Far above the biosphere and the ozone layer we have the ionosphere, where, fuelled by the sun, atoms and molecules are constantly being split into electrons and ions, which give the ionosphere its properties and its name. Since the pressure is extremely low (the entire ionosphere weighs less than one tonne!), the electrons and ions are allowed to exist for quite some time before they recombine. A gas which contains separated charged particles is called a plasma and plasma can be considered a fourth state of matter. This fourth state is utterly rare on the surface of the earth. Only where the thunder strikes the enormous discharge creates as plasma lasting for a fraction of a second, and it was not until the 20^th century that man was able to create a laboratory plasma. Although rare on earth, plasma is abundant in space. Our sun and many stars are made of plasma, and as much as 99% of the Universe is in the plasma state. Our nearest and most accessible plasma "laboratory" is the ionosphere. By using electromagnetic waves as our messengers through space, systematic studies of this plasma can be conducted.

The interactions between our terrestrial space plasma and the electromagnetic waves, both from cosmos and, as in the "SEE-experiments", from high power HF radio transmitters, are not as simple as one could imagine. In the plasma, charges interact with electric and magnetic fields. Thus the charges get accelerated by the electric field components of the radio waves. When charges in one volume move, a charge imbalance is created inducing new electric fields in the plasma which, in their turn, interact with other charges. The ionospheric plasma consist of several different kinds of molecules and consequently different kinds of ions, mainly atomic oxygen, molecular oxygen and nitric oxide. Some of these ions are slower to accelerate than others because of their different masses. Their original random motion, the plasma temperature, which in its turn can be divided into ion temperature and electron temperature, also influences the wave-plasma interactions.

At least one more major complication must be added to the equation: the ionospheric plasma is magnetised by the Earth's magnetic field. The moving charges are influenced by the magnetic fields so that they accelerate perpendicularly to the motion and the magnetic field. The accelerating charges induce magnetic fields in their turn.

How can one even begin to study a phenomena with so many faces?
One must make approximations, neglect some factors and see how far a simplified model can explain the reality observed through experiments. Where the adopted model is insufficient to explain observations, a new extended model must be found. In the studies of waves in plasma this process is taking place right now.

The Institute of Space Physics, Uppsala Division, (IRFU), has taken part in numerous campaigns over the last two decades with the aim to investigate and understand interactions between electromagnetic waves and the ionospheric space plasma. As a matter of fact the material collected by the wave group at IRFU add up to several bookshelves of printed spectra and tens of gigabytes on the database hard drive. Of course, for every campaign the data stream grows exponentially, owing to improved data processing and digital storage. For instance in the most recent campaign, Sura 4, four times as much data was collected as during all previous campaigns combined.

In the beginning of the eighties, electromagnetic wave-wave interactions in plasmas were often explained by a simple model, known as stimulated scattering, not very different from how a ball hits a wall. The ball bounces back, but with somewhat less energy, and perhaps in another direction. The bounce exerts pressure on the wall and this pressure is spread as a sound wave within the wall. Imagine the ball as a photon, a quantised electromagnetic wave, and the wall as a layer of the ionosphere. However, when it comes to radio waves, interactions are electromagnetic instead of mechanical. 

When an electromagnetic radio wave is emitted from the ground, it could for example be a high frequency (HF) radio wave from a radio station. It ascends through the atmosphere without noticeable distortions. As we gain altitude the "air" gets thinner and more and more ultraviolet rays from the sun are able to penetrate down to where we are, to split charges, and the density of electrons and ions increases. The electron density in its turn influences a property of the plasma called the plasma frequency. When charges lie within the electric field of the wave, they accelerate. When this plasma frequency happens to coincides with the frequency of our radio wave, we get a situation similar to that of a forced oscillator. An uniform acceleration of several charges is induced in the plasma. This motion results in an imbalance of charge and electric fields that strive to wipe out the imbalance. This, in turn, pulls the charges back again. When we would have had equilibrium of charge, the charges still have momentum and continue their motion. We end up with yet another charge imbalance and electric fields as a counter effect. If these interactions propagate in space in the same direction as charges move, we have a longitudinal wave. These waves are called ion acoustic waves and Langmuir waves.

In the process, part of the energy and momentum of the electromagnetic wave is transferred to the plasma waves. What is left is still an electromagnetic wave, but one that moves in a the opposite direction and with a lower frequency. This process is called stimulated backscattering, but it is no longer held to be the entire truth of wave-plasma interactions.

At the end of the seventies, huge high power HF transmitting facilities were built, transmitting electromagnetic waves with powers high enough to induce turbulence in the ionosphere strong enough to be separated from the natural background radio noise. Through radar probing, the existence of enhanced plasma waves was soon discovered, at frequencies offset from the radar frequencies. In the Langmuir waves the electron density is raised at intervals of a wavelength and those high density areas reflect the radar beam. Thus, analogously to a crystal lattice, the wavelength of the Langmuir wave can be inferred through positive interference of reflected beams. This is known as the Bragg phenomenon.

Through radar probing one could also see that in the region into which the radio wave was pumped, a turbulence, reflecting radar beams as turbulent air reflects light, was induced. That is, not only a general increase in temperature as one had expected. And this is where our story begins. 

In the early 1980's at the Institute of Space Physics, Uppsala Division, Bo Thidé, Harald Derblom and Åke Hedberg did research on plasma phenomena, and of course having heard of the powerful HF transmitter at Tromsø, Norway, they wanted to take their shot at experimenting upon the turbulent regions of the ionosphere. In August 1981, Bo Thidé and Åke Hedberg headed for Tromsø; and the "heater". The division of labour was as follows: Bo Thidé would make an attempt at detection of the backscattered remainder of the original electromagnetic HF wave, foreseen by the theory of stimulated backscattering, which had never been done before. For this purpose he had constructed a narrow band quadrature receiver.

In the same campaign Åke Hedberg was investigating the turbulent volume of the plasma from Kiruna, Sweden through, the more traditional, radar techniques. The latest technology was used. For the emission, the Tromsø "Heating" facility transmitting high power (1.4 MW) radio waves, and for detection of the much weaker back scattered waves (of the order of nanowatts since the signals were received and analysed some 17 km away from the transmitter), the two scientists, with the help of the colleague Harald Derblom, had managed to borrow a brand new analogue signal analyser, HP3585A, that had a frequency span of 0-40 MHz and a dynamic range of almost 100 dB, yielding high quality spectra of the down-coming sky wave.

Theories and earlier observations of the plasma waves had lead to predictions as to the outcome of the experiments. Of course, in the process of stimulated backscattering, as in all processes ever occurring, both energy and momentum must be conserved. The energy of a wave is proportional to its frequency. Therefore, according to the theory of stimulated backscattering, the frequencies of the plasma wave and the backscattered EM-wave were to add up to the same frequency as the original pump wave. A simple linear picture of a basically non-linear phenomenon.
If these theories were correct, the backscattered waves were to be seen in a spectrum as a peak in intensity either within 30 Hz or several MHz below the original pump wave.

But this was found not to be the case. On the contrary, Bo Thidé could see nothing out of the ordinary in these parts of the spectra, perhaps a broadening of the pump wave peak and something that looked like background noise but at much higher levels than normal. Was it a new discovery, something that would shatter current theories, or was it a total failure?

During the mere forty minutes he had at his disposal for the investigation, Bo Thidé managed to quickly rebuild his measurement setup, allowing him to study a much broader range of frequencies than originally intended to see if the back scattered waves were to be found there. And there they were, an entire spectrum of HF induced waves at different frequencies, not at a few tens of Hertz but rather shifted up to a few hundreds of kilohertz from the pump wave frequency. These waves were emitted from the ionospheric plasma upon stimulation by a high frequency radio wave, they were later to be called "Stimulated Electromagnetic Emissions" (SEE). On the 31^st of August 1981 the first "perfect" SEE spectra were observed and recorded, proving the contemporary theories inadequate.

Entirely new theories of plasma interactions had to be formed, and that needed a more solid experimental background knowledge. What was the phenomena really about and under what conditions did it occur?

Further measurements lead to the discovery that the polarisation of the wave strongly influenced the generation of stimulated electromagnetic emissions. The pump waves used in the experiments were circularly polarised and depending on the direction of rotation of the field components, clockwise (ordinary) or anticlockwise (extraordinary) the effect could or could not be observed.

This indicated that the magnetic field was an important contributor to the observed scattering effect. A magnetised plasma is birefringent, the same way as some crystals are. Both here and in crystals polarised electromagnetic waves take on different paths depending on their polarisation. But where a crystal has different diffraction indexes for light plane polarised parallel to the optical axis and light plane polarised perpendicular to it, a plasma has different plasma frequencies for circularly polarised light with the rotation in different directions. The observation thus indicated that the observed scattering effects occurred at an altitude that was never reached by the extraordinary wave, since the higher plasma frequency causes the ray to be reflected already at lower altitudes.

Another important condition that was crucial for the SEE can be explained in terms of charges in gyro motion around the Earth's magnetic field lines. Magnetic fields bend the paths of charges so they move perperdiculary to the field lines, i. e. in circles or spirals around the field lines, of which the frequency depends on the charge and the mass of the particle in question. Pump waves corresponding in frequency to these natural gyro motions of electrons tend to further accelerate the gyro motion, just like a forced harmonic oscillator. In the case of HF pumping this in the end leads to strong turbulence in the space plasma. Here we have the mother of the arsenal of complicated wave-modes and the non-linear conversion between them that later were to become the theoretical model to explain the scattering effect.

In order to obtain the strong non-linear effects the pump wave has to exceed a certain threshold level. Several comparisons between low power and high power pumping were made in these first virgin experiments. The importance of higher energy levels to increase the non-linearity of the wave - plasma interactions were evident.

It was indeed an important week when the IRFU campaign, referred to as "Heating 1," took place, in the autumn of 1981. Old theories were shattered and the ground was laid for an entirely new way of viewing the interactions between a plasma and electromagnetic waves, in space as well as in the laboratory. 

But now we have skipped quite a few steps: first the gathered observations was to be presented to and debated by colleagues and scientist all over the world. Analysis, and new experiments to clarify certain points are still, and have been since almost twenty years, taking place, of which some of it is by the Swedish Institute of Space Physics, Uppsala. As time goes by, the discussions, analyses and experiments condense to new theories and models of the wave-plasma interactions, and these too are later to give way for yet newer theories.

It is fair to say that one can regard this first "Heating" project in 1981 as the seed to a new non-linear area in of plasma physics. It was also the discovery of a new technique to investigate the space plasma called stimulated electromagnetic emissions (SEE). This technique makes it possible to study the space plasma from the ground in a new way and it is nowadays used for studies of plasma processes and can even be used as a diagnostic tool to determine local properties of different space plasma.