EVENING-TRANSEQUATORIAL-PROPAGATION-IONOSPHERIC-PLASMA-BUBBLES

EVENING TRANSEQUATORIAL VHF PROPAGATION

Roger Harrison INTRODUCTION VHF transequatorial propagation (TEP) is so called because it involves the reception of VHF signals, or the making of VHF contacts, over very long paths that cross the geomagnetic equator. The frequencies are well above those supported by the familiar F-layer propagation at HF, and signal strengths are often much higher than would be expected for the distances involved. In addition, the paths involved are not those commonly experienced with VHF sporadic-E propagation, in length, geography and seasonal characteristics. There are two types, or modes, of TEP: afternoon-type TEP (aTEP) and evening-type TEP (eTEP), based on the hours they occur during a day. eTEP PATHS Paths cross the geomagnetic equator and range from 3000 to 6000 km (1800-3700 miles) in length, but longer paths have been recorded, e.g. a 144 MHz contact between I4EAT, Northern Italy (JN54VG), and ZS3B, Namibia (JG73), about 7800 km; observation of a 144 MHz ZE2 beacon (Zimbabwe) by DC3MF (Sth Munich, Germany), about 8000 km. The paths most regularly observed cross the geomagnetic equator within a small range of angles close to 90 degrees, and very nearly symmetrically spaced about the geomagnetic equator. Paths having an obliquity of 15 degrees or more experience considerably fewer occurrences, particularly in the bands above 50 MHz. Path terminals are generally situated in a zone between 30 and 55 degrees magnetic dip angle north and south of the geomagnetic dip equator. However, a station in one hemisphere will most often contact stations in a zone close to its geomagnetic conjugate in the opposite hemisphere. The map of the Asian-Australian sector shows examples of the paths observed (contacts made and signals heard). The short-dash line running from Singapore to the Darwin-Southern Japan path indicates an observation made by a Defence Science Establishment (now DSTO) scientist who received and recorded Darwin VHF beacons in Singapore one evening using a handheld antenna pointing skyward. The intriguing reason for this is explained later.



Given that paths to 8000 km have been established in the African-European sector, evening TEP contacts on at least 50 MHz, and conceivably 144 MHz, over such distances could be achieved between Australia and Japan, e.g. from Coffs Harbour (or Armidale) in Northern NSW, or Woomera in South Australia, to JA7 on the northern coast of Honshu, Japan. MOF FREQUENCIES Maximum observed frequencies (MOFs) for evening TEP extend to at least 432 MHz. No upper limit has been established. Since TEP was first reported in 1947 (Ed Tilton, World Above 50 MHz, QST, May and October 1947), MOFs for evening TEP climbed ever higher during each decade. Over the 1950s and 1960s, Australian amateurs and DX listeners observed VHF signals in the evening at frequencies



up to 70+ MHz. In the late 1960s, Stuart Kingan ZK1AA on Raratonga, Cook Islands in the South Pacific, observed NTSC chA2-A6 TV signals from Hawaii on frequencies up to 90 MHz. Until the mid-1970s, the highest frequency observed was the 102.7 MHz Defence Science beacon in Darwin recorded in Yamagawa, Japan. On 8 November 1976, around 0040 UTC, YV5ZZ in Venezuela noticed Mode A downlink signals from Argentina via Oscar 7 some 10 minutes earlier than normal AOS. On listening to the 2m uplink frequency, YV5ZZ heard LU7DJZ direct, over a 5000 km path (reported in World Above 50 MHz, QST, January 1977). Then, on 29 October 1977, YV5ZZ contacted LU1DUA on 144 MHz, a 5044 km path. Subsequently, on 24 February 1978, VK8GB in Darwin, Northern Australia exchanged reports on 144 MHz with JH6TEW in Kyushu, Japan, a path of 4950 km. On 28 April 1978, ZE2JV in Rhodesia (now Zimbabwe) worked 5B4WR on Cyprus on 144 MHz, a path of 6300 km. It appears the first recorded instance of evening TEP on 432 MHz was the reception of the ZE2JV beacon by SV1DH in Greece, on 20 March 1979, a path of 6300 km. eTEP TIMES AND SEASONS Evening TEP has a peak diurnal occurrence between 2000-2300 LMT (at the midpath, where it crosses the geomagnetic equator). It may open a little earlier, and extend past midnight on the lower frequencies, but openings on the higher frequencies rarely extend past 2300 local time. Openings peak around the equinoctial months of March-April and September-October, particularly for the higher frequencies. All this clearly apparent in the diagram showing the diurnal and seasonal characteristics of evening TEP.



On the lower frequencies, openings may still occur through the solstitial months. Evening TEP openings are more frequent during high sunspot years, although never disappear during solar minima years. Researchers have noted that the requirement of path symmetry about the magnetic equator is more important during solar minimum than at solar maximum. eTEP SIGNAL STRENGTHS Observed signal strengths for evening TEP can range from weak to very strong, although signals of weak to fair strength are more usual. Path loss can, at times, be substantially less than the freespace path loss for the distances involved, but is generally 20 dB to 50 dB greater than that. For a 6400 km path, for example, free-space path loss is about 150 dB. Signals as strong as -66 dBm (S9 being -73 dBm) have been observed (e.g. in the South American sector), while signals in the range -85 dBm to -110 dBm seem to be typical in the JA-VK sector. Signals rise in strength rapidly from the start of an opening, typically reaching a peak within 15-20 minutes, after which they plateau for a period before declining slowly.



For a given opening, signals are generally of higher strength at the lower frequencies. Evening TEP exhibits a characteristic 'fluttter' fade at rates up to 15 Hz on 50 MHz and rather faster fade rates on higher frequencies, often giving 144 MHz signals a 'raspy' quality. The fading can also be 'choppy', fading out completely for fractions of a second. The chart recordings of the Australian Defence Science beacons at Darwin, on 48, 72, 88 and 102 MHz, as received in Yamagawa, southern Japan, over 16-17 October 1970, illustrate many of the general signal characteristics of evening TEP. The time axis proceeds from right to left; the vertical markers are at 10 minute intervals. Note the steep rise as the path opens, the lower frequencies opening earlier than the higher frequencies, and the lower frequencies continuing after midnight, while the higher frequencies decline after 2300. Note also the rapid fading of about 5-10 dB, ranging up to about 15 dB, and the longer period fades of about 40 minutes to an hour between peaks (Kuriki et al, Journal of Radio Research Labs, Japan, Vol. 19, 1972).



Doppler shifts of 20 to 50 Hz on 50 MHz are observed, and similar Doppler 'smearing' or spreading of signals. On 144 MHz, Doppler shifts vary from +50 Hz ranging up to -350 Hz have been recorded, generally averaging around -100 Hz. However, Doppler spreading of more than 1 KHz has been recorded, giving rise to remarks that the signals "sound like auroral or moonbounce" reception.



The diagram shows the pattern of Doppler shift on the 144 MHz signal of ZE2JV in Harare (Zimbabwe) recorded by SV1DH in Greece for two eTEP openings during October 1980.



PROPAGATION MODE The propagation mode of evening TEP is via 'ducting' or 'guiding' through field-aligned 'bubbles' of depleted ionisation which thread the equatorial ionosphere and extend symmetrically north-south across the geomagnetic equator. The general details are illustrated in the diagram below.



Ionisation inside the bubbles can be very sparse - a 10th to a 1000th less than the surrounding F layer, while the bubble wall has a sharp boundary, where the ion (and electron) density changes rapidly between the inside and the outside of the bubble. A signal from a VHF transmitter at a favourable location in one hemisphere enters the end of the bubble at a grazing angle to the magnetic field at that point and then skids around the top wall until it exits the bubble in the opposite hemisphere where a station at a favourable location can receive the signal. These equatorial plasma bubbles ('EPBs') appear in the base of the ionosphere about an hour after the sun has set at that height (around 250-350 km). They rise up and extend rapidly along



the curved magnetic field lines, drifting eastward at speeds of 25-125 metres/second. Their upward motion is typically 125-350 metres/second; some have been measured rising at supersonic speeds of more than 2 km/sec! They rise to peak heights of 1500 km or more at the geomagnetic equator and their ends can extend to more than 40 degrees dip angle; i.e. south of Darwin and north of Yamagawa in southern Japan. The reason for the observed flutter fading and the Doppler shift and spreading is now abundantly clear! The diagram illustrates the general development of the EPBs and the consequent influence on signal elevation angles and path lengths.



The bubbles may be 40-350 km in diameter and successive bubbles may be 40-100 km apart. The walls are not smooth, except for the early phase of their development, are often bifurcated (e.g. like looking up an elephant's trunk), and are sometimes remain open at the bottom. When this latter feature is present, the propagation can 'leak' to locations below and to the side of the bubble. Indeed, there have been reports of such events. For example, as mentioned in Part 1, I saw a presentation given by Doug Fyfe, a Defence Science Establishment (now DSTO) scientist, who received and recorded Darwin VHF beacons in Singapore one evening using a handheld antenna pointing skyward. EPBs have diurnal and seasonal characteristics that match those of evening TEP, which is why eTEP is observed to behave as it does. The diurnal characteristics of eTEP at 44-48 MHz and 144 MHz over the JA-VK path are compared to that for EPBs (global average of satellite measurements) in the diagram below.



EQUATORIAL PLASMA BUBBLES To tell when eTEP propagation is likely on any given evening, you can look for the precursors, or the 'signatures' of equatorial plasma bubbles. There's the time-hounoured method of monitoring upper-HF and/or lower-VHF signals from suitable locations. For example, monitoring the JA2IGY NCDXF/IARU time-share 5-band HF beacon in Japan from around 1900 LMT (see http://www.ncdxf.org/beacons.html). Developing EPBs will cause multipath and and off-great circle propagation on the 21 MHz and 28 MHz beacon signals (or whatever else is being monitoring). Look out for characteristic sudden, deep fading patterns and phase distortion. When flutter fading arrives, you know there may be bubbles in the equatorial ionosphere. Fortunately, the Ionospheric Prediction Service Radio and Space Services vertical incidence ionosonde at Vanimo, on the north coast of Papua New Guinea, is ideally placed to detect the presence of EPBs. They produce characteristic range-spreading, or 'spread-F', on ionograms. Extra 'traces' appear suddenly on the ionogram, as shown in the examples in the attached set of images. The presence of spread-F on Vanimo ionograms is a necessary, but not sufficient condition of itself for VHF evening TEP.



The two left-hand ionograms show typical night time F-layer traces, while the two right hand ionograms show the start of spread-F five minutes later. In the upper right hand ionogram, a weak extra trace at the lower frequency end of the first F-layer reflection is just visible. In the lower right hand ionogram, the range-spreading is very obvious. The IPS ionogram viewer is online at: http://www.ips.gov.au/HF_Systems/1/3 LINKS AND REFERENCES Geoscience Australia maps http://www.ga.gov.au/map/ IPS ionogram viewer http://www.ips.gov.au/HF_Systems/1/3 IPS foF2 near real-time ionospheric map http://www.ips.gov.au/HF_Systems/1/4



50 MHz F2 Propagation Mechanisms J. R. Kennedy K6MIO/KH6 Gemini Observatory ∗ Hilo, HI Introduction The possibility of worldwide F-layer propagation is a particularly intriguing part of the challenge of six-meter operation. Even the casual six-meter operator will soon notice that there are some fairly mysterious things going on when it comes to ionospheric propagation. The more seasoned operator will notice that there are a number of patterns that are prevalent, but it is still very difficult to predict when the band will open, especially on a day-to-day basis. Unfortunately, there are no simple answers to this dilemma. Nevertheless, there are some pieces to the F2 puzzle that are known and understood, and some clues to those that remain mysterious. In order to understand (however imperfectly) when the band will open, it is essential to have some understanding of why the band will open. A discussion of why signals propagate has to begin with some basic facts about how radio waves behave in the ionosphere. There are three basic elements that critically effect this propagation: 1. The amount of ionization present, 2. The angle of attack of the incoming signal to the ionosphere, and 3. The presence of large or small irregularities in the ionization. These factors play key roles in the success or failure of a communications path via either E or F layers. Although there are many external things that influence the status of the three conditions, in the end, it is the combination of these three that make or break any path. The way in which external events effect these three parameters determines what kind of propagation will occur. Six-meter F2 propagation is a very improbable event, from a statistical point of view. While this may seem obvious, it has a very important consequence. "Unlikely" events in complex physical systems are often the result of a combination of factors, some of which also may be fairly unlikely. This is certainly the case with most six-meter F2 activity, where propagation is almost always at or very near the ultimate edge of what is possible. The task of predicting band openings generally involves predicting not just one event, but the coincidence of several events, and not always the same ones nor in the same combinations. In truth we do not yet know what all of the factors are, much less how they interact. On the other hand, there are a number of things that are known to be significant and most of them involve the Sun in one way or another. The Ionosphere and the Sun The Earth's atmosphere extends from the surface to heights well in excess of 1,000 km. The density of the static atmosphere is highest near the surface and decreases progressively as one goes upwards. Most of the atmospheric mass is located very near the Earth's surface, with more than half of the mass contained within just the first 6 km. The Sun, on the other hand, pours its radiation down on the atmosphere from the topside. Thus, the upper regions of the atmosphere receive the full brunt of the Sun's ultraviolet, x-rays, cosmic



rays, and so forth. This radiation interacts with the atmosphere on its way down, and as it does, portions of the radiation are absorbed at different levels, producing ionization in the process. The interactions between solar rays and the air molecules and atoms are very complicated. There is a whole complex of chemistry that can take place at these rarefied heights that normally has no importance at lower levels. What wavelengths are absorbed at what levels is determined in part by the chemical components and particle densities found at each level. For example, generally ultraviolet radiation is absorbed fairly high in the atmosphere, while x-rays penetrate somewhat lower, and cosmic rays go still lower. When solar photons get absorbed it is a result of a collision with an atom or molecule. Often such collisions occur with sufficient force to knock off one or more electrons from the atom or molecule struck. This leaves behind a missing (absorbed) photon and positively and negatively charged ions. Usually the positive ions are the comparatively heavy cores of the atoms or molecules and the negative ions are the relatively light electrons. It is the electrons that play the dominant role in radio propagation. 400



300



200



100 Height (km)



Electron Density, N (m-3 ) Figure 1. A typical plot of the daytime electron density as a function of height above the ground, showing the C, D, E, F1, and F2 layers. Since different wavelengths are absorbed at different heights, the solar radiation leaves behind several distinct layers that are characterized by enhanced levels on ionization. The F2 layer is the highest of these, both in terms of distance above ground and in terms of electron density. This layer extends from roughly 250 km above the Earth to well over 500 km on occasion, with a peak daytime electron density in excess of 1012 e/m3 around 350 km. Extreme-ultraviolet radiation (EUV) from the Sun is the main source of ionization in the F region. Table 1 below shows typical characteristics of the various layers. It should be noted that the “boundaries” between the layers are ill defined and for sake of brevity the list of ionizing radiation is not entirely complete.



TABLE 1 Ionospheric Layers Layer Height Density Ionizing (km) (e/m3) Radiation C D E F1 F2 30 60 90 120 250 60 5 x 109 Cosmic Rays 90 1 x 1010 Hard X-Rays 120 8 x 1010 Soft X-Rays 250 5 x 1011 Extreme Ultraviolet 500+ 3 x 1012 Extreme Ultraviolet



Ionospheric Radio-Wave Propagation The Amount of Ionization – When an upward moving radio wave reaches the ionosphere, the electric field in the wave forces the electrons in that layer into a sympathetic oscillation at the same frequency as the passing wave. A certain amount of the wave energy is given up to this mechanical vibration of the electron cloud. As a result, the passing wave gets weaker. At this point there are two sorts of things that can happen. In the lower atmosphere, the total number of particles may be dense enough that the oscillating electrons collide with other particles almost immediately (say, in less than one RF cycle). When that happens, the wave energy that was converted to the mechanical electron oscillation is now converted as heat to the atmosphere before anything else can happen. This energy is lost forever as far as the wave is concerned. This is another way of saying that the radio wave energy is absorbed. This is precisely what happens to signals below about 10 MHz when they encounter the daytime D layer. The average daytime frequency of collisions between particles in this region during the day is about 10 million collisions per second. So, it strongly absorbs radio frequencies below 10 MHz, but has a progressively smaller effect as the frequency gets higher. At the other extreme, if the collision frequency within the layer is significantly less than the wave frequency, and if the electron density in the vibrating cloud is greater than a certain critical value, then the whole cloud can act more like a static reflector. Almost all of the wave energy is given up to the vibration in a very short distance, all the electrons vibrate in phase, and together they reradiate the original wave energy back downward again. Thus, the wave skips off the ionospheric layer, hopefully to be received at some distant point. It should be noted that in most cases what actually occurs is something intermediate to these two extremes. Some amount of absorption always occurs. Moreover, as Figure 1 shows, when the electron density does exceed the critical skip value, it does so gradually. Rather than a discontinuous reflection, there is a refraction of the wave, a more gradual bending back around toward the ground (sporadic E is an exception, it is usually very close to a true reflection). Even when the critical density is not reached (and the wave passes through the ionosphere and escapes into space), a certain small fraction of the wave is reradiated by the vibrating cloud and some of this new signal goes downward as a form of ionospheric scattering.



Notwithstanding the fact that the F2 layer retains some of its ionization even at night, all of the foregoing should make it clear that six-meter F2 propagation is normally a daytime affair, unless one lives in the tropics (but more on that later.) The Angle of Attack – The above example only tells us what happens if the path of the wave strikes the ionosphere with an angle of attack of 90° (i.e. going straight up). In the more general case, if skip is to occur, the Maximum Usable Frequency (MUF) that will bounce is determined by both the maximum electron density that the wave encounters in the ionosphere and the angle at which the wave hits the reflecting/refracting layer. If a signal that is sent off very near the horizon (that is, with a zero angle of radiation), due to the curvature of the ionosphere around the Earth, the signal will normally hit the ionosphere at an angle of attack between 10° to 20°. The exact value depends on the height of the ionospheric layer and exact angle of radiation.



Radio propagation is not the only place one sees grazing-incidence reflection effects. For example, everyone knows that if you toss a rock in a lake it will break through the surface of the water and sink, never to be seen again. The same thing happens to a radio wave when it collides with the ionosphere and its frequency is above the MUF – it breaks through and disappears. However, even a rock can be made to bounce off the surface of the lake – any child can do it. The secret, of course, is that it must hit the surface at a very shallow angle, that is to say, at grazing incidence. Skipping radio waves is essentially the same as skipping stones. If the ionosphere were a smooth sphere, simple geometry would show that M ≈ 3.4 at the F2 layer. Since Table 1 shows that the average E-layer ionization is 40 times less than the F2 layer, one might rightly wonder why sporadic E propagation is so much more common and frequently produces much higher MUF’s than the F-layer. Half of the answer is that the E layer is closer to the Earth and this makes the angle of attack much smaller than for the F layer. At E-layer heights, Figure 2 shows that M ≈ 5.4. Thus, the Elayer MUF will be nearly 60% higher than the F layer for the same number of electrons. (The other half of the ES story is that the “sporadic” process also increases the amount of ionization in thin localized regions to levels considerably higher than the average in the E layer as a whole. Taken together, the two effects can produce very high E-layer MUF’s.) It is important to note that the angle of attack is also effected by the radiation angle of the antenna. Especially at six meters where one is struggling to get the MUF up high enough, a low angle of radiation from the antenna can be very important. Here, it is not just a matter of getting the longest skip distance, but one of getting any skip at all. F-Layer Propagation for All Seasons (Except Summer) The amount of ionization in a given layer at a given time depends on the dynamic balance between those processes causing ions to be produced and those causing ions to be lost (e.g. by returning to their original neutral state). Put another way, the ion density depends on the amount of radiation arriving from the Sun causing the production of ions, minus the loss of ions due to electrons being recaptured by positive ions. The rates and mechanisms of these processes vary widely from layer to layer.



For example, the underlying (neutral) density of the D layer is much higher than that of the F layer. Consequently, in a given length of time, electrons and positive ions there are much more likely to collide and recombine. As a result, by comparison to the F layer, the maximum D-layer ionization is held down during the day by collisional losses, and the D layer disappears in minutes when the Sun goes down. By contrast, the underlying particle density in the F layer is much less than the D layer and ions last a lot longer in the F layer. Even late at night, often there are enough ions left to support HF communications. Another important effect in ion production is the angle with which the sunlight strikes on the top of the Earth's atmosphere. When the rays arrive at a large angle to the vertical the energy is spread out over a larger area, and thus the energy density at any one spot is reduced. So, fewer electrons are produced at sunset and sunrise than at high noon. In addition to this diurnal effect, there is a seasonal one as well. The local winter hemisphere is tipped partly away from the Sun and its various layers receive proportionately less radiation. Consequently, in the Northern Hemisphere fewer daytime F-layer electrons are produced at a given time of day in January than are produced in July. However, remember that production is only half the equation. There is a peculiarity in the F2 layer, not found in the other layers, called the winter anomaly. Although daytime ion production is higher in the summer, there are seasonal changes in the molecular-to-atomic ratio of the underlying (neutral) atmosphere that cause the summer ion loss rate to be even higher. The result is that the increase in the summertime loss overwhelms the increase in summertime production, and total F2 ionization is actually lower, not higher, in the local summer months. Put the other way around, daytime F2 electron densities, and thus MUF's, are higher in the local winter. Measured at the "half-width", the winter peak starts in October and lasts until May or June. Most years there is a fairly flat peak between December and April. It is interesting to note that during the peak years of solar cycles 18 through 21, October was almost always the beginning halfmaximum, whereas the ending half-maximum varied from April to July. The ending months were pretty consistent within a given cycle, but cycles with higher maxima favored May with an occasional April, while those with lower maxima favored June with an occasional July. Interestingly, the winter anomaly shows the most seasonal fluctuation near solar maximum. Here the local winter daytime MUF's are twice as high as the daytime summer values, while they are only about 20% higher during solar minimum. The central message in all of this is that, on average, F2 propagation between points on the same side of the equator will be much better in the local winter and near solar maximum.



Figure 3. A schematic plot of the Northern Hemisphere Figure 4. A schematic of the overlap of the northern and variations of MUF due to the winter anomaly. The values southern winter anomalies in the western hemisphere, of MUF shown are typical mid day figures for mid latitudes as modified by the effect of the (magnetic) equatorial near solar maximum. anomaly. The shaded areas show the most likely times for mid latitude north-south transequatorial multihop. If one is interested in multihop along generally north-south paths, then the winter anomaly comes into play in another way. If it is wintertime in one hemisphere, it will be summertime in the other. So, for example, in January the first hop of a two-hop path from North to South America might make it, only to have the second hop fail due to the absence of the winter effect at the second skip point. Obviously, the best times of year for such a path would seem to be in the Fall and Spring when the winter anomaly effects overlap a bit in both hemispheres. The winter anomaly is not the only seasonal effect. Many two-hop (or more) north-south openings on six meters seem to have no evidence of stations at the end of the first hop. This is often due to the ionospheric equatorial bulge known as the equatorial anomaly. Within ±20° of the Earth's magnetic equator there is a pronounced outward bulge in the ionosphere. Though generally regarded as an afternoon or early evening phenomenon, it occurs at other times as well. It is thought to be produced by a combination of a persistent thickening of the F layer near the equator and a daily fountain effect. This afternoon fountain apparently is the result of a build up of west-toeast electric fields in the equatorial E layer. In combination with the Earth's magnetic field and ionospheric winds, these fields pump electrons upward from the E and lower F layer into the upper F2 layer. Thus, significantly enhancing F2 layer electrons. It was pointed out earlier that the angle with which the radio wave encounters the layer is also a factor in determining the MUF. The equatorial bulge produces two regions, one north of the equator and the other south, where the ionosphere is systematically tilted. Of particular interest are the points at the cor-ners where the generally spherical ionosphere is bent upwards to form the bulge.



Figure 5. A diagram of a transequatorial chordal hop off the tilted north and south skip points. These points lie about 20° north and south of the Earth's magnetic equator and can be major contributors to daytime transequatorial six-meter propagation from mid latitudes and clearly cause nighttime TEP in the tropics. This upward tilt is such that an upcoming wave hits the near corner at a shallower angle of attack to the tilted layer than it would to the usual spherical layer. This means that it will have a higher MUF for the same value of electron density. That is, the M factor is larger than 3.4, perhaps by quite a bit. The wave is not bent all the way back toward the ground. However, it is bent enough to cross the equator and hit the tilted layer on the far side, without ever coming back to Earth. This so-called "chordal hop" to the second tilted region produces another small-angle reflection that can be just enough to return the signal to the Earth on the far side of the equator. The effect is a kind of double hop with nothing in between, and at a higher MUF than could otherwise be supported. Since the "bulge" effect does produce higher MUF's, this is often the cause of multihop north-south propagation of six meters. Moreover, it is a fairly low loss path. The wave never comes down at the path midpoint, so it avoids two passes of D-layer absorption that normal double hop would have encountered. In order for this form of propagation to function, both the north-side and south-side tilted regions need to be ionized enough to make the path work. Again, if either one is insufficient to skip then the whole path fails. At HF there is often a fair amount of margin, but not at six meters. Here usually the best chance occurs when both sides of the equatorial bulge are equally illuminated by solar radiation, and this situation only occurs around the autumn and spring equinoxes when the Sun is most nearly over the equator. Normally one would think that this would occur in late September and late March. However, the Earth's magnetic field is tilted with respect to the geographical coordinates. In the Western Hemisphere the magnetic equator is as much as nearly 11° south of the geographic equator. This means that the "magnetic equinoxes" occur about a month or more earlier (August and February) for paths between North and South America. From the point of view of a station in North America, as the path of interest swings around further east or west, the magnetic equinoxes are more nearly the same as the geographical ones. Consequently, the date of the magnetic equinox is



dependent on the amount of east or west component included in the north-south path, and where you happen to be on the Earth. There is also an interaction between the balanced illumination at the magnetic equinoxes and the double-hop winter anomaly effect (which is geographical, not magnetic). For example, consider a path between North and South America. Here it should be noted that, in part as a consequence of the location of the magnetic equator, MUF's are frequently higher over South America than North America. During the Fall equinox period, on the northern side of a path, August is too early for much help from the (northern) winter anomaly. By contrast, the southern skip point is bolstered by both the winter and equinox effects and has a substantially higher MUF than the northern point. If this path is going to succeed it will likely have to wait until October when the northern skip point is in better shape. Put another way, by October the northern and southern skip point MUF's will be about the same, as they pass each other going in opposite directions. During the Spring period, February, as well as March and April, are all still in the peak for the north end. Even though the southern point is in the summertime, it has a higher MUF to begin with, and the equinox effect adds even more. The result is that the southern point MUF may well be higher than the northern point MUF, despite the absence of help from of the winter anomaly. In the meantime the northern point is as good as it will ever get, because of the winter anomaly. Consequently, for generally north-south paths across the magnetic equator, periods more nearly centered on October and March are generally the best, and often the Spring gives more consistent daytime transequatorial performance. Of course, unless one has an E-layer link up, or some other funny business, another requirement is that the stations each be close enough to the nearest ±20° tilted skip point to illuminate that point by radio line of sight. Based purely on geometry, the southern half of the U.S. has an advantage for South American paths and the southwestern states for the South Pacific. (Early Spring sporadic E can make a big difference on the mainland US.) The Solar Cycles For reasons that are still unknown, the general background magnetic field of the Sun reverses polarity every 11 years or so. Thus, the Sun experiences a 22-year magnetic polarity cycle of north to south to north again. This effect is accompanied by a cycle of solar activity that reaches a peak approximately every eleven years. The peak itself can be fairly broad, having significant effects for three or four years. The solar activity cycle is seen in virtually every kind of signal we can receive from the Sun, from radio waves to x-rays. Not surprisingly then, the amount of ionizing radiation impinging on the atmosphere varies with this same pattern, including the EUV that is the principal source of the F2 layer. Consequently, propagation is decidedly better near solar maximum, but the seasonal effects are still superposed on the general enhancement seen during the solar maximum. There is a second kind of variation due to the fact that the Sun rotates on its axis every 27 days or so, coupled with the fact that "activity" on the solar surface is generally confined to a few specific regions at any one given time. As a result, if the Sun is active at all, it is quite common for one side to be active and the other side relatively quiet. As the Sun rotates there is often a very pronounced 27-day cycle in the radiation reaching the Earth. It should be noted that the active solar longitudes change over time. The 27-day cycle of activity commonly repeats for several cycles and then briefly is interrupted as old solar active regions fade and others emerge. When new active regions do develop, typically at some other longitude, the



cycle will be reestablished, but with a different phase. In other words, knowing that a particular two-week period was active last month is a pretty good predictor that the same two-week period this month also will be active. However, it is a very poor predictor of activity during the corresponding period six months from now. There is no doubt that during solar maximum, and especially during periods of high activity, the amount of EUV reaching the ionosphere increases substantially. In principle, this should mean better propagation. People have tried for some time to get direct measurements of the EUV radiation with an eye toward making short-range predictions of propagation conditions, but so far these have not been very successful. Very little of the F2 producing EUV reaches the Earth's surface, precisely because it is absorbed making ions in the F layer. A number of spacecraft have carried EUV sensing instruments, but generally these detectors are susceptible to damage from the very radiation they wish to measure. As a result, their sensitivity changes in time, making accurate, long-term, absolute measurements very difficult to get. For many years scientists have been using the 10.7-cm solar radio flux as a proxy for EUV emission. This radiation is formed at about the same level in the Sun as the EUV, and has a similar temperature sensitivity. Under relatively quiet solar conditions the 10-cm and EUV fluxes track pretty well. However, there is a component of the 10-cm emission that is also sensitive to other forms of activity including those that produce x-rays. Consequently, during periods of high activity 10 cm is not a good linear indicator of the strength of the EUV radiation reaching the F layer. In fact, it tends to overestimate the EUV considerably. It must be said that while the 27-day effect definitely influences propagation through flares and such, many people (including the author) think it is not as profound as is generally thought. There is a good correlation between the long-term average of the 10-cm flux and F2 propagation (as there is with sunspots, flare counts, and many other activity measures). Thus, if the flux is high on av-erage, month after month, propagation will probably be good. However, vertical incidence ionograms show virtually no day-to-day correlation between 10-cm fluctuations and the measured critical frequencies (fc’s) that would signal the expected MUF’s. This is not to discount keeping track of the 10-cm flux; it is a useful indicator of the general level of solar activity. Unfortunately, during solar maximum one never knows whether high flux numbers mean high EUV and high F-layer MUF's, or high x-rays and high D-layer absorption. It is certainly true that long periods of time with high flux numbers will contain periods of good propagation, but it is very hard to say whether the openings will come on individual days when the flux is 300 or on days when it is 150. The individual days with high fluxes are a lot less important than whether there were some days with high fluxes within the last 30 to 40 days. Other Solar Effects Perhaps the most talked about solar events are flares. These more or less random high-energy outbursts produce a variety of effects, and no two flares are ever exactly alike. Flares generally occur in active regions and, if they are to effect the Earth, the active side of the Sun must face the Earth. To this extent they are weakly predictable. There is also a mysterious 157-day period



associated with flares, and most other solar activity measures, that so far has been ignored by most propagation prognosticators. Some suspect that this is related to a periodic effect in the emergence of new active regions on the Sun and the resetting of the phase of the 27-day cycle. The impact of solar flares is very unpredictable. Generally there is a large outburst of x-rays associated with big flares. This usually produces an almost immediate increase in both the amount and maximum frequency of D-layer absorption. This often results in widespread "blackouts" of the HF spectrum on the daylight side of the Earth that may last for many hours. There may also be outbursts of EUV in some flares that produce a very rapid response in the F layer. However, while the x-ray flux may change by a factor of 100 or even a 1,000 in a major flare, the EUV may only go up by a factor of 2 to 5. Sometimes excellent six meter openings do result from a flare, provided that the D layer does not get in the way (it generally does not). At other times flares seem to hurt rather than help. Of course, another possible effect of flares is geomagnetic disturbances. All flares blow some material away from the surface of the Sun. If the trajectory of this material takes it to the Earth (it often misses), then it will arrive within a few hours to a couple of days and may produce profound disturbances in the geomagnetic field. This can be either good or bad for F2 propagation. Whatever the conditions were before the particles arrived, it often means things will then change. Storms effecting the F layer often have a positive and then negative effect on the electron densities, and hence the MUF, on the time scale of several hours. At mid latitudes, electron densities can climb as much as 20% above the ambient and then drop to 30% below the ambient all over a period of 24 hours or less. On the other hand, in equatorial regions a general enhancement of 5 - 10% is often seen, with no pronounced negative effect. Transequatorial Propagation (TEP) Large-Scale Irregularities – One of amateur radio's many propagation discoveries was that stations located close to the magnetic equator are often able to communicate at 50 MHz by way of the F layer in the dead of night, over long paths that cross the magnetic equator. The basic mechanism is the equatorial anomaly discussed earlier. That is, the afternoon fountain effect causes enhanced electron densities and tilted layers to form within 20° of the magnetic equator late in the day or early in the evening. These conditions can persist long into the night with some contacts taking place long after local midnight. They readily provide near grazing-incidence chordal hops at six meters. For operators who have the good fortune to be in the TEP zone, the paths themselves do not have to be especially north-south. In the simplest case, the two stations are on opposite sides of the magnetic equator, although they can be at a considerable angle to the north-south line. All that is necessary is that the two ±20° corners be at usable chordal skip points. In reality, many contacts are made using TEP a form of side-scatter mode. If the two stations are substantially east or west of each other (in geomagnetic coordinates) their signals will enter the region between the two chordal skip points at a considerable angle to a north-south line. When this happens, the signals can bounce back and forth within the north and south walls of the equatorial bulge, using the bulge as a duct. As depicted in Figure 6, signals can be thought of as zigzagging north and south in the short term, but generally moving along in an east-west direction under the bulge until they find a weak point and break out. From there, they may go either north or south, depending on which side they find



the “door” out, irrespective of whether the signal originally entered the duct from the north side or the south side. SOME COMMON TEP PATHS Typical examples of across-the-equator TEP include the nighttime pipeline that often exists between Hawaii and Australia. Geomagnetically, this is essentially a north-south path. Usually signals are pretty clean, and quite strong. On the other hand, it is not that uncommon for Hawaiians to hear Japan at the same time – and on the same beam headings as Australia – on what sounds like backscatter. Japan and Hawaii are on the same side of the equator and mostly east-west of one another. Finally, propagation across the equator, but largely along the TEP zone, can produce very strong signals, such as the link between Hawaii and South America. However, many times these signals are strongly modulated indicating significant scattering within the duct. Long Path Magic Nothing illustrates the effect of the angle of attack better than long-path (wrong-way) propagation. At first glance one might expect that long-path links would be doing it the hard way. However, this is not always the case. No matter which direction one points an antenna, if the path goes at least half way around the world, one is assured of crossing the magnetic equator at least once. Consequently, there is at least one opportunity for a chordal hop, with its elevated MUF and absence of D layer absorption. TEP ASSISTED LONG PATH



Figure 7. TEP can provide the launching points for shallow attack angle grazing hops that cover long distances, with higher than normal MUF’s and low absorption. At such times, long-path can be a superior mode of propagation



More importantly, if the chordal hop is at the beginning of the path, then the signal may be injected into the ionosphere at the end of the hop at a very shallow angle. If it is shallow enough, it will continue skipping around the ionosphere in a series of short grazing-incidence hops, as described in Figure 7. Although it can happen at any latitude (especially if aided by, say, a sporadic-E link up to the TEP zone), stations within the TEP zone itself have the most frequent opportunities to experience this kind of propagation. Consider the case of a path starting in Hawaii and ending in Spain. The long-path link passes southwest from Hawaii, over Australia, Antarctica, Africa, and finally to Spain. The key factor here is that the first and last hops are off the equatorial bulge. If conditions are right at both ends (and in the middle) the chordal hop might be shallow enough that, when bouncing off the southern edge of the anomaly, it never comes down to Earth. Instead it continues to bounce like a rock skipping across a lake as the curving ionosphere keeps coming back to meet it again. If the same conditions exist at the magnetic equator over Africa, as those south of Hawaii, the shallow skipping wave will finally be bounced down out of the ionosphere by the northern edge of the bulge, landing in Spain. Since there is little D layer absorption and the MUF’s are very high due to the angle, the long path is actually possible, while the short path, with its completely traditional earth-sky-earth hops, is completely out of the question. The effect does not always require the equatorial bulge to launch or retrieve the signal. Any condition that produces a tilted layer, or even intense scattering, could produce the same effect – the equatorial anomaly is just the most dependable. Gray Lines and More Bumps By now it is clear that, at six meters, it often takes all you can get to produce an MUF high enough to support communications. Another point that should be clear is that, at the margins, tilted ionospheric layers can be critical to producing a band opening. Another way of achieve a tilted layer is simply the effect of the rising and setting of the Sun on the amount of ionization in the F layer. Figure 8 shows a cartoon of this phenomenon. On the night side of the twilight zone, the reflecting level for a given frequency tends to be higher up – since the particle density is lower, collisions are less frequent and thus the ionization lasts longer without the Sun. On the daylight side, the Sun has begun replenishing the ions so that the amount required to skip is now available at a lower level. The net effect is two tilted twilight layers that constantly move around the Earth once a day. For example, at times contacts are made between Hawaii and South Africa. The path goes right over Australia and the band is commonly open to Australia at the same time. To their great frustration, the Australians hear nothing (but excited KH6’s). This condition typically occurs around 0800Z (2200 Hawaii, 1800 Australia, and 0900 South Africa). On the Hawaii end, the signals are launched and retrieved by the equatorial anomaly. The curved surface of the southern edge of the bulge sprays some of the signal down a “low-road” toward Australia, and some of the signal stays up close to the ionosphere and continues on westward on a “high road”. At the same time, the sunset gray-line bump is over eastern Australia. The high-road signal makes a second chordal hop off the gray line, thus firing the signal into the daylight to the west. By the



time the high-road signal comes to Earth the first time, it will have traveled more than 10,000 km – more than 60% of the way to Africa without ever touching the ground. The Australians don’t have a chance because the African signals never came to Earth there (most of the time, anyway).



Figure 8. This sketch shows a possible long-path circuit with grazing incidence hops around the night time side of the Earth. The signals are launched and retrieved from the grazing hops by gray line bumps on both ends of the path. In the case of the Hawaii – South Africa path described above, and the Hawaii - Spain path described earlier, there may well be another tilted layer or ionospheric bump that plays a role. Having come 10,000 km on two chordal hops, if nothing else strange happens, the signal going west still has 6,000 km to go, requiring at least two conventional daytime hops. However, it will also pass relatively near the South Magnetic Pole in the process. Satellite measurements of the F layer near the poles can show significantly tilted layers there, as ions organized themselves along the magnetic field lines that dip nearly to the vertical near the pole. Some of these structures resemble the equatorial bulge, but on a somewhat smaller scale. It is conceivable that these layers might produce a third chordal hop and take the signal all the way to the end of its path before it comes to Earth. Scatter Small-Scale Irregularities – Large-scale bumps and ducts are clearly irregularities in the ionosphere and, as noted, they can cause skip conditions that are rather different from the simple picture of earth-skyearth hops. It is also true that smaller-scale irregularities can produce interesting effects as well, particularly when they occur in great numbers.



Despite the effort to separate the different effects in this presentation and deal with them one at a time, there already have been several references to scatter in the preceding material. This is a consequence of the fact that long-range propagation on six-meters often is the result of a combination of effects. Scattering plays a role in many paths; some effects are positive and others negative. In what follows it will become apparent that scattering is inexorably connected to the issue of tilted layers. In traditional picture of skip, there is a single reflecting or refracting layer of very large horizontal size. Ionospheric scatter differs from skip in that the size of the reflecting or refracting “layer” is usually rather small, but there are many of them. Scattering occurs when a signal encounters a large number of “scattering centers” that are larger than about half a wavelength across. Each of the individual scatterers can be thought of as a ball-shaped bubble of ionized gas. The sizes of these bubbles might be anywhere from a few tens of meters up to several hundred kilometers. When a radio wave encounters a round bubble, due to the shape the wave is bounced in all directions (not just in one direction as a flat layer would), thus the use of the word “scatter”. If there is a large number of the scattering centers, then enough of the signal might be reflected to be readable at some distant point. Of course, since these bubbles are all at different distances from the transmitting (and receiving) site, the scattered signals arrive at the receiving site with different phases, leading to a lot of garbling due to massive multipath interference. Moreover, as will be seen, in the ionosphere, these centers are normally moving, adding Doppler shift to the witch’s brew of funnies. The ball shape of the scatterers not only reflects the signal in many directions at once, but it also means that, depending on the exact direction of the reflection, there is wide variation in the angle of attack. Everything from straight out and straight back to grazing incidence will be present. Signals going in some directions will have low MUF’s and others will have very high MUF’s. Thus, while the quality of the signal may suffer due to scattering, it may also make communication possible at VHF. There are two magnetogeographical regions where ionospheric scatter is fairly common. One is in the magnetic tropics and the other is near the magnetic poles. In the tropics, this effect is intimately associated with the equatorial anomaly and the afternoon fountain. The great ionospheric updrafts that move electrons from the E and F1 regions up into the F2 region produce enormous plumes of turbulent plasma that become aligned with the magnetic field lines. These plumes are composed of a large number of rising bubbles of plasma, in a way looking like a bottle of soda water just after the cap has been removed. In reality, this is what causes the equatorial bulge. The region within the walls of the bulge is filed with scattering centers. While coherent skip can occur at the lower corners of the bulge, signals entering the bulge itself have significant scattering opportunities. Driven by the afternoon fountain, as one would expect this is basically a nighttime effect. Generally, speaking the ionospheric winds in this region cause the plasma to drift eastward. This often adds a systematic Doppler shift to the signals. When this region is probed with an ionosonde (a transmitter that sends signals straight up to measure the critical frequency) instead of the returned signal showing a sharp single F layer echo, the display shows a huge diffuse region of echoes ranging from the expected floor of the F layer up to above 800 km. This condition is referred to as Spread F.



Spread F scattering in the tropics is seasonally enhanced near the equinoxes and negatively impacted by magnetic disturbances. Magnetic storms will completely suppress the effect. It was mentioned earlier that there are tilted layers near the poles and that some of these look rather like small copies of the equatorial bulge (although their magnetic field alignment in nearly vertical, instead of horizontal like the equatorial region). Here again, one finds Spread F scattering regions. Near the poles, the scattering centers are often found during the day, as well as the night. The condition is present most of the time, although the equinoxes are favored. It is less pronounced in the winter and summer months. It seems to be enhanced near solar maximum. This effect is responsible for the heavily modulated signals often encountered in polar crossing paths. The effect is very rare between 20 and 40 degrees in latitude (north and south). However, it does appear in the mid latitudes poleward of 40 degrees. In this region it is almost always associated with a magnetic storm. One variation of the scatter picture doesn’t involve scattering from ionospheric irregularities at all. Ground backscatter can occur when a traditional ionospheric reflecting layer causes a signal to come back to earth at the end of, say, its first hop. If is strikes a large irregular surface on the ground, for example a mountain range, it may then send some of the signal to skip back in the direction from which it came. This will allow it to be heard by stations that are generally near the transmitting station. Some Final Comments and Conjectures The three basic ingredients for a propagation path are ionization, angle, and irregularities, in the right combinations. If we restrict our thinking to the HF spectrum it is easy to explain the bulk of the propagation in terms of ionization alone. This is simply because there are generally enough ions to make some path work at almost any angle. As stated earlier however, at six meters, F2 usually relies on combinations, and combinations of combinations. While the bulk of the previous discussion has been focussed on F2, there are a number of ways that a signal can get to, or from, the F layer. For example, if the F-layer ionization was not quite enough to support the angle for an F2 hop directly from the ground, the partial bending of the wave on the way up by an E-layer cloud (itself not sufficient to cause a reflection) might cause the wave to hit the F layer at a shallower angle and thus produce a high enough M factor to make the F hop work after all. Similarly, a tilted E layer could produce the same effect. On the other hand, a real reflection from an E layer cloud might skip a signal, originating from a station far from the magnetic equator, and bring it close enough to the equator to take advantage of the equatorial anomaly. Another possibility is the lengthening of a two (or more) hop F2 path by a reflection from the topside of an E cloud in between the two F hops. This avoids the two intermediate passes of D-layer absorption, as well as any scattering of the signal by the ground at the path midpoint. The role of medium-scale F-layer distortions is a very poorly understood area. While we often construct pictures of the ionosphere as a smooth spherical surface, this is an idealized picture that cannot account for much of what is commonly seen in six-meter F-layer propagation. The F layer is a convoluted surface affected by the motion of the subsolar point, ionospheric wind patterns, and even tropospheric weather.



There are a variety of traveling-wave-like structures that are known to exist that can produce moving "ripples" in the ionospheric surface. Similarly, there is evidence to suggest that from time to time there are smaller localized bulges or "domes" not associated with the equatorial anomaly. Any effect such as these that produces tilting in the F layer has the potential to locally raise the MUF because of the angle of attack M factor effect. The sources of these distortions are not clear. When the other contributing factors to the 50 MHz MUF are not quite enough to open the band, these subtle effects, no doubt, can play an important role. But since they relate to a kind of ionospheric weather situation, they are very hard to predict with our present knowledge. Even random irregularities in the E or F layer can have an effect. These can produce scattering that again can inject (usually weak) signals into the F layer at angles more suitable for subsequent skipping. Those who listen for 48 and 49 MHz television carriers as band condition indicators often hear signals that arise from the combination of high power and scattering injection. Summary When F2 propagation occurs on six meters it is usually the result of the coincidence of a number of different phenomena that, together, produce suitable combinations of ionization levels and radio-wave angles of incidence. Solar extreme-ultraviolet radiation is the principal source of ionization. Consequently, the Sun is the source of many of these effects. However, the full range of operational factors and their modes of interaction is not known. The diurnal cycle, the solar activity cycle, solar rotation, solar flares, the equatorial anomaly, and a variety of terrestrial effects are all known to contribute. Evidence suggests that there are a number of other poorly understood factors. Various phenomena that affect the angle of incidence may make more significant contributions to the basic occurrence of F2 at 50 MHz than at lower frequencies. So, when is the band open? Well, it is somewhat like playing a slot machine with lots of wheels. There are certain combinations that come up winners to some degree or another. There is an important difference from a slot machine though. The odds of some of the wheels coming up favorable are not entirely random. Within limits, the daily wheel, the solar cycle wheel, and the 27day wheel are all predictable to some degree. And just to make the game interesting, there are several more wheels we do not know about. In time, we will figure out more of them. We can say that for stations at magnetic mid latitudes the favored times are: Same Hemisphere (North or South) Daytime, local winter (November to May in the north), near solar maximum, perhaps during the peak two weeks of the current 27-day cycle, plus unknown factors. Transequatorial Paths Daytime, October-November and March-April, near solar maximum, perhaps during the peak two weeks of the current 27-day cycle, plus unknown factors. Not withstanding these patterns, it is obvious that this is truly an unfinished story. Even though the list above addresses the most probable times, it is clear that F2 can pop up at any time, even as a result of a flare at solar minimum. Many openings have happened when none of these conditions were known to be present. This only goes to underscore the importance of the unknowns and to present a challenge to the community to keep listening. One thing that is known for certain: you can’t work them if your radio is off. Some References



Ionospheric Radio, Kenneth Davies, Peter Peregrinus Ltd., London, 1990 Introduction to Ionospheric Physics, Henry Risbeth and Owen K. Garriott, Academic Press, New York, 1969 Ionospheric Radio Waves, Kenneth Davies, Blaisdell Publishing Company, Waltham MA, 1969



COPEX - Conjugate Point Equatorial Experiment in Brazil: a brief outline An outstanding ionospheric problem that calls for global scale concerted efforts on the part of the scientific community is related to the day to day variability of Equatorial F region irregularities that are driven from processes below the ionosphere. Also known (for historical reasons) as equatorial Spread F (ESF), it is a phenomena that produces large turbulent like variations of electron density at F region heights producing large index of refraction variations. ESF occurs in association with the plasma depleted flux tubes, widely known as plasma bubbles which develop at the dusk hours into vertically extended formations extending to 1500 km over the magnetic equator and thousands of kilometers (?25?) into the low latitude ionosphere on either side. They consist of irregularity structures of scale sizes varying from 10’s of centimeters to 100’s of kilometers. These affect HF communications, as well as the satellite signals used for the many practical applications in our daily life: point to point communications, satellite to ground communications, navigation systems based on GPS satellites, geodesy and over-the-horizon radars (drug traffic control). Despite the progress made in the last decades to understand the physical mechanism responsible for the formation of F-region irregularities, we have not made any progress in achieving the capability of predicting its occurrence, not even in a retrospective way. ESF is initiated by plasma instability processes operating at the F layer bottomside that grow into the topside ionosphere. It is recognized, both empirically and theoretically, that the altitude reached by the F-region peak and the bottom-side density gradient of this region have an important impact on its occurrence. Its annual variability, which varies with longitude, is understood in terms of these parameters and in terms of the alignment of the solar terminator with respect to the magnetic field lines. Nevertheless, there still exist unknown factors with a day to day variability that are not accounted for. The source of perturbation in the form of gravity waves originating from lower atmospheric regions, the integrated electrical conductivities along the flux tubes connecting the conjugate E layers are among the factors responsible for the ESF variability. Progress will be made only if one can achieve a full description of the state of the magnetic field tube before it goes unstable. Its state should be described along its full geographical extent, including the "feet" of the lines making contact with lower altitude regions (E-region) were the integrated transverse conductivity is defined. This requires the coordination of many different instruments along a large span of latitudes involving many geographical regions and, particularly, observation at magnetically conjugate points. The potential key factors recognized as responsible for the day-to-day variability of the ESF are: 1- A day to day variability of the E-region and lower F-region densities and conductivities; 2Variability of equatorial F layer height; 3- Zonal and meridional components of thermospheric winds; and 4- day to day variability in gravity wave effects at ionospheric altitudes that produce a seeding of density fluctuations that would be eventually amplified by the instabilities to form the large F-region fluctuations which requires also the assessment of possible weather effects in the low latitude zone. Motivated by the need to improve our understanding of the ESF day-to-day variability based on the above factors, it is planned to conduct a conjugate point observational campaign of two-month duration in Brazil that will involve intensive ionospheric sounding by digisondes complemented by other relevant diagnostic instruments. The magnetic conjugate points should be located such that



the conjugate E layers are field line mapped to the F layer peak, or the bottomside, over the magnetic equator. It is noted that Brazilian land territory in the western region of the country satisfies such magnetic conjugacy conditions. The three selected locations are: Campo Grande (MS): Lat. -20:26:34; Long. -54:38:47 (southern conjugate point); Boa Vista (RO): Lat. 02:49:11; Long. -60:40:24 (northern conjugate point; and Cachimbo (PA): Lat.-9: 28:00; Long. -54:50:00 magnetic equatorial point. A schematic of the Conjugate Point Equatorial Experiment (COPEX) is shown in the Figure below. The experiment is scheduled to take place during the period of October-November 2002, which is in the middle of the spread F occurrence season in Brazil. The Aeronomy group at the Brazilian National Institute for Space Research (Instituto Nacional de Pesquisas Espaciais- INPE), in collaboration with international groups, is coordinating the COPEX. The specific instruments to be operated are: Digital Portable Sounders (such as DPS-4), optical imagers and GPS recievers, and possibly other instruments, such as magnetometers, to be decided upon. The international scientific groups interested in this interesting problem are welcome to collaborate in the COPEX campaign. They are also urged to conduct coordinated experiments in other longitude sectors of the globe to make a global effort to investigate the ESF variability under the umbrella of the EPIC prog



Super-Mode VHF Transequatorial Propagation

Leo F McNamara, Boston College/AFRL Background Transequatorial Propagation (TEP) is the name given to the phenomena in which signals at VHF frequencies are observed to propagate on long circuits across the magnetic equator at frequencies well above the normal Maximum Observed Frequency (MOF), and with much higher than expected signal strengths. The observations go back to the 1950s or even earlier. Some of the first observations were made by radio amateurs, and are described by Southworth (1960). There are two types of TEP, which have different propagation modes that rely on unique features of the equatorial ionosphere. These are called the afternoon-type TEP and the evening-type TEP, or, alternatively super-mode and ducted-mode TEP. Both types have maximum occurrence rates in the equinoxes at solar maximum. The evening type of TEP is highly correlated with the presence of range spreading on ionograms at the point where the transmitted signals reach the ionosphere (McNamara, 1973). Propagation of this type of TEP is via ducting through equatorial bubbles aligned along lines of force of the Earth's magnetic field. The afternoon type of TEP involves a chordal (or super) propagation mode, with reflections from both crests of the equatorial anomaly, without an intervening ground reflection. Circuits lengths can reach up to 8000 km. MOFs typically reach 50 to 60 MHz. The super mode is illustrated in Figure 1. [See also Figure 6.24 of Davies, 1989]. Figure 1. The super (or chordal or FF) mode of transequatorial propagation



Heron (1980) provided the last known review of TEP, and included references to earlier reviews. It is the super-mode TEP that is the subject of this paper. Super-mode TEP has the following characteristics: 1. An MUF greater than the normal 2F MUF, i.e., greater than about 40 to 50 MHz, and possibly reaching 70 MHz



2. A peak occurrence at around 1700 to 1900 local time, around the equinoxes and around solar cycle maximum 3. Typical path lengths from 5000 to 6500 kilometers. 4. Normally strong steady signals with a low fading rate and distortion (from multipathing or Doppler spread) Gibson-Wilde (1967, 1969) showed examples of the equatorial anomaly in the Australasian longitudes, and illustrated the high level of correlation between the morphology of the anomaly peaks and the existence of super modes. McNamara (1991, Figures 9.3 and 9.4) showed that the anomaly peaks have higher densities and are more widely separated, and the F2 layer peak at the magnetic equator is higher, for days during which super-mode TEP was actually observed on the Yamagawa (southern Japan) to Townsville (northern Australia) circuit in August 1970. It has been well established by raytracing simulations that the propagation mode for the afternoontype TEP is a super mode. It is essential that the model of the ionosphere adopted for raytracing purposes be representative of the actual ionosphere along the circuit at the time of the TEP observations. Monthly median models fail to support the super mode in simulation. Figure 6.24 from Davies (1989) shows some of the raypaths given by the Jones raytracing program (Jones and Stephenson, 1975) when tracing through the equatorial cross-section given by Wright (1959). Figure 6-24. North-south chordal ray path across the magnetic equatorial region. The path shows two ionospheric reflections without an intervening ground reflection.



Equatorial Propagation

Ray Cracknell, G2AHU

Background Measurements of solar flux up to and including October 2000, and the scarcity of long distance F-layer QSOs on 50 MHz, seem to suggest that solar cycle 23 will go down in the records (together with cycle 20) as years when the solar flux maximum was considerably lower than in cycles 17,18,19, and 21,22. In cycles 20 and 23, in the temperate regions of the southern and northern hemispheres in particular, F2-layer ionisations were too low to support regular world-wide radio propagation at 50 MHz. After the International Geophysical Year in the peak years of cycle 19 (1957 to 1959), nobody was in the least surprised that cycle 20 was an anticlimax. However, after cycles 21 and 22, which were also exceptionally good, very few predictions were made to prepare us for a similar anticlimax in cycle 23. Nevertheless, most of Britain is only one sporadic-E hop north from the Mediterranean, the northern limit of tropical propagation. In the days of the 50 MHz permit holders, in October 1985, there was a fine opening from Botswana, centred on Yorkshire, during which Botswana was worked from southern Scotland and the northern half of England. Both ends of the QSOs were monitored by an amateur in Malta. This occurred near sunspot minimum, between the maxima of cycles 21 and 22. The Tropical Ionosphere It needs an understanding of the tropical F region to appreciate how that Botswana 50 MHz QSO occurred. An alternative to our findings might have been multi-hop sporadic-E (Es) as that mode of propagation is not directly dependent upon solar activity, but strangely enough, in equatorial regions, Es is not usable at 50 MHz as it forms a smooth layer which only reflects lower frequency HF signals. Thus we can eliminate any possibility of multi-hop Es being responsible, although the first hop out of Britain could only have been by sporadic-E.



Fig.1: The effect of the Earth’s magnetic field on free electrons reaching the tropical ionosphere.



All forms of solar radiation, like radiant heat, are at maximum intensity at the equator and at a minimum at the poles. As solar radiation, especially ultra-violet rays, causes ionisation of rarefied air, we expect to find higher electron density in tropical regions and the solar wind with its flow of free electrons adds to the process. Free electrons tend to follow the magnetic lines of force of the Earth's magnetic field, which also acts as a magnetic shield to protect us from bombardment by harmful radiation. The density of free electrons in the F region of the ionosphere determines its ability to reflect, or more strictly to refract radio signals back to ground at a distant point. The effect of the Earth's magnetic field on the F layers of the ionosphere is illustrated in Fig. 1. The force exerted on the electrons at any given point is a combination of the vertical component attracting electrons towards the centre of the Earth and the horizontal component attracting them towards the geomagnetic poles. At the magnetic poles there is no horizontal component and they are only pulled down towards the surface. In equatorial regions we find the geomagnetic equator, or more accurately the zero dip equator, where the magnetic lines of force are parallel to the Earth's surface and the vertical and horizontal components cancel themselves out. As one may imagine, the effects on the ionosphere over the zero-dip equator are quite dramatic and of special interest in explaining how VHF and UHF signals encountering the ionosphere near the magnetic equator are influenced. Like the measurement of latitude, the dip angle changes from 0° at the zero-dip equator to 90° at the magnetic poles (see Fig. 2). Although the measurements are similar, the magnetic poles are situated away from the geographical poles, and the zero-dip equator correspondingly differs considerably from the geographical equator. At the magnetic north and south poles where there are not any horizontal components, free electrons are attracted downwards and auroras result, especially during magnetic storms. Elsewhere, free electrons follow the magnetic lines of force as the dip angle varies from the equator to the poles. The vertical component prevents rapid dispersal of the F regions, but after sunset over the dip equator, where the dip angle is zero, night-time dispersal is rapid with no magnetic force to prevent it. This occurs regularly in a belt which is approximately five degrees either side of the dip equator; but is beyond the apparent five-degree barrier, the vertical component is sufficient to hold the F-region in place, often until the early hours of the morning. At the line of zero magnetic dip, that is the zero-dip equator, it is important to understand the changes that take place in the morning, afternoon, evening and night, with the rising, setting and disappearance of the sun on the F-region of the equatorial ionosphere. They are illustrated in Fig. 3. After sunrise, ionisation rapidly builds up, mainly due to ultra-violet radiation and the increasing force of the solar wind, until such time as the sun reaches its zenith and builds up electron density as illustrated in Fig. 3(a). In Fig. 3(b), after mid-day, the ionising influence decreases but the flow of electrons continues. This gradually reduces the electron density around the zero dip equator, until it reaches the area where the vertical component of the magnetic field exerts sufficient force to arrest it, thereby forming an area of very dense ionisation both to the north and south of the zero dip equator.



Fig. 3(a): Typical F-region ionisations before noon local time.



Fig. 3(b): Changes during the afternoon.



Fig. 3(c): Typical ionisations after 2000 local time. Figs. 3 a, b & c: The varying effect of the sun during the delay on the tropical ionosphere. After sunset, in Fig. 3(c), the lower density around the zero-dip equator continues to fall until approximately 20.00 hr local sun time, when it begins to break up and drift away. But the high-density zones remain firmly bound by the vertical attraction of the earth's magnetic field until they finally disappear by recombination into uncharged atoms.



Fig. 4: Dr Aarons’ 50 MHz radar backscatter measurements during the evening from Jicamarca, Peru at 19.00-22.00 local standard time, October 16-17 1976. Dispersal of charged particles over the dip equator against time is illustrated in Fig. 4, which was published by Dr Jules Aarons in 1977. It is very important to note that the horizontal axis is against time, not distance from the zero-dip line, and illustrates changes from 19.10 to 22.00 hours local time. In private correspondence with Dr Aarons of Boston University in 1991, he expressed the opinion that the high-density zones tended to split into field-aligned sausage-like structures travelling from west to east. This had been also suggested to us by SV1DH's Doppler shift observations on 144 MHz from ZE2JV in 1980. When the diagram was published by Basu and Aarons (1977) some amateurs jumped to the conclusion that these irregularities were responsible for transequatorial VHF propagation. But although the regions above 700km are not illustrated, the density is very unlikely to be sufficient, as an altitude of over 1000km is required for one-hop propagation over the distances worked by TEP. Nevertheless, the dispersal of electrons illustrated is undoubtedly responsible for the flutter fading that frequently develops after 20.00 local time. The final theoretical concept is that reflections from the ionosphere are only symmetrical if a layer is smooth and also of consistent altitude and density, but tilts play a very important part in HF, VHF and UHF DX modes of propagation. If tilts in virtual height or density occur, then the angle of arrival at the layer will be different from the angle back towards the ground, and if the distance is increased thereby a higher frequency is able to be used. Tilts are operative at the Grey Line and in the use of sporadic-E, which sometimes propagates frequencies up to 144 MHz although the maximum frequency from a vertical sounding may be as low as 5 MHz. Also, transequatorial propagation (TEP) uses the tilts between the high-density zones on both sides of the deep bite-out over the line of zero magnetic dip in the late evenings. The use of tilts in TEP is illustrated in Fig. 5. The tropical high-density zones are also usable for normal F2 propagation, and with the maximum frequency returned to earth from vertical sounders rising at times to nearly 20 MHz, one-hop 50 MHz propagation can take place throughout the solar cycle.



Fig 5 (a): The geometry of a tilt at the nose of a high quality density region.



Fig 5 (b): The ray path followed by TEP (the billiard ball mode). Fig. 5: Transequatorial Propagation (TEP) showing the use of apparent tilts from high and low density zones. East-west contacts along the high-density zones can be multi-hop and the circuit from Greece to South America has often been utilised. North to south TEP circuits are likewise sometimes available at 50 MHz, throughout the solar cycle, between favourably located stations about 6000 km apart. Nevertheless seasonal conditions do vary widely, particularly at solar minimum. In general, it is evident that conditions are best near the equinoxes when the high-density zones are balanced on both sides of the zero-dip equator.



Fig. 6: The TEP zones as visualised from Limmasol, Cyprus by 5B4WR and by ZE2JV from Salisbury, Southern Rhodesia (now Harare, Zimbabwe) in 1958. Bearing the above considerations in mind, it is possible to appreciate how a sporadic-E transmission from Yorkshire could be bounced off the Mediterranean and carried on with successive F2 reflections to Botswana even at solar minimum. The reasons for recommending the exercise were based on results during the International Year of the Quiet Sun (IQSY), published by ZE2JV (now G2AHU) and 5B4WR in June 1965 (RSGB Bulletin pp 367370) aided by reports on the reception of our 70 MHz beacon ZE1AZD in England, also during sunspot minimum conditions. The persistence of Eric Parvin, G2ADR in 1987 in monitoring the Botswana 50 MHz signal and alerting other operators was most commendable (see page 32 for more on Eric's historic QSO). The northern and southern TE zones as observed in 1958 from Cyprus in the north and Salisbury (now Harare) in the south are illustrated in Fig 6. It was originally published in "The transequatorial propagation of VHF Signals" by ZE2JV in QST in December 1958, pp 11 to 17, and subsequently in the first edition of the ARRL VHF Manual, p 21 and several subsequent editions. History Edward Tilton, W1HDQ, the long-time VHF editor of QST, described TEP as a unique amateur achievement and a brief description of its development over the Europe to Africa circuit should be a pleasant change from mere theory. The earliest known use of transequatorial propagation took place in October 1947 almost simultaneously in South America and Africa, when XEIKE worked LU6DO and several other Argentine stations on 50 MHz, and G5KW (operating as MD5KW) worked VQ2PL on several occasions. We all sat up and took notice when PA0UN and PA0UM, G6DH and others worked ZS1P and ZS1T on 50 MHz and Fred Anderson, ZS1LA received audio and video TV from British 40 MHz TV. The first professional research was published in 1957 by Prof. O G Villard of Stanford University, who was also a well-known W6 amateur. He and his co-authors used ground-backscatter experiments south from the West Indies and proposed that opposing tilts in the virtual height of the F-region on either side of the equator allowed propagation without a ground reflection between them. The transequatorial mode was, as a result, nicknamed 'the billiard ball mode' (see again Fig. 5).



Fig. 7: An extract from Ray Cracknell’s paper from the proceedings of the Science Convention at Salisbury, Rhodesia, May 1960, p 6. He was right, except that the tilt was not due so much to physical tilts in the virtual height of the layer as to the pronounced variations in electron density between the high density zones approximately five degrees of latitude either side of the zero-dip line and the low density bite-out in between, which produces the apparent tilts required for Prof. Villard's calculations. It is now difficult to appreciate the pressures and interest in international communications in the days of the Sputniks and propagation research. Transequatorial propagation was no exception. In 1959, Prof. Obayashi published a paper entitled "Long distance HF propagation along exospheric field-aligned ionisations". He used backscatter soundings south from Tokyo on 28 MHz. This was published soon after my article in QST and a deputation from Boulder flew out to Salisbury to see me. They wanted to set up one end of a TEP experiment using high power sweep frequency sounders and rhombic antennas. The government turned down the project and I told Dr Davis that I was quite certain it was ionospheric propagation and not the extra-spherical mode. I was in daily contact with the late Chalky Whiting, ZC4WR in Cyprus and we agreed that all we had to do was to measure time delays. I borrowed a CRO and built a pulsing unit to key the TX on 50 MHz, which ZC4WR received and rebroadcast as modulation on 29.5 MHz. I displayed the outgoing and returned pulses together with time pulses on the twin beam CRO and the results appear in Fig. 7, which is an extract from the proceedings of a science congress held in Salisbury in May 1960. After it nothing further was heard of the extra-spherical mode.



Of special interest in the results reported in Fig. 7 is that there were three modes of propagation, namely: 1. Normal two hop F layer (2F2); 2. TEP without flutter-fading in the early evening (F type TE); 3. Pure TE in the later evening with its flutter-fading, diffused characteristics and very high MUFs. In 1980, contacts on 144MHz and 432 MHz were made with Cyprus (5B4WR) and Greece (SVIDH). Conclusions The details of work during cycle 21 were published in QST, November and December 1981 by ZE2JV, ZS6PW, and SV1DH but we were mainly concerned with 144 MHz work and confirming our propagation theory. On 144 MHz we did not succeed in working beyond the sharp cut-off line across northern Italy, in spite of a long series of tests from ZE2JV with HB9QQ in Switzerland. Many, including GJ4ICD, listened for the regular test transmissions in the UK without success, but as TEP across Africa on 144 MHz did not commence until nearly 20.00z and we should have been very fortunate to find 144 MHz sporadic-E operative to the Mediterranean at that time of the evening. The TEP system works best when the high-density zones on both sides of the zero-dip line are balanced, when the vertical sun at noon is over the line of zero dip. It is complicated, as over Africa the zero-dip line is located up to 11 degrees north of the geographical equator (see Fig. 6) and it crosses South America well to the south of it. Nevertheless optimum conditions occur at or near the equinoxes (actually in October over Africa) and is poorest at the solstices. Variations between solar activity maximum and minimum are pronounced, although 50 MHz propagation from the high-density zones is still possible but with reduced reliability during periods of sunspot minimum. The presence of high electron density areas in the F-layer has other effects as well as boosting transequatorial propagation. These regions are formed approximately 5° to 15° from the zero-dip line and can provide propagation at higher frequencies than are available in temperate zones. They are also interestingly convenient for amateurs working multi-hop QSOs in a line roughly parallel to the zero-dip line. But transequatorial propagation (TEP) is best when working over optimum distances, as near as possible at right angles to the zero-dip line. Seasonal and solar cycle variations are pronounced but, as we saw from our Es plus TEP experiences, TEP from the most favourable locations can also take place during a solar activity minimum. Another aspect which was a major concern of Dr Aarons was the effect upon communications of the 'plumes of ionisation' illustrated in Fig. 4 when the path to satellites was passing through the ionosphere above tropical areas. Fortunately solar cycle 23 has produced lower solar activity than was predicted; this has resulted in much diminished flutter-fading and the effect on communications with satellites has been minimal. This should be some consolation to those of us hungry for more DX on 50 MHz. Terminology It is very difficult to say who first used the term 'TEP' instead of transequatorial propagation. In the early days it was referred to as 'anonymous propagation', because it was clearly a mode differing from normal multi-hop F layer propagation. It became 'Transequatorial Propagation' in the title and 'TE propagation' in the text of the 1959 QST article and I can only guess that it was Ed Tilton, W1HDQ, the VHF Editor of QST, who initiated the use of 'TEP'. It would be of interest if someone who still has 1959 and 1960 QSTs searched his articles for its first use. It should always be confined to meaning, 'anonymous' F-layer propagation of VHF radio signals across the equator' and care should be taken to apply it only to the 'billiard ball mode'. It is correct to label the Botswana to Yorkshire contacts as Es + TEP. West African QSOs without crossing the equator should be labelled Es + F and contacts with northern Brazil should be Es + 2F as the path runs parallel



to the geomagnetic equator. Contacts further south might be either multi-hop F (nF2) or Es + nF2. Long distance HF contacts with Australia which cross the equator would not be considered as TEP although they may encounter TEP-type flutter-fading at night and may at times miss out an earth reflection due to encountering a Grey Line at dawn or sundown.