h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

kea: I'm not up on the subject enough to understand the effects of depth on how an earthquake is felt. But, for the record, the Kobe epicenter was 14km deep, which is considered shallow: http://www.tfhrc.gov/pubrds/fall96/p96au17.htm

While I was looking for that info, I came across several references which discussed Kobe's geology and the effect it had on magnifying the amplitude of the energy waves. Apparently Kobe sits on the very edge of a basin of dense rock, filled in with softer sediments. The energy waves travel at higher speed in the denser rock, and slower in the sediments. I'm not sure if I understand the effect of this correctly, but I would hazard to guess that the leading edge of the deeper waves arrives first, and then the surface waves arrive. That difference in timing causes them to add together in a way, making the ground move up and down much more violently at that point.

As an illustration, If you filled your bathtub and made a wave that travelled the length of it, and made another to follow it, you'd see that when the first wave bounces off the far end and collides with the new one, you see one much bigger wave at the moment of collision. That's the sort of thing I'm talking about. If you had the means to make one wave which travelled quicker than the other, you'd see the same effect whe slow one was overtaken. That's the concept.

The articles I found on the subject seemed to have found the same effect in Kobe and in Santa Monica, at the edge of the LA basin, during the 1994 Northridge quake.

The principle of superposition of waves could well explain an increase in intensity. Are you supposing that the wave increases were localised, so that they weren't measured?

The wave increases were localized, and therefore not measured. However, with seismograph data from multiple sources it is possible to calculate local intensities. Basically what you do is input the numbers into a computer system that has all the geological data, and run a simulation. You need quite a lot of computing power to do it. We know how fast waves will travel through various types of rock, and what effects the velocity will have on the properties of the wave. The results of the simulations for Northridge and Kobe are consistent with the data, and explain why certain areas were more heavily damaged than others.

If you're really interested, and are not on dial-up: http://www.psc.edu/research/graphics/gallery/earthquake.html

Apparently a month after the earthquake 400 measurement points were used for a *microtremor measurement* which taken from February 1995 to February 1996 with a *Portable Intelligent Collector*:

http://www.sdr.co.jp/papers/local_site_effect_on_kobe.pdf

They showed a narrow (1-2 km) belt of increased damage that ran East-West, undoubtedly created by local effect wave convergence or *local site effect*. This was undoubtedly the primary cause of the cutting off of transportation, gas, etc. since the belt was across the narrow area described by the Researcher's piece.

Re the depth thing. New Zealand is a series of islands right on top of where the pacific and australasian plates meet. We have alot of shallow quakes. Geologists here often talk about the depth of big quakes as being important. If I get the time I'll see if I can find out more.

Here's how they record recent quakes, which includes the depth as well as the magnitude and distance from the nearest town: http://www.geonet.org.nz/recent_quakes.html



The geology, and wave convergence info is interesting. The earthquake I was in had a major mountain range between the epicenter and any towns. There was alot of landslide damage along the coast between the epicentre and the mountains and within the mountain range itself, but not on the other side of the mountains. I always wondered if the mountains had some kind of buffering effect, but I haven't seen that discussed in NZ.

I'm thinking too, does it depend on what kind of faultline the earthquake happens on?

It does. Different fault lines produce different types of motion, which can have different effects on the surface.

Here's an easy primer, with visualizations: http://earthquake.usgs.gov/image_glossary/fault.html

And apparently the largest earthquakes are generated by thrust faults.

So given the *depth* factor and localised severity due to wave attenuation the Richter scale is irrelevant in deciding the destructive force of the earthquake in regard to Japanese history?

Yes and no. Depth shouldn't be a factor, because seismographs record the effects of an earthquake at the surface, and that's all we're really concerned with. But the basin effect means there was greater force in a localized area than the Richter scale leads us to believe. So the scale gives us a common reference, but doesn't tell the whole story.

As for "in regard to Japanese history," I stand by my original statement. Relativism doesn't apply here. A 6.9 earthquake represents a tremendous release of force, and is never a minor event, especially when it is close to a major urban center. However many stronger earthquakes Japan may have experienced does not change that fact in any way.

Incidentally, earthquake magnitudes calculated by seismographs are not exact, and initial measurements are often updated over the course of the same day. Final determination of magnitude on the Richter scale requires months of study at the epicenter and along the fault line. Several of the sources I found while investigating the basin effect referred to the Kobe quake as a 7.2 (including the one I cited up there in post 21). My experience with California earthquakes tells me that usually the first number is lower, and they are revised upwards. So I looked it up to figure out which is which, and it turns out that 6.9 is the measurement on the moment magnitude scale, meaning it was the initial number coming off the seismographs. Using that number as well as other factors, the JMA assigned the Kobe quake a magnitude of 7.2 on the Richter scale.

http://www.ce.berkeley.edu/Programs/Geoengineering/research/Kobe/Somerville/qnews.html

A 7.2, as I stated earlier, qualifies as 'Major' by USGS guidelines.

So, do we know what kind of fault/earthquake happened at Kobe?

zoomer, I thought magnitude, depth, and distance are *all* important. Is it that you have to know the magnitude and then place it in context of the locale?

>>>
When an earthquake occurs, two things need to be determined quite quickly: the epicentre and the magnitude.

The actual source of an earthquake, called its focus, is usually many kilometres deep in the Earth, where the rock as been strained and has finally reached its breaking point.  The epicentre is the place on the Earth's surface directly above the earthquake's focus.

The magnitude number describes how big the earthquake is, not the sense of how badly you get shaken, but how big the disruption was within the Earth.  If you are near a large earthquake, you will be shaken very strongly.  If you are far enough away, you might not feel it at all.  To calculate magnitude seismologists use a seismograph to measure how much the ground moved, and then they take into account the distance from the focus.  This means that no matter what distance you are from the earthquake, you always get the same answer for the magnitude.

...The intensity of an earthquake, on the other hand, is its destructiveness due to the amount of ground movement at a particular place.  Intensity is measured accotrding to the Modified Meercalli scale which has ten steps for New Zealand earthquakes.  The intensity of an earthquake does not relate directly to the amount of energy release in an earthquake (which dictates magnitude) because intensity decreases with distance from the event, and because natural features such as rock type, soli type and the amount of water in the area, modify its destructiveness from place to place.

...Civil Defence needs to know the location of the earthquake and its magnitude quite quickly in order to judge whether the effects are likely to be very severe.  If the magnitude is small, or if the epicentre is offshore, it is likely no damage has been caused.  But if the magnitude is large, and especially if the epicentre is near a city of any size, many people could have been affected and Civil Defence may need to be involved.
<<<
http://www.mtaspiring.school.nz/Tephra/Thehowwhatandwhereofanearthquake.htm

The author qualified his description as I mentioned in the second post here:

"And the author twice mentioned that fact:

*earthquakes with a magnitude of 6.9 on the Richter scale are common in Japan*

*a minor earthquake...(footnote)..Minor for Japan, anyway.*"

So first off he was not referring to *USGS guidelines* but Japanese statistics and secondly he was using the data he had, not conflicting data you found. Lastly the statements were relative in nature, not universal. Statistically there are more earthquakes in Japan higher than this one _as_it_was_measured_ therefore it was statistically and historically below average.

Consider, if I say I present at a major earthquake as far as Victoria, British Columbia is concerned I would be correct. The earthquake was minor by most standards, hardly a china rattler but it was major in that in recent recorded history it was the largest. That is to say since records have been kept which is a very short time. If I was being universal or absolute I could mention that ancient earthquakes have been detected through geology that were much, much higher. All the data I have at my disposal that is direct and measured since the invention of the Richter scale still shows that I was present at a major earthquake...for Victoria and in fact for the entire province if you don't count quakes off the coast.

Sorry, missing *was*..present.

kea, the point is that the statements the author of this piece made were in regard to measurements. The data he found conflicted the measurements and the damage done, in that there was more damage done than one would expect from a RELATIVELY (for Japan) minor quake. YES depth can be a mitigating factor and YES wave attenuation can be another one. The point was that a country like Japan had a city that was almost destroyed by what was measured as a statistically lesser quake. This was explained by the natural *corridor* that Kobe was in and in fact that corridor could be multipliers for both the depth and *positive* wave overlaps. At the same time the corridor prevented relief since it included a major rail line and supply runs. The micro measurements done after the fact were donated by the very rail company that this quake was devastated by.

Yes I understand that about the entry I thought we were exploring the reasons why magnitude may not reflect actual damage done.

<<*earthquakes with a magnitude of 6.9 on the Richter scale are common in Japan*>>

And allow me to point out, once again, that the 6.9 was not on the Richter scale, but on the moment magnitude scale. The measurement on the Richter scale is 7.2.

<>

Even at 6.9, it would be FAR above average. Tiny earthquakes are happening constantly along fault zones, meaning the "average" is well below 3, which is where people begin noticing them. For reference, I'm sure Japan has more activity than this: http://quake.wr.usgs.gov/recenteqs/Quakes/quakes0.html

<>

Earthquakes, like any natural disaster, are measured in severity by the energy released, damage done, and lives lost. In every one of those categories, the Kobe quake qualifies as major. And in every one of those categories, your Victoria quake was minor. The Kobe quake destroyed lives, and the Victoria quake provided something to talk about over afternoon tea.

You may be confusing Moment Magnitude Scale with Local Magnitude Scale as I did in my first post here. These sources show that 6.9 was the measurement for Richter or it's equivalent LMS:

http://en.wikipedia.org/wiki/Great_Hanshin_earthquake
http://www.zephryus.demon.co.uk/geography/resources/earth/kobe.html


Here's that says it was 6.9:
http://encyclopedia.lockergnome.com/s/b/Great_Hanshin_earthquake


7.1:
http://www.kajima.co.jp/tech/hanshin_earthquake/

7.2:
http://www.japan-101.com/history/hanshin_earthquake.htm
http://www.cnr.colostate.edu/avprojects/98proj/world_volc/web_docs/kobe.html
http://www.factbites.com/topics/Great-Hanshin-earthquake
http://www.jpri.org/publications/occasionalpapers/op2.html
http://gees.usc.edu/GEES/Reports/Report3/japan/KOBE.HTML#Introduction
That one from the Civil Engineering Department of the University of Southern California, and F. Oka, M. Sugito, and A. Yashima, Department of Civil Engineering, Gifu University, Gifu, Japan.
http://www.artsci.wustl.edu/~copeland/kobe.html


7.3:
http://www.kkr.mlit.go.jp/en/topics_hanshin.html
http://www.answers.com/topic/great-hanshin-earthquake

So given that there are multiple methods of measuring earthquakes and the most common is the Richter scale with it's identical equivalent Local Magnitude Scale I'd say that the 6.9 the author quoted was off but not by much. And given that the damage was intensified by depth, wave attenuation, (and now I find the fact that it was very close to a subduction) the damage was much greater than a 6.9 or even a 7.2 magnitude quake would cause. SO, given the below average state of a quake between 6.9 to 7.2 on the LMS scale, a quake which was relatively minor using that scale in regard to Japan (hisorically) created much more damage than one would expect. Which is what he said. I would have been more impressed if he had included the _reason_ that the narrow *corridor* Kobe is situated in was more intensely damaged, but I had to go find out for myself.

Damage:
http://www.shinsai.or.jp/hrc-e/publish/lessons_ghe/lghe00.html
(That's just the preface, the rest is alternately heavy with statistics and fascinating bits.)


This one explains the difference between intensity and magnitude:
http://www.seismo.unr.edu/ftp/pub/louie/class/100/magnitude.html

And this one has some great photos and a map showing how the subduction action *cut off* the peninsula that Kobe is on:
http://www.seismo.unr.edu/ftp/pub/louie/class/100/effects-kobe.html

Ooops. Just remove that *Here's that says it was 6.9:* and nudge that third link up to it's brother and sister.

And once again as you were so kind to point out:

In the Richter scale- *Each whole number is TEN TIMES the amplitude, and THIRTY TIMES the energy level.*

So even if statistically there are fewer quakes over 7.2 on the Richter their incredibly increased magnitude and intensity would tip the scales and make 7.2 *less than average*.

What mathematical formula are you using that tells you that 7.2 is below average?

The formula you pointed out, which logrithmically increases the force of the earthquake. I'm unable to show the dramatic nature of the graph but this may help:

1.0               30 pounds   Large Blast at a Construction Site
1.5              320 pounds
2.0                1 ton      Large Quarry or Mine Blast
2.5              4.6 tons
3.0               29 tons
3.5               73 tons  
 4.0            1,000 tons     Small Nuclear Weapon
4.5            5,100 tons     Average Tornado (total energy)
5.0           32,000 tons
5.5           80,000 tons     Little Skull Mtn., NV Quake, 1992
6.0        1 million tons     Double Spring Flat, NV Quake, 1994
6.5        5 million tons     Northridge, CA Quake, 1994
7.0       32 million tons     Hyogo-Ken Nanbu, Japan Quake, 1995; Largest
                                Thermonuclear Weapon
7.5      160 million tons     Landers, CA Quake, 1992
8.0        1 billion tons     San Francisco, CA Quake, 1906
8.5        5 billion tons     Anchorage, AK Quake, 1964
9.0       32 billion tons     Chilean Quake, 1960
10.0       1 trillion tons      (San-Andreas type fault circling Earth)
12.0     160 trillion tons     (Fault Earth in half through center,
                               OR Earth's daily receipt of solar energy)

As you can see, anything between 7 and 8 is far, far less than 2/3 of the force it takes to split the planet in half. My point was that with such earthquakes as these:
 
http://www.reliefweb.int/rw/rwb.nsf/AllDocsByUNID/bb65e4f4cafe323249256dad00128a93

*Source: Agence France-Presse (AFP)
Date: 26 Sep 2003


Massive 8.0 Richter quake hits northern Japan; 479 people reported hurt
by Shingo Ito

HIROO, Japan, Sept 26 (AFP) - Two powerful earthquakes, one of them measuring 8.0 on the Richter scale, rocked northern Japan on Friday, injuring around 480 people and forcing thousands more to evacuate their homes, officials and news reports said.*

A quake in the 7 range is not unusual in Japan. What other country has 8 and higher quakes?

<< 7.5 160 million tons Landers, CA Quake, 1992>>

That one was fun. We had a 6.7 (or so) in Big Bear that same morning. They were both on different fault lines, so there was no foreshock-aftershock relationship. Just two major quakes within the same geographic area, a few hours apart. Good times.

They were major in force, but not in damage or lives, because luckily, both areas are sparsely populated. The biggest problem was that the Landers quake disrupted a rail line, and a passenger train came along later.

Anyway...

<>

This and your chart do not answer my question. You were talking about averages... well, the average quake is down there below the small nuke area. That's because little ones happen constantly. You can have an earthquake below 3 happen in a particular area 50 times in one day (as that USGS map shows... 72 quakes yesterday, and only one above 3). The whole of Japan experiences an earthquake of 7 or greater every 18 months. The preponderance of smaller quakes drags down the average. 6.9 is FAR above average.

Average annual occurance of quakes worldwide by intensity, again courtesy of the USGS:

Descriptor Magnitude Average Annually
Great 8 and higher 1
Major 7 - 7.9 18
Strong 6 - 6.9 120
Moderate 5 - 5.9 800
Light 4 - 4.9 6,200 (estimated)
Minor 3 - 3.9 49,000 (estimated)
Very Minor < 3.0
Magnitude 2 - 3: about 1,000 per day
Magnitude 1 - 2: about 8,000 per day

Now, if you only want to consider newsmaking earthquakes, which are pretty much always natural disasters of varying degrees, then yes... for catastrophic earthquakes that destroy lives and property, Kobe was fairly minor. But how stupid is that?