Radio Emissions

Observed Radio Emissions

James LaBelle, P. I. and Simon Shepherd

The aurora borealis (or northern lights) is known for its spectacular green and red light, and these light emissions have long been studied by scientists for clues about the origin and nature of the aurora. Although unseen by the eye, the aurora also produces electromagnetic radiation at radio frequencies, which in turn provide a wealth of further information about auroral physics in much the same way that radio astronomy complements optical observations of stars and galaxies.

While very low frequency radio waves from the aurora have been described long ago, the frequency band from 100 kHz to many MHz has been relatively neglected, and for several years the experimental Geoplasma Physics Group at Dartmouth has operated remote radio observatories from various sites around the world in order to discover what auroral radio emissions occur in this frequency range. Sites at which Dartmouth currently maintains radio receivers are shown in Figure 1. At each site is located a programmable stepped frequency receiver (PSFR), magnetic loop antenna, and a personal computer. In the standard mode the PSFR sweeps from 30-5000 kHz every 2 seconds. The computer controls the receiver and logs the data. Once a month the data is backedup and sent to Dartmouth for processing.

Figure 2 shows a spectrogram of one hour of data from one of our sites located in Arviat, Northwest Territories. Each 2 second sweep is displayed as a vertical stripe of grayscale with the darker pixels corresponding to more intense signals.

Much of the noise in this frequency range is man-made; for example, AM radio stations broadcast on frequencies between 550 and 1600 kHz. The most common natural signals are atmospherics, radiowave pulses generated by lightning which creates static when received on AM radio. However, of interest to our group are more unusual radio waves generated by the aurora. We have observed three basic types of these waves, which are called auroral hiss, auroral roar, and MF burst. Among the observational discoveries made by our group in the last few years are:

1. The auroral roar emission was discovered in the late 1970's. It consists of a relatively narrow-band wave (relatively pure tone) near 2.8-3.0 MHz, depending on the latitude of the observing site. We discovered in 1993 that there is a twin type of auroral roar which occurs at higher frequencies, near 4-4.5 MHz, which we call "harmonic roar."

2. In 1994, we discovered a new type of radio emission never documented before, which we call "MF-burst," because it occurs at Medium Frequencies (MF) and it is bursty in nature. It is observed very commonly in association with aurora, and there is as yet no theory to explain it.

3. Auroral roar was presumed to have a bandwidth of a few hundred kiloHertz, centered near 2.8-3.0 MHz. In 1994, we trained a high-frequency and -time resolution receiver on an auroral roar and discovered that it is not one wave, but a train of scores of very pure tones which turn on and off and change frequency in a complicated pattern. This discovery of auroral roar fine structure (part I) throws all existing theories of auroral roar into doubt. No theory has been proposed for the roar which explains these data.

4. In 1996, we operated the same high-frequency and -time resolution receiver (a downconverting receiver) in Churchill, Manitoba and captured hundreds of minutes of auroral roar fine structure (part II). A great variety of spectral and temporal features were observed adding to the complexity of any theory explaining these emissions. In addition the bandwidth of auroral roar fine structure was measured to be less than 6 Hz, which is very difficult to explain with current theories.

5. The observed polarization of auroral emissions places constraints on the mode in which the observed waves can propagate in the ionosphere. In 1996 Dartmouth verified that auroral hiss was right-elliptically polarized and discovered that auroral roar and MF-bursts were both left-elliptically polarized.

1.	The auroral roar emission was discovered in the late 1970's. It consists of a relatively narrow-band wave (relatively pure tone) near 2.8-3.0 MHz, depending on the latitude of the observing site. We discovered in 1993 that there is a twin type of auroral roar which occurs at higher frequencies, near 4-4.5 MHz, which we call "harmonic roar."
2.	In 1994, we discovered a new type of radio emission never documented before, which we call "MF-burst," because it occurs at Medium Frequencies (MF) and it is bursty in nature. It is observed very commonly in association with aurora, and there is as yet no theory to explain it.
3.	Auroral roar was presumed to have a bandwidth of a few hundred kiloHertz, centered near 2.8-3.0 MHz. In 1994, we trained a high-frequency and -time resolution receiver on an auroral roar and discovered that it is not one wave, but a train of scores of very pure tones which turn on and off and change frequency in a complicated pattern. This discovery of auroral roar fine structure (part I) throws all existing theories of auroral roar into doubt. No theory has been proposed for the roar which explains these data.
4.	In 1996, we operated the same high-frequency and -time resolution receiver (a downconverting receiver) in Churchill, Manitoba and captured hundreds of minutes of auroral roar fine structure (part II). A great variety of spectral and temporal features were observed adding to the complexity of any theory explaining these emissions. In addition the bandwidth of auroral roar fine structure was measured to be less than 6 Hz, which is very difficult to explain with current theories.
5.	The observed polarization of auroral emissions places constraints on the mode in which the observed waves can propagate in the ionosphere. In 1996 Dartmouth verified that auroral hiss was right-elliptically polarized and discovered that auroral roar and MF-bursts were both left-elliptically polarized.