Study: Earthquakes on one side of Earth can trigger new ones on opposite side

Research finds quakes can systematically trigger other ones on opposite side of Earth

CORVALLIS, Ore. – New research shows that a big earthquake can not only cause other quakes, but large ones, and on the opposite side of the Earth.

The findings, published Aug. 2 in Nature Scientific Reports, are an important step toward improved short-term earthquake forecasting and risk assessment.

Scientists at Oregon State University looked at 44 years of seismic data and found clear evidence that temblors of magnitude 6.5 or larger trigger other quakes of magnitude 5.0 or larger.

It had been thought that aftershocks – smaller magnitude quakes that occur in the same region as the initial quake as the surrounding crust adjusts after the fault perturbation – and smaller earthquakes at great distances – were the main global effects of very large earthquakes.

But the OSU analysis of seismic data from 1973 through 2016 – an analysis that excluded data from aftershock zones – using larger time windows than in previous studies, provided discernible evidence that in the three days following one large quake, other earthquakes were more likely to occur.

Each test case in the study represented a single three-day window “injected” with a large-magnitude (6.5 or greater) earthquake suspected of inducing other quakes, and accompanying each case was a control group of 5,355 three-day periods that didn’t have the quake injection.

“The test cases showed a clearly detectable increase over background rates,” said the study’s corresponding author, Robert O’Malley, a researcher in the OSU College of Agricultural Sciences. “Earthquakes are part of a cycle of tectonic stress buildup and release. As fault zones near the end of this seismic cycle, tipping points may be reached and triggering can occur.”

The higher the magnitude, the more likely a quake is to trigger another quake. Higher-magnitude quakes, which have been happening with more frequency in recent years, also seem to be triggered more often than lower-magnitude ones.

A tremblor is most likely to induce another quake within 30 degrees of the original quake’s antipode – the point directly opposite it on the other side of the globe.

“The understanding of the mechanics of how one earthquake could initiate another while being widely separated in distance and time is still largely speculative,” O’Malley said. “But irrespective of the specific mechanics involved, evidence shows that triggering does take place, followed by a period of quiescence and recharge.”

Earthquake magnitude is measured on a logarithmic 1-10 scale – each whole number represents a 10-fold increase in measured amplitude, and a 31-fold increase in released energy.

The largest recorded earthquake was a 1960 temblor in Chile that measured 9.5. The 2011 quake that ravaged the Fukushima nuclear power plant in Japan measured 9.0.

In 1700, an approximate magnitude 9.0 earthquake hit the Cascadia Subduction Zone – a fault that stretches along the West Coast of North American from British Columbia to California.

Collaborating with O’Malley were Michael Behrenfeld of the College of Agricultural Sciences, Debashis Mondal of the College of Science and Chris Goldfinger of the College of Earth, Ocean and Atmospheric Sciences.

The paper:

Earthquakes are part of a cycle of tectonic stress buildup and release. As fault zones near the end of
this seismic cycle, tipping points may be reached whereby triggering occurs and small forces result in
cascading failures. The extent of this effect on global seismicity is currently unknown. Here we present
evidence of ongoing triggering of earthquakes at remote distances following large source events. The
earthquakes used in this study had magnitudes ≥ M5.0 and the time period analyzed following large
events spans three days. Earthquake occurrences display increases over baseline rates as a function of
arc distance away from the epicenters. The p-values deviate from a uniform distribution, with values for
collective features commonly below 0.01. An average global forcing function of increased short term
seismic risk is obtained along with an upper bound response. The highest magnitude source events
trigger more events, and the average global response indicates initial increased earthquake counts
followed by quiescence and recovery. Higher magnitude earthquakes also appear to be triggered
more often than lower magnitude events. The region with the greatest chance of induced earthquakes
following all source events is on the opposite side of the earth, within 30 degrees of the antipode.
0 0 vote
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
August 13, 2018 9:33 am

I’ll need an office globe to check that out. How wide is 30 degrees?

Reply to  ResourceGuy
August 13, 2018 9:36 am

Just use 60 north or south latitude as a reference.

Reply to  ResourceGuy
August 13, 2018 11:18 am

The circumference of Earth is about 25,000 mi. (24,901 miles) a +/- 30 degree sector (60 degrees) is one sixth of that or 4150 miles. This yields a radius of over 2000 miles from the center of the antipode. A rather big bullseye. A 30 degree sector is half that, or 1000 mile radius.
(if you really want to crunch the geometry:

Robert W. Turner
Reply to  ResourceGuy
August 13, 2018 5:52 pm

Check out the swarm of earthquakes on the North Slope AK right now, I don’t remember seeing that on the north side of the state in the last 10 years of watching daily EQs.

The antipode is East Antarctica but West Antarctica is within the 30 degrees.

Reply to  Robert W. Turner
August 14, 2018 3:29 am

been 3 middlin large 5ish ones down sth last few weeks
i note quite often after those we get one in aus
we’ve had one in sth aus and one in west aus last 7 days roughly
but quakes are rare for us

Tom Halla
August 13, 2018 9:35 am

Interesting study. 30 degrees of latitude or longitude does seem a bit large to use for warnings, though.

August 13, 2018 9:36 am

There were a number of strong earthquakes in northern Alaska this past weekend, including a M 6.4. Might this be an opportunity to test the results of this study?

Reply to  PaulH
August 13, 2018 11:26 am

The Alaskan quake up on the North Slope has triggered an amazing amount of aftershocks, over 200+ just in the last 24 hours.

Reply to  goldminor
August 14, 2018 3:30 am

yeah but odder was the cessation of at least one day of all quakes in Hawaii, when there were 100+most days prior for near 2 mths straight.

Reply to  PaulH
August 13, 2018 12:38 pm

This is an unusual area for strong EQ (SW of Kaktovik). I didn’t look at the M6.4 but there was a M6.0 reported today at only 12km. Most of the smaller shocks are less than 10km and even less than 1km. Odd and interesting.

Reply to  PaulH
August 13, 2018 2:05 pm

This is another mere speculation that can never be proved or disproved. Yawn.

Ian Magness
August 13, 2018 9:40 am

Scary stuff!
The same “literally” applies to AGW tales – start a BS panic story about, say, coral reefs dying and sea levels rising in the South Pacific, or Antarctic ice melting and penguin populations crashing, and watch the green brigade and and MSM hype the story in Northern Europe, except this time it’s polar bears, sandeels and puffins.
Happens frequently.

Reply to  Ian Magness
August 13, 2018 1:00 pm


As you know, I’m the most committed of climate change sceptics, but I think this is an obvious effect of an earthquake. So obvious it barely needs scientific study.

Grab the edge of your duvet and flick it as though making the bed. Presto, the other end of the duvet moves. You never get that perfectly made bed syndrome without a few flicks because you can’t anticipate every factor involved.

In the same way, flick the end of a continental mass and, somewhere, somehow, a ripple occurs and someone, somewhere feels the effects.

I don’t know much about science but I’m damn sure a lever is probably fundamental to most of it, and if you lever something at one end, the other end feels the effect. The real question to me is, where is the fulcrum?

Reply to  HotScot
August 13, 2018 1:12 pm


the 30 degree bit, meh.

Reply to  HotScot
August 13, 2018 1:30 pm

It’s not a lever.
They are energy waves, both compression and transverse. Some refer to transverse waves as gravity waves, as here on earth they are usually driven by gravitational forces attempting to stabilize a disturbance parallel to their vector (a vertical displacement).
Compression waves travel exactly the same way as sound waves (sound waves are compression waves). They are a pressure disturbance which causes local volumetric reductions and expansions in whatever working medium is affected. The speed at which a compression wave travels through a substance is called the speed of sound for that substance. Media (substances) that are very dense and relatively incompressible (metals, rock, water) transmit compression waves very fast, and have a higher speed of sound. Compressible substances such as gases (air) both slow and dissipate the compression energy faster due to high compression (hysteresis) losses.

Reply to  rocketscientist
August 13, 2018 2:55 pm


In all sincerity, thank you for that.

But in my simplistic, uneducated way, I perceive a lever as being fundamental to everything. A door, a can opener, an IC engine, a birds beak, a rocket; they all employ leverage of some description.

To me, tectonic plate movement invokes leverage, it may be from compressive force, but that in itself is invoked by leverage of some description. It might be the eruption of a volcano, which, whilst a geological phenomenon, invokes leverage, the energy comes from somewhere to eject volumes of magma and ash and move an entire island a millimetre.

Try to live one minute of your life without leverage, it’s not possible. Your lungs uses leverage, your heart. Try making a cup of tea without leverage. In my opinion, it’s one of the most important principle of natural existence.

Is it even possible that atoms exist without leverage? They couldn’t bounce of one another without it. Surely? Does magnetism utilise leverage? I don’t know, but it doesn’t seem unreasonable that it does. Well, at least to me.

Reply to  HotScot
August 14, 2018 8:02 am

Perhaps we are talking past each other as you seem to be using “leverage” as synonym for influence. In Science and engineering it has a specific meaning that stem from the definition of “simple machines”.
All complex machines are created by using many variations and combinations of 4 simple machines:
Lever and fulcrum
Wheel and axle (some argue that this is simply a continuous lever)
Inclined plane
Piston and cylinder

Some also add:
Pulley (wheel and axel), Wedge (inclined plane), and Screw (inclined plane wrapped around an axel)

What exactly are your lungs or heart prying apart? In this instance I would argue that the piston (diaphram or heart muscle) rather than a lever and fulcrum were working.

John Johnston
Reply to  HotScot
August 13, 2018 5:45 pm

C’mon, the more “obvious” a so-called effect is, the greater the need of scientific study. In fact Newton’s laws of motion, relativity and other theories and discoveries could only have been made by people intelligent enough and committed enough to question the “obvious”.

August 13, 2018 9:42 am

Re-check those bristlecone pines. I’m sure the data for this are hidden in there, somewhere.

Reply to  BallBounces
August 13, 2018 9:53 am

Surely you don’t mean plural for this?

Reply to  ResourceGuy
August 13, 2018 1:02 pm


A question that’s always puzzled me. Did Mann rely on a single tree for his hockey stick theory?

Joel O'Bryan
Reply to  HotScot
August 13, 2018 1:43 pm

Please don’t abuse the word “theory”. Mann is to the scientific method what Bernie Madoff was to investment broker ethics.

Reply to  Joel O'Bryan
August 13, 2018 2:57 pm


My sincere apologies. I lose control of my digits on occasions.


August 13, 2018 9:42 am

So if pressure waves are sent out in all directions from the original earthquake, I would think they merge on the other side of the globe, assuming they travel approximately the same speed, and would reinforce each other. This could produce enough energy even with losses to cause another event.

There has to be a fault or built up stresses in the area where these waves hit each other to cause another earthquake which is possibly why 30 degrees makes sense. My theory anyway.

Reply to  rbabcock
August 13, 2018 10:41 am

+/- 30° from the “antipode”.

The theroy suggests that the pressure waves all arrive at the opposite side of the globe simultaneously and then act upon the closest fatcture zone within a 30° cone (or is it +/- 30° which is a 60° cone). The larger area is very big considering it covers 60° of latitude. That would equate in round numbers to about a 2000 mile radius from the antipode.
…yeah , that nails it down to a continental sized area…

Clay Sanborn
Reply to  rbabcock
August 13, 2018 3:40 pm

Constructive interference of waves… Two or more smaller earthquake waves colliding and forming a temporary large wave that subsequently causes the Mother of All Earthquakes; caused of course, by CAGW. /sarc

Reply to  rbabcock
August 14, 2018 1:12 am

Here is a paper that is based on empirical measurement of the focussing effect:

And, no, not all seismic waves arrive simultaneously, far from it. And the focussing is actually fairly tight, the spread is apparently mostly due to the fact that the Earth is not perfectly spherical, that the interior is not quite isotropic and that the crust is on the average thicker in the northern hemisphere (many of the waves bounce off the bottom of the crust)

Johnny Cuyana
August 13, 2018 9:44 am

Since, long ago, understanding that “larger scale” earthquakes and/or explosive volcanic phenomena which cause seismic waves — such as the Krakatoa event — can reverberate through the entire planet for several weeks, or longer, I have wondered whether such tectonic events, if happening in “swarms”, could have some impact on earth rotational, and/or orbital performance; and, by extension … even then on some measurable climate changes.

I was always confident that research level earth science academy has had this same notion, much before it ever crossed my brain, but, upon reading this article it made me rethink of this hypothesis.

Comments, regarding this matter, from any “pros” … will be appreciated greatly.

Reply to  Johnny Cuyana
August 13, 2018 10:03 am

Earthquakes can impact the earth’s rotational speed since earthquakes can move large masses up or down. Depending on the nature of the fault zone.
If memory serves, the Fukushima quake sped up the earth’s rate of rotation by a micro second or two.

There is no way for an earthquake to affect the earth’s orbit. The only way a volcano could affect the orbit would be if the initial explosion was strong enough to eject material at escape velocity or higher.

Alan Tomalty
August 13, 2018 9:48 am

Are we sure that sequestering CO2 in the ground won’t lead to an increase of earthquakes?

Reply to  Alan Tomalty
August 13, 2018 10:04 am

I know, I know.
Since CO2 traps heat, injecting CO2 into the ground will cause the crust to melt.

Reply to  MarkW
August 13, 2018 10:54 am

That would be the “greenbasement” effect.

Reply to  Jim
August 13, 2018 11:09 am

That seems to be a rather ‘moldy’ thought!

Reply to  Jim
August 13, 2018 1:35 pm

OMG, the stuff you find in grandma’s root cellar!

Reply to  MarkW
August 13, 2018 2:09 pm

Well since it’s “millions of degrees” down there already just think of how hot it will get when the magic molecule gets there.

August 13, 2018 9:49 am

Yea, but how is CO2 causing it? There has to be a way to tie every observation back to CO2 or the researchers will lose funding.

August 13, 2018 9:57 am

I recall that there was some speculation that the Deccan Traps event might have been triggered by the Chicxulub event via a similar mechanism.

Rud Istvan
Reply to  Phil.
August 13, 2018 10:22 am

Correct that there was that speculation. Has since been fairly well disproven two ways. 1. Timing is off. 2. India’s location then is off. Current thinking is that Deccan Traps resulted from the hot spot plume that is now the Reunion hotspot.

Robert W. Turner
Reply to  Rud Istvan
August 13, 2018 6:07 pm

I haven’t actually been able to actually find good information on the iridium layer in relation to the Deccan Traps. You got a link?

Reply to  Robert W. Turner
August 14, 2018 1:25 am

As yet nobody has been able to find the iridium layer there. Remember that outside North/Central America it is rarely preserved except in deep water deposits. On land a few millimeters or centimeters of dust is quickly washed away, particularly when there is no vegetation, which there usually wasn’t in the immediate aftermath of Chicxulub.

Reply to  Rud Istvan
August 14, 2018 1:17 am

No, timing isn’t off. The Deccan traps started erupting well before Chicxulub but the main eruptive pulse (which comprised 70% of the total extruded basalt) cannot be separated chronologically from the Chicxulub impact.

August 13, 2018 9:57 am

WUWT’s policy against discussion of the Electric Universe precludes me from discussing the most logical inference.

Reply to  Controversies of Science
August 13, 2018 3:12 pm

Having now gone through all of the responses so far, I’m very much surprised that nobody even mentioned this considerably related observation at

“Researchers in Taiwan monitored 144 earthquakes between 1997 and 1999, and they found that for those registering 6.0 and higher the electron content of the ionosphere changed significantly one to six days before the earthquakes.”

And this stunning development as well! From

“Widespread tsunami-like waves of 23-27 June in the Mediterranean and Black Seas generated by high-altitude atmospheric forcing

A series of tsunami-like waves of non-seismic origin struck several southern European countries during the period of 23 to 27 June 2014. The event caused considerable damage from Spain to Ukraine. Here, we show that these waves were long-period ocean oscillations known as meteorological tsunamis which are generated by intense small-scale air pressure disturbances. An unique atmospheric synoptic pattern was tracked propagating eastward over the Mediterranean and the Black seas in synchrony with onset times of observed tsunami waves …”

Could it possibly be any clearer that WUWT needs to reconsider its policy against discussions of electricity in space?

Robert W. Turner
Reply to  Controversies of Science
August 13, 2018 6:22 pm

Looks like your comments cleared but I don’t see what the electric universe thing has to do with those papers.

Reply to  Robert W. Turner
August 13, 2018 7:08 pm

You’ve got an enhanced ionospheric electron density and earthquakes occurring on opposite ends of the Earth. Treat the atmosphere and crust like a leaky capacitor; treat the ionosphere and core as conducting plasmas.

The ionosphere is a plasma double layer. Is it possible that some plasma events penetrate the double layer such that an electrical connection is made from one end of the Earth to the other, through the core?

If you are needing an example, then consider a Venus-Earth-Mars alignment. Venus’ magnetotail is just barely long enough to touch the Earth’s magnetosphere. If the same thing is happening at that moment with Earth and Mars, it seems that you’d have all of the right ingredients to induce a discharge through the Earth’s core.

The following video presents a similar idea, but instead of direct electrical connections, considers the role of electromagnetic resonances due to planetary alignments …

It’s possible that one or both is happening (?).

Whether or not earthquakes result from planetary alignments or is of course a considerably more complex matter.

Reply to  Controversies of Science
August 14, 2018 3:36 am

sprites;-) not all charges go down or sideways
haarp was used to create a plasma ball for a dew minutes in ionosphere, they were pretty happy about it.
bet thepower use was massive.

Pop Piasa
August 13, 2018 10:04 am

Odd that there appears to be no record of any quakes on the opposite side of the globe coinciding with the largest earthquake documented in NA, the New Madrid quake of 1811-1812.
The best I can find is this Pdf at
It calls the quakes “enigmatic”.

D. J. Hawkins
Reply to  Pop Piasa
August 13, 2018 10:36 am

Anywhere on the continental US puts you in the Indian Ocean on the other side.

Pop Piasa
Reply to  D. J. Hawkins
August 13, 2018 11:42 am

Thanks, I didn’t bother to look at that. My bad.
…LOL, so much for Wile E Coyote digging to China!

Reply to  D. J. Hawkins
August 13, 2018 1:38 pm

Poor Diego Garcia Island will pop off into outer space if the big one ever hits LA!

Reply to  rocketscientist
August 13, 2018 2:32 pm

If it doesn’t capsize first.

Robert W. Turner
Reply to  Pop Piasa
August 13, 2018 6:26 pm

The Reelfoot rift system is not enigmatic. Large quakes happen there irregularly but on an average of 160 years per dated sand blowouts that occur during.

August 13, 2018 10:11 am

I have a very similar theory to this. USGS used to have a way to watch a date range of earthquakes on the map sequentially. I would center it on the pacific and it would seem (no proof) that large eqs would trigger other eqs. Inner connectivity in a sphere is not out of the range of believable for me. I think a great many things could be explained if the scope of the science was widened. Its the trouble with blinder science… sometimes you cant see the big picture.

August 13, 2018 10:25 am

It has to be the first quake that causes the 2nd, it couldn’t possibly be a common cause that is responsible for both…{facepalm}

Ron Long
August 13, 2018 10:25 am

There are two basic problems with this proposed scenario. The first is that stress cannot be transmitted through the earths molten core and almost not through the plastic mantle, stress must be transmitted through the earths lithosphere. The second problem is Newtons Gravity formula, which is also the Law of Inverse Squares, that is, a force attenuates as the square of the separation increases (center of mass to center of mass for gravity). Earthquakes are produced when a fault moves and that fault must have accumulated stress near to the point of overcoming drag on the fault plane. So a large earthquake should add stress to faults in rapidly diminishing distance from the epicenter, which should produce an aftershock pattern decreasing as the distance from the epicenter increases. Antipode effect? How? I have an MS in Geology from Oregon State University, did I miss something?

Smart Rock
Reply to  Ron Long
August 13, 2018 12:03 pm

Ron: I think the idea is that there are fault zones all over the place, and they are in various stages of being stressed. A big quake sends out shock waves, and these will shake up the crust everywhere, and faults that are stressed close to the point where they will move, leading to the secondary quake, a bit earlier than they would without the shake-up.

The antipodean aspect is quite reasonable; if shock waves radiate out omnidirectionally from the site of a quake, travelling mainly through the crust, they will start to form interference patterns on the far side of the globe. You can do similar things with ripples in a bathtub.

Stress is not propagated directly, but seismic waves can trigger stress relief at distant points. The stress that is relieved was already there.

Reply to  Smart Rock
August 13, 2018 1:34 pm

Ron Long, Smart Rock

Every day’s a schoolday at WUWT for me.

Probably fundamental geology to you guys but things I had never thought of.

Thank you.

Robert W. Turner
Reply to  Ron Long
August 13, 2018 6:33 pm

The next time an 8.0+ earthquake happens anywhere in the world, look at your local seismograph, then get back to us how that’s impossible.

Or you can just look at this.

What did they teach you at OSU about earthquakes, that man causes them by fracking?

John Bell
August 13, 2018 10:29 am

The big rock that hit Mexico some 66 million years ago, what did it feel like to be standing at the antipode when those shock waves came around? that must have been an earth quake big enough to throw an animal up in the air.

Reply to  John Bell
August 13, 2018 1:50 pm

That animal would necessarily have to have been a whale or fish as the antipode for that event would have been in the middle of the Indian Ocean.
As to how it would have felt, I would say wet.

Reply to  rocketscientist
August 13, 2018 2:10 pm

Correct as to location, but of course no whales 66 Ma:

Whether the antipodean giant marine reptiles at the end of the Cretaceous felt anything or not, they were promptly about to go extinct.

Reply to  John Bell
August 13, 2018 5:44 pm


I believe India was much further south and close to the antipodal node 66MYA. Of course we had the huge Deccan Plateau volcanism at that time continuing for hundrerds to thousands of years. It may well have been the last straw for the dinos and lots of other life. Enough lava to cover the entire landmass of the Earth by up to 1m according to some reports.
Makes the present day activity at Hawaii look a bit pathetic(unless you are unlucky enough to live there).

They don’t make volcanoes like they used to!

Charles Higley
August 13, 2018 10:33 am

Finally, a real case for tipping points, when the stress builds up to the point that something gives or triggers an earthquake or two or three on more.

Tom O
August 13, 2018 10:35 am

Did I misread something here?

“”Each test case in the study represented a single three-day window “injected” with a large-magnitude (6.5 or greater) earthquake suspected of inducing other quakes, and accompanying each case was a control group of 5,355 three-day periods that didn’t have the quake injection.””

This doesn’t sound like data analysis, it sounds like they were using a “model” of some type. Sorry, but I really don’t have much faith in models any more. This is another area where there isn’t enough knowledge to warrant the term “model.”

August 13, 2018 10:40 am

Crazy idea: the earthquakes are precipitated by exotic objects, like mini-blackholes, that fly through the Earth at high speed. Sometimes, but not always, both the entry and exit locations are disrupted and have an earthquake within a few hours or days.

Reply to  Seattle
August 13, 2018 12:15 pm

If that were true, earthquakes would occur randomly around the globe instead of being concentrated mostly at plate boundaries.

Secondly, the article stated that secondary quakes at the antipodes were sometimes delayed by as much as three days. A “high speed “black hole”” wouldn’t take three days to transit the earth.

Reply to  MarkW
August 13, 2018 1:40 pm


You’re to darn logical. 🙂

Although, a black hole hit might be ‘instant’ but the shock wave would take no less time to travel then from a non black hole event.

And before I disappear up my own black hole, I think your first point is entirely convincing.

Reply to  Seattle
August 13, 2018 1:54 pm

Those are called neutrinos, but they don’t cause earthquakes.
Why would they need to pass through the center of the planet?

Reply to  rocketscientist
August 13, 2018 2:34 pm

In order to collect $200?

August 13, 2018 10:41 am

Excerpt from this article:
“In 1700, an approximate magnitude 9.0 earthquake hit the Cascadia Subduction Zone – a fault that stretches along the West Coast of North American from British Columbia to California.”
[end of excerpt]

And the tsunami it raised in January 1700 on Vancouver Island crested at ~200 feet vertical height, about ten times higher than the ~20 foot-high tsunami that killed hundreds of thousands of people in SE Asia in 2005. The 1700 tsunami was so huge that it caused destruction all the way across the Pacific Ocean in Japan, and is recorded in their Court records.

As an observation, we are ill-prepared for such an event, which has a much greater probability of occurrence than anything to do with dangerous man-made global warming. For people who need to worry about something, worry about a mega-tsunami, caused by another major quake along the Cascadia Fault.

Glad to see the January 1700 Cascadia tsunami mentioned. The disastrous 2005 SE Asia tsunami vertical height was about 20 feet. The 1700 tsunami vertical height was about 200 feet. Yup.

I posted this March 12, 2011 in the Vancouver Sun.

During this (2005 major tsunami) event, Japan residents got 10 to 15 minutes warning to move to higher ground. The tsunami alarm was issued 3 minutes after the earthquake occurred.

If the Cascadia fault ruptured, the folks at Tofino would get about that much warning, assuming our tsunami warning system worked perfectly.

In Vancouver, they would get about 30 minutes warning. Ever try to evacuate Surrey and Delta in 30 minutes? Yeah, that’ll work.

A tsunami wave moves over the open ocean at speeds over 700 km/h (500 mph).

The tsunami warning system worked Friday, with the agency alerting people to imminent tsunamis within three minutes of the quake, and the first waves struck 10 to 15 minutes later. The alert may have saved hundreds of lives, as some residents were able to flee to higher ground.

Read more:

August 13, 2018 12:49 pm

The tsunami occurred on 26Dec2004.

August 13, 2018 1:51 pm


I understand the inevitable collapse of the west coast of Tenerife into the sea will cause a major tidal wave to reach the east coast of the US, swamping Bermuda in it’s path.

Perhaps urban myth as my sister lives in Bermuda and the story circulates.

Reply to  HotScot
August 13, 2018 2:13 pm

Probably not for tens of thousands more years, although you never know:

Reply to  Theo
August 13, 2018 3:02 pm


No one thought of global warming 10,000 years ago, but apparently it’s here now.

As you say, you never know.

Reply to  HotScot
August 13, 2018 9:40 pm

HI HotScot,

I’ve heard that rumour about a huge future Atlantic tsunami, but never investigated it.

I have given some thought to another Cascadia Fault 9.0 event, and the current civil defense drills remind me of those practiced during the Cold War era – you know the one where you get under your desk, lace your hands at the back of your head, push your head down between your knees, then press down a little further, and kiss your ass goodbye!

The current contingency plans are not going to work for low-lying areas like Surrey and Delta, imo. No smiley face this time.

Best, Allan

August 13, 2018 10:49 am

Here’s a diagram showing how shock waves propagate through the Earth. Note that there are no direct paths through the core.

A shock originating on one side of the planet will travel by many paths. Only on the opposite side of the globe will all the shock waves arrive at about the same time. That’s where you get the maximum effect, except of course right near the original earthquake.

Rud Istvan
August 13, 2018 10:49 am

I find this result worse than doubtful on three grounds.
1. Methodology. P values from a small sample,of large earthquakes are dubious at best. Earthquakes happen all the time. Some will happen somewhere within three days of any large earthquake. Result is likely a statistical junk artifact because of points 2 and 3.
2. Crustal rock shockwave energy dissipates as a rough function of d^2, so mostly +/- 30 degree antipodal triggering would involve about the least possible energy. Any Triggering from shockwave energy should probabalistically happen more the closer to the major event.
3. 3 day window timing. Earthquake triggering shockwaves would be S or P waves, not very likely surface waves since major earthquakes are almost always at some significant depth. The slower S waves 10k km travel time is ~20 minutes. Faster P wave 10k km time is ~10 minutes. Both are ‘averages’, because depends on depth and intervening rock types. So there is no shockwave energy to trigger anything anywhere after less than an hour given slowest furthest case equatorial circumference of ~40k km. The arbitrary three day window just enables the non physical statistical junk of point 1.

Reply to  Rud Istvan
August 13, 2018 12:18 pm

At the antipode, shock waves from all directions arrive at about the same time. This negates the d^2 affect.

Reply to  MarkW
August 13, 2018 12:37 pm

True. Past that, things become complicated to the point of being intractable. The ‘simple’ behavior of a thin sphere is a case in point.

The behavior of a thin sphere is way complicated. link The link is interesting because it shows how quasi periodic phenomena arise. The waveforms observed on a thin sphere are very much like what happens with the Earth’s climate. Of course, compared with the Earth with its oceans and atmosphere and mantle and core, a thin sphere is simple. LOL

Robert W. Turner
Reply to  MarkW
August 13, 2018 6:50 pm

And here’s a fun fact, the 2004 Sumatra EQ vertically displaced the crust by at least 1 cm everywhere on Earth. But the further away you were the longer the period on which this occurred so you didn’t notice.

And yes, the antipode +/- 30 degrees is where the waves would all come together and interfere with each other.

Bob Weber
August 13, 2018 10:50 am

It happened twice last night, and once on the 5th, from ANSS (&GSN):

comment image

comment image

Joel O'Bryan
Reply to  Bob Weber
August 13, 2018 1:54 pm

Now that looks scary.
Are the bottom two images of the Southern Hemisphere?

Bob Weber
Reply to  Joel O'Bryan
August 13, 2018 2:17 pm

They’re sets from two nights a week apart. They’re actually not ordered the same. The top pair, 5Aug18 are ANSS & GSN, and the bottom two are opposite, GSN & ANSS from last night, 12Aug18.

ANSS is USA, and GSN is global. Last night there were two strikes a few hours apart felt worldwide. Each heliplot is for a different location for the last 24 hrs, all synced in time, and continually updated on each website, ie, here’s Lake Yellowstone at this hour.

Bob Weber
Reply to  Bob Weber
August 14, 2018 7:55 am

The EMSC data says Northern Alaska was hit first on 8/12 then again with a large aftershock, with the rest of the earthquakes following both times.

2018-08-12 14:58:54.7 69.62 N 145.25 W 10 6.4 NORTHERN ALASKA
2018-08-12 21:15:02.5 69.62 N 144.36 W 10 6.0 NORTHERN ALASKA

The GSN duplicates a number of American sites from ANSS.

Non Nomen
August 13, 2018 11:02 am

If the first earthquake triggers a second one on the other side of the globe and it happens to be that this one is amplified by some sort of a local predisposition, then some sort of echo ought to occur and this echo might re-trigger another earthquake where the first took place. Are there observations of this kind?

Bob Weber
Reply to  Non Nomen
August 13, 2018 12:03 pm

The timing among individual heliplots can be visually compared. The systematic way would be to first find the earliest largest earthquake among the bunch, which would be easy to do with regular filterable data from EMSC.

August 13, 2018 11:52 am

There’s a youtuber called Dutchsinse that has been accurately forecasting earthquakes for some time now. His live feed has apparently moved to Twitch now because of the latest youtube actions.

Ed Zuiderwijk
August 13, 2018 12:03 pm

After the great Andaman Boxing day quake, the one causing the tsunami that drowned a hundred thousand, the Earth reverberated for hours like a struck bell. No surprise that such big quakes can destabilise faults at the far end of the globe.

Reply to  Ed Zuiderwijk
August 13, 2018 1:08 pm

There is something else to consider in aftereffects from large quakes. The current grouping up in North Alaska is a good example of how a strong teleseismic quake can lead to a dampening of the overall global quake count for days or many days after the initial large quake.

The huge quake count at Hawaii also had a global effect while it was running. Interesting to note that the 24/hr global rate of 2.5+ mag quakes rose around 50% last week when the series of quakes in Hawaii finally came to an end early last week, after a 4 month run. The last 24 hours is the first day without a 2.5+ mag quake in Hawaii since early April. Then this 6.4 quake in Kaktovik Alaska which struck almost 24 hours ago also dampened the global count around 40%, excluding the several hundred plus quakes per 24/hr at the Alaskan location.

Reply to  goldminor
August 13, 2018 2:38 pm

There is no mechanism for an earthquake in one part of the globe to suppress earthquakes in another part of the globe.

Rich Davis
Reply to  goldminor
August 14, 2018 3:21 am

Are you suggesting that pressure waves emanating from low magnitude quakes are acting to relieve stresses non-catastrophically so that minor quakes are suppressed, but higher amplitude waves due to major quakes have the opposite effect, triggering other major quakes?

If it is more than a coincidence then there should be thousands of similar cases associated with volcanic activity. Excluding low-magnitude quakes within some small radius of volcanic activity should cause those time periods to show a decreased count globally, compared to periods without major volcanic activity.

Reply to  Rich Davis
August 14, 2018 1:23 pm

I first took note of this effect during the March 2011 quake in Japan. For many weeks after that quake there were very few quakes anywhere else around the globe greater than 2.5 mag. Unfortunately, most of my notes from that time period were all lost as I kept a daily count at the time for over a year.

Any quake rated as a teleseismic quake can affect the total global rate from my observations. The global rate is still a bit below average with the exception of the rate in Alaska of 80/24hrs and globally 20/24hrs, sixty hours after the strong quake.

Crispin in Waterloo
August 13, 2018 12:25 pm

Waves traveling from one spot around the world will arrive at the antipode at the same time, adjusting for the density and thickness of the media through which they pass. This means the arrival point where the incoming waves “sum” will be offset to the same extent there is a signal delay through the materials

It appears the biggest difference one can find moves the summation point 30 degrees.

It sounds like a rogue wave: the sum of multiple wavelets at one point to give a brief, significant jarring. Wherever there are predisposed concentrations of stress within that window, the coincidence of waves would push it over the edge.

Very interesting, and with sufficient knowledge about the Earth’s interior, a viable prediction mechanism. Cool!

Reply to  Crispin in Waterloo
August 13, 2018 2:39 pm

Only if you know how much stress has built up in the faults at the antipode and how close any of them are to breaking.

Gary Pearse
August 13, 2018 1:27 pm

Mods, I had an interesting post on the big Peruvian quake of 2007 maybe triggering the cave in of the Crandall Canyon coal mine in Utah that killed a number of miners and rescuers. For some reason it was disappeared. I tried to advise of this quake/mine cave in in an email to the inquiry, but got no acknowlegement back. My point was, it doesnt have to be at the antipodes. The big quake in Qom, Iran that killed 25,000 people also was followed by several notable quakes near and far. You only need a situation where an incipient quake, or cave in or bridge collapse etc. needs only a nudge. Common sense is enough data for this sort of thing.

Joel O'Bryan
August 13, 2018 1:38 pm

Clearly from the figure above, the Earth is flat.

August 13, 2018 1:53 pm

The earthquake’s wave propagation is function of the earth’s internal structure. Analysis of a particular type of a wave that penetrates the inner core (shown in green in the animation below) from data associated with numerous major earthquakes has concluded that the Earth’s inner core is lopsided and it rotates at a slightly different angular velocity than the rest (the inner core’s differential rotation). It is thought that it may be responsible for the slow drift of the South Atlantic anomaly westwards, as well as gradual reduction in the magnetic field intensity across Americas and the increase in the eastern hemisphere.

August 13, 2018 1:59 pm

I sure would like to see similar verbiage, like this:

“The understanding of whether human activity is warming the planet is still largely speculative.”

August 13, 2018 2:36 pm

“Each test case in the study represented a single three-day window “injected” with a large-magnitude (6.5 or greater) earthquake suspected of inducing other quakes, and accompanying each case was a control group of 5,355 three-day periods that didn’t have the quake injection.”

What is this “injected”? Thought this was based upon observations.

August 13, 2018 3:37 pm

It sounds as though, in theory, if someone exploded a large nuclear device, or several, on the floor of the Indian Ocean off the coast of South Africa, they could induce a large earthquake in California, sending it into the Pacific.

Not likely, but perhaps the plot line of a future James Bond movie….but I wonder how many would be rooting for the bad guys in this scenario.

Reply to  Jtom
August 14, 2018 10:41 am

It would be rather difficult to implement. To get good energy transfer the bomb should be underground which is hard in deep ocean. An air explosion where the fireball does not reach the ground does not have much of a seismic footprint. The 50 MT “Tsar Bomba” where the fireball did not quite reach the ground only had a magnitude of about 5.

August 13, 2018 3:41 pm

‘The findings, published Aug. 2 in Nature Scientific Reports, are an important step toward improved short-term earthquake forecasting and risk assessment.’

Not really. This is gross speculation.

Reply to  Gamecock
August 14, 2018 1:55 am

Not quite speculation. The focussing effect is quite real. But it would be very short term forecasting. The seismic waves arrive at the antipodes 20 to 40 minutes after the original quake.

Reply to  tty
August 14, 2018 11:01 am

Could enough time for people to evacuate sea coasts and get to higher ground.

I would if on the Pacific coast when warning arrived. Of course, I’d run into everyone else trying to do the same.

Which is why, were I to live on the coast, I’d buy an electric helicopter.

John Johnston
August 13, 2018 5:40 pm

Correlation ain’t causation. It may be that some states of Earth’s magma have greater likelihood of triggering major quakes at widely separated points on the surface than others. That seems at least as likely as attributing the cause of the latter quakes to the earlier ones.

Geoff Sherrington
August 13, 2018 7:36 pm

The paper is speculation because of a major impediment, namely, that earthquakes can be caused by different mechanisms. For example, the Hawaii island chain pushing up through the Pacific can be envisioned as having coincident earthquake noise. OTOH earthquakes happen in the middle of stable, old, tectonic plates on land. Further, there are shallow earthquakes that likely happen with rock metamorphism such as dewatering/dehydration of sediment piles. Even isostatic adjustment is probably noisy with quakes.
There is no compelling logic to implicate the antipode. (There is, unrelatedly, an interesting story from the cold war missile era, that ballistic missiles launched in any direction from the main site then in Russia would converge on the antipode S-E of New Zealand , a good location to shoot them down.
Shooting down is the best way to treat this earthquake paper. Geoff

Reply to  Geoff Sherrington
August 14, 2018 1:48 am

“OTOH earthquakes happen in the middle of stable, old, tectonic plates on land.”

Very rarely though (except in connection with (de)glaciation). And isostatic adjustment is actually remarkably quiet seismically except immediately after deglaciation. Check Fennoscandia or the Canadian Shield. Lots of isostatic adjustment, very little seismic activity.

And the seismic focussing effect is definitely real:

And I don’t see this as being much help in forecasting earthquakes. Let’s say there is a really big one in the Indian Ocean. What do you do? Send out a warning “there might be a quake in California/Cascadia in ten minutes”?

Also note that the seismic energy by itself is far too weak to cause earthquakes. But it may be enough to trigger earthquakes when the strain on a fault is very close to breaking point, and cause a quake to happen slightly earlier than it would have done otherwise.

Very large impacts (much larger than Chicxulub) can focus enough energy at the antipodal point to directly cause tectonic effects. The chaotic terrain antipodal to the Caloris basin on Mercury is a prime example.

And why should anyone send ICBM:s from Russia to the US by way of New Zealand? It is much easier by way of the Arctic.

Geoff Sherrington
Reply to  tty
August 14, 2018 5:32 am

What is up? You are normally smarter than this.
Ballistic missiles are not steered after liftoff. They are launched at various directions to hit targets. If left to go far enough, they all cross the antipode. Nowhere else to go.
Re quakes, I gave but a few examples of types. Does not matter if a type is rare, for the paper discussed. To me, the critical bit is energy. The trigger energy, by analogy to a gun, need not be large. The destructive energy is in the cordite. Meaning, you can have lots of quake triggers over a time, but unless the quake locality has cordite potential, it is a non event. The art of prior measurement of such potential is far from adequate, not from lack of effort, but because it is so difficult.
So the paper really deals with nothing useful. Geoff.

Reply to  Geoff Sherrington
August 14, 2018 10:29 am

“Ballistic missiles are not steered after liftoff.”

ICBM:s most definitely are. The evolution of inertial navigation was very largely driven by the need to precision steer ICBM:s. It is true that early ICBM:s were purely ballistic after burnout, but for more modern MIRV:ed missiles the “bus” was steered until all the MIRV:s had separated.

And if you think a bit you will realize that there are two great circle courses between any two points. One (the longer) will pass the antipodal point. The other, shorter, will not. ICBM trajectories almost invariably use the shorter route for obvious reasons (less energy needed, better precision, shorter flight time).

The Soviet Union did develop the R36ORB missile with FOBS (Fractional Orbit) capability which theoretically could have used “the long way around” to attack from the south, but it is doubtful if this was ever implemented. The missile was retired in 1983.

Reply to  tty
August 14, 2018 11:03 am

ICBMs also have blow out panels to stop them or change direction before MIRVed warhead separation.

Geoff Sherrington
Reply to  tty
August 15, 2018 4:04 am

OK, tty, good to see the thinking cap is still on.
You know enough to have realized that I was talking about the early days before steering. We used to distinguish between ballistic missiles and guided missiles. It is not important, it was but an interest comment.
Corporately, in our mineral exploration days, we were involved in the technology of inertial and other navigation types via the need to be able to know the path of a diamond drill hole going at times up to 3,000 metres below the surface. Those strong looking, heavy drill rods actually behave more like spaghetti al dente. If, as often happens, the path goes through or near magnetic rocks like those with magnetite or pyrrhotite for example, then using a compass technique to get a bearing is out of the available technology. We were into laser ring methods but this was before they could be made small enough to prevent catastrophic laser light loss on tight bends. Life can be interesting when you have earned the scientific freedom to imagine “What if?”
We also used to wonder what newly-engineered printing device for our 1960s computers would replace the ASR33 tty Later we learned with fascination about the accuracy of guidance of cheap plastic ink jet printer heads, far better control than we had imagined possible. We learned a lot about miniaturization.
Do keep writing enjoyable, correct comments, tty Geoff.

August 14, 2018 12:57 am

Not unexpected. It is well established that seismic energy is focussed near the antipodal point. Google “Caloris Basin” and “chaotic terrain” for a well-known extreme example.
It has even been suggested that the huge main pulse of the Deccan flood basalt was triggered by the (antipodal) Chicxulub impact

Dr. Strangelove
August 14, 2018 6:51 am

The authors of this paper need a tutorial on probability. The annual frequency of M5.0 or greater earthquakes is around 1,600. How many earthquakes to expect in any 3-day period?
1600/(365/3) = 13
30 deg. latitude = 2070 mi.
surface area = (3.1416) 2070^2 = 13.46 million sq. mi.
Now spread that 13 earthquakes evenly over Earth’s total surface area (197 million sq. mi.)
197/13 = 15.15 million sq. mi.
That’s bigger than 13.46 million sq. mi.
Therefore, you expect to get an M5.0 or greater earthquake in your target area by chance alone. The probability is P = 1

Dr. Strangelove
Reply to  Dr. Strangelove
August 14, 2018 6:56 am

P < 1 = 13.46/15.15 = 0.89

Bob Weber
August 14, 2018 7:38 am

Once again from today, on 14Aug18:

comment image

August 14, 2018 12:25 pm

In Greece since the 1970s, someone appeared very observant and connected the earthquakes in West Central America with earthquakes in Greece. He was particularly successful in determining the time of the earthquake in Greece (+/- one day) and size. For the epicenter, additional in situ work was needed to identify precursor phenomena. The methodology was very successful but the creator was remained unknown because he was not a scientist, he was a skier teacher!!

August 15, 2018 7:06 am

It is apparently happening on the Sun too.
Comet hitting Sun triggers eruption on the opposite side of Sun.

%d bloggers like this: