Content uploaded by Paul W.J. Glover
Author content
All content in this area was uploaded by Paul W.J. Glover on Dec 04, 2014
Content may be subject to copyright.
1
How to make artificial earthquakes
Guillaume Cyr and Paul Glover
Deep within the mountainous regions of
KyrgyzstanÎ ground-breaking experiments
are being carried out. The Russians are making
earthquakes. In a research project more akin to
that visited by Tintin in Syldavia than the real
world, scientists are using large pulsed MHD
generators to inject thousands of amperes of
current into the ground. The current causes
earthquakes to occur up to 150 km away. No
one knows the real mechanism that causes the
artificial earthquakes, but they are linked
inextricably to the injection of the electrical
current. Paul Glover, the professor of
petrophysics at Université Laval (Québec), has
suggested a mechanism, and it is the job of his
student Guillaume Cyr to model the
mechanism in order to see if it is capable of
producing pore fluid pressures sufficient to
trigger an earthquake. Pulsed magneto-hydrodynamic (MHD)
generators tap into the extremely high
magnetic fields generated by a moving
plasma to produce extremely high electrical
currents. The generators used to trigger
earthquakes can produce 2800 amperes at
1350 volts for up to 12.1 seconds, in other
words energies as high as 23 megajoules.
In the image Í, the three long tubes
generate the plasma and fire it through a
non-conducting cavity that is surrounded
by electrical coils housed in the large
circular enclosures. Effectively, this is
three pulsed MHD generators in parallel.
When the plasma moves at high speed
through the non-conducting cavity it
generates an extremely high magnetic field
perpendicular to the movement. The
magnetic field then generates a large
electrical current in the coils.
The Kyrgyz mountains south of Bishkek
in Kyrgyzstan.
A 1500 MA pulsed MHD generator at the
Kyrgyzstan site.
2
Viktor Novikov and his colleagues of the Institute for High Energy Densities of the Russian Academy of
Sciences have carried out a large number of current injection experiments using an approximately 5 km long
dipole at the Bishkek Research Station of the Russian Academy of Sciences in the Chu valley area of the Kyrgyz
mountains (northern Tien Shan). They found that the number of earthquakes in the region within 150 km of the
injection site increased by over 10 standard deviations (σ) of the background seismicityÐ. To put this in context,
statistics tells us that the number of earthquakes would exceed 3σ only once in every 400 samples (99.75%) if
the earthquakes are not related to the current injections. The probability of the signal being more than 10
standard deviations from the mean by chance is so minuscule, it would be expected to occur only once in every
1015 measurements (i.e., for only a few microseconds since the beginning of the universe, about 10 billion
years!).
The increase in the
earthquakes starts a few days
after each injection and
continues for about 5 more
daysÎ. The earthquakes
generally occur along
previously known fault zones
showing that the current
injection is triggering
slippage where there is
already accumulated strain
energy. Where earthquakes
are not distributed along
known fault zones, new fault
zones may be mapped. The
artificial earthquakes had
magnitudes up to mb=5
(Gutenberg-Richter).
What are the applications of artificial earthquakes? Each time an earthquake occurs some of the accumulated
strain energy is released. Hence if small controlled earthquakes can be generated in an area the strain
accumulation may be allayed, making the occurrence of a large destructive earthquake less likely. This is
analogous to an inoculation : the artificial generation of an attenuated earthquake or earthquakes protects against
a large quake just as the presence of a weak form of a disease allows the body to develop antibodies to fight the
attack of a dangerous disease. The difference is that here we are inoculating the earth! However, the logarithmic
nature of the scales for earthquake measurement implies that it would need over 172,000 artificial earthquakes of
mb=5 to protect against one destructive earthquake (mb=7), and that is clearly impractical. Maybe then the
technique could be used to trigger a large earthquake that is overdue, giving at least the advantage of knowing
when the earthquake will occur. Unfortunately this application is probably politically impossible. The real
advantage of the technique may be found at smaller scale, in the mapping of fault zones, and the triggering of
rock and mud slides that would otherwise remain a danger, and of course, in the understanding of the earthquake
process in general.
The smoking gun! Earthquakes occurring 2 to 7 days
after current injection.
3
At least, it would give us more of an insight into the earthquake process if we knew the mechanism that caused
the artificial earthquakes.
Professor Paul Glover of Université Laval in
Québec has suggested that an electro-kinetic
mechanism may be the missing causal link. In his
theory the injected current creates a three-
dimensional electric field in the subsurface. The
electro-kinetic mechanism uses the electric field
to move the pore fluid at depth. If the pore fluid
flows into a fault zone it may accumulate and
raise the pore fluid pressure within the fault zone.
It is known that increases of pore fluid pressure
within fault zones more than a critical pressure of
0.05 MPa are sufficient to trigger an earthquake if
the fault has sufficient accumulated strain. In the
graph of pore fluid pressure against time Í the
pore fluid pressure increases very quickly at a rate
τ+ because it is being driven by the electro-
kinetic drive and then decreases more slowly at a
rate τ- because the pressure dissipates passively.
Earthquakes are possible while the pore fluid
pressure is over a certain critical level.
While the electro-kinetic drive has
been confirmed in the Petrophysics
Laboratory of Université Laval and a
few other laboratories around the
world, it is uncertain if the mechanism
can provide fluid pressures sufficient
to trigger earthquakes up to 150 km
from the injection point. Guillaume
Cyr, a student at Université Laval has
been modelling the process
numerically. His two dimensional
models of the subsurface are created in
a software package Comsol
MultiphysicsÎ.
The theoretical variation of pore fluid pressure
caused by current injection
Model showing pore fluid pressures over 2 MPa (red) are generated to
the right of the first layer (500 m thick). Areas in white have pressures
greater than 2 MPa and surround each electrode of the injection dipole.
4
After defining the structure of the model he is able to solve the differential equations that describe fluid flow,
electrical flow, electro-kinetic coupling, mass balance and thermal effects in parallel on a finite element grid.
The modelling is still a work in progress. However early steady state solutions in a horizontally layered earth
indicate that pore fluid pressures over 3 MPa can be achieved easily with a current injection of 1500 amperes,
and that the pressure remains higher than the critical value of 0.05 MPa up to 150 km from the injection point.
We now know that sufficient fluid pressures can be generated using a steady state differential equation solution.
So far the modelling does not contain any information about how the fluid pressures vary with time. This is
actually extremely important because it may be that activating the pulsed MHD generator for only 10 seconds is
insufficient to obtain the steady state values. Solving these complex differential equations as a function of space
and time is an extremely complex task and represents the next step for M. Cyr. It is hoped that we will have
some initial solutions by December 2009.
Until then, we can confirm that plugging your electricity supply into your garden is not only dangerous, it will
not cause an earthquake – please do not try it at home.