Breaking Up with the Richter Scale

One of my New Year's blogolutions was to clear out my to-blog folder, and bring closure to my unfinished drafts by simply posting them as-is. This is one of those drafts. Disorganized paragraphs, unfinished sentences, and general incoherence enhance the natural character and beauty of a half-written blog post and should not be considered flaws or defects.

Draft date: January 23, 2008

A powerful earthquake measuring 7.7 on the Richter scale ripped through northern Chile today, causing panic, power outages, damage to airport tarmacs and road closures, according to news reports.

I wince every time I see a newspaper article talking about the Richter scale (here it's from the New York Times last November). Don't get me wrong, I love the Richter scale - but I love it like you will always love the ex who taught you how to have a real relationship, even though it didn't work out in the end. Seismology's relationship with the Richter scale was deep and meaningful, and they will always think fondly of their time together, but they are no longer right for one another.

Here is how seismology and the Richter scale met and fell in love. Prior to the mid-1920s, the business of making seismometers was largely carried out by individual seismologists, who all had their own clever ideas for design improvements. As a result, each seismometer was unique, and there was no good way to make reliable comparisons between seismograms recorded at different stations.

In 1925, a couple of guys named Wood and Anderson had some clever ideas and built a seismometer. It was nice enough on its own, but the real innovation was that they made 20 of them. These instruments were installed throughout Southern California.

By 1935, this array of Wood-Anderson seismometers had collected enough data that Charles Richter and Beno Gutenberg were able to come up with the world's first-ever measure of the "bigness" of an earthquake - a measure that is independent of where you are standing when the earthquake happens. Why Beno Gutenberg's name didn't make it onto the concept, I don't know.

Anyway, the Richter or Richter-Gutenberg magnitude is a function of the maximum amplitude of the waves recorded on a Wood-Anderson seismometer, and the distance between that seismometer and the earthquake. The fact that the distance away from the earthquake is part of the equation is tremendously important. It is only after you account for the fact that seismic waves get weaker with distance that you can start to say anything meaningful about what is happening at the fault. Figuring out what faults do and why is one of the major goals of seismology, and I hope it's obvious that characterizing an earthquake in terms of what's happening at the fault is a useful perspective.

I harp on this point because it's related to another annoying misconception, albeit one that is less frequently perpetrated by major newspapers: Magnitude is not the same as intensity. People will sometimes say things like yeah, but over at the Dark Tower, that earthquake was only a magnitude 3, so the orcs didn't get upset at all. An earthquake's magnitude doesn't change when you move from one place to another. What changes is the intensity of the earthquake experience, which is measured using a different scale, the Mercalli Intensity Scale. This scale is based on the severity of effects observed at a particular location, and any given earthquake will produce a range of intensities.

But back to the story! We can already see the seeds of discontent in this relationship. The Richter scale was too demanding - it needed a particular seismometer, and it needed to be in Southern California, or everything would go wrong. Similar amplitude/distance relationships were found for other regions; these are now known as local magnitudes, and are closest in spirit to the original Richter magnitude. Seismologists also grew up from using just the maximum amplitude of an earthquake to find the magnitude, and now for large earthquakes we use the whole seismogram (er, well, the whole seismogram with the high frequencies filtered out) to determine a moment magnitude. And so on and so forth.

Using a modern broadband seismic record, it's possible to determine what an earthquake would have looked like on a Wood-Anderson seismometer, and calculate a real Richter magnitude. This process has very little physical meaning outside of Southern California, but you can do it, just for old time's sake.