This headline on Drudge caught my eye: “The quake even disturbed the earth’s rotation”.
:dubious:
Applicable quote:
Question: Is this possible? If this quote came from some politician I would probably just dismiss it, but coming from someone who’s the head of a national geophysics institute makes me wonder if it’s true.
Is it?
No.
Since the cause of the quake is internal to the earth, the center of gravity won’t shift. Basic physics.
Quakes are limited to the thin outer crust of the planet, which doesn’t really count as ‘internal’.
So colour me sceptical. Any effect on the Earth’s rotation would be miniscule to the extreme. So, if true, I’d say it was said more for effect than relevance. It sounds more like a reporter getting the wrong end of the stick.
Still, a terrible display of geological forces.
But the moment of intertia most certainly can, which would affect the rotational speed.
This is a very interesting question. The answer in the end turns out to be yes, theoretically, but the effect may not be detectable.
The earth is almost a rigid body. If it were completely rigid, the question would be moot because earthquakes, which are the release of stored energy, cannot occur in a rigid body. A rigid body cannot store the strain energy that is needed to drive an earthquake. Also, the mass of a rigid body is fixed in its distribution.
On the other hand, if a body were not at all rigid, like a blob of liquid or a cloud of gas, then the question would also be moot. The points in a non-rigid body can all do their own thing, so as a whole, the body has no unique angular velocity. It would be impossible to talk about changes in the rotation because the rotation could not be meaningfully defined.
But the earth is flexible enough that it can have internal strain energy that can lead to internal rearrangements, yet stiff enough that it still keeps its basic shape, which permits a reasonable, though not exact, definition of its rotation (angular velocity). Let the “rotation of the earth” be defined as the length of the day at a particular location.
The only way to change the angular velocity of an (almost) rigid body is to apply an external torque or change the distribution of mass. An earthquake is a release of internal forces, so it can’t generate an external torque. Any forces involved would be balanced out in equal and opposite pairs. However, an earthquake would cause movement of the crust and oceans, which would redistribute the mass of the earth and could cause a change or irregularity in the length of the day. Whether this would be detectable or even above the “noise level” generated by other routine rearrangements of the earth’s mass like the tides, I don’t know.
Another way an earthquake could affect the rotation is by generating local angular momentum through movement of the crust or ocean water. The angular motion would bounce around and eventually damp out.
In short, I see no theoretical reason why a powerful earthquake couldn’t affect the rotation of the earth, taking “rotation” to mean the exact length of the day.
I wonder whether the guy who was quoted had a particular observation or piece of data in mind when he said the rotation of the earth was affected. I’m not a geophysicist, so I don’t know exactly which assumptions they work under, how they exactly define “the rotation of the earth”, etc.
Every time you walk up a flight of stairs, the rotation of the earth slows down, and when you come back down it speeds up again. The effect is too small to measure, but angular momentum must be conserved. On a larger scale, the ocean’s daily tides produce a daily oscillation in the planet’s rate of rotation of roughly 80 microseconds. Atmospheric movement, and el Nino can increase day length by as much as 0.6 milliseconds. The reservoirs behind dams also alter day length by shifting water from one latitude and altitude to another.
Will an earthquake shift day length? The answer depends on whether the quake shifts a significant mass with respect to the earth’s axis of rotation. A quake on a fault in which the two sides slip past each other won’t have much effect, but if one side of the fault thrusts over the other, ther will be an effect on the planet’s rotation.
Thanks for the informative replies. I’m hopelessly ignorant of these things, so I should have known better than doubt the words of a leading geophysicist.
Squink, I was reading your links, and I noticed one was to the Bad Astronomy message board. I knew he* had a site, but not a board. I started looking around, and found this thread. There’s a link to a news article from New Zealand that is somewhat related to this.
*Does the Bad Astronomer still hangout here?
Hasn’t visited for a month, hasn’t posted for nine months.
I hope it doesn’t stop the Earth’s rotation, otherwise we’ll have to tunnel to the core in a vessel made of Unobtanium and detonate hydrogen bombs to get it spinning again.
I suspect that remark wasn’t so much a testament to the power of earthquakes to affect Earth’s rotation (even an earthquake as devastating as this one was) so much as to the fantastic degree of precision with which we can measure Earth’s rotation these days.
Please pardon the tangent: how much of a factor is surface friction (earth’s surface vs. troposphere) in stabilizing the earth’s rotational speed?
What if some large country erected a series of 1,000-foot-tall square rigid sails; would this have any effect on the earth’s speed?
I’m assuming not, since so many large cities contain giant structures with no noticeable lengthening of the day, but I wondered: why not?
It’s not “external” like a meteor collision or something.
Raza, the atmosphere is more or less moving with the Earth. That’s why we don’t experience a constant 1000mph wind.
That’s too bad - he’s wicked smaht.
[/Mainer]
I’ve been following the thread over at his site, and sure enough, B A himself joined in and offered the following:
The force of this earthquake is absolutely stunning.
Some relevant news stories on the quake, it’s power and effect on the planet:
http://www.cnn.com/2004/TECH/science/12/27/quake.seismic.ap/index.html
http://sg.news.yahoo.com/041228/1/3pim1.html
[sub]My apologies if quoting someone from another board is verboten for some reason[/sub]
Just updating with an article from CNN online. Complete article at the link.
I wonder about this statement. Suppose I assume that the earth is a loose collection of dust particles held together only by their mutual gravitation with no cohesion between particles.
I now compute the reduction in the gravitation resulting from the rotational central acceleration. I find that the reduction in gravitational acceleration and hence the force on an increment of the earth’s mass at the equator results in a diameter that is 27 miles greater than would be the case without rotation. This means that the equator is lifted 13.5 miles in this hypothetical. The actual equatorial bulge is about 13 miles.
I’m not sure of the meaning of this but it looks like the earth is pretty close to a collection of loose material held together by mutual gravitation.
Any geologists care to comment on my method and result?
By “almost rigid”, I mean only that the earth holds its shape well enough that calculating its angular velocity based on the length of the day at some geographical location gives a meaningful result. You can pick a location to do your measurements from and then make later measurements and comparisons under the assumption you’re still at the same latitude and radius. This only works because under typical forces, the earth’s configuration, at least as defined by the geography of the crust, hardly changes.
I repeated your calculation and got the same result. One possibility is that the strain (13/7925=0.16% radial, which is too much for solid rock) is all in the mantle and not the crust, but then the crust would still have to sustain “hoop” strain of about the same magnitude. A more plausible scenario is that the crust didn’t solidify until the the earth was already spinning and bulging at the equator, so the crust doesn’t have to sustain any strain relative to its stress-free configuration. That is consistent with an almost-rigid crust, providing geographic stability, and with a bulge at the equator representative of no cohesion in the earth.
This raises the question of how we know that the crust doesn’t slide around on the mantle or get “left behind,” with the mantle spinning around underneath it. Is there a method for detecting strain in the mantle by radar or sonar or something? I’m trying to form a mental picture of the earth’s structure with little success due to my near-complete ignorance of the subject - the density is probably a function of the radius, because the temperature, pressure and chemistry keep changing, and I have no idea what the viscosity of molten rock at high temperature and pressure is, whether it is purely viscous or has a “yield stress”, and how the crust interacts with the mantle. What’s at the center of the earth? How the hell would you find out? Why is the earth’s spin axis stable? Does it have to do with the magnetic field generated by the mantle currents? I’m sure a geophysicist could address these points without even cracking a book.
Actually “solid rock” is considered to have zero tensile strength. Although it does have some, structural designs don’t put stone in tension. So as the circumferance of the earth grows because of centrifugal effect, the rock layers in the crust would just fracture and pull apart while offering little resistance to the growth.
And, incidently stone is about 7.5 times more flexible than steel in flexure and 3 times in compression. Thissite and this one give the modulus of elasticity of stone. In flexure it (with quite a bit of variability) runs around 4,000,000 and in compression about 9,000,000. Steel is 30,000,000.
I’m not sure we have evidence that the rotation rate of the earth when it was still molton was the same as now.
Geologists study the interior of the earth by means of both natural and man-made seismic events.