*THREAD* Okay let's talk about magnitude. For non-specialists, the important point is that magnitude reflects overall earthquake size, whereas the shaking that you feel at any given place depends a lot on what happens to waves after they leave the fault.
The question is, what do we mean by "size." The seismology community has adopted moment magnitude as the "best measure" on an earthquake size; whether we say so or not, we're almost always reporting moment magnitudes, at least for big earthquakes.
Fundamentally, moment magnitude reflects the size of the fault patch that moved in an earthquake and the average movement (slip) along that patch. It is not a direct reflection of radiated energy.
Radiated energy depends on other variables, like whether motion along the fault patch was smooth or jerky. Two earthquakes with the same moment magnitude can have quite different shaking, at least over the frequency range that matters for most buildings.
On average, moment magnitude correlates with energy release, but I see two problems with the statement, "magnitude is a measure of the energy of an earthquake."
First, even if most people don't care about gory details, I don't think scientists should say things that are actually wrong. "Magnitude is a measure of the overall size of an earthquake" is, I think, clear, and not wrong.
Second, I'm not sure that seismologists have thought all of this through as much as we should. It was eye-opening for me to crank up baby equations to produce Figure 6 in this 2014 pub: pubs.geoscienceworld.org/ssa/bssa/artic…
In short, a magnitude 4.6 quake can produce the same shaking at frequencies of 1-10 Hz as a magnitude 5.4 quake, depending on how the fault moves. The GFZ group has indeed written some good papers on these issues, 1st discussed in the 80s by peeps like Hanks & Boore. & Johnston.
Also, as I've talked about, there are major implications for investigations of historical earthquakes, for which we often have *only* intensity estimates, which reflect shaking at about 1-10 Hz...and then we try to estimate moment magnitude.
Which is kind of like listening to only the high tones in a piece of music and trying to say what the low tones are. We do this with equations based on averages, but the uncertainties are enormous...and not, ime, fully appreciated.
Stepping off my soapbox now. Thanks for making it this far :)
