I make no claim to be an expert on seismology, but my understanding is that
earthquake "magnitudes" were always intended to be a measure of energy,
starting with Richter's first 1935 version. In fact, several of the
magnitude formulas start with what are (apparently) energy units, such as
dyne centimeters (ergs) or newton meters (joules). Apparently, the reason
they keep coming up with new definitions of magnitude--tweaking the formulas
and adding fudge factors--is to give a more honest measure of energy.
Why don't they use power (energy per time)? I suppose it's because they are
interested in the total amount of work (energy) that the quake can do.
Quakes happen very quickly, so trying to divide energy by time might lead to
wildly different power values for the same quake. Kind of like lightning
bolts, I suppose, where I see figures in joules but not watts.
As for the observed affects (local acceleration, "shaking", damage, etc.),
these vary greatly from spot to spot, depending on the soil, rock structure,
depth of focus, building type, direction to epicenter, types of waves, etc,
so are not useful as an intrinsic measure for the quake as a whole. As you
say, there are other (often quite subjective) scales for the observed
"intensity", such as the Mercalli scale (which is analogous to the Beaufort
scale for winds).
As I understand it, there are two basic ways of determining the magnitude
(energy) of a quake:
(1) Days to years after the event, by measuring the displacement and area of
the fault slippage.
(2) During the event, by measuring the amplitude, frequency, and timing of
various kinds of quake waves.
The first method is simple in theory, though often hard to do in practice,
especially if the fault scarp isn't exposed. You just multiply the force
required to break the rock by the distance it moved, which is the definition
of work (energy). Specificially,
E = µ A d
where
E is the energy (work), in joules, or "moment" as seismologists call it
(moment of what? there are lots of different "moments" in physics); µ is the
"sheer modulus" or pressure required to sheer [break] the rock (30 GPa or
more, depending on the type of rock); A is the area of the fault over which
breakage took place (which can be many km2); and d is the displacement
(never more than a few meters; usually much less)
Since Pa = N/m2, the SI units are (N/m2) (m2) (m) = N.m = J
For big quakes, this multiplies out to energies in the exajoules.But I see
that the energies that websites show by this method are thousands of times
larger than the energies they show in their magitude vs. energy tables. I
can't explain this discrepancy. Does anyone know a seismologist who can make
sense of this for us?
Method 2 involves all kinds of qualifiers--different formulas depending on
how far you are from the quake; which waves are measured; which time
frequencies to meaures; and other stuff, much of it derived empirically, I
gather.
> -----Original Message-----
> From: Bruce Raup
>>
> I definitely agree with that, especially when it comes to water heater
> specs, food energy, and any other case where the energy is the base
> quantity being measured.
>
> However, the feeling I got from reading various web sites on the
> definition of earthquake magnitudes is that the translatation from
> magnitude to energy is not a precise science, and that the magnitude is
> the more primary measurement. So rather than throw away the magnitudes
> and report the energy of a quake with a large uncertainty, it's probably
> better to report the magnitude first, then say something such as "this is
> estimated to correspond to x joules, subject to further refinement." Of
> course, there are different definitions for the magnitude, so there's
> probably need for standardization there.
>
> Additionally, there is not a direct correspondence between the energy
> released in a quake and the intensity of shaking locally, which is
> probably what most people care about more. In Japan, the media report not
> the Richter-like magnitude, but a "shindo" ("degree of shaking") number in
> the range of 0 to 7 (http://www.minc.ne.jp/kics/kbt/spring96/eqprep.html),
> similar to the Modified Mercalli scale
> (http://wwwneic.cr.usgs.gov/neis/general/handouts/mercalli.html) used in
> the US.
>
> (Re total energy vs intensity: I lose the same amount of gravitational
> potential energy whether I hike down the mountain or fall off the cliffy
> side. But there's a big difference in the effects on my body!)
>
> Other parameters I've heard talked about in Japan are horizontal ground
> accelerations and ground displacements, which are of course in SI units.
> An article about the Kobe quake at
> http://www.newscientist.com/nsplus/insight/quakes/kobe1.html quotes
> some of these numbers.
>
> I'll try to ask some experts about this in the near future.
>
> I saw a couple of web sites using an exponent of 1.44M rather than 1.5M
> (e.g. http://lasker.princeton.edu/ScienceProjects/curr/eqmag/eqmag.htm).
>
> Cheers,
> Bruce
>
> --
> Bruce Raup
> National Snow and Ice Data Center Phone: 303-492-8814
> University of Colorado, 449 UCB Fax: 303-492-2468
> Boulder, CO 80309-0449 [EMAIL PROTECTED]
>
>
