Drop the Richter Scale

January 13th, 2010

richterscale2.jpg

Whenever news of an earthquake hits, we are told that the quake had a magnitude of, say, 3.2 or 5.0. Or 7.0, as was the case yesterday in Haiti. We all understand that 7 is worse than 5, of course, but I fear that few of us really understand or appreciate the degree of that difference.

Popularly (but inacurately) called the Richter scale, today’s seismologists measure an earthquake’s energy according to what is technically called the moment magnitude scale.

The magnitude scale is logarithmic: Each increase in the magnitude number actually represents more than 30 times the magnitude of energy released. A 7.0 magnitude quake is 32 megatons of seismic energy, where a 6.0 is only 1 megaton of seismic energy.

I imagine Mr. Richter and his successors prefer to use such a logarithmic scale because it permits them to communicate a quake’s magnitude extremely efficiently. They’re mathematicians, they intuitively understand that 5.4 is actually 5 times bigger than 5.0.

But for the rest of us, even those of us who are numerically literate, the logarithmic scale isn’t something we use every day. To most of us, 5.4 is only a little more than 5. Indeed, 7 isn’t that much more than 5 for us, either — yet in earthquake terms, 7.0 is one thousand times as destructive as 5.0.

I appreciate that the media thinks enough of us, the public, to report using the same technical jargon that scientists use. But in this case, I suspect they are doing the public — and the victims — a disservice. What if, instead of simply calling Haiti the victim of a “7.0 earthquake”, they called it a “32 megaton earthquake”? Or a “32,000 kiloton earthquake”? This would permit people who understand what a 4.0 earthquake feels like (and lots of people do understand this) understand that Haiti is today the victim of an experience thirty-two thousand times worse.

Some ideas on how to help the victims in Haiti:

  • Help Haiti with your mobile phone: text “HAITI” to 90999 and a $10 donation to the American Red Cross will be charged to your phone bill. It takes no time at all.
  • Nearly half of Haiti is under the age of 18. Make a donation to UNICEF to help.
  • Donate to Doctors Without Borders

There are 17 notes on this page:

  1. As a survivor of a few earthquakes 6.5 and above, and a couple F4 tornadoes in Oklahoma, I can attest that these scales are relatively useless unless you’ve felt the distinction between the numbers. A 5.4 is exciting. A 6.5 can feel like a never-ending apocalypse underfoot. A 7.0 is basically the worst thing you can experience. I can’t even fathom an 8 or a 9.

    So yeah, let’s get rid of the Richter scale.

    And while we’re at it, the Wind Chill scale is pretty useless, too. It means almost nothing.

    January 13th, 2010 | 6:17 pm
  2. Do you really think that the average American would find 32,000 kiloton and 32 megaton any more meaningful?

    January 13th, 2010 | 6:31 pm
  3. @Colin: It’s all about the relative magnitudes. They don’t need to know what a kiloton really means (although 1000 tons of TNT is pretty vivid explanation of it), as long as they always use the same measure (I nominate kilotons) every time, I think it would help.

    January 13th, 2010 | 6:49 pm
  4. I guess a new scale would need to factor in closeness to the surface as well? So then you’d have a unitless scale between 1(truck outside) …. 10 (worst possible)

    January 14th, 2010 | 7:43 am
  5. I disagree, by way of agreeing with Colin. If you haven’t experienced a 32 megaton earthquake, or if you don’t know that the earthquake you experienced was 32 megatons, it doesn’t mean anything to the lay person. “Bigger means more” is as much as we can expect.

    Further, I disagree with your characterization of mass media treatment of us by reporting with the Richter Scale. The news has to answer two questions about earthquakes: “How does that number translate to human experience?” and “What can I do about it?” The choice of scale doesn’t do anything to make answering either of those questions easier. As the figure accompanying this post illustrates, the Richter Scale at least has easily-explainable relationships between the differences in numbers and the differences in qualitative effect.

    The Richter Scale is only “scientific jargon” when news media stop using it. Mass media audiences don’t really understand what a “megaton” is either, only that 32,000 megatons is probably more than 32 megatons. The Richter Scale only means something because it’s in common circulation, because that’s how the news talks about it. You could just as easily call the IQ scale scientific jargon (see below). If a curious newswatcher decides to look into the scientific literature to satisfy their own curiosity and aren’t scared off by such “jargon” as the Richter Scale, that much the better!

    Even if I did agree that mass media should report earthquake magnitude with a linear scale, trying to make it happen is a losing bet. People in earthquake-prone areas already understand the Richter Scale as well as it matters to them, they will not appreciate your efforts, and the media that service them will have to abide. Just look at the intelligence-quotient scale. Modern intelligence test scores aren’t quotients at all; they’re quantile ranks, but we map the percentile scores to a spot on a normal distribution with a mean of 100 and a standard deviation of 15ish anyway because that’s what we’re used to.

    January 14th, 2010 | 6:54 pm
  6. So basically, you’re looking for a linear representation of something that can have an exponential spread?

    January 14th, 2010 | 8:09 pm
  7. @David: Earthquake magnitudes can be measured linearly. The logarithmic representation is a shorthand, that’s all. Earthquakes are not inherently logarithmic or anything. You could just as well represent a city’s population logarithmicly: a city with a population of a thousand could have a “population magnitude” of 4 while a city of a million could have a population magnitude of 6, for example.

    @Jeremy: I think the Richter scale *does* help people understand what an earthquake’s effect is, Not just because we’ve heard of other quakes before and might recall the magnitude numbers, but more saliently (as Clay pointed out) a great many Americans *have experienced earthquakes first hand*. But when a person who has experienced a 4 hears about a far-off land who just experienced a 6, I don’t think the number adequately expresses the destruction, terror, and desperation those who just endured the 6 truly went through. Imagine if the news in Southern California regularly said “we’ve had a 10 megaton earthquake” or “we had a 125 megaton earthquake”, but then this week said that Haiti just experieced “a 150 million megaton earthquake”. No matter how mathematically educated anyone is, I think they will recognize that 150 million is way more than 10, and that the people who just got hit aren’t merely the victims of two or three notches higher on the meter.

    You seem to simultaneously think people *do* understand logarithms and that they *don’t* understand, or barely so, that 32,000 is way more than 32. I don’t get your argument at that level — it seems like you’re saying they should jast use th words “severe”, “strong”, etc, an drop the numbers altogether. IMHO, most folks don’t get big numbers, but the kiloton scale would easily make the jump from the average American tremblor to a true disaster look like something anyone can see - we all grok that millions are waaaay more than hundreds.

    Think of it this way: in real life, we *never* measure anything with a logarithmic scale. We use linear scales for speedometers, populations, weights and measures.

    Instead of arguing about why getting rid if it might not work, maybe try arguing what good it does to you an me to have a logarithmic scale at all? I’m sure there is a reason seismologists use it, but I cannot imagine why we do.

    January 14th, 2010 | 9:14 pm
  8. For the purpose of talking about earthquake magnitudes, numbers express reproducible measurements. Numbers do not express destruction, terror, or desperation; nor should they try. That’s what photos, quotes, and stories of victims are for. (The research of Paul Slovic and associates is tangentially related to this.)

    You’d like earthquake magnitude to be reported in a scale that makes relationships between earthquakes by interval comparison (where the difference between 1 and 2 is the same as the difference between 2 and 3) more obvious, since that’s what many of us assume we’re supposed to do to compare numbers.

    Most people still suck at it, even when you make the inputs easier, and especially when you want to inspire a proportional emotional response. (Again, that’s what photos are for.) For mass media purposes, it’s enough to compare the earthquake in Haiti to local experience on an ordinal scale (where 2-1 doesn’t have to equal 3-2, as long as 3>2>1). Ordinal comparisons in the Richter Scale are just as intuitive as in kilotons, and the Richter Scale is already in the common vocabulary.

    January 14th, 2010 | 11:14 pm
  9. @Jeremy: I can’t agree that numbers can’t or don’t represent suffering. Of course pictures and words can be more vivid, but sheer numbers — of the dead and displaced, of acres ruined, of homes destroyed — have deep resonance in the telling if almost any emotionally challenging story in all of history… this earthquake is a case in point: we knew that this quake was a 7.0 almost a full day before any photos or reports from the ground had come through. Every minute counts in a disaster. For almost half a day, the only news we had at all was that number: 7.0. If the number were more viscerally massive, I suspect some people may have moved just a little bit more quickly and decisively on Tuesday night.

    For what it’s worth, I’m not at all arguing that ‘kiloton’ is a more familar and meaningful term to people than a Richter number. We could measure quakes in tremblions or vibratons or whatever word we make up for it, as long as the scale is linear. As long as a given disaster that is a hundred thousand times worse than another given diaster is assigned a magnitude number a hundred thousand times bigger. If this were the case, we might have been able to intuit at 5:00 on Tuesday night, when the only information we had was “7.0″, that tens of thousands of people might be dead or dying, instead of waiting for the photos and stories to trickle in.

    As for your last paragraph, you are essentially saying the opposite of every word I wrote (you’re saying, I think, that logarithmic and linear scales are to most people basically the same thing, that we should stick with what we have because it’s familiar, even if misunderstood, and that there’s nothing substantively wrong with that; while I say that this is a profound, inexplicable, and entirely unnecessary problem), so naturally I disagree.

    I’m quite surprised at the negative responses to this suggestion. I don’t understand it. I ask again: Why have a logarithmic scale?

    January 15th, 2010 | 2:06 am
  10. …Instead of arguing about why a linear scale might enhance the oh-my-god-that’s-a-big-number effect compared to a logarithmic scale that everyone uses and few fully understand, maybe try arguing what good it does to be talking about the energy output of earthquakes at all? Seismologists do it because that’s part of their subject. News uses it because the seismologists use it (”Because all the cool kids are doing it” is actually a very defensible reason), and because measuring the earthquake in dollars and lives is too slow. Reporters don’t have the resources to count any of the three for themselves, and seismologists are faster than accountants.

    If you really want to use different numbers to describe earthquakes, start by throwing energy out the window (right after the sensationalist comparisons to explosives), and quantify the damage, followed by the surface intensity. The ground I’m standing on moved this many inches, my collapsing house killed my two goldfish and amputated my dog, and it will cost me $X to rebuild — not that audiences will be any less numb to the regional totals so astronomically outside their personal experience. The correlation between the damage and the horizontal displacement or energy doesn’t make either a proxy for the other.

    January 15th, 2010 | 2:25 am
  11. Alternately, you could try the Jack Bauer technique, cruising back and forth on the streets with a megaphone crying “Thousands of lives are at stake!” But I wont be doing that, and it wouldn’t work if I did.

    If the purpose is to report a measurement of what happened, one scale is as good as the next as long as we stick to the same one and thereby compare apples to apples. If the purpose is to advocate action, and to support your case with evidence that appeals to our post-Age of Reason sensibilities, choosing a scale with some shock value might make sense… if big numbers were actually any good at inspiring action. It doesn’t work that way. Paul Slovic says so:

    http://journal.sjdm.org/7303a/jdm7303a.htm

    January 15th, 2010 | 2:47 am
  12. @Jeremy: Thanks for the continued explanation and thoughts, and the Slovic link is fascinating. I agree with you insofar as my suggestion isn’t likely to make a HUGE difference. And I can see your point that perhaps it’s easier to (a) have a society trained to understand that when it comes to earthquakes 7.0 is horrific and 4.0 isn’t too bad than it is to (b) have a society whose eyes don’t glaze over at numbers higher than ten or twenty.

    This is where I would love to do some user testing! :-)

    January 15th, 2010 | 6:31 pm
  13. “7.0 earthquake”, they called it a “32 megaton earthquake”? Or a “32,000 kiloton earthquake”?

    What does one megaton feel like? Or even a 1 ton earthquake?

    that Haiti is today the victim of an experience thirty-two thousand times worse

    How can anyone imagine the definition of 32000 times more?

    A pin weighs 1 gramme, so in this reference, a pin would weigh 32kgs…

    ok, so now thats in your head, an earthquake that wobbles houses and chimneys fall down (a level 4), 32000 times more than that would not just affect haiti, it would destroy the World!

    January 20th, 2010 | 9:00 am
  14. In truth, I don’t think magnitude scales were ever intended to be used to inform the public about how “big” an earthquake is. (Richter devised his system to describe quakes that occur in southern California, with its own unique geology and system of faults) I’ve always thought of magnitude scales as a way for seismologists and geophysicist to keep an accounting of energy released at a location on a fault system.

    Keeping in mind that the magnitude is meant to represent energy released *at the focus,* simply changing to the actual energy released doesn’t solve the problem as it doesn’t account for depth and local geology.

    The Haiti earthquake and the 1989 Loma Prieta quake were fundamentally the same quake, in terms of mechanism and energy released (both approximately Mw=7). Taking into account only the relative depths (13 km in Haiti, 17 km in Santa Cruz) the energy released at the epicenter in Haiti was ~70% greater than in Santa Cruz.

    My suggestion would be to revert to reporting an earthquake’s Intensity based on surface wave amplitude and maximum ground acceleration…

    January 22nd, 2010 | 12:13 am
  15. Thanks for this post it, i was googling for a better understanding of quakes. The analysts describing the 8.8 quake in chile this morning were confusing.

    February 27th, 2010 | 10:35 am
  16. Chris, since you’ve posted this, I’ve referenced it a handful of times in heated conversations with people who were trying to minimize the magnitude of the Haiti earthquake and its after effects (which is just…amazing, and sad), especially when they compare it to smaller earthquakes in North America (which to me, smacks a little too much of Munchausen-by-proxy syndrome). This morning, Chile had an earthquake registering at 8.8…this post will continue to serve me in visualizing and explaining the severity of this event, and help me win a few arguments along the way. So thanks ;)

    February 27th, 2010 | 12:07 pm
  17. Glad to help. I just did the math: Today’s 8.8 in Chile is, by the terms of the moment magnitude scale itself, almost 500 times bigger than the 7.0 in Haiti.

    February 27th, 2010 | 7:23 pm

Add a note