## Sunday, July 31, 2011

### A, B, C's of Dx Fundamentals of the Art of DXing IV

#### What are RF Waves?

To begin, it is helpful to understand what are radio waves and how they move (propagate) from A to B. Since we all know a bit about water waves, they can serve as a beginning point. Consider a pebble dropped (kerplunk!) onto the calm surface of a pond. As the pebble breaks the tension of the water surface, it pushes aside (disturbs) the surface water around it, which then collapses back upon itself as the pebble sinks below. Because the surface tension acts as a sticky coupling between adjacent water molecules, any motion in one area will disturb nearby areas as well, and these will then disturb others beyond, and so forth, as illustrated in the 2-dimensional diagram below:

After the pebble breaks the surface, the disturbance propagates radially outward from the point of impact as a surface wave. This illustrates the fundamental definition of all wave motion:
Wave motion is the propagation of a disturbance through an elastically-coupled medium.
The term "elastically coupled" means that if you disturb one area of the medium, it will affect the adjacent areas. It is easy to understand this for various waves in media, such as water waves (water surface) and sound waves (air, solids, liquids), but what about radio waves? Although it is certainly true that they propagate through air in our atmosphere, we know that they also propagate through space between the heavenly bodies, where there is essentially nothing but a vacuum - so what "elastically coupled medium" are they using???? Hmmm....!!!!
Historically, the study of radio waves required a new concept of what the medium of propagation might be and scientists in the 19th century, grappling with this difficult question, laid the groundwork for our current understanding of what are now called Fields. While investigating the new phenomena of electricity and its possible relation to the older known phenomenon of magnetism, the electro-magnetic effect was discovered in which a current flowing in a wire produced a magnetic field around the wire; also that changing magnetic fields would produce electric fields that could cause a current to flow, that is:
A changing Electric field will create a Magnetic field, and a changing Magnetic field creates an Electric field.
Studies of this phenomenon led to the startling discovery that high-frequency alternating electrical currents in wires would produce "energy" waves that could travel through the air without the need for wires. It was soon understood that these waves were of an electromagnetic (EM) nature; that is, they consisted of time-varying electric and magnetic fields. For many years, early scientists struggled to extend their understanding of ordinary waves in solids, liquids, and gases to that of the "wireless" EM waves. The big question was: what was the elastic medium that supported the disturbance known as EM waves? Finally, they conceived of a medium that they called "ether" (no relation to the anesthetic!), which was thought to be an extremely low-density gaseous substance that permeated all of space and allowed the propagation of EM waves. The idea of a substantive ethereal medium was eventually shown to be erroneous and was replaced by concept of "vector fields" - that is, the potential for a directional force (electrical, magnetic, gravitational, nuclear) to arise at a point in space as the result of the presence of a physical object. As this abstraction of a "potential force at every point" is difficult for many, it will be easier for our purposes to imagine all of space as being permeated by an invisible 3-D grid of "field-lines" consisting of invisible lines that intersect at every point in space. In the figure below illustrated in 2-D, the lines exist at all points in space, within all atoms and molecules, and extending throughout the universe. These are not actually things; rather, they are an intrinsic attribute of space and time, representing the potential for directional forces (e.g., EM, gravity, or nuclear) to arise in the event of a disturbance to space-time caused initially by a physical body. For example, when a solid body is placed at a point in space, it gives rise to a gravitational force field (and if it accelerates, it will produce a Gravity Wave ... but we digress!).
Of interest to us is the fact that a high-frequency oscillatory motion of electrical current flow in a wire (our antenna), caused by a time-varying E-field in the wire, will simultaneously give rise to a time-varying M-field around the wire, and this amounts to a disturbance in the surrounding space. As the E-field in the wire rises and falls, it produces an M-field encircling the wire that also builds and collapses. The disturbance of the surrounding space caused by the changing M-field around the wire creates another E-field farther out around the wire, which in turn collapses, producing another changing M-field extending farther outward, and therefore a changing E-Field even farther out, and so on ... . One might be inclined to imagine this as a repeating sequence in which the changing E- and M-fields "leap-frog" outward .
In the illustration above, consider the charge to be part of an RF current oscillating back & forth along the wire, creating the outward ripple of EM fields. Each complete back-and-forth cycle of the current produces one wavelet around the wire that propagates outward. So it is that an EM wave disturbance is formed, rippling away from the antenna.
To better visualize the actual wave, imagine the sinusoidal representation as being rotated around the antenna in 3-D, forming an expanding doughnut shape with the antenna through its center, propagating outward at the speed of light as a spherical wave-front of increasing diameter. Each current cycle produces a wavelet that leaves the antenna, as seen in a top-view cross-section in the left figure below (dipole antenna at center), and in an end-view of the antenna in the right figure.
Notice that because the antenna is horizontal to the ground, all of the E-field vectors are formed horizontally. By definition, the "polarization" of the EM-wave is determined by the E-field vector. In this case, the wave is horizontally polarized. Can you guess what would be the polarization of a vertical antenna?
If we were to be able to actually see the wave as it emerged from the antenna, we would expect that there would be countless interactions with the environment at play within a very wide area around the antenna, including the effects of the ground, buildings, trees, and metallic conductors like house wiring, powerlines and fences. All of these factors in the near-field region of the antenna would be expected to affect the final shape of the wave as it expanded outwards. At some distance from the antenna, beyond the near-field region, we would expect to eventually see the "final form" of the wave-front. In fact, this theoretical region at a distance of many wavelengths from the antenna is called the far-field region.
We will discuss more of these issues in the chapter on "Antennas", but for our purpose here, let's just ask "What would the wave-front of the dipole antenna look like if there were absolutely no interacting structures, including the ground and the antenna supports?" Hmmm ... you might then ask ... "Where in the world would you go to get away from ALL structures?? ... Space???". Good guess!!! That is exactly what is done - of course, only theoretically at this point (although one day we may be able to do it in reality!). Using antenna modeling programs based upon the theory of how RF waves emerge from a conductor, we can have the computer software "draw" the predicted pattern of an antenna that is placed out in space, free from any interactions with physical objects. These abstract models are called "Free-space" patterns and are very useful for apples-to-apples comparisons of different antenna designs.
The theoretical free-space plot of the dipole far-field radiation pattern is seen shown below.
Free-space plots are mainly used in comparing antenna designs without concern for unknown local variables (antenna height, ground conductivity, structures, etc). As a rough rule-of-thumb, ground effects will "slice off" the bottom of the pattern below the antenna, absorbing some of the energy while reflecting some back to interact with emerging wave-fronts and alter the final pattern. As we shall see in the chapter on "Antennas", depending upon the height above ground (and surrounding objects), the actual radiation patterns may be considerably altered from that seen in the theoretical plots (see the chapter on "Antennas").
A nicely illustrated elementary lesson on EM waves is available online from the University of Colorado, Boulder (references). It is important to note that although the outwardly propagating wavefront is increasing in diameter, the frequency of the "ripple" (and therefore wavelength) do not change; however, since the energy is being spread over an increasing volume of space, it follows that the amount of energy that will be received in any given direction will be decreasing with distance from the source. In fact, the wave energy received in any given direction will be reduced by one-fourth with each doubling of distance. Also, we should always remember the inverse relationship between wavelength () and frequency (f):
where V is the speed of light in the the medium in which it is traveling. In free-space, V is approximately 300,000,000 m/s; however, it is a bit less in air and even less in solids. In fact, since we must so frequently calculate antenna dimensions, it is very useful to remember a convenient approximation of the above equation for wavelengths in copper wire:
Now that we have an idea of what radio waves are, let's take a look at the things that may affect their propagation, especially in the HF bands, between us and the DX!

# HF Propagation

Revised: 12-Feb-2007

Despite the fact that it involves only a small band of the total electromagnetic wave spectrum, HF propagation is a complex topic. Despite all attempts at brevity, this chapter contains a large amount of material, discussed in greater detail, than most of the others (with the possible exception of "Antennas"). However, without propagation there would be no HF DXing, so it is the most important topic of all and for that reason it is as you see it here. If the material seems to be more than you want to tackle, or if it does not interest you at present, then just skip to the summary at the end for a distillation of the key issues. You are always welcome back at anther time!
Note: Figures are clickable for pop-up enlargement; some are animated.

#### What Affects EM-Wave Propagation?

After operating a ham station for awhile, one soon learns that there are many factors that affect propagation and the mix is continually changing by night/day, month, season, and long-term changes in solar activity. However, all of the factors are related to one thing - our sun, Old Sol. The Sun, a huge nuclear-fusion-powered furnace, continually spews out charged particles and radiation of all almost all wavelengths from the microwave and up. In its annual rotation about the Sun, the Earth is continually bathed in this Solar flux, also referred to as the "Solar wind".
One of the first things that we learn as radio aficionados is that the different bands vary over the course of the daily 24-hour period in their ability to support communication. When the sun rises each morning, the solar radiation from the extreme ultra-violet (EUV) to the X-ray wavelengths begin to bombard the atoms and molecules in the upper atmosphere, causing them to temporarily give up some of their electrons, and resulting in a region of electrically-charged particles consisting of the negatively charged electrons and their now positively-charged parent atom or molecule , as illustrated in the figure below (Note: A collection of charged particles is also called a "plasma"). These charged particles are referred to as "ions" and the process of their production is known as "ionization". This process begins at the extreme outer limits of the atmosphere and progresses downwards toward the surface as the sun rises. As can be seen in the profile graph below called an "ionogram", the density of the free-electron ions is not uniform, but increases smoothly in altitude, with some peaks (E,F) and valleys.
So it is that within a portion of our atmosphere, roughly 100 to 350 Km (60 - 220 miles) above the Earth, is a region of charged particles called the ionosphere. As the day progresses and Solar radiation intensity increases, the density of the ions increases and the depth of the ionosphere extends farther downward towards the Earth's surface. Studies have shown that as the ionization depth increases, density differences may be found within the regions of ionization at different altitudes (see profile graph above). The different ion-regions are formed because the shorter (X-ray) wavelengths are less-readily absorbed and penetrate more readily into the denser lower atmospheric regions, while the longer (EUV) interact readily with the lense-dense upper regions. The different regions are given alphabetical letters to indicate their relative height, beginning with the highest, formed by EUV radiation and known as the F-region, followed later and at a lower altitude by the E-region formed by "soft" X-rays, and finally by late morning, the "shorter wavelength "hard" X-rays form the D-region near the surface. Because the atmospheric density increases with descending altitude, all of the high-energy (X-ray and EUV) radiation is completely absorbed with the formation of the ionosphere. This is fortunately for most living things (marine life and exoskeletal insects like cockroaches would probably be unaffected), as in time, these high-energy wavelengths would be deadly!
While it is often convenient to illustrate and refer to the different ion regions as "layers", as done in some of the figures here and elsewhere, one should understand that there are no distinct boundaries; rather, one region simply builds in density, then "fades", rising again at higher altitude into another into another defined region. This can be seen clearly in the figure below of an Ion Density Profile, or "ionogram" which is just a graph of the altitude vs. the number of electrons per cubic meter.
Descending in altitude, one first sees the bulge of the F-region, the largest of the regions. As it grows in the course of the day, most of the EUV radiation is absorbed in its formation, and it begins to taper off; however, by late-morning, the most energetic of the EUV may cause a small secondary increase in ionization that can be seen in the above graph as a very slight "bump" in the shape of the line between the arrows indicating the "F" and the "E" regions. This new region of the "bump" is more like the F- than the next level E-region. For this reason, the F-region is said to have two components: the higher, larger F2-region, and the lower, smaller, F1-region. Just below the day-time F1-region at a lower altitude, one can see the smaller "nose" of the E-region, and finally, at ~60 Km, just before the graph indicates no more electrons, on can see the very slight "bump" of the D-region. Below the D-region is the non-ionized lower neutral atmosphere.
Later, as the day wanes and the solar influence lessens on the dayside of the globe, the attraction between the ions of different charge, freed of the ionization effects of the solar radiation, begin to recombine into their previously neutral atoms and molecules, literally going to sleep for another night. By early evening, the electron densities of the D- and F1-regions are usually reduced to the point where these regions have little or no affect upon communications on the amateur bands, and the ever-present F2-region contracts to relatively high altitudes during the night hours. During the night hours, sufficient remnants of the E-region remain, although reduced in electron density, and this becomes the primary factor contributing to propagation on 80m and 160m.
In addition to the diurnal (daily) variations due to the Sun, there are other phenomena that influence the ionospheric distributions, including high-altitude winds that can cause shifts in the electron densities, especially the E-region. Other factors will be discussed below.
Since radio waves are electromagnetic phenomena, it is reasonable to assume that they could interact with the ionosphere. In fact, the free electrons readily interact with HF waves and this is of prime interest to us as it is what allows us to work DX! It should be in our best interests to understand more about it.
Solar Variations & Sunspots
Although the daily solar variation is one that is certainly obvious, there are other cyclic events of the Sun that are of great importance to the propagation of radio waves. The Earth-Sun relationship is continually going through four periodic cycles that affect propagation at any given location on the globe:
• 24-hour diurnal (night/day) cycle: rotation of the Earth on its axis, producing daily ionospheric changes.
• Solar axial rotation: cycle of approximately 27 days, resulting in the likelihood of a repeat of the current conditions 27 days later.
• Earth's seasonal cycle: as the Earth orbits the Sun, due to the change in the angle at which sunlight illuminates the globe as a result of the tilt of the Earth's axis with respect to the plane of orbit, the level of solar exposure rises and falls. As seen in a schematic of this, summertime occurs in the Northern hemisphere when that portion of the globe is
"tilted" towards the Sun and therefore under greatest degree of illumination; the least amount of illumination (and coldest temperature range) occurs approximately 6 months later when the Earth is at the opposing point of orbit. Seasonal changes affect the length of daylight hours and therefore the extent of atmospheric ionization: during the summer months, the higher frequency bands have good propagation, even late into evening and night hours, while winter months offer shorter days but the least absorption, providing better day-time high-frequency band conditions than in the Summer and excellent low-band conditions. The Southern hemispheric seasons (and therefore seasonal propagation conditions) are the reverse of those in the North.
• The Sun's 11-year (approximate) seasonal Cycle. The Sun, a ball of extremely hot gases, has equatorial "storm seasons" much the same as occur in the tropical regions of the Earth, except that the increase and decrease in the solar storm season spans approximately 11 Earth-years. During this period, the solar activity ranges from very little, all the way to peak periods in which the solar surface is roiled with large cyclonic areas of intense, tangled magnetic fields. These solar cyclones are cooler than the rest of the surface, so they appear to observers as dark spots, called sunspots, as seen in the figure below of spots and goups that appeared in October, 2003, some years after the peak of cycle 23.
The quantity of sunspots can vary from none to hundreds and records have been kept for the last 200 years. The daily amount of sunspots is known to vary in approximately 11-year cycles as seen in the figure below (courtesy NASA - http://science.msfc.nasa.gov/ssl/pad/solar/images/zurich.gif); however, cycles have been as short as 9 years and as long as 14 years.
Scientists now refer to these sunspot cycles by number, beginning with the first on record from the 1700s, to the current cycle 23, expected to end in early 2007. Counting sunspots on a daily basis is a difficult task, especially since weather can obscure the sun for many hours, and the ablity, skill, and equipment of counters may vary quite a bit. Often, there are not just individual "spots" but also clustered groups. In 1848, to compensate for the difficulties in getting accurate sunspot counts, a Swiss astronomer, J.R. Wolf, developed a method for recording the "sunspots" that accounted for groups and individual spots, as well as variations in viewing conditions. The result was called a sunspot number, R, (as opposed to "... the number of individual sunspots ...") to be used to represent the daily count:
R = k( 10g + f )
where g is the number of sunspot groups and is multiplied by 10 to indicate the greater potency of groups over individuals; f is the number of individual spots, including as many as can be counted within the groups; and k is a "fudge" factor calculated to take into account the viewing conditions (type & power of telescope, etc). This is still used today and is what is meant by the "sunspot number".
Smoothed Sunspot Number (SSN)
Since the Sun affects (among other things) radio propagation, solar activity has been an object of study for many years. Because of the periodic nature of the solar phenomena, an important part of the studies has focused upon ways to predict future solar conditions and events, such as the amount of expected ionization on a given day or the future course of the next cycle. Unfortunately, there is usually a considerable amount of fluctuation in the daily number of spots over he course of a year, making it difficult to associate it with specific conditions on Earth. In order to try to reduce the "jitter", scientists average the daily values each month; even so, the monthly averages still show large variability. To smooth out the numbers even more, a method was developed using a 12-month average of the monthly averages to assign a longer-term average value to the mid-point of the period. So that the mid-point falls at the middle of a month and not at the beginning (or end), the method actually uses 1/2-month values on either end, with 11 full months between: [m1R/2 + m2R + ...+M11R + m13R/2]/12, where m1R - m13R are the averages for each of the months 1-13. The mid-point month of this sequence is the 7th month, so the calculated average would be assigned to it. This value is called the Smoothed Sunspot Number (SSN). For example, in December of 2006, the SSN for June, 2006 would be calculated using 1/2-month averages for December, 2005 and December, 2006, along with monthly averages for January-November, 2006. This means that the sunspot number for June of a given year will not be determined until January, 2007. The smoothing effect of the SSN (blue line) may be readily seen in the figure below, compared to the monthly averages (black dots & line). The figure is courtesy of NOAA Space Environment Center (http://www.sec.noaa.gov/SolarCycle/).
Using analyses of past smoothed sunspot data from previous and current cycles, predictions can be made for several years into the future, as seen in this NASA graphic below for Cycles 23-24 (background image shows huge coronal eruptions).

It is the predicted smoothed sunspot numbers that are used in the propagation prediction programs discussed below.
As the year 2007 is beginning, we are either still at the end of cycle 23 or at the beginning of cycle 24 ... it will actually be year or so before Solar scientists can decide! The estimates at his time are for the cycle minimum to occur between March and September of 2007.
Solar Flux Index
Use of the smoothed sunspot number data accumulated over the years led to improved methods of predicting the rise and fall of future solar cycles, and it is still in use today. However, accurate counting of sunspots was sometimes difficult because of weather and/or observer difficulties. In the late 1940s, scientists discovered a reasonably close correlation between the amount of high-altitude radio noise in the 2800 MHz band (10.7cm ) and the sunspot observations. Since the 10.7cm noise measurements could be standardized and easily made under any weather conditions, this led to the development of monitoring instruments, techniques, and continued daily monitoring. Because it is a measurement of solar flux radiation "noise" levels only at the 10.7cm wavelength and not a measure of all solar flux (particularly not of the Solar UV-radiation responsible for ionospheric formation), they are intended to serve as an index of Solar flux variability. For that reason, the measurement is known as an index of Solar activity, called the Solar Flux Index (SFI).
High-altitude measurements of the radio noise at 10.7 cm flux are made daily at local noon (2000UTC) at the Pentictin Radio Observatory in Penticton, BC, Canada. Measurements are reported in solar flux units (one solar flux unit = 10−22 watt per square meter-hertz) and the values range from 50 - 300, with rising numbers suggesting the possibility of greater ionization levels and possibly better HF propagation. Unfortunately, like the sunspot number, the daily SFI has high variability compared to daily propagation conditions so that it is not a very reliable daily predictor of the band conditions, but just as for the sunspot number, smoothed SFI have proven to be more useful for predictive purposes. For example, the figure below (courtesy NOAA Space Environment Center - http://www.sec.noaa.gov/SolarCycle/) shows how a plot of the monthly average SFI (black dots & line) compares with that of the smoothed SFI (blue line).
A last word about SFI: one may see reported daily values of very few or even zero sunspot numbers with corresponding daily SFI in the range of 50 - 80 or so. This underscores the difference between the two and the fact that the SFI is only a partial measure of daily Solar activity. For perspective, a crudely approximate relationship between smoothed SFI (SSFI) and the smoothed sunspot number (SSN) is
SSFI ~ 60 + SSN
From this we see that, on average, even for zero sunspots, the SSFI is not expected to be zero. While the equation does not apply to the daily solar flux index, it is true that the daily SFI is never zero even if there are no sunspots.
Now a word of caution: many hams can often be heard discussing the SFI as though it were the sunspot number, or in the belief that it was a good indicator of the day's propagation potential. Some may even use the daily SFI in their propagation programs, hoping to get "up-to-the-minute" measures for better predictions. In fact, the single daily measure of SFI is a poor indicator of the daily propagation conditions. As shown by the equation and graph above, when collected and averaged over a year, the smoothed SFI does correlate reasonably well with smoothed sunspot number, but the daily value is all but useless as a predictor of daily propagation conditions.
Geomagnetic Disturbances
We all know that the Earth has a magnetic field, similar to that of a bar-magnet, with the South-magnetic pole approximately co-located with the Earth's geographic North pole, and the N-magnetic pole at the geographic South pole. The geomagnetic field lines flow from the geographic South pole to the North (i.e., N-magnetic to S-magnetic), with maximum filed density at the poles. While the field lines are horizontal at the geo-equator, they rise from the surface below the Equator and enter the surface above the Equator, resulting in an increasing magnetic "dip" angle as one moves higher in latitude toward either pole.
An interesting question at this point is: "Does the Earth's magnetic field have any effect upon HF signals?". After all, radio waves are EM phenomena ... do they interact with the geomagnetic field? The answer to the last part of the question is that the geomagnetic field is too weak to directly interact with HF waves; however, the answer to the broader question of " ... does it have any effect?" is considerably more complex; but, in short, the answer is a resounding "Yes!". Even in its normally calm state, the geomagnetic field interacts with the ionospheric electrons, exerting a force on the electrons in a direction perpendicular to the magnetic field lines, causing the electrons to spiral around the magnetic lines and drift towards the poles. The result is that as the ions form in the course of the day, the free electrons from the Equatorial region drift along the magnetic field lines to the mid-latitudes, producing regions of charge concentrations at the North and South mid-latitudes that are greater than at the Equator, contrary to what one would expect, so it is sometimes called the "Equatorial Anomaly". In addition, other phenomena, some of which are Earth-centered (e.g., earthquakes, or possibly gravity waves) as well as Solar-centered (discussed below) can affect the geomagnetic field. Suffice to say that, in the short-term (days, months), geomagnetic influences are usually the primary daily propagation variable.
While Earth-centered causes of geomagnetic variation are less known, those of the Sun are familiar to all. Large eruptions on the Solar surface can suddenly eject substantially more than the usual amounts of charged particles towards the Earth. As the Sun's activity increases towards each cycle peak, not only is there an increase in the daily Solar flux, but also the likelihood rises for sudden eruptions called solar flares, or even super-events called coronal mass ejections (CME) , that greatly increase the stream of radiation (UV & X-ray) and ions (charged particles) in the "Solar wind" through which the Earth must travel in its solar orbit. When that happens, the charged particles that reach the Earth become entrapped in the magnetic field lines (another good thing for us or we'd be bombarded into extinction, although the cockroaches would probably still survive! ) and are then pulled to the polar regions, spiraling to where the geomagnetic field is strongest. This can cause enlargement of the auroral oval (below) and give rise to spectacular auroral displays (below) that may be seen in the northern latitudes.
Along with the auroral activity, this may result in moderate, large, or on rare occasions, severe variations in the Earth's polar magnetic field. When this happens, it is known as a geomagnetic disturbances, and in the more extreme cases, a geomagnetic storm. How significant are these events? Well, on occasion the geomagnetic storms have been powerful enough to result in damage to satellite electronics, cause complete radio blackouts for several hours, and even disable regional power grids!
Measures of the geomagnetic field have been made since the early 1930s in observation laboratories at different latitudes (see: www.gfz-potsdam.de/pb2/pb23/GeoMag/niemegk/kp_index/kp_sites.html). Currently, there are 13 observatories worldwide using magnetometers to measure any changes in the field intensity at their site in units of nano-Tesla (nT) and reporting the maximum observed results every 3-hours. In the US, the measurements are made in Sitka, AK, Boulder, CO and in Fredericksburg, VA. The latter two are reported as "mid-latitude" values. Geomagnetic field variations, when they occur due to solar influences, may be several orders of magnitude above normal (0 - 500nT or more), so the nT measurements are transformed using a logarithmic scale with a "fudge factor" that removes the latitudinal differences. The log-transform results in a "compression" of the scale and the results are reported as an index of the field values, known as the K-index, with a range of 0 - 9. Since the magnetometer readings (in nT) are corrected for latitude when converted to K-values, all of the observatory K-values are equivalent worldwide regardless of observatory latitude. A K-index of zero means that the geomagnetic field strength is unchanged or normal, while values > 0 mean that it is disturbed, and a rising K-index indicates there is an increasing likelihood for possible geomagnetic storms, which are said to exist when K > 4 (see "Miscellaneous Notes" appendix for storm alert codes). Geomagnetic storms can cause ionospheric changes that result in disruptions of HF propagation, so they are of great concern to DXers.
The K-index is a short-term indicator of magnetic field conditions, reported every 3-hours. It would be useful as well to have some measure of the recent geomagnetic activity, for example, an "average K" for the last 24-hours. However, since the K-values are logarithmic, the 8 daily values for each 3-hour period cannot be averaged (for the same reason that you can't average dB values). In order to obtain an average measure of the daily geomagnetic report, the logarithmic K-scale is essentially "decompressed" into "a-values" every 3-hours at each observatory using the table below:
 K-to-A Transformation Scale K 0 1 2 3 4 5 6 7 8 9 a 0 3 7 15 27 48 80 132 208 400
Since the K-values have already been adjusted for latitude, it follows that the a-values are independent of latitude as well. Each of the observatories reports its a-values and the 24-hour average of all worldwide a-values is reported as the daily Planetary A-Index (Ap), or just the "A-index" (See "Misc. Notes" appendix for an example). The A-index gives us an indication of the average condition of the geomagnetic field over the last 24-hours. Like sunspots, geomagnetic activity is also periodic with an approximate 6-year cycle (re: http://www.ips.gov.au/Educational/3/1/3), so that peaks seem to occur at the low points of the sunspot cycle. Because of the shorter period, geomagnetic influences, on average, occur throughout the sunspot cycle. This means that, with the exception of the ionization extremes at very high and very low SFI, it is the geomagnetic field activity described by the K,A indices that causes most of the variations in our day-to-day DXing.
Solar Indices
Today, there are many observed and recorded solar measurements that are available for describing and predicting what is happening on - or in - the Sun, but for DXers, the three that were described above are readily accessible and particularly useful for gauging the solar influence upon the ham bands on a daily basis. Reported every 3-hours, eight times per day, they can give a pretty good daily picture of what to expect on the bands, and we refer to them collectively as the Solar Indices. In summary, the Solar indices are:
• Solar Flux Index (SFI): Daily high-altitude measure of radio noise at 10.7 cm wavelength (2800MHz) indicative of the amount of Solar flux intensity. Measured daily at 2000UTC at the Pentictin Radio Observatory in Canada. Scale is linear and range is 50 - 300. Rising numbers indicate the potential for greater ionization levels.
• K-index: Measure of variation from daily norm of the geomagnetic field, made at 13 laboratories worldwide and reported every 3-hours. In the US, the measurements are made in Boulder, CO and in Fredericksburg, VA, and reported as the "mid-latitude" K-index. Range is 0 - 9 and rising numbers mean increasing current level of polar magnetic field disturbance, with possible adverse affects on HF propagation for K>4.
• A-index: Also known as the Planetary A-Index (Ap), it is a daily average representation of the worldwide observations of the K-index over the past 24-hours. Range is 0 - 400 and A>27 mean HF propagation over last 24-hours may have been poor.
The solar indices may be interpreted in the following context of their typical time-frames of variation:
 SFI slow change ~ days large rise is good A intermediate change ~ past 24-hours A > 27 was bad K rapid change ~ 3-hours K > 4 is bad
The solar indices are readily available via WWV broadcasts and on the Internet from the NWS Space Environment Center www.sec.noaa.gov). Below is an example of the bulletin they provide in text format:
Eight updated bulletins are issued every 24-hours at 0000, 0300, 0600, 0900, 1200, 1500, 1800, and 2100 UTC unless impending severe conditions warrant additional reports. For a description, see www.sec.noaa.gov/Data/info/WWVdoc.html. Note a few important things about the reports:
1. Date and time of issue is at the top.
2. Date for SFI and A-index is stated; if the report is issued before the current day SFI and A-index measurements are available, then the prior-day values (and date) appear. In this report, the SFI & A-index are for the prior day.
3. K-index is for day of the report (date, time re-stated).
4. Expected geomagnetic disturbance parameters [Geomagnetic Storms (G1-G5), Solar Radiation Storms (S1-S5), Radio Blackouts (R1-R5)] are stated.
In this particular report for 1800 UTC on 05-Dec-06, the A-index of the previous day, 04-Dec-06, was 0, but K = 2 at 18:00 UTC of the current day, indicating that events over the last 18-hours were more unsettled than the day before. Remember that the A-index is a 24-hr average for the previous day and therefore trails the K-index which is a current measurement. From 0000 – 1800UTC, the reported A-index is for the previous 24hours and does not change in the reports. For the 2100UTC report, the last of the day, an estimate of the A-index is used for the current day, based on the 7 measurements collected from 0000 – 1800UTC. At 0000 UTC, the solar flux and K-index for the next day are reported and the actual A-index is for the previous day, derived from all 8 measurements, is reported, dropping the qualification of “estimated”. See the Miscellaneous Notes appendix for more on this.

And so we end our brief review of Solar influences upon atmospheric propagation in anticipation of the new solar cycle 24 and with the hope of improving conditions for working DX in the next few years! Now, we'll explore how all of this is of particular interest to DXers. Now let's look at how the ionosphere allows us to talk to the world.

Ionospheric "Skip"
If it were not for the ionosphere, all RF radiation would propagate outward into the far reaches of the universe and only line-of-sight radio communication would be possible, limited to a few tens-of-miles. However, as illustrated in the figure below, depending upon the frequency, electromagnetic fields of RF waves and the free-electrons in the ionosphere can interact in several ways.
In (a) and (b), the interaction causes the wave to be "bent", a wave phenomena known as refraction; and in (c), the energy of the wave is completely dissipated in the course of the interaction (more on this later).
A point of clarification is in order: when waves traveling in one medium reach an interface (boundary) with another medium, the interaction that ensues may have two different components: a reflected wave that rebounds from the interface boundary and returns to the primary medium, along with a refracted wave that enters and propagates through the new medium (a nice graphical animation of this is available at www.phy.ntnu.edu.tw/java/propagation/propagation.html). While it may appear that the RF waves in case (b) are "reflected" from the ionosphere, that is not the case as there is really not a distinct "boundary" of the neutral atmosphere and ionosphere. In fact, the waves travel for some distance into the increasingly denser regions of the ionosphere (more below about what happens) before the final outcome of their propagation path is determined: whether they only "bend" a bit, then exit into "space"; or whether they are bent sufficiently to return to Earth and allow us to work DX! For this reason, we really should refer to the process as "refraction" and not "reflection". Nevertheless, one frequently encounters the term "reflection" in discussions of wave-ionosphere interactions, just as you will see use of the term "layers".
Of course, DXers are interested in case (b) in which HF-wavelengths are refracted enough to return to the Earth's surface. This occurs usually most effectively by the upper-most F-region, but also on occasion by the E-region, especially below 10 MHz. This process of refraction of the radio waves is called "skip",
and the angle of refraction in this case is generally equal to the angle of incidence [by convention, the angles are measured from the line of vertical incidence]. The radiation angle shown in the figures is that for whatever wave energy is radiated in the direction of the desired communication path; in actuality, depending upon antenna design, energy may be radiated in other directions but does not contribute to the communication link (see discussion of radiation angle in "Antennas" chapter). As the radiation angle decreases, the incident angle increases, allowing the skip distance to increase, up to a maximum of approximately 4,000Km (2,500mi).
Although this is of great benefit to those operating in the HF spectrum, what about waves in the VHF spectrum and above - don't they travel only by line-of-sight? This question brings us to a good spot to pause and take a conceptual look at just how radio waves interact with the ionosphere. In the figure below, we see that the E-field component of the oscillating RF waves will "excite" the free electrons that they encounter, giving up some of the wave's energy by causing the electrons to oscillate in resonance.
The lower frequencies have longer wavelengths, so they will interact with greater numbers of free electrons per cycle than do higher frequencies with shorter wavelengths. Because of the extremely low density of the upper regions of the ionosphere (you could not survive without a breathing source of air), the free electrons rarely encounter other particles, so as the wave goes by, the affected electrons re-radiate the energy from their momentarily "excited" state, returning virtually all of it to the wave. The energy returned by re-radiation from the free electrons may be in a slightly different direction than the incident wave direction, causing the wave to deviate or refract, depending upon the initial incident angle. If the wave encounters enough of the free electrons along its course through the ion region, the amount of the refraction can be enough to "bend" it back to Earth - hence, a "skip". It's not too difficult to understand that as the wavelength decreases (frequency goes up), fewer ions will be encountered per cycle since the shorter wavelengths occupy less volume of space per cycle. For this reason, there is a point at which the wavelengths are too short to interact with enough of the free electrons be effectively refracted back to Earth, so they simply proceed outward into space - no skip!
In the lower ionospheric regions, the density of particles is much higher, with a greater proportion of neutral atoms & molecules compared to free electrons. For this reason, when RF waves excite free electrons in the lower regions, not all of the energy is returned to the wave due to the greater likelihood of the electrons colliding with neutral atoms & molecules and losing their excitation energy (see "Absorption" below).
There isn't a sharp cutoff frequency for skip - it depends upon the ion-density at any given time. Indeed, during times of high solar flux, there may be enough ionization to produce F2 skip as high in frequency as the 6m (50.0 MHz) band and above, allowing sudden openings for several thousand miles and even world-wide. Also, the mid-latitude regions of enhanced electron concentrations formed as a result of the "Equatorial Anomaly" mentioned previously may give rise to an interesting mode of propagation in which signals in the higher-HF and VHF bands may enjoy minimal-loss equatorial transits for significant distances. This is called Trans-Equatorial Propagation (TEP) and has the following characteristics:
• typically occurs from mid-afternoon to early evening local time;
• most frequent at equinoxes and when sunspot number is high;
• MUF above 50 MHz with path-lengths up to 6,000 Km.
In addition, there are interesting phenomena that sometimes occur in the E-region of the ionosphere that can produce skip at VHF and also as low as the 10m band. One such sporadic occurrence is when ion-dense patches form in the E-region resembling ion "clouds" that can produce localized signal skip over hundreds of miles on 6m and may even extend upwards to 2m. The cause of these sporadic "clouds" is unknown but thought to be high-altitude wind-shear. This phenomenon, known as "Sporadic-E" or Es, is one of the reasons why 6m is called the "Magic Band". In summary, most HF skip, and rare worldwide VHF skip, is produced by the F2-region, although for frequencies from ~10 MHz and below there may sometimes be refraction from the E-region; and occasionally, the E-region can produce Es VHF skip.
It should be evident from the figures that the radiation angle at which the signal leaves the Earth is critical to the skip phenomenon. Like the skipping stones on a pond surface, lower radiation angles will result in larger angles-of-incidence and refraction.
So, it appears that the smaller the angle of radiation, the greater will be the skip distance, as illustrated in the figure above, and this is generally true. That is why for DXing we want as much of our radiated signal to leave the antenna at the lowest possible angle from the horizontal in order to achieve the maximum skip. If possible, one would like to have a radiator that emits a signal with most of the energy leaving the antenna parallel to the ground surface, thereby causing it to be incident upon the ionosphere at the maximum angle and so enjoy maximum skip. On the other hand, as the incident angle approaches 0-degrees or vertical incidence (i.e., a so-called "cloud warmer" antenna), the waves "bounce" almost straight down or, as we'll see later, they may just go right out into space (see "MUF" and "fo" below). From this we conclude that a low angle of antenna radiation is very desirable for DX communication. Angles of 30o or less are considered to be low.
OK, so far, so good, but if the maximum skip distance is limited to about 4,000Km (2,500mi) by the geometry of Earth's curvature, how do we manage to work global DX stations over greater distances? For this we can be thankful of the fact that the Earth's surface has conductive properties (remember, you use it to "ground" your station!), so radio waves skipping back to the surface may be refracted back into the atmosphere, allowing for another - or even several more - skips. This is known as "multi-hop" and allows round-the-globe HF communication, illustrated below.
Unfortunately, interactions with the surface are not nearly as energy-efficient as with the ionosphere and signal attenuation occurs with each surface interaction, as well as the non-refractive interactions within the ionosphere (see "Absorption" below). For this reason, when conditions are not ideal, it may be difficult for low-power stations with less-than-optimal antennas to expect to make very long-range contacts. An interesting point to remember is that since salty sea water is a much better conductor than soil, oceans are more effective EM-wave "reflectors" than soil, so that one can expect better multi-hop propagation over long ocean expanses. DXers should appreciate the fact that 70% of the Earth's surface is covered by oceans!
The wave-ionospheric interaction in the F-region that produces HF "skip" is a very efficient process with very little energy loss, but that is not to say that interactions with other ion-regions are as efficient. In fact, as the lower-most regions (E, D) form in the course of the day, they - and especially the D-region - increasingly absorb energy from HF waves.
Absorption
As is the case with most things, there is a good and bad aspect of the ionosphere. The good is obvious: skip, and multi-hop propagation. The bad is that the lower ion regions, especially the D layer, tend to absorb much of the RF energy, attenuating signals rather that allowing them to skip back to Earth. At the lower altitudes, there is a significantly higer density of neutral (non-ionized) atoms and molecules, along with the ionized free electrons. When the waves interact with the free electrons in the lower altitude regions, the some of the excited electrons are more likely to collide with an neutral atom and lose some or all of their momentarily excited energy. In that case, the energy given to the electron cannot be returned to the wave, so the wave loses that bit of energy. The loss of wave energy due to excited electron collisions with neutral atoms is known as ionospheric absorption. As the peak ionization levels rise at midday, ionospheric absorption may rob all of a low power signal energy in the course of a single skip. However, as discussed and illustrated previously, as the wavelength decreases, fewer ions will be encountered per cycle since the shorter wavelengths occupy less volume of space, so there will be less absorption at higher frequencies. In fact, the absorption is inversely proportional to the square of the frequency so that at any given time, if you jump to a band that is twice the frequency of the first, the absorption at the higher frequency will be one-fourth that at the lower frequency:
absorption ~ 1/f2
This means that the absorption at 28 MHz is one-quarter that one would expect at 14 MHz! Quite a large benefit to using the higher frequency if propagation is available! Since D-layer absorption is greatest in the midday hours, this explains why the frequencies below 14MHz are less useful during the daytime than the higher frequencies and, because ionization of the D-layers is least during the hours between dusk and dawn, the lower frequency bands are then good for DXing. So why don't we forget about intermediate frequencies and just operate on our highest frequency bands at night and the lowest frequency bands during the night? Well, indeed, isn't that what we usually do?
Maximum Usable Frequency (MUF)
As a result of the hourly altitudinal density variations that occur in the ion levels, some frequencies may not be usable for communications over a particular global path at certain times of the day. At any given time of a 24-hour period, depending upon the extent of atmospheric ionization and the capability of your antenna to provide low-angle radiation, there is a maximum usable frequency (MUF) that marks the upper-most extent of signal frequency for which maximum skip may be expected. Of course, we can usually easily determine the local MUF at the current time by tuning on increasingly higher frequency bands until it is observed that there are no longer any signals being heard (although sometimes the higher bands have no detectable activity simply because no one is on!). However, the local MUF simply describes the highest useable frequency for first-hop skip. DXers aren't only interested in the first 4,000Km - we would like to be able to work stations around the globe. The problem is that when the local MUF is 28 MHz, it may only be 10 MHz in HZ (Saudi Arabia). What a DXer needs is to know when to expect the optimal conditions along an entire communications path. Not only is it useful to know this for the current time, but there are also times when it would be nice to be able to predict what the path MUF might be on some future date. For example, suppose there was an up-coming DXpedition to a rare entity such as Bouvet Island (3Y) - What would be the highest (least absorption loss) frequency band to expect to be open for propagation to 3Y from your location during the time of the scheduled operation next year? If you had this information, perhaps you could then plan on when to take off from school or work in order to get 'em in the log at an optimal time and band, or whether you should cancel your vacation plans! Can this be done? The answer is yes. By using statistical analysis of accumulated historical data on solar indices and measurements of ionization region heights, values for expected future smoothed solar indices can be estimated and used to predict the MUF for future time periods. Here is a very brief summary of the basics of how it's done.
- Critical Frequency (fo): Measurements of the height and density of the ionospheric layers that reflect signals back to Earth are periodically made at various sites, generally every 15 minutes, using an ionosonde, a device that sends a signal straight up ("vertical incidence") and listens for the echo return. By sweeping through the HF spectrum until the return echo is lost, one can determine the height and ion density of each of the refracting regions (figure below).
The frequency above which the echo is lost is known as the critical frequency (fo). The critical frequency of the different refracting ion regions are designated as foF2, foF1, and foE. These data for fo, h, ion density, time-of-day, and corresponding smoothed solar indices are archived for incorporation into the prediction models described below.
- Maximum Usable frequency (MUF): Contrary to what you might think, the critical frequency, fo, is not the maximum usable frequency (MUF)! If a signal just above fo enters the ionosphere at an incident angle of greater than 0-degrees, it will travel a slightly greater distance through the region than it will at vertical incidence, so it will have a little more time to interact with ions and there is a greater chance for refraction to occur before it exits into space. For this reason, non-vertical (oblique) incidence signals experience a MUF that is higher than fo. In fact, it is equal to some multiple of fo that depends upon the angle of incidence with the ionosphere:
MUF = fo / cos(i)
where "i" is the angle of incidence of the signal upon the ionosphere, measured between vertical incidence and the signal ray. We can call 1/cos(i) the "MUF-factor", or "M-factor", so that MUF = M x fo and note that, since cos(i) is always less than or equal to 1, the M-factor will always be equal to - or greater than 1. For example, for vertical incidence, i=0 degrees and cos(0) =1, so M=1.0; for i = 75 degrees, cos(75) = 0.26, and M = 3.8. Therefore we conclude that the MUF will always be equal to or greater than fo, depending upon incidence angle. Since it is the MUF that is of interest to DXers, we rarely hear or speak of the critical frequency and it is discussed here solely to provide an understanding of its role in estimating the MUF.
Illustrated in the figure below, for skip-mode communication to occur between two points (A,B) using an antenna with radiation angle "a"<90o, the signal frequency cannot exceed the Maximum Usable Frequency (MUF).
Knowing the critical frequency fo and the height of the reflecting ion region, the radiation angle, the distance between two points of interest, and using a bit of geometry and trigonometry, the MUF may be calculated for communication between the two points. Since the fo has been correlated with the historical data of smoothed sunspot numbers, future estimates of the fo, and therefore the MUF, may be calculated. This is what propagation prediction programs do for us.
Before continuing, it may be of interest to point out a curious aspect of the MUF phenomenon. At first glance, the illustration above might lead one to conclude that as the ion-region altitude increases, so will the skip distance and the MUF increase. While this is true for the skip distance, it is just the opposite for the MUF! This may be demonstrated graphically by drawing the above figure with ion layer arcs of increasing radius (increasing "h") and observing that the incident angle "i" becomes nearer to vertical (0 degrees) incidence. Since the angle of incidence will decrease, then according to the MUF = fo/cos(i) equation, because the value of the cosine approaches 1 for smaller angles, the MUF will decrease towards the value of fo! Tsk, tsk! ... whatever propagation giveth, it also taketh away! (To see some simple calculation details of this argument, based upon a 1997 WorldRadio magazine column by Carl, K9LA, see "Propagation notes" in the "Miscellaneous Notes" appendix).
Propagation Prediction Software
Mathematical models of how the solar-terrestrial interactions occur to produce varying ionospheric propagation conditions have been developed and refined over the years. Most are based upon smoothed sunspot numbers derived from monthly averages "smoothed" over a 12-month period (see SSN above), therefore they make predictions averaged over a period of a month. For this reason, the predictive results of the ionospheric propagation models are expressed as median monthly values: for a predicted MUF over a given path during a future month, the actual MUF may be expected to less on half of the days of the month, and greater on half of the days.
Computers have greatly expanded the capability of the models by speeding up the calculations that produce the predictive output, and allowing more variables to be included, such geomagnetic conditions and signal-to-noise ratio (SNR) estimations. Today, DXers have the choice of a number of software propagation programs that can provide, among other useful things, predictions of the monthly MUF for any chosen communications path, along with the expected SNR, given the Tx power and the Tx/Rx antenna gains (see example below). Typically the user specifies the working conditions (i.e., beginning and end-points of the path of interest, antenna type, power, etc.) for a specified future date and the program then estimates the MUF for a period of days bracketing the day of interest. In the past, one had to get the future estimates of the future smoothed sunspot numbers from a graph or chart; today, computer program use online access to solar information databases to download the values for you (e.g., SSN at http://www.ngdc.noaa.gov/stp/SOLAR/ftpsunspotnumber.html).
Things to remember about MUF propagation predictions:
• The estimates are derived by using statistical odds based upon past measurements of fo vs. smoothed sunspot number (or smoothed SFI), so the results are typically median values of the MUF during a month. This means half of the time, the MUF prediction may be higher or lower than the estimate on the day of interest.
• Because MUF calculations may be done only for a single skip at a time, in order to estimate conditions for longer paths it is necessary to make estimations for each of the sequential signal hops required between the two points of interest and then use the lowest estimated MUF value of the sequence as the predicted MUF for the entire path.
• The time-of-day of occurrence of MUFs are dependent upon the chosen path, and times for optimal SP or LP may differ.
• The MUF is important because absorption decreases with increasing frequency, so the best bang for the buck is usually at or near the MUF.
However, one should not get the impression that there are no lower usable frequency bounds since, as one descends in frequency, absorption increases. In fact, it may increase to a point where reliable communication paths cannot be sustained; so there is also a Lowest Usable Frequency (LUF). Experienced DXers are frequently interested in working DX stations on all of the bands for enjoyment of the challenge and possibly in working towards one of more of the various awards. Knowledge of MUF and LUF limitations are useful in knowing where and how to look for DX of interest.
There is an abundance of propagation prediction software available and some programs are free (see References). The figure below illustrates HamCap by VE1NEA, one of the freeware packages listed in the References. HamCap is actually an "envelope" program that uses another freely available application, VOACAP, as the calculation engine for the predictions. Both must be downloaded and installed for use.
In (a) latitude and longitude of the home location are shown (default values entered initially upon installation) and one selects available options; screen (b) allows entry of the DX coordinates (or callsign), along with the anticipated power level of the station (with least power) and the path (SP or LP) for the predictions; the Solar parameter used is the smoothed sunspot number (accessed automatically online), and there is an option to use the K-planetary; also, selecting the "Ant" button (lower-right) provides a list of antenna options; (c) the predicted MUF for the day of interest; and (d) map of SP or LP route, grayline, sun, and predicted signal level conditions (light = good; dark = poor). In (c), the white line shows current time (far-left; 00:00 UTC), and placing the cursor on any spot will show predicted conditions at that time (cursor shows as faint black cross in center-lower yellow square on the 14 MHz line; conditions for the time corresponding to 13:30 UTC, the time corresponding to the x-axis position of the cursor, appear in information bar at bottom). Note in the center-left of the information bar there is a value for predicted signal-to-noise ratio (SNR). Just as one needs to know the useable frequencies, it is also important to have some estimate of the quality of the communication path. Prediction software usually provides estimates for expected SNR based upon historical data, expected noise environment (rural, urban, etc), and signal attenuation for a given power level. In addition, estimates of path reliability as a statistical factor (e.g., "80% of the time") may be available. For a detailed discussion of propagation prediction parameters and a review of the features and availability of several packages (including shareware), see the webpages by ON4SKY at www.astrosurf.com/luxorion/qsl-review-propagation-software.htm.
DXers should have one of more of these programs and spend some time in learning how to use them. It can be very interesting and often illuminating to run predictions for each day in order to become familiar with how they work and how reliable they are. Having said this, I must add a word of caution: one must understand the odds-limitations of the methodology - it is still highly advisable to get up early and tune until late, checking ALL of the bands, no matter what the predictions may say, if you really want to enjoy the best chances of working DX!
Beam Headings: Short Path vs. Long Path
When using omni-directional antennas (low dipole or vertical) one usually just tunes for any audible signal and then proceeds to adjust receiver settings to optimize reception. The direction of optimal signal propagation is rarely considered by most hams until they begin using a directional antenna system. When a directional antenna (e.g., Yagi, quad, phased-vertical array) is used for DXing, it is then necessary to know where to "point" the antenna. Many believe at first that this is too obvious to be of real concern, as initial experiences with new beam antennas are usually for contacts within the immediate geographic region of one or two thousand miles or less. For contacts that are within a single skip range (2500 miles), the sense of RF signal path is based on our innate 2-dimensional ("flat") map concept, and we just point the antenna in the direction of the other station as indicated on the usual "flat" regional maps. However, when one is interested in global DX, the question of the beam heading for the shortest path is a bit more complex. For example, someone in the central USA wishing to contact a station in central China may look at a two-dimensional Mercator projection world-map in an atlas (figure below) and conclude that the beam heading should be almost due East, or 90 degrees in terms of standard compass directions.
(Map courtesy Alabama Maps, Cartographic Research Laboratory, University of Alabama, http://alabamamaps.ua.edu/)
In fact, the proper heading for the shortest path from the central US to central China would be closer to a NNW direction, or 334 degrees! How can this be? The answer has been known to cartographers, navigators, and aircraft pilots for years - following a great circle route is the shortest distance between two points on the globe. The shortest distance, or short path (SP), from your location to another on the far side of the world is a great circle route and great circle beam headings are unique to each point on the globe, so your headings will be different than someone else who lives elsewhere. Don't worry, it's all been figured out for you already! There are countless ham software applications that provide the correct beam heading from your particular location to any of the DXCC entities, or they can be readily obtained for a given prefix by a few mouse-clicks on a DX Cluster. Also, see for example the online program provided on the Magnolia DX Association and the New Jersey DX Association websites (Appendix: DX Organizations). A rule of thumb is that for two locations in the northern hemisphere, the short path will "arc" in a northerly direction; for two points in the southern hemisphere, the short path would "arc" in a southerly direction. For locations that are on the same side of the Equator and are separated by more than 1/4 of the Earth's circumference, the short path between them is most likely over the polar region.
Since we live on a sphere, it isn't too difficult to understand that in traveling from one point to another, one could take the short path, or alternatively, travel in the opposite direction and eventually arrive at the same distant point over a longer route. The longer route has the fancy technical name of long path (LP). Long-path beam headings are found by adding 180 degrees to the short path directions (Note: if the sum is >360, then subtract 360). For the case of a communications path from the central USA to central China, both paths are illustrated in the figure below at approximately mid-day in the USA. Both paths have combinations of daylight and dark segments, although the SP has the shorter daylight leg.
Note that in this case, both the short path and the long path traverse very high latitudes of a polar auroral region. For polar paths, particular consideration must be given to geomagnetic conditions, as geomagnetic storms usually mean significant polar absorption, resulting in weak signals that have a "watery" or "fluttery" sound. Illustrated on a "flat" map (as in the figure above), the curvature of the Earth is represented by the curvature of the paths. Beginning DXers would do well to develop a better understanding of great circle routing (both SP & LP) using a small globe and stretching the shortest possible length of string between the home-location and other distant points .
Skip conditions may persist late into the local nighttime hours due to residual ionization of the upper F-region and minimal remnants of the day's lower (D) absorption region. For this reason, at the time of day (mid-day) in the above figure, one would expect that the SP would be more effective, because of less signal loss on the longer dark leg of the path. However, if one can envision the above figure as it would look a few hours later in the day as the sun moves to the West (i.e., towards the left in the figure) as it approaches sunset in the central US, then the day-night terminus would be closer to the central US and most of the LP would then be in darkness, thereby possibly offering a more favorable path. However, there is no way of knowing which path will actually be best - one should always try both to determine the optimal one.
Long path propagation takes advantage of low absorption conditions and is especially useful for low-power stations. In fact, low-power long-path contacts over distances well beyond the antipodes are common when conditions are right. Because of the low absorption on the night-side of the globe, one often encounters very interesting propagation conditions along the long path in which signals do multi-hops with less signal loss (i.e., stronger signal strength) than via the short path! Finally, many believe that long path propagation is useful only to those with directional antennas - not true! Although a directional antenna does allow optimal use (and recognition, therefore better appreciation) of long path conditions, since isotopic antennas receive and radiate energy in all directions, they are also able to transmit and receive along long path directions. Of course, with an omni-directional antenna, there is no certain way to determine WHEN you're actually using long path propagation.

#### Gray line and Propagation

Ionospheric density, absorption, MUF, and skip change as the Earth rotates on its axis and the region exposed to sunlight changes over a 24 hour period. In general, higher frequencies provide better propagation during the daytime, while low-frequencies are better at night. However, some particularly interesting propagation phenomena can occur during the two short intervals of changeover from dark-to-light at dawn and from light-to-dark at dusk.
DXers have found that during the night/day and day/night transitions of dawn and dusk, excellent opportunities for working DX may sometimes be found on ALL frequencies up to the MUF. A projection of the Day/Night boundary on the world globe is called the grayline which can be seen in the figure below as a faint region of "grayness" along the day-night terminus, which, as in the illustration of SP & LP, appears as a curved line on a "flat" map.
During twice-daily period of night/day transition, HF propagation in areas within or near the grayline may undergo sudden significant changes because:
• at sunrise (SR), the sun begins to build the upper (skip) F-layers, but has not yet produced the lower (absorption) E, D-layers; and
• at sunset (SS), the setting sun has a diminished effect on the lower absorption layers, while much of the upper skip layers may remain.
There are two different aspects of grayline propagation:
• Grayline path propagation: In addition to grayline enhancement of signals from Eastern sunrise-to-dark areas, another very unique aspect of grayline propagation can sometimes occur when the two stations at each end of a path are both in, or near, their local grayline. That is, one is in or near the dawn terminus while the other is in or near the dusk terminus. As the sun rises or sets, the sudden change (increase or decrease) in atmospheric ionization results in an ion-density gradient from high to low altitude - a pinching or "tilting" of the upper (F) region of the ionosphere at either end of the grayline. Instead of skipping back down to the surface, signals encountering a "tilted" F-region may do so at extremely large incidence angles and may be refracted at an angle that takes them along a chord of the Earth's curvature following the North-South path of the grayline. At this point, the signal is enjoying what is essentially a very low angle of radiation. Subsequent interactions with the F-region may result in additional chordal refractions ("hops") until the ray encounters another ionospheric "tilt", at which point it is deflected back to the surface (figure below). This phenomenon is analogous to "whispering galleries" for sound waves and is known as chordal hops.

The advantage of the chordal hop mode of propagation is that, unlike surface interactions during normal skip, there is significantly less signal attenuation by the ionospheric interactions. Because of the likelihood of chordal hop conditions that may be found along the entire length of the grayline, it is possible for signals to experience preferential propagation by traveling along the the actual path of the terminus rather than along either short- or long-path routes. For this reason, grayline propagation conditions are exceptionally advantageous for low-power communications. Since the solar path parallels the equator, the grayline (except for the extreme northern latitudes) runs N or NW to S or SW, crossing the equator. For that reason, the dawn/dusk ionospheric "tilts" straddle the equatorial region, so that chordal hops are often described as being a trans-equatorial propagation mode. Also, since the absorptive D-region changes lag the SR/SS events by some period of time, it may be possible to experience chordal hops for some time after the actual SR/SS grayline period.
Grayline path propagation can result in strange beam headings that are sometimes referred to as a "crooked path". For example, the activation of the Andaman Islands (VU4) in December, 2004, was of great interest to me as it was the last of the DXCC entities that I needed at the time. The intrepid group that went to this very rare entity, last activated in 1987, had numerous obstacles to overcome before obtaining permission to operate on the Indian possession off the coast of Thailand (indeed, they subsequently found themselves in the midst of the terrible tsunami that caused wide-spread destruction and deaths in Indonesia, Thailand, India, Sri Lanka, and East Africa at the end of December '04). The combination of the need to use very modest equipment and antennas for their operation and the fact that it took place near the bottom of sunspot cycle 23 meant that many of us in North America had a very difficult time in hearing them. I was fortunate to work Ms. Bharathi, VU4RBI, one evening on SSB via short-path (335 deg), so early the next morning I began to tune for the CW operator. Years ago I learned that if you cannot hear a station along either the short- or long-paths, you should not hesitate to try different beam headings. Long before I knew anything about chordal hop propagation, I had discovered the "crooked path" route that allowed me, using only 100w and a modest tri-bander at 40ft, to work many Indian, Sri Lankan, and Thailand stations just after my dawn and into my early morning hours. At that time of year, the orientation of my morning grayline runs approximately NNE-SSW, so I turned my quad to the SSW as the grayline came over my location. Sure enough, I began to hear VU4NRO on 20m CW and within a short time, the signals rose nicely to 559 allowing me to make a contact! Afterwards, I swung the quad over 360 degrees and back again to 225 degrees, and I found that the grayline path provided the only copy.
Summary
Here is a summary of the most important aspects of HF propagation discussed above:
• Daily Solar Flux Index (SFI) is an estimate of Solar flux levels over the last 24-hours - higher values may suggest openings on higher bands; however, it is generally a poor indicator of daily conditions.
• Daily A-index is a summary of the geomagnetic conditions over the past 24-hrs - higher values indicate prior-day geomagnetic disturbances; values of A>27 mean geomagnetic storm conditions were present.
• K-index (reported 8 times per day) is a 3-hour summary of the current geomagnetic conditions - rising values signal the potential for geomagnetic storms; K > 4 means storm conditions exist and HF-propagation may be poor.
• Remember: the A is what already happened; the K is what's happening now; if A is high and K is rising it may mean increasingly poor HF - but possibly good VHF - propagation.
• Estimated future smoothed sunspot number or smoothed-SFI (available online) are the propagation predictors of choice.
• Propagation prediction software uses monthly-averaged smoothed SN or SFI, therefore communications path predictions are median values over a month.
• Learn to use propagation prediction software for optimal signal paths, but remember that this is an odds-based estimation of median values.
• Always be aware of the time vs. location of the Sun with respect to the Day/Night regions, so you will pay attention to the terminus (grayline) and the worldwide sunrise/sunset locations.
• Local sunrise may bring enhanced signals from the Dark (West) region; Eastern region sunrise may provide enhanced signal reception in the Dark (West) region.
• Always check both SP & LP routes as possibilities, especially during the hours after local sunrise and sunset.

# HF Propagation

Despite the fact that it involves only a small band of the total electromagnetic wave spectrum, HF propagation is a complex topic. Despite all attempts at brevity, this chapter contains a large amount of material, discussed in greater detail, than most of the others (with the possible exception of "Antennas"). However, without propagation there would be no HF DXing, so it is the most important topic of all and for that reason it is as you see it here. If the material seems to be more than you want to tackle, or if it does not interest you at present, then just skip to the summary at the end for a distillation of the key issues. You are always welcome back at anther time!
Note: Figures are clickable for pop-up enlargement; some are animated.

#### What Affects EM-Wave Propagation?

After operating a ham station for awhile, one soon learns that there are many factors that affect propagation and the mix is continually changing by night/day, month, season, and long-term changes in solar activity. However, all of the factors are related to one thing - our sun, Old Sol. The Sun, a huge nuclear-fusion-powered furnace, continually spews out charged particles and radiation of all almost all wavelengths from the microwave and up. In its annual rotation about the Sun, the Earth is continually bathed in this Solar flux, also referred to as the "Solar wind".
One of the first things that we learn as radio aficionados is that the different bands vary over the course of the daily 24-hour period in their ability to support communication. When the sun rises each morning, the solar radiation from the extreme ultra-violet (EUV) to the X-ray wavelengths begin to bombard the atoms and molecules in the upper atmosphere, causing them to temporarily give up some of their electrons, and resulting in a region of electrically-charged particles consisting of the negatively charged electrons and their now positively-charged parent atom or molecule , as illustrated in the figure below (Note: A collection of charged particles is also called a "plasma"). These charged particles are referred to as "ions" and the process of their production is known as "ionization". This process begins at the extreme outer limits of the atmosphere and progresses downwards toward the surface as the sun rises. As can be seen in the profile graph below called an "ionogram", the density of the free-electron ions is not uniform, but increases smoothly in altitude, with some peaks (E,F) and valleys.
So it is that within a portion of our atmosphere, roughly 100 to 350 Km (60 - 220 miles) above the Earth, is a region of charged particles called the ionosphere. As the day progresses and Solar radiation intensity increases, the density of the ions increases and the depth of the ionosphere extends farther downward towards the Earth's surface. Studies have shown that as the ionization depth increases, density differences may be found within the regions of ionization at different altitudes (see profile graph above). The different ion-regions are formed because the shorter (X-ray) wavelengths are less-readily absorbed and penetrate more readily into the denser lower atmospheric regions, while the longer (EUV) interact readily with the lense-dense upper regions. The different regions are given alphabetical letters to indicate their relative height, beginning with the highest, formed by EUV radiation and known as the F-region, followed later and at a lower altitude by the E-region formed by "soft" X-rays, and finally by late morning, the "shorter wavelength "hard" X-rays form the D-region near the surface. Because the atmospheric density increases with descending altitude, all of the high-energy (X-ray and EUV) radiation is completely absorbed with the formation of the ionosphere. This is fortunately for most living things (marine life and exoskeletal insects like cockroaches would probably be unaffected), as in time, these high-energy wavelengths would be deadly!
While it is often convenient to illustrate and refer to the different ion regions as "layers", as done in some of the figures here and elsewhere, one should understand that there are no distinct boundaries; rather, one region simply builds in density, then "fades", rising again at higher altitude into another into another defined region. This can be seen clearly in the figure below of an Ion Density Profile, or "ionogram" which is just a graph of the altitude vs. the number of electrons per cubic meter.
Descending in altitude, one first sees the bulge of the F-region, the largest of the regions. As it grows in the course of the day, most of the EUV radiation is absorbed in its formation, and it begins to taper off; however, by late-morning, the most energetic of the EUV may cause a small secondary increase in ionization that can be seen in the above graph as a very slight "bump" in the shape of the line between the arrows indicating the "F" and the "E" regions. This new region of the "bump" is more like the F- than the next level E-region. For this reason, the F-region is said to have two components: the higher, larger F2-region, and the lower, smaller, F1-region. Just below the day-time F1-region at a lower altitude, one can see the smaller "nose" of the E-region, and finally, at ~60 Km, just before the graph indicates no more electrons, on can see the very slight "bump" of the D-region. Below the D-region is the non-ionized lower neutral atmosphere.
Later, as the day wanes and the solar influence lessens on the dayside of the globe, the attraction between the ions of different charge, freed of the ionization effects of the solar radiation, begin to recombine into their previously neutral atoms and molecules, literally going to sleep for another night. By early evening, the electron densities of the D- and F1-regions are usually reduced to the point where these regions have little or no affect upon communications on the amateur bands, and the ever-present F2-region contracts to relatively high altitudes during the night hours. During the night hours, sufficient remnants of the E-region remain, although reduced in electron density, and this becomes the primary factor contributing to propagation on 80m and 160m.
In addition to the diurnal (daily) variations due to the Sun, there are other phenomena that influence the ionospheric distributions, including high-altitude winds that can cause shifts in the electron densities, especially the E-region. Other factors will be discussed below.
Since radio waves are electromagnetic phenomena, it is reasonable to assume that they could interact with the ionosphere. In fact, the free electrons readily interact with HF waves and this is of prime interest to us as it is what allows us to work DX! It should be in our best interests to understand more about it.
Solar Variations & Sunspots
Although the daily solar variation is one that is certainly obvious, there are other cyclic events of the Sun that are of great importance to the propagation of radio waves. The Earth-Sun relationship is continually going through four periodic cycles that affect propagation at any given location on the globe:
• 24-hour diurnal (night/day) cycle: rotation of the Earth on its axis, producing daily ionospheric changes.
• Solar axial rotation: cycle of approximately 27 days, resulting in the likelihood of a repeat of the current conditions 27 days later.
• Earth's seasonal cycle: as the Earth orbits the Sun, due to the change in the angle at which sunlight illuminates the globe as a result of the tilt of the Earth's axis with respect to the plane of orbit, the level of solar exposure rises and falls. As seen in a schematic of this, summertime occurs in the Northern hemisphere when that portion of the globe is
"tilted" towards the Sun and therefore under greatest degree of illumination; the least amount of illumination (and coldest temperature range) occurs approximately 6 months later when the Earth is at the opposing point of orbit. Seasonal changes affect the length of daylight hours and therefore the extent of atmospheric ionization: during the summer months, the higher frequency bands have good propagation, even late into evening and night hours, while winter months offer shorter days but the least absorption, providing better day-time high-frequency band conditions than in the Summer and excellent low-band conditions. The Southern hemispheric seasons (and therefore seasonal propagation conditions) are the reverse of those in the North.
• The Sun's 11-year (approximate) seasonal Cycle. The Sun, a ball of extremely hot gases, has equatorial "storm seasons" much the same as occur in the tropical regions of the Earth, except that the increase and decrease in the solar storm season spans approximately 11 Earth-years. During this period, the solar activity ranges from very little, all the way to peak periods in which the solar surface is roiled with large cyclonic areas of intense, tangled magnetic fields. These solar cyclones are cooler than the rest of the surface, so they appear to observers as dark spots, called sunspots, as seen in the figure below of spots and goups that appeared in October, 2003, some years after the peak of cycle 23.
The quantity of sunspots can vary from none to hundreds and records have been kept for the last 200 years. The daily amount of sunspots is known to vary in approximately 11-year cycles as seen in the figure below (courtesy NASA - http://science.msfc.nasa.gov/ssl/pad/solar/images/zurich.gif); however, cycles have been as short as 9 years and as long as 14 years.
Scientists now refer to these sunspot cycles by number, beginning with the first on record from the 1700s, to the current cycle 23, expected to end in early 2007. Counting sunspots on a daily basis is a difficult task, especially since weather can obscure the sun for many hours, and the ablity, skill, and equipment of counters may vary quite a bit. Often, there are not just individual "spots" but also clustered groups. In 1848, to compensate for the difficulties in getting accurate sunspot counts, a Swiss astronomer, J.R. Wolf, developed a method for recording the "sunspots" that accounted for groups and individual spots, as well as variations in viewing conditions. The result was called a sunspot number, R, (as opposed to "... the number of individual sunspots ...") to be used to represent the daily count:
R = k( 10g + f )
where g is the number of sunspot groups and is multiplied by 10 to indicate the greater potency of groups over individuals; f is the number of individual spots, including as many as can be counted within the groups; and k is a "fudge" factor calculated to take into account the viewing conditions (type & power of telescope, etc). This is still used today and is what is meant by the "sunspot number".
Smoothed Sunspot Number (SSN)
Since the Sun affects (among other things) radio propagation, solar activity has been an object of study for many years. Because of the periodic nature of the solar phenomena, an important part of the studies has focused upon ways to predict future solar conditions and events, such as the amount of expected ionization on a given day or the future course of the next cycle. Unfortunately, there is usually a considerable amount of fluctuation in the daily number of spots over he course of a year, making it difficult to associate it with specific conditions on Earth. In order to try to reduce the "jitter", scientists average the daily values each month; even so, the monthly averages still show large variability. To smooth out the numbers even more, a method was developed using a 12-month average of the monthly averages to assign a longer-term average value to the mid-point of the period. So that the mid-point falls at the middle of a month and not at the beginning (or end), the method actually uses 1/2-month values on either end, with 11 full months between: [m1R/2 + m2R + ...+M11R + m13R/2]/12, where m1R - m13R are the averages for each of the months 1-13. The mid-point month of this sequence is the 7th month, so the calculated average would be assigned to it. This value is called the Smoothed Sunspot Number (SSN). For example, in December of 2006, the SSN for June, 2006 would be calculated using 1/2-month averages for December, 2005 and December, 2006, along with monthly averages for January-November, 2006. This means that the sunspot number for June of a given year will not be determined until January, 2007. The smoothing effect of the SSN (blue line) may be readily seen in the figure below, compared to the monthly averages (black dots & line). The figure is courtesy of NOAA Space Environment Center (http://www.sec.noaa.gov/SolarCycle/).
Using analyses of past smoothed sunspot data from previous and current cycles, predictions can be made for several years into the future, as seen in this NASA graphic below for Cycles 23-24 (background image shows huge coronal eruptions).

It is the predicted smoothed sunspot numbers that are used in the propagation prediction programs discussed below.
As the year 2007 is beginning, we are either still at the end of cycle 23 or at the beginning of cycle 24 ... it will actually be year or so before Solar scientists can decide! The estimates at his time are for the cycle minimum to occur between March and September of 2007.
Solar Flux Index
Use of the smoothed sunspot number data accumulated over the years led to improved methods of predicting the rise and fall of future solar cycles, and it is still in use today. However, accurate counting of sunspots was sometimes difficult because of weather and/or observer difficulties. In the late 1940s, scientists discovered a reasonably close correlation between the amount of high-altitude radio noise in the 2800 MHz band (10.7cm ) and the sunspot observations. Since the 10.7cm noise measurements could be standardized and easily made under any weather conditions, this led to the development of monitoring instruments, techniques, and continued daily monitoring. Because it is a measurement of solar flux radiation "noise" levels only at the 10.7cm wavelength and not a measure of all solar flux (particularly not of the Solar UV-radiation responsible for ionospheric formation), they are intended to serve as an index of Solar flux variability. For that reason, the measurement is known as an index of Solar activity, called the Solar Flux Index (SFI).
High-altitude measurements of the radio noise at 10.7 cm flux are made daily at local noon (2000UTC) at the Pentictin Radio Observatory in Penticton, BC, Canada. Measurements are reported in solar flux units (one solar flux unit = 10−22 watt per square meter-hertz) and the values range from 50 - 300, with rising numbers suggesting the possibility of greater ionization levels and possibly better HF propagation. Unfortunately, like the sunspot number, the daily SFI has high variability compared to daily propagation conditions so that it is not a very reliable daily predictor of the band conditions, but just as for the sunspot number, smoothed SFI have proven to be more useful for predictive purposes. For example, the figure below (courtesy NOAA Space Environment Center - http://www.sec.noaa.gov/SolarCycle/) shows how a plot of the monthly average SFI (black dots & line) compares with that of the smoothed SFI (blue line).
A last word about SFI: one may see reported daily values of very few or even zero sunspot numbers with corresponding daily SFI in the range of 50 - 80 or so. This underscores the difference between the two and the fact that the SFI is only a partial measure of daily Solar activity. For perspective, a crudely approximate relationship between smoothed SFI (SSFI) and the smoothed sunspot number (SSN) is
SSFI ~ 60 + SSN
From this we see that, on average, even for zero sunspots, the SSFI is not expected to be zero. While the equation does not apply to the daily solar flux index, it is true that the daily SFI is never zero even if there are no sunspots.
Now a word of caution: many hams can often be heard discussing the SFI as though it were the sunspot number, or in the belief that it was a good indicator of the day's propagation potential. Some may even use the daily SFI in their propagation programs, hoping to get "up-to-the-minute" measures for better predictions. In fact, the single daily measure of SFI is a poor indicator of the daily propagation conditions. As shown by the equation and graph above, when collected and averaged over a year, the smoothed SFI does correlate reasonably well with smoothed sunspot number, but the daily value is all but useless as a predictor of daily propagation conditions.
Geomagnetic Disturbances
We all know that the Earth has a magnetic field, similar to that of a bar-magnet, with the South-magnetic pole approximately co-located with the Earth's geographic North pole, and the N-magnetic pole at the geographic South pole. The geomagnetic field lines flow from the geographic South pole to the North (i.e., N-magnetic to S-magnetic), with maximum filed density at the poles. While the field lines are horizontal at the geo-equator, they rise from the surface below the Equator and enter the surface above the Equator, resulting in an increasing magnetic "dip" angle as one moves higher in latitude toward either pole.
An interesting question at this point is: "Does the Earth's magnetic field have any effect upon HF signals?". After all, radio waves are EM phenomena ... do they interact with the geomagnetic field? The answer to the last part of the question is that the geomagnetic field is too weak to directly interact with HF waves; however, the answer to the broader question of " ... does it have any effect?" is considerably more complex; but, in short, the answer is a resounding "Yes!". Even in its normally calm state, the geomagnetic field interacts with the ionospheric electrons, exerting a force on the electrons in a direction perpendicular to the magnetic field lines, causing the electrons to spiral around the magnetic lines and drift towards the poles. The result is that as the ions form in the course of the day, the free electrons from the Equatorial region drift along the magnetic field lines to the mid-latitudes, producing regions of charge concentrations at the North and South mid-latitudes that are greater than at the Equator, contrary to what one would expect, so it is sometimes called the "Equatorial Anomaly". In addition, other phenomena, some of which are Earth-centered (e.g., earthquakes, or possibly gravity waves) as well as Solar-centered (discussed below) can affect the geomagnetic field. Suffice to say that, in the short-term (days, months), geomagnetic influences are usually the primary daily propagation variable.
While Earth-centered causes of geomagnetic variation are less known, those of the Sun are familiar to all. Large eruptions on the Solar surface can suddenly eject substantially more than the usual amounts of charged particles towards the Earth. As the Sun's activity increases towards each cycle peak, not only is there an increase in the daily Solar flux, but also the likelihood rises for sudden eruptions called solar flares, or even super-events called coronal mass ejections (CME) , that greatly increase the stream of radiation (UV & X-ray) and ions (charged particles) in the "Solar wind" through which the Earth must travel in its solar orbit. When that happens, the charged particles that reach the Earth become entrapped in the magnetic field lines (another good thing for us or we'd be bombarded into extinction, although the cockroaches would probably still survive! ) and are then pulled to the polar regions, spiraling to where the geomagnetic field is strongest. This can cause enlargement of the auroral oval (below) and give rise to spectacular auroral displays (below) that may be seen in the northern latitudes.
Along with the auroral activity, this may result in moderate, large, or on rare occasions, severe variations in the Earth's polar magnetic field. When this happens, it is known as a geomagnetic disturbances, and in the more extreme cases, a geomagnetic storm. How significant are these events? Well, on occasion the geomagnetic storms have been powerful enough to result in damage to satellite electronics, cause complete radio blackouts for several hours, and even disable regional power grids!
Measures of the geomagnetic field have been made since the early 1930s in observation laboratories at different latitudes (see: www.gfz-potsdam.de/pb2/pb23/GeoMag/niemegk/kp_index/kp_sites.html). Currently, there are 13 observatories worldwide using magnetometers to measure any changes in the field intensity at their site in units of nano-Tesla (nT) and reporting the maximum observed results every 3-hours. In the US, the measurements are made in Sitka, AK, Boulder, CO and in Fredericksburg, VA. The latter two are reported as "mid-latitude" values. Geomagnetic field variations, when they occur due to solar influences, may be several orders of magnitude above normal (0 - 500nT or more), so the nT measurements are transformed using a logarithmic scale with a "fudge factor" that removes the latitudinal differences. The log-transform results in a "compression" of the scale and the results are reported as an index of the field values, known as the K-index, with a range of 0 - 9. Since the magnetometer readings (in nT) are corrected for latitude when converted to K-values, all of the observatory K-values are equivalent worldwide regardless of observatory latitude. A K-index of zero means that the geomagnetic field strength is unchanged or normal, while values > 0 mean that it is disturbed, and a rising K-index indicates there is an increasing likelihood for possible geomagnetic storms, which are said to exist when K > 4 (see "Miscellaneous Notes" appendix for storm alert codes). Geomagnetic storms can cause ionospheric changes that result in disruptions of HF propagation, so they are of great concern to DXers.
The K-index is a short-term indicator of magnetic field conditions, reported every 3-hours. It would be useful as well to have some measure of the recent geomagnetic activity, for example, an "average K" for the last 24-hours. However, since the K-values are logarithmic, the 8 daily values for each 3-hour period cannot be averaged (for the same reason that you can't average dB values). In order to obtain an average measure of the daily geomagnetic report, the logarithmic K-scale is essentially "decompressed" into "a-values" every 3-hours at each observatory using the table below:
 K-to-A Transformation Scale K 0 1 2 3 4 5 6 7 8 9 a 0 3 7 15 27 48 80 132 208 400
Since the K-values have already been adjusted for latitude, it follows that the a-values are independent of latitude as well. Each of the observatories reports its a-values and the 24-hour average of all worldwide a-values is reported as the daily Planetary A-Index (Ap), or just the "A-index" (See "Misc. Notes" appendix for an example). The A-index gives us an indication of the average condition of the geomagnetic field over the last 24-hours. Like sunspots, geomagnetic activity is also periodic with an approximate 6-year cycle (re: http://www.ips.gov.au/Educational/3/1/3), so that peaks seem to occur at the low points of the sunspot cycle. Because of the shorter period, geomagnetic influences, on average, occur throughout the sunspot cycle. This means that, with the exception of the ionization extremes at very high and very low SFI, it is the geomagnetic field activity described by the K,A indices that causes most of the variations in our day-to-day DXing.
Solar Indices
Today, there are many observed and recorded solar measurements that are available for describing and predicting what is happening on - or in - the Sun, but for DXers, the three that were described above are readily accessible and particularly useful for gauging the solar influence upon the ham bands on a daily basis. Reported every 3-hours, eight times per day, they can give a pretty good daily picture of what to expect on the bands, and we refer to them collectively as the Solar Indices. In summary, the Solar indices are:
• Solar Flux Index (SFI): Daily high-altitude measure of radio noise at 10.7 cm wavelength (2800MHz) indicative of the amount of Solar flux intensity. Measured daily at 2000UTC at the Pentictin Radio Observatory in Canada. Scale is linear and range is 50 - 300. Rising numbers indicate the potential for greater ionization levels.
• K-index: Measure of variation from daily norm of the geomagnetic field, made at 13 laboratories worldwide and reported every 3-hours. In the US, the measurements are made in Boulder, CO and in Fredericksburg, VA, and reported as the "mid-latitude" K-index. Range is 0 - 9 and rising numbers mean increasing current level of polar magnetic field disturbance, with possible adverse affects on HF propagation for K>4.
• A-index: Also known as the Planetary A-Index (Ap), it is a daily average representation of the worldwide observations of the K-index over the past 24-hours. Range is 0 - 400 and A>27 mean HF propagation over last 24-hours may have been poor.
The solar indices may be interpreted in the following context of their typical time-frames of variation:
 SFI slow change ~ days large rise is good A intermediate change ~ past 24-hours A > 27 was bad K rapid change ~ 3-hours K > 4 is bad
The solar indices are readily available via WWV broadcasts and on the Internet from the NWS Space Environment Center www.sec.noaa.gov). Below is an example of the bulletin they provide in text format:
Eight updated bulletins are issued every 24-hours at 0000, 0300, 0600, 0900, 1200, 1500, 1800, and 2100 UTC unless impending severe conditions warrant additional reports. For a description, see www.sec.noaa.gov/Data/info/WWVdoc.html. Note a few important things about the reports:
1. Date and time of issue is at the top.
2. Date for SFI and A-index is stated; if the report is issued before the current day SFI and A-index measurements are available, then the prior-day values (and date) appear. In this report, the SFI & A-index are for the prior day.
3. K-index is for day of the report (date, time re-stated).
4. Expected geomagnetic disturbance parameters [Geomagnetic Storms (G1-G5), Solar Radiation Storms (S1-S5), Radio Blackouts (R1-R5)] are stated.
In this particular report for 1800 UTC on 05-Dec-06, the A-index of the previous day, 04-Dec-06, was 0, but K = 2 at 18:00 UTC of the current day, indicating that events over the last 18-hours were more unsettled than the day before. Remember that the A-index is a 24-hr average for the previous day and therefore trails the K-index which is a current measurement. From 0000 – 1800UTC, the reported A-index is for the previous 24hours and does not change in the reports. For the 2100UTC report, the last of the day, an estimate of the A-index is used for the current day, based on the 7 measurements collected from 0000 – 1800UTC. At 0000 UTC, the solar flux and K-index for the next day are reported and the actual A-index is for the previous day, derived from all 8 measurements, is reported, dropping the qualification of “estimated”. See the Miscellaneous Notes appendix for more on this.

And so we end our brief review of Solar influences upon atmospheric propagation in anticipation of the new solar cycle 24 and with the hope of improving conditions for working DX in the next few years! Now, we'll explore how all of this is of particular interest to DXers. Now let's look at how the ionosphere allows us to talk to the world.

Ionospheric "Skip"
If it were not for the ionosphere, all RF radiation would propagate outward into the far reaches of the universe and only line-of-sight radio communication would be possible, limited to a few tens-of-miles. However, as illustrated in the figure below, depending upon the frequency, electromagnetic fields of RF waves and the free-electrons in the ionosphere can interact in several ways.
In (a) and (b), the interaction causes the wave to be "bent", a wave phenomena known as refraction; and in (c), the energy of the wave is completely dissipated in the course of the interaction (more on this later).
A point of clarification is in order: when waves traveling in one medium reach an interface (boundary) with another medium, the interaction that ensues may have two different components: a reflected wave that rebounds from the interface boundary and returns to the primary medium, along with a refracted wave that enters and propagates through the new medium (a nice graphical animation of this is available at www.phy.ntnu.edu.tw/java/propagation/propagation.html). While it may appear that the RF waves in case (b) are "reflected" from the ionosphere, that is not the case as there is really not a distinct "boundary" of the neutral atmosphere and ionosphere. In fact, the waves travel for some distance into the increasingly denser regions of the ionosphere (more below about what happens) before the final outcome of their propagation path is determined: whether they only "bend" a bit, then exit into "space"; or whether they are bent sufficiently to return to Earth and allow us to work DX! For this reason, we really should refer to the process as "refraction" and not "reflection". Nevertheless, one frequently encounters the term "reflection" in discussions of wave-ionosphere interactions, just as you will see use of the term "layers".
Of course, DXers are interested in case (b) in which HF-wavelengths are refracted enough to return to the Earth's surface. This occurs usually most effectively by the upper-most F-region, but also on occasion by the E-region, especially below 10 MHz. This process of refraction of the radio waves is called "skip",
and the angle of refraction in this case is generally equal to the angle of incidence [by convention, the angles are measured from the line of vertical incidence]. The radiation angle shown in the figures is that for whatever wave energy is radiated in the direction of the desired communication path; in actuality, depending upon antenna design, energy may be radiated in other directions but does not contribute to the communication link (see discussion of radiation angle in "Antennas" chapter). As the radiation angle decreases, the incident angle increases, allowing the skip distance to increase, up to a maximum of approximately 4,000Km (2,500mi).
Although this is of great benefit to those operating in the HF spectrum, what about waves in the VHF spectrum and above - don't they travel only by line-of-sight? This question brings us to a good spot to pause and take a conceptual look at just how radio waves interact with the ionosphere. In the figure below, we see that the E-field component of the oscillating RF waves will "excite" the free electrons that they encounter, giving up some of the wave's energy by causing the electrons to oscillate in resonance.
The lower frequencies have longer wavelengths, so they will interact with greater numbers of free electrons per cycle than do higher frequencies with shorter wavelengths. Because of the extremely low density of the upper regions of the ionosphere (you could not survive without a breathing source of air), the free electrons rarely encounter other particles, so as the wave goes by, the affected electrons re-radiate the energy from their momentarily "excited" state, returning virtually all of it to the wave. The energy returned by re-radiation from the free electrons may be in a slightly different direction than the incident wave direction, causing the wave to deviate or refract, depending upon the initial incident angle. If the wave encounters enough of the free electrons along its course through the ion region, the amount of the refraction can be enough to "bend" it back to Earth - hence, a "skip". It's not too difficult to understand that as the wavelength decreases (frequency goes up), fewer ions will be encountered per cycle since the shorter wavelengths occupy less volume of space per cycle. For this reason, there is a point at which the wavelengths are too short to interact with enough of the free electrons be effectively refracted back to Earth, so they simply proceed outward into space - no skip!
In the lower ionospheric regions, the density of particles is much higher, with a greater proportion of neutral atoms & molecules compared to free electrons. For this reason, when RF waves excite free electrons in the lower regions, not all of the energy is returned to the wave due to the greater likelihood of the electrons colliding with neutral atoms & molecules and losing their excitation energy (see "Absorption" below).
There isn't a sharp cutoff frequency for skip - it depends upon the ion-density at any given time. Indeed, during times of high solar flux, there may be enough ionization to produce F2 skip as high in frequency as the 6m (50.0 MHz) band and above, allowing sudden openings for several thousand miles and even world-wide. Also, the mid-latitude regions of enhanced electron concentrations formed as a result of the "Equatorial Anomaly" mentioned previously may give rise to an interesting mode of propagation in which signals in the higher-HF and VHF bands may enjoy minimal-loss equatorial transits for significant distances. This is called Trans-Equatorial Propagation (TEP) and has the following characteristics:
• typically occurs from mid-afternoon to early evening local time;
• most frequent at equinoxes and when sunspot number is high;
• MUF above 50 MHz with path-lengths up to 6,000 Km.
In addition, there are interesting phenomena that sometimes occur in the E-region of the ionosphere that can produce skip at VHF and also as low as the 10m band. One such sporadic occurrence is when ion-dense patches form in the E-region resembling ion "clouds" that can produce localized signal skip over hundreds of miles on 6m and may even extend upwards to 2m. The cause of these sporadic "clouds" is unknown but thought to be high-altitude wind-shear. This phenomenon, known as "Sporadic-E" or Es, is one of the reasons why 6m is called the "Magic Band". In summary, most HF skip, and rare worldwide VHF skip, is produced by the F2-region, although for frequencies from ~10 MHz and below there may sometimes be refraction from the E-region; and occasionally, the E-region can produce Es VHF skip.
It should be evident from the figures that the radiation angle at which the signal leaves the Earth is critical to the skip phenomenon. Like the skipping stones on a pond surface, lower radiation angles will result in larger angles-of-incidence and refraction.
So, it appears that the smaller the angle of radiation, the greater will be the skip distance, as illustrated in the figure above, and this is generally true. That is why for DXing we want as much of our radiated signal to leave the antenna at the lowest possible angle from the horizontal in order to achieve the maximum skip. If possible, one would like to have a radiator that emits a signal with most of the energy leaving the antenna parallel to the ground surface, thereby causing it to be incident upon the ionosphere at the maximum angle and so enjoy maximum skip. On the other hand, as the incident angle approaches 0-degrees or vertical incidence (i.e., a so-called "cloud warmer" antenna), the waves "bounce" almost straight down or, as we'll see later, they may just go right out into space (see "MUF" and "fo" below). From this we conclude that a low angle of antenna radiation is very desirable for DX communication. Angles of 30o or less are considered to be low.
OK, so far, so good, but if the maximum skip distance is limited to about 4,000Km (2,500mi) by the geometry of Earth's curvature, how do we manage to work global DX stations over greater distances? For this we can be thankful of the fact that the Earth's surface has conductive properties (remember, you use it to "ground" your station!), so radio waves skipping back to the surface may be refracted back into the atmosphere, allowing for another - or even several more - skips. This is known as "multi-hop" and allows round-the-globe HF communication, illustrated below.
Unfortunately, interactions with the surface are not nearly as energy-efficient as with the ionosphere and signal attenuation occurs with each surface interaction, as well as the non-refractive interactions within the ionosphere (see "Absorption" below). For this reason, when conditions are not ideal, it may be difficult for low-power stations with less-than-optimal antennas to expect to make very long-range contacts. An interesting point to remember is that since salty sea water is a much better conductor than soil, oceans are more effective EM-wave "reflectors" than soil, so that one can expect better multi-hop propagation over long ocean expanses. DXers should appreciate the fact that 70% of the Earth's surface is covered by oceans!
The wave-ionospheric interaction in the F-region that produces HF "skip" is a very efficient process with very little energy loss, but that is not to say that interactions with other ion-regions are as efficient. In fact, as the lower-most regions (E, D) form in the course of the day, they - and especially the D-region - increasingly absorb energy from HF waves.
Absorption
As is the case with most things, there is a good and bad aspect of the ionosphere. The good is obvious: skip, and multi-hop propagation. The bad is that the lower ion regions, especially the D layer, tend to absorb much of the RF energy, attenuating signals rather that allowing them to skip back to Earth. At the lower altitudes, there is a significantly higer density of neutral (non-ionized) atoms and molecules, along with the ionized free electrons. When the waves interact with the free electrons in the lower altitude regions, the some of the excited electrons are more likely to collide with an neutral atom and lose some or all of their momentarily excited energy. In that case, the energy given to the electron cannot be returned to the wave, so the wave loses that bit of energy. The loss of wave energy due to excited electron collisions with neutral atoms is known as ionospheric absorption. As the peak ionization levels rise at midday, ionospheric absorption may rob all of a low power signal energy in the course of a single skip. However, as discussed and illustrated previously, as the wavelength decreases, fewer ions will be encountered per cycle since the shorter wavelengths occupy less volume of space, so there will be less absorption at higher frequencies. In fact, the absorption is inversely proportional to the square of the frequency so that at any given time, if you jump to a band that is twice the frequency of the first, the absorption at the higher frequency will be one-fourth that at the lower frequency:
absorption ~ 1/f2
This means that the absorption at 28 MHz is one-quarter that one would expect at 14 MHz! Quite a large benefit to using the higher frequency if propagation is available! Since D-layer absorption is greatest in the midday hours, this explains why the frequencies below 14MHz are less useful during the daytime than the higher frequencies and, because ionization of the D-layers is least during the hours between dusk and dawn, the lower frequency bands are then good for DXing. So why don't we forget about intermediate frequencies and just operate on our highest frequency bands at night and the lowest frequency bands during the night? Well, indeed, isn't that what we usually do?
Maximum Usable Frequency (MUF)
As a result of the hourly altitudinal density variations that occur in the ion levels, some frequencies may not be usable for communications over a particular global path at certain times of the day. At any given time of a 24-hour period, depending upon the extent of atmospheric ionization and the capability of your antenna to provide low-angle radiation, there is a maximum usable frequency (MUF) that marks the upper-most extent of signal frequency for which maximum skip may be expected. Of course, we can usually easily determine the local MUF at the current time by tuning on increasingly higher frequency bands until it is observed that there are no longer any signals being heard (although sometimes the higher bands have no detectable activity simply because no one is on!). However, the local MUF simply describes the highest useable frequency for first-hop skip. DXers aren't only interested in the first 4,000Km - we would like to be able to work stations around the globe. The problem is that when the local MUF is 28 MHz, it may only be 10 MHz in HZ (Saudi Arabia). What a DXer needs is to know when to expect the optimal conditions along an entire communications path. Not only is it useful to know this for the current time, but there are also times when it would be nice to be able to predict what the path MUF might be on some future date. For example, suppose there was an up-coming DXpedition to a rare entity such as Bouvet Island (3Y) - What would be the highest (least absorption loss) frequency band to expect to be open for propagation to 3Y from your location during the time of the scheduled operation next year? If you had this information, perhaps you could then plan on when to take off from school or work in order to get 'em in the log at an optimal time and band, or whether you should cancel your vacation plans! Can this be done? The answer is yes. By using statistical analysis of accumulated historical data on solar indices and measurements of ionization region heights, values for expected future smoothed solar indices can be estimated and used to predict the MUF for future time periods. Here is a very brief summary of the basics of how it's done.
- Critical Frequency (fo): Measurements of the height and density of the ionospheric layers that reflect signals back to Earth are periodically made at various sites, generally every 15 minutes, using an ionosonde, a device that sends a signal straight up ("vertical incidence") and listens for the echo return. By sweeping through the HF spectrum until the return echo is lost, one can determine the height and ion density of each of the refracting regions (figure below).
The frequency above which the echo is lost is known as the critical frequency (fo). The critical frequency of the different refracting ion regions are designated as foF2, foF1, and foE. These data for fo, h, ion density, time-of-day, and corresponding smoothed solar indices are archived for incorporation into the prediction models described below.
- Maximum Usable frequency (MUF): Contrary to what you might think, the critical frequency, fo, is not the maximum usable frequency (MUF)! If a signal just above fo enters the ionosphere at an incident angle of greater than 0-degrees, it will travel a slightly greater distance through the region than it will at vertical incidence, so it will have a little more time to interact with ions and there is a greater chance for refraction to occur before it exits into space. For this reason, non-vertical (oblique) incidence signals experience a MUF that is higher than fo. In fact, it is equal to some multiple of fo that depends upon the angle of incidence with the ionosphere:
MUF = fo / cos(i)
where "i" is the angle of incidence of the signal upon the ionosphere, measured between vertical incidence and the signal ray. We can call 1/cos(i) the "MUF-factor", or "M-factor", so that MUF = M x fo and note that, since cos(i) is always less than or equal to 1, the M-factor will always be equal to - or greater than 1. For example, for vertical incidence, i=0 degrees and cos(0) =1, so M=1.0; for i = 75 degrees, cos(75) = 0.26, and M = 3.8. Therefore we conclude that the MUF will always be equal to or greater than fo, depending upon incidence angle. Since it is the MUF that is of interest to DXers, we rarely hear or speak of the critical frequency and it is discussed here solely to provide an understanding of its role in estimating the MUF.
Illustrated in the figure below, for skip-mode communication to occur between two points (A,B) using an antenna with radiation angle "a"<90o, the signal frequency cannot exceed the Maximum Usable Frequency (MUF).
Knowing the critical frequency fo and the height of the reflecting ion region, the radiation angle, the distance between two points of interest, and using a bit of geometry and trigonometry, the MUF may be calculated for communication between the two points. Since the fo has been correlated with the historical data of smoothed sunspot numbers, future estimates of the fo, and therefore the MUF, may be calculated. This is what propagation prediction programs do for us.
Before continuing, it may be of interest to point out a curious aspect of the MUF phenomenon. At first glance, the illustration above might lead one to conclude that as the ion-region altitude increases, so will the skip distance and the MUF increase. While this is true for the skip distance, it is just the opposite for the MUF! This may be demonstrated graphically by drawing the above figure with ion layer arcs of increasing radius (increasing "h") and observing that the incident angle "i" becomes nearer to vertical (0 degrees) incidence. Since the angle of incidence will decrease, then according to the MUF = fo/cos(i) equation, because the value of the cosine approaches 1 for smaller angles, the MUF will decrease towards the value of fo! Tsk, tsk! ... whatever propagation giveth, it also taketh away! (To see some simple calculation details of this argument, based upon a 1997 WorldRadio magazine column by Carl, K9LA, see "Propagation notes" in the "Miscellaneous Notes" appendix).
Propagation Prediction Software
Mathematical models of how the solar-terrestrial interactions occur to produce varying ionospheric propagation conditions have been developed and refined over the years. Most are based upon smoothed sunspot numbers derived from monthly averages "smoothed" over a 12-month period (see SSN above), therefore they make predictions averaged over a period of a month. For this reason, the predictive results of the ionospheric propagation models are expressed as median monthly values: for a predicted MUF over a given path during a future month, the actual MUF may be expected to less on half of the days of the month, and greater on half of the days.
Computers have greatly expanded the capability of the models by speeding up the calculations that produce the predictive output, and allowing more variables to be included, such geomagnetic conditions and signal-to-noise ratio (SNR) estimations. Today, DXers have the choice of a number of software propagation programs that can provide, among other useful things, predictions of the monthly MUF for any chosen communications path, along with the expected SNR, given the Tx power and the Tx/Rx antenna gains (see example below). Typically the user specifies the working conditions (i.e., beginning and end-points of the path of interest, antenna type, power, etc.) for a specified future date and the program then estimates the MUF for a period of days bracketing the day of interest. In the past, one had to get the future estimates of the future smoothed sunspot numbers from a graph or chart; today, computer program use online access to solar information databases to download the values for you (e.g., SSN at http://www.ngdc.noaa.gov/stp/SOLAR/ftpsunspotnumber.html).
Things to remember about MUF propagation predictions:
• The estimates are derived by using statistical odds based upon past measurements of fo vs. smoothed sunspot number (or smoothed SFI), so the results are typically median values of the MUF during a month. This means half of the time, the MUF prediction may be higher or lower than the estimate on the day of interest.
• Because MUF calculations may be done only for a single skip at a time, in order to estimate conditions for longer paths it is necessary to make estimations for each of the sequential signal hops required between the two points of interest and then use the lowest estimated MUF value of the sequence as the predicted MUF for the entire path.
• The time-of-day of occurrence of MUFs are dependent upon the chosen path, and times for optimal SP or LP may differ.
• The MUF is important because absorption decreases with increasing frequency, so the best bang for the buck is usually at or near the MUF.
However, one should not get the impression that there are no lower usable frequency bounds since, as one descends in frequency, absorption increases. In fact, it may increase to a point where reliable communication paths cannot be sustained; so there is also a Lowest Usable Frequency (LUF). Experienced DXers are frequently interested in working DX stations on all of the bands for enjoyment of the challenge and possibly in working towards one of more of the various awards. Knowledge of MUF and LUF limitations are useful in knowing where and how to look for DX of interest.
There is an abundance of propagation prediction software available and some programs are free (see References). The figure below illustrates HamCap by VE1NEA, one of the freeware packages listed in the References. HamCap is actually an "envelope" program that uses another freely available application, VOACAP, as the calculation engine for the predictions. Both must be downloaded and installed for use.
In (a) latitude and longitude of the home location are shown (default values entered initially upon installation) and one selects available options; screen (b) allows entry of the DX coordinates (or callsign), along with the anticipated power level of the station (with least power) and the path (SP or LP) for the predictions; the Solar parameter used is the smoothed sunspot number (accessed automatically online), and there is an option to use the K-planetary; also, selecting the "Ant" button (lower-right) provides a list of antenna options; (c) the predicted MUF for the day of interest; and (d) map of SP or LP route, grayline, sun, and predicted signal level conditions (light = good; dark = poor). In (c), the white line shows current time (far-left; 00:00 UTC), and placing the cursor on any spot will show predicted conditions at that time (cursor shows as faint black cross in center-lower yellow square on the 14 MHz line; conditions for the time corresponding to 13:30 UTC, the time corresponding to the x-axis position of the cursor, appear in information bar at bottom). Note in the center-left of the information bar there is a value for predicted signal-to-noise ratio (SNR). Just as one needs to know the useable frequencies, it is also important to have some estimate of the quality of the communication path. Prediction software usually provides estimates for expected SNR based upon historical data, expected noise environment (rural, urban, etc), and signal attenuation for a given power level. In addition, estimates of path reliability as a statistical factor (e.g., "80% of the time") may be available. For a detailed discussion of propagation prediction parameters and a review of the features and availability of several packages (including shareware), see the webpages by ON4SKY at www.astrosurf.com/luxorion/qsl-review-propagation-software.htm.
DXers should have one of more of these programs and spend some time in learning how to use them. It can be very interesting and often illuminating to run predictions for each day in order to become familiar with how they work and how reliable they are. Having said this, I must add a word of caution: one must understand the odds-limitations of the methodology - it is still highly advisable to get up early and tune until late, checking ALL of the bands, no matter what the predictions may say, if you really want to enjoy the best chances of working DX!
Beam Headings: Short Path vs. Long Path
When using omni-directional antennas (low dipole or vertical) one usually just tunes for any audible signal and then proceeds to adjust receiver settings to optimize reception. The direction of optimal signal propagation is rarely considered by most hams until they begin using a directional antenna system. When a directional antenna (e.g., Yagi, quad, phased-vertical array) is used for DXing, it is then necessary to know where to "point" the antenna. Many believe at first that this is too obvious to be of real concern, as initial experiences with new beam antennas are usually for contacts within the immediate geographic region of one or two thousand miles or less. For contacts that are within a single skip range (2500 miles), the sense of RF signal path is based on our innate 2-dimensional ("flat") map concept, and we just point the antenna in the direction of the other station as indicated on the usual "flat" regional maps. However, when one is interested in global DX, the question of the beam heading for the shortest path is a bit more complex. For example, someone in the central USA wishing to contact a station in central China may look at a two-dimensional Mercator projection world-map in an atlas (figure below) and conclude that the beam heading should be almost due East, or 90 degrees in terms of standard compass directions.
(Map courtesy Alabama Maps, Cartographic Research Laboratory, University of Alabama, http://alabamamaps.ua.edu/)
In fact, the proper heading for the shortest path from the central US to central China would be closer to a NNW direction, or 334 degrees! How can this be? The answer has been known to cartographers, navigators, and aircraft pilots for years - following a great circle route is the shortest distance between two points on the globe. The shortest distance, or short path (SP), from your location to another on the far side of the world is a great circle route and great circle beam headings are unique to each point on the globe, so your headings will be different than someone else who lives elsewhere. Don't worry, it's all been figured out for you already! There are countless ham software applications that provide the correct beam heading from your particular location to any of the DXCC entities, or they can be readily obtained for a given prefix by a few mouse-clicks on a DX Cluster. Also, see for example the online program provided on the Magnolia DX Association and the New Jersey DX Association websites (Appendix: DX Organizations). A rule of thumb is that for two locations in the northern hemisphere, the short path will "arc" in a northerly direction; for two points in the southern hemisphere, the short path would "arc" in a southerly direction. For locations that are on the same side of the Equator and are separated by more than 1/4 of the Earth's circumference, the short path between them is most likely over the polar region.
Since we live on a sphere, it isn't too difficult to understand that in traveling from one point to another, one could take the short path, or alternatively, travel in the opposite direction and eventually arrive at the same distant point over a longer route. The longer route has the fancy technical name of long path (LP). Long-path beam headings are found by adding 180 degrees to the short path directions (Note: if the sum is >360, then subtract 360). For the case of a communications path from the central USA to central China, both paths are illustrated in the figure below at approximately mid-day in the USA. Both paths have combinations of daylight and dark segments, although the SP has the shorter daylight leg.
Note that in this case, both the short path and the long path traverse very high latitudes of a polar auroral region. For polar paths, particular consideration must be given to geomagnetic conditions, as geomagnetic storms usually mean significant polar absorption, resulting in weak signals that have a "watery" or "fluttery" sound. Illustrated on a "flat" map (as in the figure above), the curvature of the Earth is represented by the curvature of the paths. Beginning DXers would do well to develop a better understanding of great circle routing (both SP & LP) using a small globe and stretching the shortest possible length of string between the home-location and other distant points .
Skip conditions may persist late into the local nighttime hours due to residual ionization of the upper F-region and minimal remnants of the day's lower (D) absorption region. For this reason, at the time of day (mid-day) in the above figure, one would expect that the SP would be more effective, because of less signal loss on the longer dark leg of the path. However, if one can envision the above figure as it would look a few hours later in the day as the sun moves to the West (i.e., towards the left in the figure) as it approaches sunset in the central US, then the day-night terminus would be closer to the central US and most of the LP would then be in darkness, thereby possibly offering a more favorable path. However, there is no way of knowing which path will actually be best - one should always try both to determine the optimal one.
Long path propagation takes advantage of low absorption conditions and is especially useful for low-power stations. In fact, low-power long-path contacts over distances well beyond the antipodes are common when conditions are right. Because of the low absorption on the night-side of the globe, one often encounters very interesting propagation conditions along the long path in which signals do multi-hops with less signal loss (i.e., stronger signal strength) than via the short path! Finally, many believe that long path propagation is useful only to those with directional antennas - not true! Although a directional antenna does allow optimal use (and recognition, therefore better appreciation) of long path conditions, since isotopic antennas receive and radiate energy in all directions, they are also able to transmit and receive along long path directions. Of course, with an omni-directional antenna, there is no certain way to determine WHEN you're actually using long path propagation.

#### Gray line and Propagation

Ionospheric density, absorption, MUF, and skip change as the Earth rotates on its axis and the region exposed to sunlight changes over a 24 hour period. In general, higher frequencies provide better propagation during the daytime, while low-frequencies are better at night. However, some particularly interesting propagation phenomena can occur during the two short intervals of changeover from dark-to-light at dawn and from light-to-dark at dusk.
DXers have found that during the night/day and day/night transitions of dawn and dusk, excellent opportunities for working DX may sometimes be found on ALL frequencies up to the MUF. A projection of the Day/Night boundary on the world globe is called the grayline which can be seen in the figure below as a faint region of "grayness" along the day-night terminus, which, as in the illustration of SP & LP, appears as a curved line on a "flat" map.
During twice-daily period of night/day transition, HF propagation in areas within or near the grayline may undergo sudden significant changes because:
• at sunrise (SR), the sun begins to build the upper (skip) F-layers, but has not yet produced the lower (absorption) E, D-layers; and
• at sunset (SS), the setting sun has a diminished effect on the lower absorption layers, while much of the upper skip layers may remain.
There are two different aspects of grayline propagation:
• Grayline path propagation: In addition to grayline enhancement of signals from Eastern sunrise-to-dark areas, another very unique aspect of grayline propagation can sometimes occur when the two stations at each end of a path are both in, or near, their local grayline. That is, one is in or near the dawn terminus while the other is in or near the dusk terminus. As the sun rises or sets, the sudden change (increase or decrease) in atmospheric ionization results in an ion-density gradient from high to low altitude - a pinching or "tilting" of the upper (F) region of the ionosphere at either end of the grayline. Instead of skipping back down to the surface, signals encountering a "tilted" F-region may do so at extremely large incidence angles and may be refracted at an angle that takes them along a chord of the Earth's curvature following the North-South path of the grayline. At this point, the signal is enjoying what is essentially a very low angle of radiation. Subsequent interactions with the F-region may result in additional chordal refractions ("hops") until the ray encounters another ionospheric "tilt", at which point it is deflected back to the surface (figure below). This phenomenon is analogous to "whispering galleries" for sound waves and is known as chordal hops.

The advantage of the chordal hop mode of propagation is that, unlike surface interactions during normal skip, there is significantly less signal attenuation by the ionospheric interactions. Because of the likelihood of chordal hop conditions that may be found along the entire length of the grayline, it is possible for signals to experience preferential propagation by traveling along the the actual path of the terminus rather than along either short- or long-path routes. For this reason, grayline propagation conditions are exceptionally advantageous for low-power communications. Since the solar path parallels the equator, the grayline (except for the extreme northern latitudes) runs N or NW to S or SW, crossing the equator. For that reason, the dawn/dusk ionospheric "tilts" straddle the equatorial region, so that chordal hops are often described as being a trans-equatorial propagation mode. Also, since the absorptive D-region changes lag the SR/SS events by some period of time, it may be possible to experience chordal hops for some time after the actual SR/SS grayline period.
Grayline path propagation can result in strange beam headings that are sometimes referred to as a "crooked path". For example, the activation of the Andaman Islands (VU4) in December, 2004, was of great interest to me as it was the last of the DXCC entities that I needed at the time. The intrepid group that went to this very rare entity, last activated in 1987, had numerous obstacles to overcome before obtaining permission to operate on the Indian possession off the coast of Thailand (indeed, they subsequently found themselves in the midst of the terrible tsunami that caused wide-spread destruction and deaths in Indonesia, Thailand, India, Sri Lanka, and East Africa at the end of December '04). The combination of the need to use very modest equipment and antennas for their operation and the fact that it took place near the bottom of sunspot cycle 23 meant that many of us in North America had a very difficult time in hearing them. I was fortunate to work Ms. Bharathi, VU4RBI, one evening on SSB via short-path (335 deg), so early the next morning I began to tune for the CW operator. Years ago I learned that if you cannot hear a station along either the short- or long-paths, you should not hesitate to try different beam headings. Long before I knew anything about chordal hop propagation, I had discovered the "crooked path" route that allowed me, using only 100w and a modest tri-bander at 40ft, to work many Indian, Sri Lankan, and Thailand stations just after my dawn and into my early morning hours. At that time of year, the orientation of my morning grayline runs approximately NNE-SSW, so I turned my quad to the SSW as the grayline came over my location. Sure enough, I began to hear VU4NRO on 20m CW and within a short time, the signals rose nicely to 559 allowing me to make a contact! Afterwards, I swung the quad over 360 degrees and back again to 225 degrees, and I found that the grayline path provided the only copy.
Summary
Here is a summary of the most important aspects of HF propagation discussed above:
• Daily Solar Flux Index (SFI) is an estimate of Solar flux levels over the last 24-hours - higher values may suggest openings on higher bands; however, it is generally a poor indicator of daily conditions.
• Daily A-index is a summary of the geomagnetic conditions over the past 24-hrs - higher values indicate prior-day geomagnetic disturbances; values of A>27 mean geomagnetic storm conditions were present.
• K-index (reported 8 times per day) is a 3-hour summary of the current geomagnetic conditions - rising values signal the potential for geomagnetic storms; K > 4 means storm conditions exist and HF-propagation may be poor.
• Remember: the A is what already happened; the K is what's happening now; if A is high and K is rising it may mean increasingly poor HF - but possibly good VHF - propagation.
• Estimated future smoothed sunspot number or smoothed-SFI (available online) are the propagation predictors of choice.
• Propagation prediction software uses monthly-averaged smoothed SN or SFI, therefore communications path predictions are median values over a month.
• Learn to use propagation prediction software for optimal signal paths, but remember that this is an odds-based estimation of median values.
• Always be aware of the time vs. location of the Sun with respect to the Day/Night regions, so you will pay attention to the terminus (grayline) and the worldwide sunrise/sunset locations.
• Local sunrise may bring enhanced signals from the Dark (West) region; Eastern region sunrise may provide enhanced signal reception in the Dark (West) region.
• Always check both SP & LP routes as possibilities, especially during the hours after local sunrise and sunset.