Part I of this series dealt with the various means of ground-based
radio probing used to measure the condition of the ionosphere, in par-
Rocket shots into Es regions were first made in the late 1940's. What
they found was not too unexpected, i.e., a sharp rise in the electron
density at the height of the Es patch as shown on the ionogram. Figure 1
is a typical example, though the shape of the peak varies in case to case
and often exhibits much 'fine-structure'.
Figure 1 104 -
Rocket 102 -
probe of |
Es 100 -
Ne 1 x 104 2 5 1 x 105 2 per cc
Up to now I've avoided the mention of electron densities. The reason is
that there is some disagreement in what rockets find and ionograms imply.
Classical electro-dynamics shows that an electromagnetic wave at a
vertical incidence to an ionized region will undergo full reflection when
the electron density per cubic centimeter (cc) is: *
Ne = 1.24 × 104 · f2
with f (frequency) in MHz. For the more inquiring, this is when the re-
fractive index reaches zero. From Snell's Law in optics the solution for
oblique incidence (the Secant Law) is found. (See July 1972 VUD VHF Radio,
Simple calculations using the table there enable one to see that for a
Ch 2 MUF we need Ne's on the order of 106 (a million free electrons/cc).
The big stumble occurs in that the rocket probes are always finding peaks
in the range 105, a factor of ten too low.
In 1968 it was determined what these rockets were finding agreed very
well with what had been labelled as fbEs (the blanketing frequency) on the
ionograms, not foEs etc. Thus the preference of calling the highest Es
reflection from the ionosonde fEs instead of foEs as it appeared not to
behave like a critical frequency should. Part III on theory will get into
this dispute more.
Rocket probes can use two main methods to determine the ambient electron
density in flight. One involves what is known as a Langmuir probe where a
fixed-voltage is maintained across two points (one often being the rocket
body itself) and the current read. The other method requires the use of two
harmonically related signals transmitted from the rocket with the accurate
recording of the phase differences as received on the ground. The latter
system has the difficulty of Es often soaking up (or shielding) the lower
frequency signal from the rocket after it is in (or above) the Es region.
* for those not familiar with scientific notation
102 = 100 103 = 1000 104 = 10,000 etc.
Temp, °K per cubic cm
sea level 288 2.6 x 1019
(Es height) 200 7.8 x 1012
(F2 region) 1480 2.8 x 108
As can be seen in the above table, the particle number density at
Es heights is about one ten-millionth of what it is at sea level. The
temperature is down (ice melts at 273 Kelvin), though not much.
So even with a million free electrons per cc, that is but only one
out of every million particles at Es levels. Compared to F2 densities,
that leaves a lot of neutral particles for the electrons to collide with
and thus cause signal loss.
The main molecular constituents at the E-layer heights are NO
(nitric oxide) and O2 (molecular oxygen). Thru the ultraviolet (mostly
the Lyman wavelengths of hydrogen) and x-ray radiation from the sun these
are changed as:
NO + hn ---> NO+ + e-
O2 + hn'---> O2+ + e-
(h, Planck's constant; n, the frequency of the ionizing 'light'
the product is the energy required to free the electron )
So, now you see how we get our free electrons. Though radio signals
can affect the positive ions, the relatively high masses (and thus high
inertia; NO+ is some 60,000 times as 'heavy' as a free electron) limit
the results. The electric field of the signal causes the charged particle
to undergo a force, F = qE (q, charge; E, electric field intensity). Now,
from Newton we have F = ma (m, mass; a, acceleration). Invoking a little
algebra, the answer is:
a = qE
In January 1966 a rocket over New Mexico carrying a mass spectrometer
(a device using electric and magnetic fields to sort out the positive ions
by their masses) found what so far has been one of the most important dis-
coveries in Es research. In a strong Es layer high levels of metallic ions
were located, such as sodium, magnesium, iron, etc. The normal NO+ density
was curiously found to be depressed within the layer as well. The free
electron count was the 'usual' for Es. Similar rocket flights into Es since
have upheld this general finding.
The positive metallic ion species (varieties) are long-lived (i.e.,
they have a very low coefficient of recombination) with lives on the order
of days compared to magnitudes shorter for NO+. So in effect, the regions
of positive metallic ions (being so 'reluctant' to recombine) were allowing
a high concentration of free electrons to build up, which in turn was
actually decreasing the normal (NO+ ) ion species by enhanced recombination.
On some flights rockets are able to make Es measurements on the downleg
of the flight. Results there can reveal any layer tilts. Those cases show
the majority of layers have tilts well under a degree (see Smith and
Mechtly, Rocket Observations of Sporadic-E Layers, March 1972 Radio Sci.)
Often on rocket flights there is a problem of more than one peak in
the ionization density. In effect, that is multiple layers. As seen in
Figure 2 below this can cause very large errors in tilt calculations
if layer identification is not carefully maintained.
For reasons more evident in Part III, there is a great need to be able
to sample the Es structure in a horizontal direction within the layer. As
Figure 1 showed, the region is extremely thin as ionospheric layers go,
usually in the range of a kilometer (0.621 mile).
_____________ _ _______________
There have been some satellite observations of Es, using mostly the
topside ionospheric sounder (a satellite-borne ionosonde aimed down).
However, as the F region has to be penetrated first in these cases, there
are complications and limitations in interpreting the ionograms (e.g., Es
below foF2 cannot be detected). (see Cathey, Some Midlatitude Sporadic-E
Results from the Explorer 20 Satellite, Journal of Geophysical Research,
74 (1969), pp. 2240-2247)
Other satellite measurement methods are possible, such as measuring
the deviation angle of arrival of a signal from the line-of-sight path.
Faraday rotation (shifting of the signal polarization) is also useful in
indicating electron densities. However, none of these has been particularly
useful for Es investigations.
There are even more exotic modes that could be used for Es probing.
Faraday rotation from a moon-reflected radar signal is one. Also, the
observation of relatively stable extraterrestrial signal sources (i.e.,
various cosmic-noise generators) is very useful in noting the passage of
ionospheric irregularities. (See NBS Monograph 80,pp. 248-253).
Next time, Part III with notes on theory mixed with personal opinions.
The March 1972 issue of Radio Science is devoted entirely to some of
the papers presented at the Third Seminar on the Cause and Structure of
Temperate Latitude Sporadic E, held at Utah State University in September
1971. Though most of the papers are highly mathematical in nature, some
make for much easier reading. Issue copies are available for $ 5 from:
American Geophysical Union
Sixth Floor, 1707 L Street, N.W.
Washington, D.C. 20036
The February 1966 issue covering the results of the first conference
(June 1965 at Estes Park, Colorado) might be available also (originally
at $ 1). The second meeting (June 1968 - Vail, Colorado) had its results
printed in a limited number of 'Conference Proceedings', perhaps still
available from the Institute for Telecommunication Sciences, Boulder, CO
80302. A similar publication for the 3rd gathering was made and includes
several papers (one by Mel Wilson) not used in Radio Science. Best chances
for catching these in libraries would be with the science departments of
Page last modified March 5, 1999