Faults and Earthquakes
 

 Faults are approximately planar breaks in the Earth's crust and mantle.  The media often refer to faults as "fault lines", but faults are planar features that extend down into the Earth.  If faults intersect the ground surface (some do, some don't), then the line of intersection is called the fault trace.  A fault is a break between adjoining fault blocks, which have moved relative to each other.  There is no gap between the two blocks, just a planar feature (the fault) where rock has been pulverized and ground up during fault block movement.

 Faults can be very small, only a few square meters in area, or can be enormous, with lengths exceeding 1000 kilometers and surface areas up to 100,000 square kilometers.  Faults do not extend forever; they can die out at their edges, if fault block movement decreases to zero, or they can intersect or join with other faults at their edges.

 Active faults show active movement of one fault block relative to the other fault block.  Geologists say that a fault is active if there has been relative motion across the fault during the last 10,000 years.  The relative motion can be slow and constant (creeping motion), in which case no earthquakes (ground shaking) are produced.  The relative motion can be fast and episodic, in which case earthquakes can be produced.  Most active faults don't exhibit much fault block movement, then without warning, they move.

 Fast motion along a fault produces earthquakes; this is a cause and effect relationship.  The motion of the fault blocks and the shaking of the ground are two different things.  As the two fault blocks slide past each other, vibrations or waves are produced, which radiate out in all directions away from the fault and fault blocks. There are several types of seismic waves or vibrations that are created by fast fault block movement.  One of these types of waves travels along the surface of the Earth; it's these surface waves that people typically feel and that cause damage.

 The more fault block movement, the bigger the resulting ground shaking.  There is no direct relationship between the movement and shaking, but we can look at a couple of pairs of fault movements and earthquakes to see how the two compare.  For example, a fault that moved a centimeter in 10 seconds might produce an earthquake with a Richter magnitude of 4 to 5 or so.  For example, a fault that moved 2 meters in 90 seconds might produce an earthquake with a Richter magnitude of 7 to 8 or so.  These are crude comparisons.

 What about Richter magnitude, what is that?  There are three common measures of "earthquake" size, each of which is independent of the other (they have nothing in common, really, and cannot be compared directly).  The oldest measure of size is the Mercalli scale, based upon the amount of damage and human response; this scale was used before earthquake recording machines became sophisticated.  The Richter scale is based upon the amount of ground shaking as measured on a recorder.  The Richter scale is logarithmic; each Richter number represents a 10 fold increase in the amount of ground shaking.  The moment scale is based upon the amount of fault movement and the size of the broken area on the fault.  The moment scale is also logarithmic.  This scale is the one currently used by scientists, who dumped the Richter scale some time ago.  Scientists still calculate the Richter magnitude but only for consumption by the general public, which knows that a 4 is boring, but an 8 is a whopper.

 You can see why the three scales are not comparable, and why scientists use the moment scale for analyzing faults and earthquakes.  Suppose you have a certain amount of fault movement.  If houses are built poorly, you get a high Mercalli number; the opposite if construction is good.  If the ground is soft and bouncy, you get a high Richter number; the opposite if it's strong and firm.

 What causes the fault blocks to move suddenly and without warning?  This is a crucial question, and a very contentious question.  The traditional explanation is the "sticky hands" model.  Lay your hands out flat, thumbs tucked in, and put the hands side by side; these are the two fault blocks.  The concept is that the fault is sticky, stress builds up, no movement takes places, stress builds and builds until finally the stickiness is overcome, and the hands move relative to each other.  The sticky hands model is almost a 100 years old, and it has its problems.  First, faults aren't sticky; they are well lubricated by pulverized and ground up rock that resembles toothpaste.  Second, when the stresses along active faults have been measured, the faults seem to be stress free.  Third, the sticky hands model predicts that earthquakes are cyclic or periodic, that earthquakes will take place at regular intervals.  The fault moves, releases stress, builds up, moves, releases stress, builds up, etc. in a cyclic or periodic fashion.  But no fault has ever been observed to behave this way; earthquakes don't appear in regular cycles, like tides or the seasons.  Earthquakes seem to be very random and unpredictable in their occurrence.

 There are several non-traditional or new wave explanations of fault behavior, all based on some form of randomness and unpredictability.  The earthquake exercise you will work on focuses on one of these random-based theories, known as self-organized criticality.  Faults are considered to be entities that have organized themselves into a critical state.  The critical state of being is an unpredictable one; a small event may die out and have no great effect, whereas the next small event may amplify and grow and become enormous.  Neither result is predictable.

 An excellent example is given by the sandpile.  Let sand pile up one grain at a time into a cone-shaped pile.  A single sand grain falls on top of the pile; nothing happens.  Another single sand grain falls on top of the pile, and a giant sand slide is created, sweeping away half the sandpile.  The sandpile has organized itself into a critical state, where the results of a small event are completely unpredictable.  Another excellent example is auto accidents on the freeway.  The cars on the freeway are at a critical state.  A driver swerves slightly and nothing happens.  A driver swerves slightly, another driver jerks the wheel to get out of the way, bumps a car, which fishtails into another car, eventually resulting in a 17 car pileup.   A small event may decrease to nothing, or more propagate into an enormous result.  Neither one is predicatable.

 This concept of self-organized criticality has been applied to faults.  The fault is at a critical state.  A tiny amount of movement takes place along the fault.  If the initial movement dies out, then a tiny earthquake is produced.  If the initial movement propagates and grows and increases, a giant quake is produced.  There is no predicting which will happen.  Despite many decades of scientific study trying to predict earthquakes, earthquakes have been found to be unpredictable; all predictors have failed to work.

 For more information about self-organized criticality in theory and in nature, you should check out Per Bak's book, How Nature Works.

go to top of page