March 1940 QST
Table of Contents
Wax nostalgic about and learn from the history of early electronics. See articles
QST, published December 1915 - present (visit ARRL
for info). All copyrights hereby acknowledged.
Reading through this article reminds me of studying for the
amateur radio exams. In fact, the information presented in this
1940 QST piece does not seem to be lacking anything that contemporary
discussions include. My point is that a great amount of knowledge
had already been amassed about earth's upper atmosphere a mere
four decades after the first transatlantic radio communications
were accomplished by
Marconi on December 12, 1901 from Poldhu in Cornwall, England,
to Newfoundland, Canada. Considering that at the time no instrumented
sounding rockets had been launched into the extreme upper layers
(F1 & F2, beginning at around 120
mi | 200 km), a lot had been discerned about characteristics
as they pertain to radio communications. Balloons were used
for direct measurements at lower altitudes, but inferences based
on observed behavior of long distance communications provided
most of the related knowledge.
Normal and Irregular Characteristics Which Affect Wave Propagation
As time goes on, the unremitting efforts of physicists in
probing the upper atmosphere bear the fruit of continual additions
to our knowledge of the things that happen to the signals We
launch so hopefully into space. It's helpful, now and then,
to take stock of that knowledge. The accompanying article, extracted
from the Bureau of Standards' Letter Circular LC-575, is an
up-to-date summary of known ionosphere effects. "Must" reading
for the chap who wants to keep well informed.
Fig. 1 - Simplified scale drawing showing
possible paths of wave travel. (1) A low-frequency-wave reflected
by the E layer; (2) a wave of higher frequency reflected from
the F2 layer; (3) wave which passes completely through the ionosphere
because its frequency is too high to permit reflection.
In the high atmosphere, above about 50 kilometers (30 miles),
the air particles are separated so far that collisions between
them are far less frequent than in the lower atmosphere, and
when an air particle is ionized by ultraviolet radiation from
the sun it remains ionized for a considerable time. Therefore
at any given time a large proportion of the air particles are
in an ionized condition. This does not occur much below about
50 kilometers (30 miles), because the ionizing radiations from
the sun are largely absorbed in the higher regions of the atmosphere.
Likewise there is not very great ionization density above about
400 kilometers (250 miles), because the air is so rare at such
heights that there are not enough atoms to provide for great
ionization density. The region in which the ionization is great,
great enough in fact to affect radio wave transmission, is thus
between 50 and 400 kilometers (30 and 250 miles) above the earth's
surface, and this region is called the "ionosphere".
The ionization in the ionosphere is not uniformly distributed
with altitude but is stratified, and there are certain definite
layers in which the ionization density is such as to reflect
radio waves. These layers do not remain always at the same height
but vary diurnally, seasonally, and otherwise, in both height
and ionization density. There may be a considerable number of
such layers at a given time. There are two principal ones, called
the "E" and "F" layers. The E layer is at a height of 90 to
140 kilometers at different times, usually about no kilometers.
The term F-layer is ordinarily reserved for the other layer
as it exists at night; in the daytime during most of the year
it divides into two layers which are called the "F1" and "F2".
The night F layer is at a height of about 180 to 400 kilometers.
The F1 layer exists in the daytime, at a height of about 140
to 250 kilometers. The F2 layer also exists in the daytime,
at a height of about 250 to 350 or more kilometers in the summer,
and about 150 kilometers in the winter day. (The "virtual" heights,
defined later, are somewhat higher than these values.) The fourth
layer, which is semi-permanent, is the "D" layer; it exists
only in the daytime, and its height is of the order of 50 to
90 kilometers. Little has been done on the determination of
the quantitative characteristics of the D layer, its effects
being largely inferred rather than directly observed. Existing
knowledge covers mainly the E, F, F1, and F2 layers.
The structure of the ionosphere may be visualized in an elementary
way from Fig. 1, which is for a typical summer daytime condition,
the E, F1, and F2 layers all being present. This diagram is
drawn to scale, so the angles of reflection of radio waves from
the layers may be estimated correctly. The three layers are
shown as mere thin lines, for simplicity. The layers have in
fact a certain thickness, and the density of ionization varies
somewhat in this thickness. At the right of the diagram is a
rough illustration of a possible distribution of ionization
density with height.
Dotted lines indicate two of many possible paths of radio
waves from a transmitter to a receiver as transmitted by reflection
from the ionosphere layers. This picture, simple as it is, does
in fact represent the basic mechanism of radio wave transmission
over long distances. When we consider the variations of ionization
and height of the layers with time, and the effects of the ionization
upon the received intensity and the limits of transmissible
frequency at any particular time, the picture loses its simplicity.
However, most of the phenomena of long-distance radio transmission
are completely explainable in terms of the ionosphere.
The principal ionosphere characteristics which control or
determine long-distance radio transmission are the height and
the ionization density of each of the ionosphere layers. Since
each layer has a certain thickness it is necessary to define
the sense in which the term "height" is used. When a ray or
train of waves is reflected by a layer, it is slowed down as
soon as it starts to penetrate into the layer. The process of
reflection thus goes on from the point at which the waves enter
the layer until they have been fully turned down and leave the
layer. This is true whether the waves travel vertically or obliquely
to the ionosphere. It is illustrated for the oblique case in
Fig. 2. The waves follow a curved path in the layer until they
emerge at the same vertical angle at which they entered. The
time of transmission along the actual path BCD in the ionized
layer is the same as would be required for transmission along
the path BED if there were no ionized particles present. The
height h from the ground to E, the intersection of the two projected
straight parts of the path, is called the "virtual height" of
the layer. This is the important quantity in all measurements
Fig. 2 - Showing the relation of virtual
height (h) to the height actually reached by the wave in reflection
from the ionosphere.
The virtual height of a layer is measured by transmitting
a radio signal from A, and receiving at F both the signal transmitted
along the ground and the echo, or signal reflected by the ionosphere,
and measuring the difference in time of arrival of the two.
The signal is a special, very short pulse in order that the
two may be separated in an oscillograph, as the time differences
are mere thousandths of a second. The difference between the
distance (AE + EF) and AF is found by multiplying the measured
time difference by the velocity of light. From this and the
known distance AF, the virtual height h is calculated. It is
usual to make AF zero by transmitting the signal vertically
upward and receiving it at the same place (and it is for this
case that the term "virtual height" rigorously applies). The
virtual height varies somewhat with frequency.
The effectiveness of the ions in reflecting the waves back
to earth depends on the number of ions present in a unit of
volume, i.e., the ionization density The higher the frequency,
the greater is the density of ionization required to reflect
the waves back to earth. It has been shown that, for electron
ionization, the relation (for the ordinary ray, explained below)
N = 0.0124f2
where N is the number of electrons per cubic centimeter and
1 is the highest frequency, in kilocycles per second, at which
waves sent vertically upward are reflected back to earth. Waves
of all frequencies higher than this pass on through the ionized
layer and are not reflected back to earth, while waves of all
lower frequencies are reflected. This frequency is called the
"critical frequency," and measurement of it is, with the equation
just given, a means of measuring the maximum ionization density
in an ionized layer. (Waves of higher frequencies than the critical
are sometimes reflected by another mechanism - see discussion
of "Sporadic E" later.)
Measurements of critical frequency are usually made by means
of vertical or nearly vertical transmission (that is, with the
transmitter and receiver not far apart). The process is to measure
the virtual height, by the method described above, repeating
the determination at successively increasing frequencies until
the waves are no longer received back from the layer. The highest
frequency at which waves sent vertically upward are received
back from the layer is the critical frequency of that layer.
Typical results of such measurements are illustrated in Figs.
3, 4 and 5, for different times of year, day and night. They
show critical frequencies as sharp increases in virtual height.
In Fig 3, starting at a frequency below 2000 kc. the virtual
height is found (in this example) to be about 110 kilometers,
and remains at about this height until about 3300 kc. The critical
frequency of the E layer at the time of this measurement is
thus 3300 kc.; i.e., this is the highest frequency at which
vertically-incident waves are reflected back to earth. All such
waves of higher frequency penetrate through the E layer and
go on up to a higher layer, the F2. The F2 layer has a greater
ionization density and so it reflects back waves of frequency
greater than 3300 kc. It is not until frequencies greater than
11,500 kc. are used that the F2 layer fails to reflect them,
in the case illustrated.
Fig. 3 - Virtual height plotted against frequency
for a typical winter day, with critical frequencies indicated.
Near the critical frequency the waves are excessively retarded
in the ionized layer, which accounts for the rise of the curve
at the critical frequency. At the right of the curve appear
two critical frequencies for the F2 layer. This is an indication
of double refraction of the waves due to the earth's magnetic
field, two components of different polarization being produced.
One is called the "ordinary" wave and the other the "extraordinary"
wave. The symbols o and x, respectively, are used for these
components. The critical frequency of a layer n is represented
by the symbol fn, and to such symbol the o or x is added as
a superscript. Thus the critical frequencies of the F2 layer
for the ordinary and extraordinary waves are indicated by the
respective symbols, SYMBOL HERE and SYMBOL HERE.
In the case of the E layer, the ordinary wave usually predominates
and the extraordinary wave is so weak it does not affect radio
reception. At Washington the critical frequency for the extraordinary
wave is about 750 kc. higher than for the ordinary wave (for
frequencies of 4000 kc. or higher). The difference in frequency
is proportional to the intensity of the earth's magnetic field
at the place of reflection, and is therefore different at different
places on the earth. In reporting results of measurements of
critical frequency it is now customary to give the values for
the ordinary wave; practice varied in the past.
Besides the virtual heights and critical frequencies, the
absorption of the energy of radio waves by the ionosphere is
an important factor in limiting radio transmission. This absorption
exists because the ions set in motion by the radio waves collide
with air molecules and dissipate as heat the energy they have
taken from the radio waves. Consequently the energy thus absorbed
from the radio waves is greater, the greater the distance of
penetration of waves into the ionized layer and the greater
the density of ions and air molecules in the layer, and hence
the greater the number of collisions between ions and air molecules.
Absorption is especially great in the daytime, and it occurs
chiefly in the low ionosphere, in the D or E layers. It also
occurs in the high ionosphere, near critical frequencies. Much
of the low-layer absorption disappears with the decrease of
low-layer ionization at night. Higher frequencies are less affected
by absorption than are lower frequencies, for waves passing
through the same ionized layers.
Fig. 4 - Representative summer day conditions.
The extraordinary components are clearly indicated.
Regular Variations of Ionosphere Characteristics
There are three principal types of variation of critical
frequencies which are fairly regular with time. These are diurnal
variations, seasonal variations, and year-to-year variations
with the sunspot cycle.
The diurnal and seasonal variations of the critical frequencies
of the normal E layer are particularly regular. The critical
frequencies vary with the altitude of the sun, being highest
when the sun is most nearly overhead. Thus the diurnal maximum
of the E critical frequency (fE) is at local noon, and the seasonal
maximum is in mid-summer. At night this layer usually does not
reflect waves of frequencies higher than about 1000 kc.
The diurnal and seasonal variations of the critical frequencies
of the F2 layer at Washington are quite different from those
of the E layer. The winter F2 critical frequencies exceed any
regular critical frequency found during the summer. In the winter
a broad diurnal maximum occurs in the daytime, centered around
1:00 P.M. local time. In the summer a broader diurnal maximum
centers about sunset. During the night the winter critical frequencies
are usually lower than the corresponding summer values. Thus,
the highest F2-layer critical frequencies occur during the winter
day, and the lowest F-layer critical frequencies occur during
the winter night; the summer day and night values are between.
The F2 virtual heights are much lower during a winter day
than during a summer day. The F virtual heights at night are
about the same in winter as in summer.
The seasonal effects in the ionosphere synchronize with the
sun's seasonal position, not lagging a month or two as do the
seasons of weather. Winter conditions in the F2 layer obtain
during a period of several months from about October to March,
and summer conditions for a period of several months from about
April to September. Around the equinoxes, there is a transition
period of a month or two in which the change occurs between
winter and summer conditions; in these transition periods the
ionosphere characteristics (except for the effects of sporadic
E) fluctuate between winter and summer conditions and are thus
more erratic than during the rest of the year.
There are important changes in ionosphere characteristics
in the 11-year sunspot cycle. From the sunspot minimum in 1933
to the sunspot maximum in 1937 the F- and F2-1ayer critical
frequencies doubled (for most hours of the day), and the E-layer
critical frequencies became 1.25 times as great. A consequence
is that the best radio frequencies for long-distance transmission
were approximately twice as great in 1937 as in 1933 (except
for summer daytime, when they were about 1.5 times as great).
In about 1944 they will return to minimum values.
The condition of the ionosphere varies somewhat with latitude.
For all latitudes of continental United States the differences
from the Washington values appear to be negligible, but the
values in Alaska and in the Canal Zone are somewhat different.
Fig. 5 - At night, either summer or winter,
the E layer is relatively weak, while the F1 and F2 layers combine
to form a single layer, the F. Critical frequencies are considerably
lower at night than during the daytime.
Applications to Radio Transmission
From the vertical-incidence critical frequencies and virtual
heights of the ionosphere layers, at any given time, it is possible
to calculate the upper limit of radio frequency that can be
transmitted over any distance. The calculated values of maximum
usable frequency are found to agree with direct observation
of radio transmission over such distances.
When radio waves are transmitted along the earth over any
distance, they strike the ionosphere obliquely (Fig. 1). Such
obliquely incident waves can be reflected back down with lower
ionization densities than can vertically incident waves. It
results that the larger the angle of incidence (angle of wave
path with the vertical), i.e., the greater the transmission
distance, the higher is the upper limit of frequency of waves
that can be reflected from a layer of given ionization density
or critical frequency. This upper limit of frequency, for transmission
via an ionosphere layer, for a particular time and transmission
distance, is called the "maximum usable frequency." It may be
calculated roughly, to a first approximation, by multiplying
the critical frequency of the layer by the secant of the angle
The accurate calculation of maximum usable frequencies from
vertical-incidence critical frequencies is quite complicated.
For convenience, typical values for the conversion are given
in Table 1.
Table 1 - Typical Average Ratios of Maximum
Usable Frequency to Critical Frequency (For One-Hop Transmission)
* Sporadic E transmission has no critical
frequency. The values given are ratios of maximum usable frequency
to the approximate upper limit of frequency of the stronger
sporadic-E reflections at vertical incidence.
To obtain the maximum usable frequency for transmission over
a given distance by way of a given layer, multiply the critical
frequency by the ratio given in the table. Where blanks appear
in the table, and for distances over 3500 kilometers, the distance
is too great for one-hop transmission, i.e., transmission over
such distances requires multiple reflection from the ionosphere
with intervening reflection from the ground.
The distance at which a given frequency is the maximum usable
frequency is also the minimum distance over which that frequency
is receivable. This minimum distance for any frequency is called
the skip distance; at any less distance it is impossible to
receive on that or higher frequencies, except for sporadic or
The highest maximum usable frequencies (in north temperate
latitudes) occur during the winter day and the lowest during
the winter night. The summer values for both night and day lie
between these two extremes except as modified by sporadic-E
or scattered reflections.
Maximum Usable Frequencies over Long Paths
Since the local time of day, and hence the ionosphere characteristics,
may vary a large amount throughout a long transmission path,
it is necessary to consider what part of the path determines
the conditions of transmission. For single-hop transmission
(transmission by a single reflection from the ionosphere), it
is the region half-way between the transmitter and receiver
whose conditions determine the transmission, because it is
there that the reflection from the ionosphere takes place. In
multi-hop transmission, when the radio waves are reflected from
the ionosphere, then from the ground, then back to the ionosphere,
etc., the determining conditions are in the middle of each hop.
The maximum possible distance of transmission between the
transmitter and receiver whose conditions determine the transmission,
because it is there that the reflection from the ionosphere
takes place. In multi-hop transmission, when the radio waves
are reflected from the ionosphere, then from the ground, then
back to the ionosphere, etc., the determining conditions are
in the middle of each hop.
The maximum possible distance of transmission by a single
hop is limited by the geometry of the earth's surface and the
layer, and also by absorption or other limitation at the ground
of those waves which are nearly tangential to the earth's surface.
It is found in practice that the minimum angle with the ground
of the radio waves transmitted or received (over land) averages
about 3 1/2 degrees. From these considerations the geometry
indicates that the maximum distance along the earth by a single
hop is ordinarily about 3500 kilometers for the F2 layer, and
about 1700 kilometers for the E layer. Single-hop transmission
may sometimes be possible at greater distances than these while
at the same time multi-hop transmission over the same path may
be more efficient.
Because of the variation of ionosphere characteristics with
longitude, different frequencies may be necessary for transmission
in different directions from a given place. For example, around
sunset in winter lower frequencies are used in transmitting
eastward than in transmitting westward from the same location.
This does not mean, however, that different frequencies would
be necessary or desirable in opposite directions over the same
For very long paths in which widely different longitudes
(i.e., times of day) are involved, it sometimes happens that
the waves travel different parts of the way by different layers.
For such cases, it is necessary to take account of the heights
of the different layers to determine the lengths of the several
Effects of Ionosphere Irregularities
The primary effects of the ionosphere on radio wave propagation
are those already described, which are due to the normal or
regular characteristics of the ionosphere. The modes of variation
of those characteristics have been shown to be of a regular
and fairly predictable nature. There are some other ionosphere
phenomena which are irregular in their times of occurrence,
and make radio phenomena in general much less predictable. Five
types of such phenomena have been identified: sporadic-E-layer
reflections, scattered reflections, sudden ionosphere disturbances,
prolonged periods of low-layer absorption, and ionosphere storms.
While all five are irregular in time, the first two are primarily
due to irregularities in space.
It sometimes happens that waves are reflected by the E layer
on frequencies higher than that at which the E-layer waves normally
disappear and the reflection of waves by higher layers begins.
Thus, in the example shown in Fig. 3 waves may sometimes be
reflected at the E-layer height of 110 kilometers on frequencies
of 4 or 5 or more megacycles. These reflections are due to a
process different from the normal reflection in the ionized
layer; the process is probably one of reflection from a sharp
boundary of stratified ionization. The existence of these "sporadic
E" reflections necessitates a redefinition of the term "critical
frequency," previously defined as the highest frequency at which
waves sent vertically upward are received back from the layer.
When sporadic E reflections occur they may be received simultaneously
with reflections from higher layers; for example, vertical-incidence
reflections might be received at 8 Mc. from both the E and the
F2 layers. The E-layer critical frequency, more precisely defined,
is the value (3300 kc. in the example shown in Fig. 3) at which
the observed virtual height shows a sudden rise to large values
as the frequency is increased. Except for the occasional occurrence
of sporadic E reflections, all waves of higher frequency pass
through the E layer and are not reflected by it.
Sporadic E leads to interesting results in long-distance
radio transmission. It accounts for long-distance transmission
up to higher frequencies than by any other means. Strong vertical-incidence
reflections by sporadic E sometimes occur at frequencies up
to about 12 Mc. By reason of the large angles of incidence possible
with the E layer, this makes occasional long-distance communication
possible on frequencies as high as 60 Mc. Such communication
is generally for only a short time and for restricted localities.
Sporadic E is thus patchy or sporadic in both geographical distribution
Sporadic E occurs most commonly in the summer, May to August,
particularly in the morning and evening, but may occur any time
of day or night. It occurs occasionally at all seasons, particularly
in the evening. It occurs more at high latitudes than in equatorial
An irregular type of reflection from the ionosphere occurs
at all seasons and is prevalent both day and night. These reflections
are most noticeable within the skip zone, or at frequencies
higher than those nominally receivable from the regular layers.
Like sporadic E, they occur at frequencies which may exceed
the F2 critical frequencies, but are unlike sporadic E in that
they are complex and jumpy thus causing signal distortion; they
occur and disappear fitfully, and are almost useless for communication
purposes. Some types are of very weak intensity. The scattered
reflections are characterized by very great virtual heights,
usually somewhere from 400 to 1500 kilometers. Their occurrence
was for a time thought to indicate the existence of another
layer above the F2 layer which might be called the G layer.
It is now, however, thought that they are of several types,
and that some of them are due to complex reflections from small,
ephemeral, scattered patches or "clouds" of ionization in or
between the normal ionosphere layers, and thence to one or more
layers and ground by single or multiple reflection.
Sudden Ionosphere Disturbances
The most startling of all the irregularities of the ionosphere
and of radio wave transmission is the sudden type of disturbance
manifested by a radio fadeout. This phenomenon is the result
of a burst of ionizing radiation from a bright chromospheric
eruption on the sun, causing a sudden abnormal increase in the
ionization in the D layer (below the E layer), frequently with
resultant disturbances in terrestrial magnetism and earth currents
as well as radio transmission. The radio effect is the sudden
cessation of radio transmission on frequencies above about 1500
The drop of the radio signals to zero usually occurs within
a minute. The effects occur simultaneously throughout the hemisphere
illuminated by the sun, and do not occur at night. The effects
last from about ten minutes to an hour or more, the occurrences
of greater intensity in general producing effects of longer
duration. The effects are more intense, and last longer, the
lower the frequency in the high-frequency range (i.e., from
about 1500 kc. up). It is consequently sometimes possible to
continue communication during a radio fadeout by raising the
The radio, magnetic, and other effects are markedly different
from other types of changes in these quantities. The effects
are most intense in that region of the earth where the sun's
radiation is perpendicular, i.e., greater at noon than at other
times of day and greater in equatorial than in higher latitudes.
Taking due account of the variation of the effects with frequency
and distance, varying effects in differing directions can be
explained. Reception in the United States from stations in the
southern hemisphere usually exhibits greater effects than reception
from other directions (because of passing the equatorial regions).
Similarly, when the disturbance occurs at a time when it is
morning at the receiving point the effects are usually greater
in reception from the east than from the west, and vice versa
for the afternoon (because of passing the region where it is
noon). A radio fade-out sometimes occurs when it is night at
the receiving point, but only when the path of the waves is
somewhere in daylight.
Prolonged Periods of Low-layer Absorption
This phenomenon is similar to the sudden ionosphere disturbance
in its effects and characteristics except that its beginning
as well as recovery is gradual and it has a longer time duration,
commonly several hours. The intensity diminution is in general
not as severe as in the more intense fadeouts, but sometimes
the intensities fall to zero.
The low-layer absorption effect appears to be due to increased
ionization in the D layer (below the E layer), exactly as for
the sudden ionosphere disturbances. The increased ionization
is caused by an abnormally great outpouring of ultraviolet light
from the sun, but in this case it is not so sudden as in the
eruptions which cause the sudden ionosphere disturbances. The
variation of the efects with frequency, and other characteristics,
are .the same as for the sudden ionosphere disturbances.
Both phenomena occur at all seasons, but the prolonged periods
of low-layer absorption have been found to occur in a group
of several weeks' duration at periods of high sunspot activity,
the groups being separated by more or less quiet periods of
several months. They frequently but not always occur during
periods when sudden ionosphere disturbances are numerous.
An ionosphere storm is a period of disturbance in the ionosphere
in which there are great anomalies of critical frequencies,
virtual heights, and absorption. Radio transmission is poor
(except for the low frequencies, below 500 kc., which are sometimes
improved). An ionosphere storm usually lasts one or two days,
and occurs both day and night. It is usually accompanied by
a magnetic storm (i.e., a period of unusual fluctuation of terrestrial
magnetic intensity). During the first few hours of very severe
ionosphere storms the ionosphere is turbulent, stratification
is destroyed, and radio wave propagation erratic. During the
later stages of very severe storms and during the whole of more
moderate storms, the upper part of the ionosphere, principally
the F2 layer, is expanded and diffused. The critical frequencies
are much lower than normal and the virtual heights much greater,
and therefore the maximum usable frequencies are much lower
than normal. It is often necessary to lower the working frequency
in order to maintain communication during one of these storms.
There is also increased absorption of radio waves during an
ionosphere storm. Ionosphere storms are most severe in auroral
latitudes .and decrease in intensity as the equator is approached.
Ionosphere storms occur approximately simultaneously over wide
geographical areas. The condition of the ionosphere is much
less uniform from point to point than on undisturbed days.
Posted December 9, 2015