaction region about 10 degrees in latitude towards the northern hemisphere. There are sound theoretical reasons why this should be the case.
The explanation of the whistler ghost phenomenon suggests that whistlers can sometimes trigger lightning or other atmospheric discharges. It is not suggested that this effect is a common one but only that it can occur. It requires rather special conditions. Firstly, a high density of trapped particles is required such as those present after a magnetic storm. Secondly, the whistler-induced precipitation must occur in a region where thunderstorm conditions exist. The initial whistler may originate in a different place, perhaps more than 1000 km from the flash that produces the ghost.
The evidence for this effect is encouraging but the final proof of the phenomenon will depend on ghosts being found at other latitudes. Because of the different paths and travel times, ghosts at other latitudes should occur with different time delays.
