 |
The physics of crop formations By John A. Burke
Posted by Darshana Sanrakshak Shambhala on January 5, 2011 at 10:58pm in Crop
Formations
Editor’s Note: Last month Nancy Talbott discussed some of the specific
analyses of physical trace cases conducted by the BLT research team of John
Burke, W. C. Levengood, and Talbott, as well as possible links between crop
circles, UFOs, and animal mutilation cases. This month Burke discusses the force
he feels may be involved in creating the crop circles.
In 1989 the
appearance of two books on crop circles, combined with some media coverage,
alerted most Americans for the first time to the appearance of a previously
unrecognized phenomena which had no precedent: large, geometric-shaped areas of
crop which had been flattened overnight.
Writing a letter to author Pat
Delgado to ask for details on the biological studies, career biophysicist W. C.
Levengood was shocked to find that no biological studies were being conducted.
He asked for and received plant samples taken in line with his instructions in
what was to become a steady stream of plants across the Atlantic. As formations
were reported in the U.S. and other countries, samples were obtained from them
as well—always with control samples taken from unaffected parts of the same
field for comparison. Today, after meticulously analyzing tissue samples from
five countries and more than 300 formations involving many types of crops,
some clear patterns have emerged.
Whatever the force which makes crop
formations, it physically alters the tissue of the flattened plants in a number
of ways. Over time an hypothesis has emerged suggesting plasma as the active
force. None of the following effects has occurred when formations have been made
(by us and others) using all the techniques claimed by those who have
“confessed” to hoaxing the crop formations:
1. Stalks which are very
often bent up to ninety degrees without being broken, particularly at the nodes,
which are like the joints of wheat stems. Something softened the plant tissue at
the moment of flattening. This is particularly dramatic in canola (rapeseed),
which otherwise is as stiff as celery at this stage of development.
2.
Stalks which are usually enlarged, stretched from the inside out by something
which seems to heat the nodes from the inside. Sometimes this effect is so
powerful, the node literally explodes from the inside out, blowing holes in the
node walls and spewing sap outside the stalk. This has been measured in
thousands of samples to a degree of 95% to 99% probability (“significant” to
“highly significant,” in the language of science). Levengood has duplicated
this effect using microwaves.
3. Stalks which are left with surface
electric charge. We have measured this in two formations which were only a few
hours old. The degree to which the stalks were bent over was proportional to the
degree of electric charge on the stalk, strongly suggesting the force which
pushed it over was electrical.
4. The thin bract tissue surrounding wheat
seed which has had its electrical conductivity increased, consistent with
exposure to an electrical charge.
Koch fractal; Silbury Hill, Wiltshire, England; 23/7/97. C. 1997. Steve
Alexander
This formation was discovered near Silbury Hill,
Wiltshire, England, in a wheat field in July of 1997. It features a border of
126 small circles and a width of 350 feet. (Photo by Steve Alexander)
Natural causes?
As scientists we had to next ask if there is
anything in nature which shares these characteristics. The answer is yes—plasma.
Plasma here is simply electrified air. It carries electric charge, and when it
travels through a magnetic field (like the geomagnetic field which exists
everywhere on the planet) it does two things:
1.) It moves in a spiral,
the most common pattern in which crop is flattened.
2.) When it spirals
thus it emits microwaves. This is the same principle used in your microwave
oven, where electrons are spun around a magnet in the roof and emit the
microwaves which penetrate the tissue and heat from the inside by interacting
with the water in the food. The nodes, the most affected part of crop formation
samples, are the site with most of the plant’s water.
Plasma was first
hypothesized as the cause of crop formations by English meteorologist Terence
Meaden. He suggested the plasma was in the form of a vortex produced
meteorologically. Unfortunately crop formations did not seem overly dependent on
any set of weather conditions, and the model did not explain non-circular
formations.
We asked ourselves, was there any other possible source for
plasma? Lightning is an example of a very powerful, very high energy plasma. It
is caused by plasma (electrically charged air) far above ground in thunderheads
up to eight miles high being attracted to opposite charges in the ground. But
lightning is a much higher energy plasma than that which makes crop circles
(where no charring
occurs).
Low energy plasma
The ionosphere, on the other hand, is a region of low
energy plasma 40-80 miles up in our atmosphere, where most of the air is
electrified by solar wind and cosmic rays. The only time that some of this
plasma gets energetic enough to glow is when we see the Northern Lights. It was
long believed that the ionosphere and the earth’s surface were completely
separate, and that never the twain would meet. In recent years, decades of
airline pilot sightings were confirmed with scientists’ photos of electrical
flashes in the air between thunderheads (8 miles high) and the ionosphere (up to
100 miles high). The several types of these have been dubbed “sprites.” These
are apparently very common events. So there are frequent exchanges of electric
cargo between the ionosphere and a storm 90 percent of the way to the earth’s
surface.
We believe that sometimes the exchange may cover the other 10
percent of the distance as well and actually reach the ground. Something similar
is known to happen every night everywhere when plasma penetrates part way down
(causing perturbations in the geomagnetic field). Normally these attempted
penetrations are bounced back the way they came by the reflective layers of the
ionosphere. These are the same reflective layers which AM radio waves bounce off
to communicate over the horizon. At night these layers weaken and rise (which is
why you can get AM radio reception much further away late at night).
They
are weakest in the predawn hours, when most crop formations occur. The ability
of plasma to penetrate these reflective layers is directly proportional to its
“vorticity”; i.e. the tighter and faster spinning the plasma cloud, the further
it can penetrate toward the ground. The “magnetic pinch” effect insures that as
such a plasmoid descends toward the surface, it shrinks in size and spins faster
(much like spinning figure skaters accelerate by pulling in their
arms).
An increase in
“ammunition”
The amount
of “ammunition” in the ionosphere, in the form of free electrons, increases up
to 100 times between sunspot maximum and sunspot minimum. Crop formation
frequency, at least in England, has roughly paralleled sunspot numbers. The huge
outbreaks of 1988-1989 coincided with the most powerful sunspot maximum in
their 170 years of recorded history, and have declined accordingly since. This
roughly eleven-year cycle should peak again near the
millennium.
The meteoritic connection
The strongest evidence for the ionosphere
as the origin of crop formation plasma comes from microscopic particles of
meteoritic dust found in 2/3 of the 32 formations where we have been able to
obtain soil samples. The heaviest concentration ever was found in 1993 in an
English formation which appeared on the night of the largest meteor shower to
hit Europe in 30 years. This example became the basis of the second paper we
have have managed to publish on crop circles in a peer-reviewed paper (Journal
of Scientific Exploration, Vol. 9, pp. 191-199,
1995).
Sub-millimeter-sized bubbles of pure iron oxide (magnetite) coated
both the ground and the crop in that formation. To summarize a detailed and
technical investigation, the material was identical to the debris which erodes
from meteors as they burn up in the atmosphere, and which takes 7-10 days to
settle to the ground. It can be picked up with a magnet (as could some of the
wheat in which it had become imbedded). It has since appeared in the majority of
formations from 13 states and 5 countries where soil samples were obtained.
Inside formations, it appears in 20 to 100 times the normal concentration for
soil.
As plasma spirals around geomagnetic field lines it creates its own
magnetic field. This would tend to attract and carry along any magnetite dust
particles encountered as it descended from the ionosphere. The ubiquitous
presence of this material has essentially ruled out a low altitude source for
the
plasma.
Established scientific facts
These are extremely well-established and
long-established scientific facts. Nothing said so far is remotely
controversial, except for the idea of plasma reaching the ground from the
ionosphere. Plasma loves to organize itself into spirals. Most aurora are
actually arrays of tight tubes of plasma vortices seen from the side as they
spin around the geomagnetic field lines. One third of all aurora organize
themselves into gross spirals as well. One candidate—the small curl—seems a
likely candidate for crop circle formation. It is often as small as 400 meters
across where it starts in the ionosphere, but shrinks as it
descends.
Plasma might, we reasoned, be reaching the wheat fields of
England from the ionosphere, but why did so many occur in one small area of
England—and how did they form some of those incredible patterns? These are two
very distinct issues. In a search for why any plasma might be particularly
attracted to one tiny area 30 miles or so across, we eventually looked at
hydro-geological maps of England and found something remarkable.
Crop
formations in England overwhelmingly appear over shallowly-buried parts of a
giant chalk aquifer. England has the world’s deepest chalk aquifer. (The white
cliffs of Dover are a view of one side of it.) They also have some of the
world’s greatest seasonal fluctuations of water levels—up to 100 feet. Was there
anything about this which might attract plasma? As it turned out, there was.
Water percolating through porous rock—any kind of porous rock—creates electric
charge. This occurs by a process called “adsorption,” where electrons are
stripped off water droplets as they move through rock pores, leaving a net
negative electric charge behind on the rock and a net positive charge on the
water which drains through.
The "DNA formation", East field, Alton
barnes, Wilts, UK. 17/6/96. C. 1996. Steve Alexander
This "sine/co-sine"
formation was discovered at East field, near Alton Barnes, Wiltshire, England,
in June of 1996. It features 89 circles and a length of 648 feet. (Photo by
Steve Alexander) Chemical reinforcement
With calcium carbonate
(the mineral which makes up chalk) there is a chemical process when the water
dissolves some of the mineral, which further reinforces this same charge
separation. Wherever charge separation Occurs in a body which can conduct
electricity, electric current flows and generates its own magnetic fields. We
measured these ground currents and their changing magnetic fields in 1993 at
Silbury Hill, long the center of the most intense crop formation activity in the
world. Relationship with aquifer
Crop formations in southern
England overwhelmingly occur where this electrically-charged rock is
closest to the surface. The largest formations and most frequent formations
happen late in the summer when the aquifer is most run down, and the most water
has therefore run through the most rock. The beginning of the modem phenomenon
of large, spectacular formations begins in the late seventies and early
eighties, a time when over-pumping for public water supplies began to lower the
water table noticeably. Droughts have coincided with banner years for crop
formations.
In England, our team has measured the kind of magnetic fields
one would expect to accompany such electric ground currents in one field that
has nearly annual formations. Four days later a major formation occurred there.
Follow-up fluxgate magnetometer measurements four days after this sixty-foot
dumbbell formation appeared showed that the magnetic readings and the currents
which produced them had vanished. This is not unlike the discharge with that
more powerful plasma—lightning. In that case ground current attracts the
airborne plasma, and when the plasma (the bolt) hits the surface it neutralizes
the ground current.
Limestone is the chemical twin of chalk. It too is
calcium carbonate, but much less porous than chalk. It too has the ability to
generate ground currents from interaction with water, but not nearly so much as
chalk. Thus it is fascinating to note that limestone aquifers are the major
exception to crop formations occurring over chalk substrata. Formations in
England do happen a minority of the time on the large limestone aquifers
there.
In the U.S. we have no substantial chalk deposits, but huge
stretches of limestone aquifers: in Florida, on the Eastern Coastal Plain,
throughout much of the Midwest, and virtually all of the Great Plains, extending
into Canada. Finally a thin stretch runs down the West Coast. These locations
are where crop formations occur. As in England, the most active sites seem to
frequently be where an edge of the aquifer occurs or where a river valley has
cut through the aquifer to produce an edge. Proximity to water is also typical
(no surprise considering the current generated between water and the rock it ran
through).
Shape most difficult to explain
This leaves
us with the question of shape—the most difficult aspect to explain with a
natural model of crop formations. The most common patterns in the crops are the
most common patterns seen in plasma in the laboratory. It is important to
remember that plasma is scale invariant; anything which happens on a scale of
inches can and will happen on a scale of miles, etc. So it is worth noting that
plasma in the lab most commonly organizes itself into a spiral—the most common
pattern of flattened crop. Next most common in plasma is the swirled disc
surrounded by concentric rings (the “bulls-eye” or “target” pattern). This is
also the next most common in the fields. Furthermore, in both mediums the
concentric rings tend to alternately swirl clockwise and counterclockwise as you
move out from the center (or in from the edge).
Other
patterns
Other patterns seen in plasma and crops include floral
patterns, nested crescents, dumbbells, and others. The hardest to understand
using the plasma model, are straight lines and right angle shapes. It is
counterintuitive to think that air can form such patterns. However, as
electrified air, plasma behaves more like an electromagnetic fluid (and so the
physics of plasma motion is “magnetohydrodynamics”) While it is also contrary to
common sense that liquids form such shapes, in fact they do—when excited.
American physicists exciting liquids with sound waves have produced surface
ripple patterns that include squares, triangles, hexagons, and others. We must
remember that a crop formation is the two-dimensional record of the passing of a
likely three dimensional shape. The ground (2D) is likely to record only a 2D
slice of a 3D plasmoid. So even 2D patterns in the plasma could got recorded on
the ground. Deterministic chaos
Deterministic chaos is a new branch of
science which has repeatedly shown that systems which are excited or turbulent
can assume surprisingly geometric patterns. Ilya Prigogine received the
Nobel Prize for showing that 2D geometric patterns often form of their own
accord in 3D pools of liquid chemical reagents.
A ball of plasma being
drawn ground-ward by an electromagnetic hot spot is likely a turbulent system.
As such we can expect that patterns will spontaneously arise, however briefly.
If that is the moment at which the plasma impacts the ground. that is the
pattern we can expect to appear in the crop. However, with plasma there is a
positive feedback loop which might tend to refine certain patterns until they
are of the picture-perfect sort we so often encounter.
Certain shapes
called waveguides will attract plasma likes bees to honey. A rectangle is one
such shape and is a primary reason why ball lightning (a high energy plasma)
loves to enter houses through the chimney. Chimneys are rectangular tunnels.
Another commonly-used waveguide in industry is the dumbbell shape (which
happens frequently in the fields.) Still another is the “key” or “F’ shape we
often see attached to circles (called the Millman Waveguide).
Plasmas
create their own magnetic field lines. If, by random chance, the magnetic field
in a turbulent plasma takes on a waveguide shape, it could create a positive
feedback loop. More plasma will be attracted to that part of the plasma ball,
vortex, or cloud which has assumed that shape. The plasma will spiral along
those magnetic field lines as it moves. When plasma spirals around a magnetic
field line it strengthens that field line, which can now in turn attract more
plasma, etc. Thus this shape might tend to get “locked in” and even refined
until close to its ideal shape. At the moment this is a highly speculative but
stimulating hypothesis. It still strains the imagination to think how some of
the more elaborate patterns might arise from sheer plasma physics. Not
always perfect
One aspect of all this that has long bothered us was
that if this is a natural phenomenon, then it should frequently not come out
geometric at all. Nature does not always get everything perfect. As it turns
out, we now believe that most plasma impacts result in non-geometric flattening
of the crop. Of course, crops around the world are constantly being flattened by
non-plasma events like wind. However, sometimes close inspection of such ragged
downed areas reveals the same bent nodes as in formations. Sometimes a large
field which gets flattened in a non-descript pattern will have within it
small areas of spiraled lay and other lay patterns (with 180 degree opposition)
impossible if wind was the cause.
Same tissue
changes
Sampling and lab analysis of many such sites has shown a great
number to have the same tissue changes as in formations. In fact the most
dramatic node changes ever recorded have been in such ragged downed
areas—including nodes which were blown apart from the internal pressure. This is
in keeping with plasma physics. Plasma will spontaneously organize itself into a
vortex shape—if the energy level of the plasma does not get too high. When the
energy level exceeds a certain threshold, the plasma’s ability to maintain the
vortex pattern breaks down. Examination of photos of crop formations very often
shows such ragged areas of downed crop all around the formation. Our
pattern-seeking minds, however, ignore this, and we go straight to the geometric
formation, considering this to be the only “genuine” event in the
field.
We believe that plasma, in whatever shape, is probably impacting
the ground far more often than we realize. We have analyzed rings in grass which
have undergone physical changes consistent with plasma contact. If plasma were
to hit streets or buildings it would leave no visible record. A series of
concentric rings found in sand on a beach showed very high levels of ionization.
A circle in dirt in a Colorado field had some of the highest concentrations of
meteoritic dust we have ever seen—only in the top three inches of ground and
only inside the circle.
Like sprites?
We believe that the
plasma we are studying may turn out to be like sprites. Their existence,
reported for decades by airline pilots, was ignored by science until a
professional scientist took photographs of them. Now that scientists are
looking, they are discovering sprites to be incredibly common wherever there are
thunderstorms. We have one daytime photo which looks like a small plasma vortex,
and the rare eyewitness accounts of circle formation are consistent with our
model. Likely one day everyone will know of such events. In the meantime we have
those amazing patterns to admire and puzzle over.
|