Earthquakes

Most earthquakes occur along zones where the Earth's crust is

undergoing deformation. Deformation results from plate tectonic forces
and gravitational forces. The type of deformation that takes place during
an earthquake generally occurs along zones where rocks fracture to
produce faults. Before we can understand earthquakes, we first must
explore deformation of rocks and faulting.
Within the Earth rocks are constantly subjected to forces that tend to
bend, twist, or fracture them. When rocks bend, twist or fracture they are
said to deform or strain (change shape or size). The forces that cause
deformation are referred to as stresses. To understand rock deformation
we must first explore stress and strain.
Stress and Strain
Stress is a force applied over an area. One type of stress that we are all
used to is a uniform stress, called pressure. A uniform stress is where the
forces act equally from all directions. In the Earth the pressure due to the
weight of overlying rocks is a uniform stress and is referred to as
confining stress. If stress is not equal from all directions then the stress
is a differential stress. Three kinds of differential stress occur.
1.
2. Tensional stress (or extensional stress), which stretches rock;
3. Compressional stress, which squeezes rock; and
Shear stress, which result in slippage and translation.


Stages of Deformation
When a rock is subjected to increasing stress it changes its shape,
size or volume. Such a change in shape, size or volume is referred to
as strain. When stress is applied to rock, the rock passes through 3
successive stages of deformation.
Elastic Deformation -- wherein the strain is reversible.
Ductile Deformation -- wherein the strain is irreversible.
Fracture -- irreversible strain wherein the material breaks.
We can divide materials into two classes that depend on their relative
behavior under stress.
Brittle materials have a small to large region of elastic behavior, but
only a small region of ductile behavior before they fracture.
Ductile materials have a small region of elastic behavior and a large
region of ductile behavior before they fracture.

How a material behaves will depend on several factors. Among them
are:

Temperature - At high temperature molecules and their bonds can
stretch and move, thus materials will behave in more ductile
manner. At low Temperature, materials are brittle.
Confining Pressure - At high confining pressure materials are less
likely to fracture because the pressure of the surroundings tends to
hinder the formation of fractures. At low confining stress, material
will be brittle and tend to fracture sooner.
Strain rate -- Strain rate refers to the rate at which the deformation
occurs (strain divided by time). At high strain rates material tends
to fracture. At low strain rates more time is available for individual
atoms to move and therefore ductile behavior is favored.
Composition -- Some minerals, like quartz, olivine, and feldspars are
very brittle. Others, like clay minerals, micas, and calcite are more
ductile This is due to the chemical bond types that hold them
together. Thus, the mineralogical composition of the rock will be a
factor in determining the deformational behavior of the rock.
Another aspect is presence or absence of water. Water appears to
weaken the chemical bonds and forms films around mineral grains
along which slippage can take place. Thus wet rock tends to
behave in ductile manner, while dry rocks tend to behave in brittle
manner.

Evidence of Former Deformation
Evidence of deformation that has occurred in the past is very evident in
crustal rocks. For example, sedimentary layers and lava flows generally
are deposited on a surface parallel to the Earth's surface (nearly
horizontal). Thus, when we see such layers inclined instead of
horizontal, evidence of an episode of deformation is present.
In order to uniquely define the orientation of a planar feature we first
need to define two terms - strike and dip. For an inclined plane the
strike is the compass direction of any horizontal line on the plane. The
dip is the angle between a horizontal plane and the inclined plane,
measured perpendicular to the direction of strike.
In recording strike and dip measurements on a geologic map, a symbol is
used that has a long line oriented parallel to the compass direction of the
strike. A short tick mark is placed in the center of the line on the side to
which the inclined plane dips, and the angle of dip is recorded next to
the strike and dip symbol. For beds with a 90
o
dip (vertical) the short
line crosses the strike line, and for beds with no dip (horizontal) a circle
with a cross inside is used.
Fracture of Brittle Rocks
Joints
Joints are fractures in rock that show no slippage or offset along the
fracture. Joints are usually planar features, so their orientation can
be described as a strike and dip. They form from as a result of
extensional stress acting on brittle rock. Such stresses can be
induced by cooling of rock (volume decreases as temperature
decreases) or by relief of pressure as rock is eroded above thus
removing weight.
Joints are zones of weakness,so their presence is critical when
building anything from dams to highways. For dams, the water
could leak out through the joints leading to dam failure. For
highways the joints may separate and cause rock falls and
landslides.
Faults - Faults occur when brittle rocks fracture and there is an
offset along the fracture. When the offset is small, the displacement
can be easily measured, but sometimes the displacement is so large
that it is difficult to measure.

Types of Faults
Faults can be divided into several different types depending on the
direction of relative displacement. Since faults are planar features,
the concept of strike and dip also applies, and thus the strike and
dip of a fault plane can be measured. One division of faults is
between dip-slip faults, where the displacement is measured along
the dip direction of the fault, and strike-slip faults where the
displacement is horizontal, parallel to the strike of the fault.
Dip Slip Faults - Dip slip faults are faults that have an inclined fault
plane and along which the relative displacement or offset has
occurred along the dip direction. Note that in looking at the
displacement on any fault we don't know which side actually moved
or if both sides moved, all we can determine is the relative sense of
motion.
For any inclined fault plane we define the block above the fault as
the hanging wall block and the block below the fault as the footwall
block.

Normal Faults - are faults that result from horizontal tensional stresses
in brittle rocks and where the hanging-wall block has moved down
relative to the footwall block.
Horsts & Grabens - Due to the tensional stress responsible for normal
faults, they often occur in a series, with adjacent faults dipping in
opposite directions. In such a case the down-dropped blocks form
grabens and the uplifted blocks form horsts. In areas where tensional
stress has recently affected the crust, the grabens may form rift valleys
and the uplifted horst blocks may form linear mountain ranges. The East
African Rift Valley is an example of an area where continental extension
has created such a rift. The basin and range province of the western U.S.
(Nevada, Utah, and Idaho) is also an area that has recently undergone
crustal extension. In the basin and range, the basins are elongated
grabens that now form valleys, and the ranges are uplifted horst blocks.

Reverse Faults - are faults that result from horizontal compressional
stresses in brittle rocks, where the hanging-wall block has moved up
relative the footwall block.


A Thrust
Fault is
a special
case of a
reverse
fault
where
the dip of
the fault
is less
than 45
o
.
Thrust
faults can
have
considera
ble
displace
ment,
measurin
g
hundreds
of
kilometer
s, and
can result
in older
strata
overlying
younger
strata.

Strike Slip Faults - are faults where the relative motion on
the fault has taken place along a horizontal direction. Such
faults result from shear stresses acting in the crust. Strike slip
faults can be of two varieties, depending on the sense of
displacement. To an observer standing on one side of the
fault and looking across the fault, if the block on the other
side has moved to the left, we say that the fault is a left-
lateral strike-slip fault. If the block on the other side has
moved to the right, we say that the fault is a right-lateral
strike-slip fault. The famous San Andreas Fault in California
is an example of a right-lateral strike-slip fault.
Displacements on the San Andreas fault are estimated at over
600 km.


Transform-Faults are a special class of strike-slip faults. These are
plate boundaries along which two plates slide past one another in a
horizontal manner. The most common type of transform faults occur
where oceanic ridges are offset. Note that the transform fault only
occurs between the two segments of the ridge. Outside of this area there
is no relative movement because blocks are moving in the same
direction. These areas are called fracture zones. The San Andreas fault
in California is also a transform fault.
Blind Faults - are faults that does not break the surface - rocks above
the fault have behaved in ductile fashion and folded over the tip of the
fault.
Active Faults- An active fault is one that has shown recent displacement
and likely has the potential to produce earthquakes. Since faulting is part
of the deformation process, ancient faults can be found anywhere that
deformation has taken place in the past. Thus, not every fault one sees is
necessarily an active fault.
Surface Expression of Faults - Where faults have broken the surface
they are shown on maps as fault lines or fault zones. Recent ruptures of
dip slip faults at the surface show a cliff that is called a fault scarp.
Strike slip faults result in features like linear valleys, offset surface
features (roads, stream channels, fences, etc.) or elongated ridges
How Faults Develop - When tectonic forces generate stress, rocks start
to deform elastically. Eventually small cracks to form along the fault
zone. When rupture occurs, the stored elastic energy is released as
seismic waves.
Earthquakes
Earthquakes occur when energy stored in elastically strained rocks is
suddenly released. This release of energy causes intense ground shaking
in the area near the source of the earthquake and sends waves of elastic
energy, called seismic waves, throughout the Earth. Earthquakes can be
generated by bomb blasts, volcanic eruptions, and sudden slippage
along faults. Earthquakes are definitely a geologic hazard for those
living in earthquake prone areas, but the seismic waves generated by
earthquakes are invaluable for studying the interior of the Earth.
Origin of Earthquakes
Most natural earthquakes are caused by sudden slippage along a
fault zone. The elastic rebound theory suggests that if slippage along
a fault is hindered such that elastic strain energy builds up in the
deforming rocks on either side of the fault, when the slippage does
occur, the energy released causes an earthquake.
This theory was discovered by making measurements at a number of
points across a fault. Prior to an earthquake it was noted that the rocks
adjacent to the fault were bending. These bends disappeared after an
earthquake suggesting that the energy stored in bending the rocks was
suddenly released during the earthquake.
Friction between the blocks then keeps the fault from moving again until
enough strain has accumulated to overcome the friction and generate
another earthquake. Once a fault forms, it becomes a zone of weakness -
so long as the tectonic stresses continue to be present more earthquakes
are likely to occur on the fault. Thus faults move in spurts and this
behavior is referred to as Stick Slip.
If there is large displacement during an earthquake, a large earthquake
will be generated. Smaller displacements generate smaller
earthquakes. Note that even for small displacements of only a millimeter
per year, after 1 million years, the fault will accumulate 1 km of
displacement.
Fault Creep - Some faults or parts of faults move continuously without
generating earthquakes. This could occur if there is little friction on the
fault & tectonic stresses are large enough to move the blocks in opposite
directions. This is called fault creep. If creep is occurring on one part of
a fault, it is likely causing strain to build on other parts of the fault.
Seismology, The Study of Earthquakes
When an earthquake occurs, the elastic energy is released sending out
vibrations that travel throughout the Earth. These vibrations are called
seismic waves. The study of how seismic waves behave in the Earth is
called seismology.


The source of an
earthquake is
called the
focus, which is
an exact
location within
the Earth were
seismic waves
are generated
by sudden
release of
stored elastic
energy. The
epicenter is
the point on
the surface of
the Earth
directly above
the focus.
Sometimes the
media get
these two
terms
confused.

Seismic waves emanating from the focus can travel in several ways,
and thus there are several different kinds of seismic waves.

Body Waves - emanate from the focus and travel in all directions
through the body of the Earth. There are two types of body waves: P
-waves and S-waves:
P - waves - are Primary waves. They travel with a
velocity that depends on the elastic properties of the
rock through which they travel.

Where, V
p
is the velocity of the P-wave, K is the incompressibility
of the material, is the rigidity of the material, and is the density
of the material.
P-waves are the same thing as sound waves. They move through the
material by compressing it, but after it has been compressed it
expands, so that the wave moves by compressing and expanding
the material as it travels. Thus the velocity of the P-wave depends
on how easily the material can be compressed (the
incompressibility), how rigid the material is (the rigidity), and the
density of the material. P-waves have the highest velocity of all
seismic waves and thus will reach all seismographs first.
S-Waves - Secondary waves, also called shear waves. They travel with a
velocity that depends only on the rigidity and density of the material
through which they travel:
S-waves travel through material by shearing it or changing its shape in
the direction perpendicular to the direction of travel. The resistance to
shearing of a material is the property called the rigidity. It is notable that
liquids have no rigidity, so that the velocity of an S-wave is zero in a
liquid. (This point will become important later). Note that S-waves
travel slower than P-waves, so they will reach a seismograph after the P-
wave.

Surface Waves - Surface waves differ from body waves in
that they do not travel through the Earth, but instead travel
along paths nearly parallel to the surface of the Earth.
Surface waves behave like S-waves in that they cause up and
down and side to side movement as they pass, but they travel
slower than S-waves and do not travel through the body of
the Earth. Surface waves are often the cause of the most
intense ground motion during an earthquake.
Seismometer
s - Seismic
waves
travel
through the
Earth as
vibrations.
A
seismomete
r is an
instrument
used to
record these
vibrations,
and the
resulting
graph that
shows the
vibrations is
called a
seismogra
m. The
seismomete
r must be
able to
move with
the
vibrations,
yet part of it
must remain
nearly
stationary.
This is accomplished by isolating the recording device (like a pen)
from the rest of the Earth using the principal of inertia. For
example, if the pen is attached to a large mass suspended by a
wire, the large mass moves less than the paper which is attached to
the Earth, and on which the record of the vibrations is made.
Modern instruments are digital and dont require the paper.

The record of an earthquake, a seismogram, as recorded by a
seismometer, will be a plot of vibrations versus time. On the
seismograph, time is marked at regular intervals, so that we can
determine the time of arrival of the first P-wave and the time of
arrival of the first S-wave. (Note again, that because P-waves have
a higher velocity than S-waves, the P-waves arrive at the
seismographic station before the S-waves)
Locating the
Epicenters of
Earthquakes -
To determine
the location of
an earthquake
epicenter, we
need to have
recorded a
seismograph
of the
earthquake
from at least
three
seismographic
stations at
different
distances from
the epicenter.
In addition, we
need one
further piece
of information
- that is the
time it takes
for P-waves
and S-waves to
travel through
the Earth and
arrive at a
seismographic
station. Such
information
has been
collected over
the last 80 or
Thus the S-P interval tells us the distance to the epicenter from the
seismographic station where the earthquake was recorded. Thus at each
station we can draw a circle on a map that has a radius equal to the
distance from the epicenter. Three such circles will intersect in a point
that locates the epicenter of the earthquake.


Magnitude of Earthquakes - Whenever a large destructive
earthquake occurs in the world the press immediately wants to
know where the earthquake occurred and how big the earthquake
was (in California the question is usually - Was this the Big One?).
The size of an earthquake is usually given in terms of a scale
called the Richter Magnitude. Richter Magnitude is a scale of
earthquake size developed by a seismologist named Charles
Richter. The Richter Magnitude involves measuring the amplitude
(height) of the largest recorded wave at a specific distance from
the earthquake. While it is correct to say that for each increase in 1
in the Richter Magnitude, there is a tenfold increase in amplitude
of the wave, it is incorrect to say that each increase of 1 in Richter
Magnitude represents a tenfold increase in the size of the
Earthquake (as is commonly incorrectly stated by the press).
A better measure of the size of an earthquake is the amount
of energy released by the earthquake. While this is much
more difficult to determine, Richter gave a means by which
the amount of energy released can be estimated:
Log E = 11.8 + 1.5 M
Where Log refers to the logarithm to the base 10, E is the energy
released in ergs, and M is the Richter Magnitude.
Anyone with a hand calculator can solve this equation by
plugging in various values of M (magnitude) and solving for
E, the energy released. I've done the calculation for you in
the following table:

Magnitude Energy (ergs) Factor
1 2.0 x 10
13
31 x
2 6.3 x 10
14
3 2.0 x 10
16
31 x
4 6.3 x 10
17
5 2.0 x 10
19
31 x
6 6.3 x 10
20
7 2.0 x 10
22
31 x
8 6.3 x 10
23

From these calculations you can see that each increase in 1 in
Magnitude represents a 31 fold increase in the amount of energy
released. Thus, a magnitude 7 earthquake releases 31 times more
energy than a magnitude 6 earthquake. A magnitude 8 earthquake
releases 31 x 31 or 961 times as much energy as a magnitude 6
earthquake.
Although the Richter Magnitude is the scale most commonly
reported when referring to the size of an earthquake, it has been
found that for larger earthquakes a more accurate measurement of
size is the moment magnitude, Mw. The moment magnitude is a
measure of the amount of strain energy released by the earthquake
as determined by measurements of the shear strength of the rock
and the area of the rupture surface that slipped during the
earthquake.
Note that it usually takes more than one seismographic station to
calculate the magnitude of an earthquake. Thus you will hear
initial estimates of earthquake magnitude immediately after an
earthquake and a final assigned magnitude for the same earthquake
that may differ from initial estimates, but is assigned after
seismologists have had time to evaluate the data from numerous
seismographic stations.
The moment magnitude for large earthquakes is usually greater than
the Richter magnitude for the same earthquake. For example the
Richter magnitude for the 1964 Alaska earthquake is usually
reported as 8.6, whereas the moment magnitude for this
earthquake is calculated at 9.2.
The largest earthquake ever recorded was in Chile in 1960 with a
moment magnitude of 9.5, The Summatra earthquake of 2004 had
a moment magnitude of 9.0. Sometimes a magnitude is reported
for an earthquake and no specification is given as to which
magnitude (Richter or moment) is reported. This obviously can
cause confusion. But, within the last few years, the tendency has
been to report the moment magnitude rather than the Richter
magnitude.
The Hiroshima atomic bomb released an amount of energy
equivalent to a moment magnitude 6 earthquake.
Note that magnitude scales are open ended with no maximum or
minimum. The largest earthquakes are probably limited by rock
strength. Meteorite impacts could cause larger earthquakes than
have ever been observed.

Frequency of Earthquakes of Different Magnitude
Worldwide
Magnitude Number of
Earthquakes per
Year
Description
> 8.5 0.3 Great
8.0 - 8.4 1
7.5 - 7.9 3 Major
7.0 - 7.4 15
6.6 - 6.9 56
6.0 - 6.5 210 Destructive
5.0 - 5.9 800 Damaging
4.0 - 4.9 6,200 Minor
3.0 - 3.9 49,000
2.0 - 2.9 300,000
0 - 1.9 700,000

Modified Mercalli Intensity Scale
Note that the Richter magnitude scale results in one number for the size
of the earthquake. Maximum ground shaking will occur only in the area
of the epicenter of the earthquake, but the earthquake may be felt over a
much larger area. The Modified Mercalli Scale was developed in the
late 1800s to assess the intensity of ground shaking and building
damage over large areas.
The scale is applied after the earthquake by conducting surveys of
people's response to the intensity of ground shaking and
destruction.
Thus, a given earthquake will have zones of different intensity all
surrounding a zone of maximum intensity.
The Modified Mercalli Scale is shown in the table below. Note that
correspondence between maximum intensity and Richter Scale
magnitude only applies in the area around the epicenter.
Thus, a given earthquake will have zones of different intensity all
surrounding a zone of maximum intensity.
The Modified Mercalli Scale is shown in the table below. Note that
correspondence between maximum intensity and Richter Scale
magnitude only applies in the area around the epicenter.
Intensity Characteristic
Effects
Richter Scale
Equivalent
I People do not feel any
Earth movement
<3.4
II A few people notice
movement if at rest
and/or on upper floors
of tall buildings

III People indoors feel
movement. Hanging
objects swing back and
forth. People outdoors
might not realize that
an earthquake is
occurring
4.2
IV People indoors feel
movement. Hanging
objects swing. Dishes,
windows, and doors
rattle. Feels like a
heavy truck hitting
walls. Some people
outdoors may feel
movement. Parked cars
rock.
4.3 - 4.8
V Almost everyone feels
movement. Sleeping
people are awakened.
Doors swing
open/close. Dishes
break. Small objects
move or are turned
over. Trees shake.
Liquids spill from open
containers
4.9-5.4
VI Everyone feels
movement. People
have trouble walking.
Objects fall from
shelves. Pictures fall
off walls. Furniture
moves. Plaster in walls
may crack. Trees and
bushes shake. Damage
slight in poorly built
buildings.
5.5 - 6.1
VII People have difficulty
standing. Drivers feel
cars shaking. Furniture
breaks. Loose bricks
fall from buildings.
Damage slight to
moderate in well-built
buildings; considerable
in poorly built
buildings.
5.5 - 6.1
VIII Drivers have trouble
steering. Houses not
bolted down shift on
foundations. Towers &
chimneys twist and fall.
Well-built buildings
suffer slight damage.
Poorly built structures
severely damaged. Tree
branches break.
Hillsides crack if
ground is wet. Water
levels in wells change.
6.2 - 6.9
IX Well-built buildings
suffer considerable
damage. Houses not
bolted down move off
foundations. Some
underground pipes
broken. Ground cracks.
Serious damage to
Reservoirs.
6.2 - 6.9
X Most buildings & their
foundations destroyed.
Some bridges
destroyed. Dams
damaged. Large
landslides occur. Water
thrown on the banks of
canals, rivers, lakes.
Ground cracks in large
areas. Railroad tracks
bent slightly.
7.0 - 7.3
XI Most buildings
collapse. Some bridges
destroyed. Large cracks
appear in the ground.
Underground pipelines
destroyed. Railroad
tracks badly bent.
7.4 - 7.9
XII Almost everything is
destroyed. Objects
thrown into the air.
Ground moves in
waves or ripples. Large
amounts of rock may
move.
>8.0
The Mercalli Scale is very useful in examining the effects of an
earthquake over a large area, because it will is responsive not only
to the size of the earthquake as measured by the Richter scale for
areas near the epicenter, but will also show the effects of the
efficiency that seismic waves are transmitted through different
types of material near the Earth's surface.
The Mercalli Scale is also useful for determining the size of
earthquakes that occurred before the modern seismographic
network was available (before there were seismographic stations, it
was not possible to assign a Richter Magnitude).

Earthquake Risk
Many seismologists have said that "earthquakes don't kill people,
buildings do". This is because most deaths from earthquakes are caused
by buildings or other human construction falling down during an
earthquake.
Earthquakes located in isolated areas far from human population rarely
cause any deaths.
Thus, earthquake hazard risk depends on

Population density
Construction standards (building codes)
Emergency preparedness
Examples:
Worst earthquake in recorded history occurred in 1556 in Shaaxi,
China. Killed 830,000 people, most living in caves excavated in poorly
consolidated loess (wind deposited silt and clay).
Worst earthquake in the 20th century also occurred in China (T'ang Shan
Province), killed 240,000 in 1976. Occurred at 3:42 AM, Magnitude 7.8
Earthquake and magnitude 7.1 aftershock. Deaths were due to collapse
of masonry (brick) buildings.
Worst earthquake so far in the 21st Century was a magnitude 7.0
earthquake that occurred in Haiti on January 12, 2010 with an estimated
death toll of 230,000! (The death toll in this earthquake is still
debated.The Hatian government claims 316, 000 deaths, while U.S.
estimates suggest something between 46,000 and 86,000).
Contrast - In earthquake prone areas like California, in order to reduce
earthquake risk, there are strict building codes requiring the design and
construction of buildings and other structures that will withstand a large
earthquake. While this program is not always completely successful, one
fact stands out to prove its effectiveness. In 1989 an earthquake near
San Francisco, California (The Loma Prieta, or World Series
Earthquake) with a Moment Magnitude of 6.9 killed about 62 people.
Most were killed when a double decked freeway in Oakland collapsed.
About 10 months earlier, an earthquake with moment magnitude 6.8
occurred in Armenia, where no earthquake-proof building codes existed.
The death toll in the latter earthquake was about 25,000!
Similarly the Moment Magnitude 7.0 2010 earthquake in Haiti had a
huge death toll mainly because of the lack of earthquake-resistent
structures. Most buildings were made of poorly reinforced concrete.

Computer simulations for large cities, like San Francisco or Los
Angeles, California, indicate that a magnitude >8.0 earthquake would
cause between 3,000 and 13,000 deaths.
3,000 if at night, when populace is asleep in wood frame houses
13,000 if during day when populace is in masonry buildings and on
freeways.


Architecture and Building Codes
While architecture and building codes can reduce risk, it should be noted
that not all kinds of behavior can be predicted.
Although codes are refined each year, not all possible effects can be
anticipated. For example different earthquakes show different
frequencies of ground shaking, different durations of ground shaking,
and different vertical and horizontal ground accelerations.
Old buildings cannot cost-effectively be brought up to code, especially
with yearly refinements to code.
Even with construction to earthquake code, buildings fail for other
reasons, like poor quality materials, poor workmanship, etc. that are not
discovered until after an earthquake.

Hazards Associated with Earthquakes
Possible hazards from earthquakes can be classified as follows:
Ground Motion - Shaking of the ground caused by the passage of
seismic waves, especially surface waves, near the epicenter of the
earthquake are responsible for the most damage during an earthquake
and is thus a primary effect of an earthquake. The intensity of ground
shaking depends on:
Local geologic conditions in the area. In general, loose unconsolidated
sediment is subject to more intense shaking than solid bedrock.
Size of the Earthquake. In general, the larger the earthquake, the more
intense is the shaking and the duration of the shaking.
Distance from the Epicenter. Shaking is most severe near the epicenter
and drops off away from the epicenter. The distance factor depends on
the type of material underlying the area. There are, however, strange
exceptions. For example, the 1985 Mexico City Earthquake (magnitude
8.1) had an epicenter on the coast of Mexico, more than 350 km to the
south, yet damage in Mexico City was substantial because Mexico City
is built on soft unconsolidated sediments that fill a former lake (see
Liquefaction, below).
Damage to structures from shaking depends on the type of construction.
Concrete and masonry structures are brittle and thus more susceptible to
damage
wood and steel structures are more flexible and thus less susceptible to
damage.
Faulting and Ground Rupture - Ground rupture generally occurs only
along the fault zone that moves during the earthquake, and are thus a
primary effect. Thus structures that are built across fault zones may
collapse, whereas structures built adjacent to, but not crossing the fault
may survive.
Aftershocks - These are smaller earthquakes that occur after a main
earthquake, and in most cases there are many of these (1260 were
measured after the 1964 Alaskan Earthquake). Aftershocks occur
because the main earthquake changes the stress pattern in areas around
the epicenter, and the crust must adjust to these changes. Aftershocks
are very dangerous because they cause further collapse of structures
damaged by the main shock. Aftershocks are a secondary effect of
earthquakes

Fire - Fire is a secondary effect of earthquakes. Because power lines
may be knocked down and because natural gas lines may rupture due to
an earthquake, fires are often started closely following an earthquake.
The problem is compounded if water lines are also broken during the
earthquake since there will not be a supply of water to extinguish the
fires once they have started. In the 1906 earthquake in San Francisco
more than 90% of the damage to buildings was caused by fire.

Landslides - In mountainous regions subjected to earthquakes ground
shaking may trigger landslides, rock and debris falls, rock and debris
slides, slumps, and debris avalanches. These are secondary effects.


Liquefaction - Liquefaction is a processes that occurs in water-saturated
unconsolidated sediment due to shaking. In areas underlain by such
material, the ground shaking causes the grains to lose grain to grain
contact, and thus the material tends to flow.
Liquefaction, because it is a direct result of ground shaking, is a primary
effect.
You can demonstrate this process to yourself next time your go the
beach. Stand on the sand just after an incoming wave has passed. The
sand will easily support your weight and you will not sink very deeply
into the sand if you stand still. But, if you start to shake your body while
standing on this wet sand, you will notice that the sand begins to flow as
a result of liquefaction, and your feet will sink deeper into the sand.



Changes in Ground Level - A secondary or tertiary effect that is caused
by faulting. Earthquakes may cause both uplift and subsidence of the
land surface. During the 1964 Alaskan Earthquake, some areas were
uplifted up to 11.5 meters, while other areas subsided up to 2.3 meters.

Tsunami - Tsunami a secondary effect that are giant ocean waves that
can rapidly travel across oceans, as will be discussed in more detail later.
Earthquakes that occur beneath sea level and along coastal areas can
generate tsunami, which can cause damage thousands of kilometers
away on the other side of the ocean.

Flooding - Flooding is a secondary effect that may occur due to rupture
of human made dams and levees, due to tsunami, and as a result of
ground subsidence after an earthquake.
World Distribution of Earthquakes
The distribution of earthquakes is called seismicity. Seismicity is
highest along relatively narrow belts that coincide with plate boundaries.
This makes sense, since plate boundaries are zones along which
lithospheric plates move relative to one another.
Earthquakes along these zones can be divided into shallow focus
earthquakes that have focal depths less than about 100 km and deep
focus earthquakes that have focal depths between 100 and 700 km.

Earthquakes at Diverging Plate Boundaries. Diverging plate boundaries
are zones where two plates move away from each other, such as at
oceanic ridges. In such areas the lithosphere is in a state of tensional
stress and thus normal faults and rift valleys occur. Earthquakes that
occur along such boundaries show normal fault motion, have low
Richter magnitudes, and tend to be shallow focus earthquakes with focal
depths less than about 20 km. Such shallow focal depths indicate that the
brittle lithosphere must be relatively thin along these diverging plate
boundaries.

Examples - all oceanic ridges, Mid-Atlantic Ridge, East Pacific rise, and
continental rift valleys such as the basin and range province of the
western U.S. & the East African Rift Valley.
Earthquakes at Transform Fault Boundaries. Transform fault boundaries
are plate boundaries where lithospheric plates slide past one another in a
horizontal fashion. The San Andreas Fault of California is one of the
longer transform fault boundaries known. Earthquakes along these
boundaries show strike-slip motion on the faults and tend to be shallow
focus earthquakes with depths usually less than about 100 km. Richter
magnitudes can be large.
Examples - San Andreas Fault, California, South Island of New Zealand,
through Port Au Prince, Haiti.
Earthquakes at Convergent Plate Boundaries. Convergent plate
boundaries are boundaries where two plates run into each other. Thus,
they tend to be zones where compressional stresses are active and thus
reverse faults or thrust faults are common. There are two types of
converging plate boundaries. (1) subduction boundaries, where oceanic
lithosphere is pushed beneath either oceanic or continental lithosphere;
and (2) collision boundaries where two plates with continental
lithosphere collide.

Subduction boundaries -At subduction boundaries cold oceanic
lithosphere is pushed back down into the mantle where two plates
converge at an oceanic trench. Because the subducted lithosphere is cold
it remains brittle as it descends and thus can fracture under the
compressional stress.


When it fractures, it generates earthquakes that define a zone of
earthquakes with increasing focal depths beneath the overriding plate.
This zone of earthquakes is called the Benioff Zone. Focal depths of
earthquakes in the Benioff Zone can reach down to 700 km.

Examples - Along coasts of South American, Central America, Mexico,
Northwestern U.S., Alaska, Japan, Philippines, Caribbean Islands.
Collision boundaries - At collisional boundaries two plates of
continental lithosphere collide resulting in fold-thrust mountain belts.
Earthquakes occur due to the thrust faulting and range in depth from
shallow to about 200 km.
Examples - Along the Himalayan Belt into China, along the Northern
edge of the Mediterranean Sea through Black Sea and Caspian Sea into
Iraq and Iran.
Intraplate Earthquakes - These are earthquakes that occur in the stable
portions of continents that are not near plate boundaries. Many of them
occur as a result of re-activation of ancient faults, although the causes of
some intraplate earthquakes are not well understood.
Examples - New Madrid Region, Central U.S., Charleston South
Carolina, Along St. Lawrence River - U.S. - Canada Border.
Seismic Hazard and Risk Mapping
The risk that an earthquake will occur close to where you live depends
on whether or not tectonic activity that causes deformation is occurring
within the crust of that area.
Primary Effects of Volcanism
Lava Flows
Lava flows are common in Hawaiian and Strombolian type of eruptions,
the least explosive.
Although lava flows have been known to travel as fast as 64 km/hr, most
are slower and give people time to move out of the way.
Thus, in general, lava flows are most damaging to property, as they can
destroy anything in their path.
Control of lava flows has been attempted with limited success by
bombing flow fronts to attempt to divert the flow, and by spraying with
water to cool the flow. The latter is credited with saving the fishing
harbor during a 1973 eruption of Heimaey in Iceland.


Violent Eruptions and Pyroclastic Activity
Pyroclastic activity is one of the most dangerous aspects of volcanism.
Hot pyroclastic flows cause death by suffocation and burning. They can
travel so rapidly that few humans can escape.
Lateral blasts knock down anything in their path, can drive flying debris
through trees.
Ash falls can cause the collapse of roofs and can affect areas far from
the eruption. Althoughash falls blanket an area like snow, they are far
more destructive because tephra deposits have a density more than twice
that of snow and tephra deposits do not melt like snow.
Ash falls destroy vegetation, including crops, and can kill livestock that
eat the ash covered vegetation.
Ash falls can cause loss of agricultural activity for years after an
eruption, a secondary or tertiary effect.

Poisonous Gas Emissions
Volcanoes emit gases that are often poisonous to living organisms.
Among these poisonous gases are: Hydrogen Chloride (HCl), Hydrogen
Sulfide (H2S), Hydrogen Fluoride (HF), and Carbon Dioxide (CO2).
The Chlorine, Sulfur. and Fluorine gases can kill organisms by direct
ingestion, or by absorption onto plants followed by ingestion by
organisms.
In 1984, CO2 gas escaping from the bottom of Lake Monoun, a crater
lake in the African country of Cameroon, killed 37 people.


In 1986 an even larger CO2 gas emission from Lake Nyos in Cameroon
killed more than 1700 people and 3000 cattle.



Secondary and Tertiary Effects of Volcanism
Mudflows (Lahars)
Volcanoes can emit voluminous quantities of loose, unconsolidated
tephra which become deposited on the landscape. Such loose deposits
are subject to rapid removal if they are exposed to a source of water.
The source of water can be derived by melting of snow or ice during the
eruption, emptying of crater lakes during an eruption, or rainfall that
takes place any time with no eruption.
Thus, mudflows can both accompany an eruption and occur many years
after an eruption.
Mudflows are a mixture of water and sediment, they move rapidly down
slope along existing stream valleys, although they may easily top banks
and flood out into surrounding areas.
They have properties that vary between thick water and wet concrete,
and can remove anything in their paths like bridges, highways, houses,
etc.
During the Mt. St. Helens eruption of May 18, 1980, mudflows were
generated as a result of snow melt on the volcano itself, and deposition
of tephra in streams surrounding the mountain.

On November 13, 1985 a mudflow generated by a small eruption on
Nevado del Ruiz volcano in Columbia flowed down slope and
devastated the town of Armero, 50 km east of the volcano and built on
prior mudflow deposits. The town had several hours of warning from
villages higher up slope, but these warnings were ignored, and 23,000
people died in the mudflow that engulfed the town.

Debris Avalanches and Debris Flows
Volcanic mountains tend to become oversteepened as a result of the
addition of new material over time as well due to inflation of the
mountain as magma intrudes.
Oversteepened slopes may become gravitationally unstable, leading to a
sudden slope failure that results in landslides, debris slides or debris
avalanches. We will cover these types of hazards in more detail later in
the course and in the next lecture.

During the May 18, 1980 eruption of Mt. St. Helens, Washington, a
debris avalanche was triggered by a magnitude 5.0 earthquake. The
avalanche removed the upper 500 m of the mountain, and flowed into
the Spirit Lake, raising its level about 40 m. It then moved to the west
filling the upper reaches of the North Fork of the Toutle River valley
(see map above).
Debris avalanches, landslides, and debris flows do not necessarily occur
accompanied by a volcanic eruption. There are documented cases of
such occurrences where no new magma has been erupted.
Flooding
Drainage systems can become blocked by deposition of pyroclastic
flows and lava flows. Such blockage may create a temporary dam that
could eventually fill with water and fail resulting in floods downstream
from the natural dam.
Volcanoes in cold climates can melt snow and glacial ice, rapidly
releasing water into the drainage system and possibly causing floods.
Jokaulhlaups occur when heating of a glacier results in rapid outburst of
water from the melting glacier.

Tsunami
Debris avalanche events, landslides, caldera collapse events, and
pyroclastic flows entering a body of water may generate tsunami.
During the 1883 eruption of Krakatau volcano, in the straits of Sunda
between Java and Sumatra, several tsunami were generated by
pyroclastic flows entering the sea and by collapse accompanying caldera
formation. The tsunami killed about 36,400 people, some as far away
from the volcano as 200 km.
Volcanic Earthquakes and Tremors
Earthquakes usually precede and accompany volcanic eruptions, as
magma intrudes and moves within the volcano.
Although most volcanic earthquakes are small, some are large enough to
cause damage in the area immediately surrounding the volcano, and
some are large enough to trigger landslides and debris avalanches, such
as in the case of Mount St. Helens.
Volcanic Tremor (also called harmonic tremor) is a type of continuous
rhythmic shaking of the ground that is generated by magma moving
underground.
Atmospheric Effects
Since large quantities of ashand volcanic gases can be injected into the
atmosphere, volcanism can have a short-term effect on climate.
Volcanic ash can cause reflection of solar radiation, and thus can cause
the temperatures to be cooler for several years after a large eruption.
The 1815 eruption of Tambora volcano in Indonesia, was the largest in
recorded history. The year following the Tambora eruption (1816) was
called the "year without summer". Snow fell in New England in July.
Volcanic gases like SO2 also reflect solar radiation. Eruptions in 1981
at El Chichn Volcano, Mexico, and 1991 at Pinatubo, Philippines,
ejected large quantities of SO2 into the atmosphere. The effects of the
El Chichn eruption were masked by a strong El Nio in the year
following the eruption, but Pinatubo caused a lowering of average
temperature by about 1oC for two years following the eruption.
Volcanic gases like CO2 are greenhouse gases which help keep heat in
the atmosphere. During the mid-Cretaceous (about 90 to 120 million
years ago) the CO2 content of the atmosphere was about 15 times
higher than present. This is thought to have been caused by voluminous
eruptions of basaltic magma on the sea floor. Average temperatures
were likewise about 10 to 12oC warmer than present.
Famine and Disease
As noted above, tephra falls can cause extensive crop damage and kill
livestock. This can lead to famine.
Displacement of human populations, breakdown of sewerage and water
systems, cut off of other normal services can lead to disease for years
after an eruption, especially if the infrastructure is not in place to provide
for rapid relief and recovery.
Volcanic Fatalaties
Over the last 500 years, volcanoes have directly or indirectly been
responsible for over 275,000 deaths. The greatest killers have been
pyroclastic flows, tsunami, lahars, and famine.
Beneficial Aspects of Volcanism
Since this course concentrates on the damaging effects of volcanism, we
won't spend too much time on the topic of the beneficial aspects of
volcanism. We note here, that volcanism throughout Earth history is
responsible for outgasing of the Earth to help produce both the
atmosphere and hydrosphere. Volcanism helps renew the soil, and soils
around active volcanoes are some the richest on Earth. Hydrothermal
processes associated with volcanism produce rich ore deposits, and the
heat rising around magma bodies can sometimes be tapped to produce
geothermal energy.
Mitgation of Volcanic Disasters
The best mitigation against casualties from volcanic eruptions is to
provide warning based on eruption forecasts and knowledge of the past
behavior of the volcano, and call for evacuations. Little can be done to
protect property as the energy involved in volcanic eruptions is too great
and few structures will survive if subjected to volcanic processes.
As volcanic ash in the atmosphere has been known to cause problems
with airplanes, a system currently exists to keep aircraft out of ash
clouds.
This can have severe economic consequences as evidenced by the near
shutdown of European airports during the 2010 eruption of a volcano in
Iceland.
Because evacuation plans rely on knowledge of when the volcano might
erupt and how it will behave when it does erupt, we will next discussion
predicting volcanic eruptions and volcanic behavior.
Predicting Volcanic Eruptions and Volcanic Behavior
Before discussing how we can predict volcanic eruptions, its important
to get some terminology straight by defining some commonly used
terms.
Active Volcano - An active volcano to volcanologists is a volcano that
has shown eruptive activity within recorded history. Thus an active
volcano need not be in eruption to be considered active.
Currently there are about 600 volcanoes on Earth considered to be active
volcanoes.
Each year 50 to 60 of volcanoes actually erupt.
Extinct Volcano - An extinct volcano is a volcano that has not shown
any historic activity, is usually deeply eroded, and shows no signs of
recent activity. How old must a volcano be to be considered extinct
depends to a large degree on past activity.
For example, Yellowstone Caldera is about 600,000 years old and is
deeply eroded. But fumorolic activity, hot springs, and geysers all point
to the fact that magma still exists beneath the surface. Thus,
Yellowstone Caldera is not considered extinct.
Other volcanoes that are deeply eroded, smaller, and much younger than
Yellowstone, that show no hydrothermal activity may be considered
extinct.
Dormant Volcano - A dormant volcano (sleeping volcano) is somewhere
between active and extinct. A dormant volcano is one that has not
shown eruptive activity within recorded history, but shows geologic
evidence of activity within the geologic recent past.
Because the lifetime of a volcano may be on the order of a million years,
dormant volcanoes can become active volcanoes all of sudden. These
are perhaps the most dangerous volcanoes because people living in the
vicinity of a dormant volcano may not understand the concept of
geologic time, and there is no written record of activity. These people
are sometimes difficult to convince when a dormant volcano shows
signs of renewed activity.
Yellowstone Caldera would be considered a dormant volcano.
Mount St. Helens was a considered a dormant volcano, having not
erupted for 123 years, before its reawakening in 1980.
Mount Pinatubo in the Philippines had been dormant for over 400 years
before its eruption in 1991.
Mount Vesuvius, near Naples, Italy was considered an extinct volcano
prior to its devastating eruption of 79 A.D.
Long - Term Forecasting and Volcanic Hazards Studies
Studies of the geologic history of a volcano are generally necessary to
make an assessment of the types of hazards posed by the volcano and the
frequency at which these types of hazards have occurred in the past.
The best way to determine the future behavior of a volcano is by
studying its past behavior as revealed in the deposits produced by
ancient eruptions. Because volcanoes have such long lifetimes relative
to human recorded history, geologic studies are absolutely essential.
Once this information is available, geologists can then make forecasts
concerning what areas surrounding a volcano would be subject to the
various kinds of activity should they occur in a future eruption, and also
make forecasts about the long - term likelihood or probability of a
volcanic eruption in the area.
During such studies, geologists examine sequences of layered deposits
and lava flows. Armed with knowledge about the characteristics of
deposits left by various types of eruptions, the past behavior of a volcano
can be determined. Bore holes often provide data (such as you
discovered in your homework exercise)
Using radiometric age dating of the deposits the past frequency of events
can be determined.
This information is then combined with knowledge about the present
surface aspects of the volcano to make volcanic hazards maps which can
aid other scientists, public officials, and the public at large to plan for
evacuations, rescue and recovery in the event that short-term prediction
suggests another eruption.
Such hazards maps delineate zones of danger expected from the hazards
discussed above: lava flows, pyroclastic flows, tephra falls, mudflows,
floods, etc.
Short - Term Prediction based on Volcanic Monitoring
Short - term prediction of volcanic eruptions involves monitoring the
volcano to determine when magma is approaching the surface and
monitoring for precursor events that often signal a forthcoming eruption.

Seismic Exploration and Monitoring - Since seismic waves are
generated by both earthquakes and explosions, and since S-waves cannot
pass through liquids, arrays of seismographs can be placed around a
volcano and small explosions can be set off to generate seismic waves.
If a magma body exists beneath the volcano, then there will be zone
were no S-waves arrive (an S-wave shadow zone) that can be detected.
Monitoring the movement of the S-wave shadow zone can delineate the
position and movement of the magma body.

As noted above, as magma moves and deforms rocks it may be
responsible for the generation of earthquakes. Thus, there is usually an
increase in seismic activity prior to a volcanic eruption. Focal depths of
these precursor earthquakes may change with time, and if so, the
movement of magma can sometimes be tracked. In addition, volcanic
tremor, as noted above, can also be indication that magma is moving
below the surface.
Changes in Magnetic Field - Rocks contain minerals such as magnetite
that are magnetic. Such magnetic minerals generate a magnetic field.
However, above a temperature called the Curie Temperature, these
magnetic minerals show no magnetism. Thus, if a magma body enters a
volcano, the body itself will show no magnetism, and if it heats the
surrounding rocks to temperatures greater than the Curie Temperature
(about 500oC for magnetite) the magnetic field over the volcano will be
reduced. Thus, by measuring changes in the magnetic field, the
movement of magma can sometimes be tracked.
Changes in Electrical Resistivity - Rocks have resistance to the flow of
electrical current which is highly dependent on temperature and water
content. As magma moves into a volcano this electrical resistivity will
decrease. Making measurements of the electrical resistivity by placing
electrodes into the ground, may allow tracking of the movement of
magma.
Ground Deformation - As magma moves into a volcano, the structure
may inflate. This will cause deformation of the ground which can be
monitored. Instruments like tilt meters measure changes in the angle of
the Earth's surface which are measured in microradians 0.00018o.
Other instruments track changes in distance between several points on
the ground to monitor deformation.
Changes in Groundwater System - As magma enters a volcano it may
cause changes in the groundwater system, causing the water table to rise
or fall and causing the temperature of the water to increase. By
monitoring the depth to the water table in wells and the temperature of
well water, spring water, or fumaroles, changes can be detected that may
signify a change in the behavior of the volcanic system.
Changes in Heat Flow - Heat is everywhere flowing out of the surface of
the Earth. As magma approaches the surface or as the temperature of
groundwater increases, the amount of surface heat flow will increase.
Although these changes may be small they be measured using infrared
remote sensing.
Changes in Gas Compositions - The composition of gases emitted from
volcanic vents and fumaroles often changes just prior to an eruption. In
general, increases in the proportions of hydrogen chloride (HCl) and
sulfur dioxide (SO2) are seen to increase relative to the proportion of
water vapor.
In general, no single event can be used to predict a volcanic eruption,
and thus many events are usually monitored so that taken in total, an
eruption can often be predicted. Still, each volcano behaves somewhat
differently, and until patterns are recognized for an individual volcano,
predictions vary in their reliability. Furthermore, sometimes a volcano
can erupt with no precursor events at all.
After the catastrophic eruption of Mount St. Helens on May 18, 1980, a
volcanic dome began to grow in the crater. Growth of this dome
occurred sporadically, and sometimes small eruptions occurred from the
dome. After several years of dedicated monitoring, scientists are now
able to predict with increasing accuracy eruptions from this dome. An
example is shown in the graphs to the right. In the weeks prior to an
eruption on March 19, 1982, the amount of seismic energy released
increased, the amount dome expansion increased, tilt increased, and SO2
emissions increased prior to the event.
Beginning on March 12, a prediction was made that an eruption would
be likely within the next 10 days. On March 15, the prediction was
narrowed to likely within 4 days, and on March 18 scientists predicted
that an eruption would occur within the next two days. On March 19 the
eruption did occur.
Note that eruption predictions such as in this example are only possible
if constant monitoring of a volcano takes place. Monitoring is an
expensive endeavor, and not all active or potentially active volcanoes are
monitored. Still, if people living around volcanoes are aware of some of
the precursor phenomena that occur, they may be able to communicate
their findings of anomalous events to scientists who can begin
monitoring on a regular basis and help prevent a pending disaster.
Education and communication is essential in reducing risk from volcanic
hazards!
Tsunami
Up until December of 2004, the phenomena of tsunami was not on the
minds of most of the world's population. That changed on the morning
of December 24, 2004 when an earthquake of moment magnitude 9.1
occurred along the oceanic trench off the coast of Sumatra in Indonesia.
This large earthquake resulted in vertical displacement of the sea floor
and generated a tsunami that eventually killed about 230,000 people and
affected the lives of several million people. Although people living on
the coastline near the epicenter of the earthquake had little time or
warning of the approaching tsunami, those living farther away along the
coasts of Thailand, Sri Lanka, India, and East Africa had plenty of time
to move higher ground to escape. But, there was no tsunami warning
system in place in the Indian Ocean, and although other tsunami warning
centers attempted to provide a warning, there was no effective
communication system in place. Unfortunately, it has taken a disaster of
great magnitude to point out the failings of the world's scientific
community and to educate almost every person on the planet about
tsunami.
Even with heightened world awareness of tsunami, disasters still occur.
On September 29, 2009, earthquakes in the Samoa region of the
southwest Pacific Ocean killed nearly 200 people, and as a result of the
Chilean earthquake of February, 2010, at least 50 casualties resulted
from a tsunami triggered by a moment magnitude 8.8 earthquake.
On March 11, 2011 a Moment Magnitude 9.0 earthquake struck off the
northern Coast of Japan. The Earthquake generated a tsunami that rose
up to 135 feet above sea level and killed over 20,000 people. Because of
Japans familiarity with earthquakes and enforcement of earthquake
resistant building codes, there was only minor destruction from the
earthquake itself. But, despite that fact that a tsunami warning system
was in place, the earthquake was so close to the coast, that little time
was available for people to react.
Besides that high death toll, the tsunami caused one of the worst nuclear
disasters in history. The Fukushima nuclear power plant, located on the
coast was hit by a 49 ft. tsunami wave that overtopped the tsunami
protection walls that were only 19 feet high, and flooded the backup
generators for the plant that were somehow placed on the first floor in a
known tsunami zone!
hat is a Tsunami
A tsunami is a very long-wavelength wave of water that is generated by
sudden displacement of the seafloor or disruption of any body of
standing water. Tsunami are sometimes called "seismic sea waves",
although they can be generated by mechanisms other than earthquakes.
Tsunami have also been called "tidal waves", but this term should not be
used because they are not in any way related to the tides of the Earth.
Because tsunami occur suddenly, often without warning, they are
extremely dangerous to coastal communities.

Physical Characteristics of Tsunami
All types of waves, including tsunami, have a wavelength, a wave
height, an amplitude, a frequency or period, and a velocity.
Wavelength is defined as the distance between two identical points on a
wave (i.e. between wave crests or wave troughs). Normal ocean waves
have wavelengths of about 100 meters. Tsunami have much longer
wavelengths, usually measured in kilometers and up to 500 kilometers.


Wave height refers to the distance between the trough of the wave and
the crest or peak of the wave.
Wave amplitude - refers to the height of the wave above the still water
line, usually this is equal to 1/2 the wave height. Tsunami can have
variable wave height and amplitude that depends on water depth as we
shall see in a moment
Wave frequency or period - is the amount of time it takes for one full
wavelength to pass a stationary point.
Wave velocity is the speed of the wave. Velocities of normal ocean
waves are about 90 km/hr while tsunami have velocities up to 950 km/hr
(about as fast as jet airplanes), and thus move much more rapidly across
ocean basins. The velocity of any wave is equal to the wavelength
divided by the wave period.
V = /P
Tsunami are characterized as shallow-water waves. These are different
from the waves most of us have observed on a the beach, which are
caused by the wind blowing across the ocean's surface. Wind-generated
waves usually have period (time between two successive waves) of five
to twenty seconds and a wavelength of 100 to 200 meters. A tsunami can
have a period in the range of ten minutes to two hours and wavelengths
greater than 500 km. A wave is characterized as a shallow-water wave
when the ratio of the water depth and wavelength is very small. The
velocity of a shallow-water wave is also equal to the square root of the
product of the acceleration of gravity, g, (10m/sec2) and the depth of
the water, d.
The rate at which a wave loses its energy is inversely related to its
wavelength. Since a tsunami has a very large wavelength, it will lose
little energy as it propagates. Thus, in very deep water, a tsunami will
travel at high speeds with little loss of energy. For example, when the
ocean is 6100 m deep, a tsunami will travel about 890 km/hr, and thus
can travel across the Pacific Ocean in less than one day.
As a tsunami leaves the deep water of the open sea and arrives at the
shallow waters near the coast, it undergoes a transformation. Since the
velocity of the tsunami is also related to the water depth, as the depth of
the water decreases, the velocity of the tsunami decreases. The change of
total energy of the tsunami, however, remains constant.
Furthermore, the period of the wave remains the same, and thus more
water is forced between the wave crests causing the height of the wave
to increase. Because of this "shoaling" effect, a tsunami that was
imperceptible in deep water may grow to have wave heights of several
meters or more.
If the trough of the tsunami wave reaches the coast first, this causes a
phenomenon called drawdown, where it appears that sea level has
dropped considerably. Drawdown is followed immediately by the crest
of the wave which can catch people observing the drawdown off guard.
When the crest of the wave hits, sea level rises (called run-up). Run-up
is usually expressed in meters above normal high tide. Run-ups from the
same tsunami can be variable because of the influence of the shapes of
coastlines. One coastal area may see no damaging wave activity while
in another area destructive waves can be large and violent. The flooding
of an area can extend inland by 300 m or more, covering large areas of
land with water and debris. Flooding tsunami waves tend to carry loose
objects and people out to sea when they retreat. Tsunami may reach a
maximum vertical height onshore above sea level, called a run-up
height, of 30 meters. A notable exception is the landslide generated
tsunami in Lituya Bay, Alaska in 1958 which produced a 60 meter high
wave.
Because the wavelengths and velocities of tsunami are so large, the
period of such waves is also large, and larger than normal ocean waves.
Thus it may take several hours for successive crests to reach the shore.
(For a tsunami with a wavelength of 200 km traveling at 750 km/hr, the
wave period is about 16 minutes). Thus people are not safe after the
passage of the first large wave, but must wait several hours for all waves
to pass. The first wave may not be the largest in the series of waves. For
example, in several different recent tsunami the first, third, and fifth
waves were the largest.

How Tsunami are Generated
There is an average of two destructive tsunami per year in the Pacific
basin. Pacific wide tsunami are a rare phenomenon, occurring every 10 -
12 years on the average. Most of these tsunami are generated by
earthquakes that cause displacement of the seafloor, but, as we shall see,
tsunami can be generated by volcanic eruptions, landslides, underwater
explosions, and meteorite impacts.
Earthquakes
Earthquakes cause tsunami by causing a disturbance of the seafloor.
Thus, earthquakes that occur along coastlines or anywhere beneath the
oceans can generate tsunami. The size of the tsunami is usually related
to the size of the earthquake, with larger tsunami generated by larger
earthquakes. But the sense of displacement is also important. Tsunami
are generally only formed when an earthquake causes vertical
displacement of the seafloor. The 1906 earthquake near San Francisco
California had a Richter Magnitude of about 7.1, yet no tsunami was
generated because the motion on the fault was strike-slip motion with no
vertical displacement. Thus, tsunami only occur if the fault generating
the earthquake has normal or reverse displacement. Because of this,
most tsunami are generated by earthquakes that occur along the
subduction boundaries of plates, along the oceanic trenches. Since the
Pacific Ocean is surrounded by plate boundaries of this type, tsunami are
frequently generated by earthquakes around the margins of the Pacific
Ocean.
Examples of Tsunami generated by Earthquakes
Although the December 2004 Indian Ocean tsunami is by far the best
well known and most deadly (and will be featured in a video in class),
we here discuss other disastrous tsunami generated by earthquakes.
April 1, 1946 - A magnitude 7.3 earthquake occurred near Unimak
Island in the Aleutian Islands west of Alaska, near the Alaska Trench.
Sediment accumulating in the trench slumped into the trench and
generated a tsunami. A lighthouse at Scotch Gap built of steel
reinforced concrete was located on shore at an elevation of 14 m above
mean low water. The first wave of the tsunami hit the Scotch Gap area
20 minutes after the earthquake, had a run-up 30 m and completely
destroyed the lighthouse. 4.5 hours later the same tsunami reached the
Hawaiian Islands after traveling at an average speed of 659 km/hr. As it
approached the city of Hilo on the Big Island, it slowed to about 47
km/hr (note that even the fastest human cannot run faster than about 35
km/hr) and had a run-up of 18 m above normal high tide. It killed 159
people (90 in Hilo) and caused $25 million in property damage.
May 22, 1960 - A moment magnitude9.5 earthquake occurred along the
subduction zone off South America. Because the population of Chile is
familiar with earthquakes and potential tsunami, most people along the
coast moved to higher ground. 15 minutes after the earthquake, a
tsunami with a run-up of 4.5 m hit the coast. The first wave then
retreated, dragging broken houses and boats back into the ocean. Many
people saw this smooth retreat of the sea as a sign they could ride their
boats out to sea and recover some of the property swept away by the first
wave. But, about 1 hour later, the second wave traveling at a velocity of
166 km/hr crashed in with a run-up of 8 m. This wave crushed boats
along the coast and destroyed coastal buildings. This was followed by a
third wave traveling at only 83 km/hr that crashed in later with a run-up
of 11 m, destroying all that was left of coastal villages. The resulting
causalities listed 909 dead with 834 missing. In Hawaii, a tsunami
warning system was in place and the tsunami was expected to arrive at
9:57 AM. It hit at 9:58 AM and 61 people died, mostly sightseers that
wanted to watch the wave roll in at close range (obviously they were too
close). The tsunami continued across the Pacific Ocean, eventually
reaching Japan where it killed an additional 185 people.
March 27, 1964 - The Good Friday Earthquake in Alaska had a moment
magnitude of 9.2. This earthquake also occurred along the subduction
zone, and as we saw in our study of earthquakes, caused deformation of
the crust where huge blocks where dropped down as much as 2.3 m.
Because the coastline of Alaska is sparsely populated, only 122 people
died from the tsunami in Alaska. With a tsunami warning system in
place in Crescent City, California, all the townspeople moved to higher
ground. After watching four successive waves destroy their town, many
people returned to the low lying areas to assess the damage to their
property. The fifth wave had the largest run-up of 6.3 m and killed 12
people.
September 2, 1992 - A magnitude 7 earthquake off the coast of
Nicaragua in Central America occurred along the subduction zone below
the Middle America Trench. The earthquake was barely felt by the
residents of Nicaragua and was somewhat unusual. A 100 km-long
segment of the oceanic lithosphere moved 1 m further below the over
riding plate over a period of two minutes. Much energy was released
but the ground did not shake very much. Seawater apparently absorbed
some of the energy and sent a tsunami onto the coast. Residents had
little warning, 150 people died and 13,000 people were left homeless.
Volcanic Eruptions
Volcanoes that occur along coastal zones, like in Japan and island arcs
throughout the world, can cause several effects that might generate a
tsunami. Explosive eruptions can rapidly emplace pyroclastic flows into
the water, landslides and debris avalanches produced by eruptions can
rapidly move into water, and collapse of volcanoes to form calderas can
suddenly displace the water.
The eruption of Krakatau in the Straights of Sunda, between Java and
Sumatra, in 1883 generated at least three tsunami that killed 36,417
people. It is still uncertain exactly what caused the tsunami, but it is
known that several events that occurred during the eruption could have
caused such tsunami.
A large Plinian eruption column blasted pumice and ash up to 40 km
into the atmosphere. This Plinian eruption column likely collapsed
several times to produce pyroclastic flows, any of which could have
generated a tsunami.
A loud explosive blast was heard as far away as Australia. This blast
was likely caused by a phreatic explosion that occurred as a result of
seawater coming in contact with the magma. The explosion could have
generated at least one of the tsunami.
At some point during the eruption a caldera formed by collapse of the
volcanic island. Areas that were once more than 300 m above sea level
were found 300 m below sea level after the eruption. The sudden
collapse of the volcano to form this caldera could have caused one or
more tsunami.
Earthquakes were felt throughout the eruption. Any one of these
submarine earthquakes could have caused a tsunami.
One of the tsunami had a run-up of about 40 m above normal sea level.
A large block of coral weighing about 600 tons was ripped off the
seafloor and deposited 100 m inland. One ship was carried 2.5 km
inland and was left 24 meters above sea level, with all of its crew swept
into the ocean.
Landslides
Landslides moving into oceans, bays, or lakes can also generate tsunami.
Most such landslides are generated by earthquakes or volcanic
eruptions. As previously mentioned, a large landslide or debris
avalanche fell into Lituya Bay, Alaska in 1958 causing a wave with a
run-up of about 60 m as measured by a zone completely stripped of
vegetation.
Underwater Explosions
Nuclear testing by the United States in the Marshall Islands in the 1940s
and 1950s generated tsunami.
Meteorite Impacts
While no historic examples of meteorite impacts are known to have
produced a tsunami, the apparent impact of a meteorite at the end of the
Cretaceous Period, about 65 million years ago near the tip of what is
now the Yucatan Peninsula of Mexico, produced tsunami that left
deposits all along the Gulf coast of Mexico and the United States.
Mitigation of Risks and Hazards
The main damage from tsunami comes from the destructive nature of the
waves themselves. Secondary effects include the debris acting as
projectiles which then run into other objects, erosion that can undermine
the foundations of structures built along coastlines, and fires that result
from disruption of gas and electrical lines. Tertiary effects include loss
of crops and water and electrical systems which can lead to famine and
disease.
Within the last century, up until the December 2004 tsunami, there were
94 destructive tsunami which resulted in 51,000 deaths. Despite the fact
that tsunami warning systems have been in place in the Pacific Ocean
basin since 1950, deaths still result from tsunami, especially when the
source of the earthquake is so close to a coast that there is little time for
a warning, or when people do not heed the warning or follow
instructions associated with the warning. These factors point out the
inadequacy of the world in not having a tsunami warning system in
place in the Indian Ocean, where in one event, the death toll from
tsunami was increased by a factor of 5 over all previous events.
Prediction and Early Warning
For areas located at great distances from earthquakes that could
potentially generate a tsunami there is usually plenty of time for
warnings to be sent and coastal areas evacuated, even though tsunami
travel at high velocities across the oceans. Hawaii is good example of
an area located far from most of the sources of tsunami, where early
warning is possible and has saved lives. For earthquakes occurring
anywhere on the subduction margins of the Pacific Ocean there is a
minimum of 4 hours of warning before a tsunami would strike any of the
Hawaiian Islands.
The National Oceanic and Atmospheric Administration (NOAA) has set
up a Pacific warning system for areas in the Pacific Ocean, called the
Pacific Tsunami Warning Center. It consists of an international network
of seismographic stations, and tidal stations around the Pacific basin that
can all send information via satellite to the Center located in Hawaii.
When an earthquake occurs somewhere in the region, the Center
immediately begins to analyze the data looking for signs that the
earthquake could have generated a tsunami. The tidal stations are also
monitored, and if a tsunami is detected, a warning is sent out to all areas
on the Pacific coast. It takes at least 1 hour to assimilate all of the
information and issue a warning. Thus for an average velocity of a
tsunami of 750 km/hr, the regional system can provide a warning
sufficient for adequate evacuation of coastal areas within 750 km of the
earthquake.
In order to be able to issue warnings about tsunami generated within 100
to 750 km of an earthquake, several regional warning centers have been
set up in areas prone to tsunami generating earthquakes. These include
centers in Japan, Kamchatka, Alaska, Hawaii, French Polynesia, and
Chile.


These systems have been very successful at saving lives. For example,
before the Japanese warning system was established, 14 tsunami killed
over 6000 people in Japan. Since the establishment of the warning
system, up until March 2011, 20 tsunami have killed 215 people in
Japan.
Like all warning systems, the effectiveness of tsunami early warning
depends strongly on local authority's ability to determine that their is a
danger, their ability to disseminate the information to those potentially
affected, and on the education of the public to heed the warnings and
remove themselves from the area.
Tsunami Safety Rules
A strong earthquake felt in a low-lying coastal area is a natural warning
of possible, immediate danger. Keep calm and quickly move to higher
ground away from the coast.
All large earthquakes do not cause tsunami, but many do. If the quake is
located near or directly under the ocean, the probability of a tsunami
increases. When you hear that an earthquake has occurred in the ocean
or coastline regions, prepare for a tsunami emergency.
Tsunami can occur at any time, day or night. They can travel up rivers
and streams that lead to the ocean.
A tsunami is not a single wave, but a series of waves. Stay out of danger
until an "ALL CLEAR" is issued by a competent authority.
Approaching tsunami are sometimes heralded by noticeable rise or fall
of coastal waters. This is nature's tsunami warning and should be
heeded.
A small tsunami at one beach can be a giant a few miles away. Do not
let modest size of one make you lose respect for all.
Sooner or later, tsunami visit every coastline in the Pacific. All tsunami -
like hurricanes - are potentially dangerous even though they may not
damage every coastline they strike.
Never go down to the beach to watch for a tsunami! WHEN YOU CAN
SEE THE WAVE YOU ARE TOO CLOSE TO ESCAPE. Tsunami can
move faster than a person can run!
During a tsunami emergency, your local emergency management office,
police, fire and other emergency organizations will try to save your life.
Give them your fullest cooperation.
Homes and other buildings located in low lying coastal areas are not
safe. Do NOT stay in such buildings if there is a tsunami warning.
The upper floors of high, multi-story, reinforced concrete hotels can
provide refuge if there is no time to quickly move inland or to higher
ground.
If you are on a boat or ship and there is time, move your vessel to deeper
water (at least 100 fathoms). If it is the case that there is concurrent
severe weather, it may safer to leave the boat at the pier and physically
move to higher ground.
Damaging wave activity and unpredictable currents can affect harbor
conditions for a period of time after the tsunami's initial impact. Be sure
conditions are safe before you return your boat or ship to the harbor.
Stay tuned to your local radio, marine radio, NOAA Weather Radio, or
television stations during a tsunami emergency - bulletins issued through
your local emergency management office and National Weather Service
offices can save your life.
Flood Stage
The term stage refers to the height of a river (or any other body of
water) above a locally defined elevation. This locally defined
elevation is a reference level, often referred to as datum. For
example, for the lower part of the Mississippi River, reference
level or datum, is sea level (0 feet). Currently the Mississippi
River is at a stage of about 3 feet, that is 3 feet above sea level.
Other river systems have a reference level that is not sea level.
Most rivers in the United States have gaging stations where
measurements are continually made of the river's stage and
discharge. These are plotted on a graph called a hydrograph,
which shows the stage or discharge of the river, as measured at the
gaging station, versus time.
When the discharge of a river increases, the channel may become
completely full. Any discharge above this level will result in the
river overflowing its banks and causing a flood. The stage at
which the river will overflow its banks is called bankfull stage or
flood stage. For example, the flood stage of the Mississippi River
at New Orleans is 17 feet. Discharge that produces a stage over 17
feet will result in the water nearing the top of the levee with
potential flooding of the city of New Orleans (the top of the levee
is actually at 25 feet above sea level). (Note that for the
Mississippi River and other large rivers in Louisiana, the current
stage and flood stage are published on a daily basis in the weather
section of the Times-Picayune newspaper).
Discharge is not linearly related to stage because discharge depends
on both the depth and width of the stream channel, or more
precisely, on the cross-sectional shape of the channel. Stage refers
only to the height of the water above some reference level. For
example, the graph below is a hydrograph of the Mississippi River
at St. Louis, Missouri during the time period of the 1993 flood.
Discharge is plotted on the Y-axis, and dates are plotted on the x-
axis. Note that stages corresponding to various discharges are
shown on the left-hand y-axis, and that the spacing between equal
units of stage are not equal along the y-axis.
Note that for the
1993
Mississippi
River Flood,
the river
reached flood
stage of 30
feet above
datum on
about June 26
and peaked (or
crested) at just
under 50 feet
above datum
on August 1.
The sudden
drops seen in
discharge
around July 15
and July 20
corresponded
to breaks in
the levee
system
upstream from
St. Louis that
caused water
to flow onto
the floodplain
upstream, thus
reducing both
the stage and
discharge
measured at
St. Louis.
Factors that Affect Flooding
As discussed previously the main factors that cause flooding are heavy
rainfall, sudden or heavy snow melt, and dam failure. Now that we
understand something about levees and floodplains, we can add to this
list the possibility of levee failure. All of these factors can suddenly
increase discharge of water into streams, within streams, and out of
streams. Furthermore, as we have just seen, when the discharge causes
the river to rise above flood stage water runs onto the floodplain. Here
we discuss the main cause of flooding, that is heavy rainfall over a short
period of time.
When rain falls on the surface of the Earth, some of the water is
evaporated and returns to the atmosphere, some of it infiltrates the soil
and moves downward into the groundwater system, and some is
intercepted by depressions and vegetation. What remains on the surface
of the Earth and eventually flows into streams is called runoff. In
general, then:
Runoff = Precipitation - Infiltration - Interception - Evaporation
Evaporation tends to be the least of these quantities, particularly over
short periods of time, and thus precipitation, infiltration, and
interception are the most important variables that determine runoff and
eventual discharge into streams.
Rainfall Distribution
If rainfall is heavier than normal in a particular area and infiltration,
interception, and evaporation are low then runoff can be high and the
likelihood of flooding will increase. Heavy rainfall can be depicted
on maps that show curves of equal rainfall. Such curves are called
isohyets, and the resulting maps are called isohyetal maps.
Lag Time - The
time difference
between when
heavy
precipitation
occurs and
when peak
discharge
occurs in the
streams
draining an
area is called
lag time.
Lag time depends
on such factors
as the amount
of time over
which the rain
falls and the
amount of
infiltration and
interception
that takes
place along the
path to a
stream.

If the amount of rain is high over a short time period, lag
time is short.
If the amount of rain is high over a longer time period, lag
time is longer.
Lack of infiltration and interception reduce lag time.
Upstream flooding and flash floods
In areas where large amounts of rain fall over a short period of time
within a small area, streams in the local area may flood, with little or no
effect on areas downstream. Such floods are referred to as upstream
floods. In such floods, water rises quickly and flows away quickly after
the storm has passed. Lag times are measured in days.

Flash floods occur when the rate of infiltration is low and heavy rains
occur over a short period of time. They are upstream floods with very
little lag time (lag times may be only a few hours). Because they come
with little warning, flash floods are the most dangerous to human lives.
Downstream flooding
If large amounts of rain fall over an extended period of time over a
large region, downstream floods (also called regional floods) may
occur. Lag times are usually longer as tributary streams
continually increase the discharge into larger streams. Such floods
extend over long periods of time and affect the larger streams as
well as tributary streams. The 1993 flood on the upper Mississippi
River is considered a downstream flood. Water levels rise slowly
and dissipate slowly (in the case of the 1993 flood, the increase in
discharge to the peak occurred over several weeks after several
weeks of intense rainfall, and it took several months for river
stages to return to normal levels).
Infiltration

Infiltration is controlled by how readily the water can seep into the soil,
be absorbed by the soil, and work its way down to the water table.
Several factors determine the rate of infiltration:
Extent of water saturation of the soil If the soil is already saturated
with water and the water table has risen as a result of rainfall prior
to a heavy storm, then little further water can infiltrate the soil, and
the rate of infiltration will be highly decreased.
Vegetation cover
Vegetation can aid infiltration by slowing the flow of water over the
surface and providing passageways along root systems for water to enter
the soil. In desert regions or areas that have recently been deforested
either by fires or humans, infiltration will be reduced, thus increasing
the rate of runoff and decreasing the lag time.
Soil types (dependent on climate)
Different soil types have different capacities to absorb moisture. Soil
type is to a large extent dependent on climate. For example a type of
soil that forms in dry, desert-like environments has a thin layer of poorly
developed soil overlying a crust of caliche. Caliche is calcium
carbonate that has precipitated out of water infiltrating though the thin
soil. The caliche zone acts as an impermeable layer though which water
can only penetrate with difficulty. Such soils in deserts, combined with
the lack of vegetation make flash flooding in desert areas more
common.
Frozen ground
If the ground is frozen little water can penetrate. Thus rainfall after
a period of cold temperatures may not be able to infiltrate through
the frozen ground.
Human construction
Humans tend to pave the Earth with such things as parking lots,
highways, sidewalks, and plazas that prevent infiltration of water into
the soil. Furthermore they tend to channel the water into storm sewer
systems and concrete lined drainages, all of which increase runoff and
decrease infiltration.
Interception
Interception involves anything that traps rainwater and prevents it from
contributing to rnnoff. This includes water that is stored on leaves and
braches of trees until it evaporates and water that is gets stored in ponds
or lakes. Thus, removal of vegetation decreases interception and results
in more runoff. Increasing vegetation or construction of retention ponds,
increases interception and results in less runoff.

Levee Failures
Natural levees are constructed as a result of flooding, as we saw in the
discussion last lecture. But, natural levees tend to be relatively low and
do not offer much protection from large discharge because they can
easily be overtopped. Human made levees, such as we see on the
Mississippi River along much of its length, are much higher and are
constructed to prevent flooding from high discharges on the River.
Most levees are constructed of piles of dirt (rock and soil) with a
concrete cover on the river side of the levee. Such levees often give a
false sense of security for those living on the floodplain the levee was
built to protect, because failure of such levees can lead to flooding,
either because discharge can become great enough to overtop the levees
or the levees can become weakened and fail. Levees can fail for three
main reasons.
2. Overtopping of levees
If high discharge in the river leads to a river stage that is higher than
any point on a levee, the water will overtop the levee and start to
flow onto the floodplain. Because the initial gradient from the
river to flood plain is relatively high, the velocity of the stream as
it overtops the levee will be high. High velocities can result in
high rates of erosion, and thus the levee that is initially overtopped
will soon become eroded and a channel through the levee will soon
be created.

3. Undercutting and slumping of levee
Higher discharge in the river will lead to higher velocities with the
stream trying to increase its width and depth. Higher velocities can
lead to higher rates of erosion along the inner parts of levees and
thus lead to undercutting and slumping of the levee into the river.
Heavy rainfall or seepage into the levee from the river can increase
fluid pressure in the levee and lead to slumping on the outer parts
of the levee. If the slumps grow to the top of the levee, large
sections of the levee may slump onto the floodplain and lower the
elevation of the top of the levee, allowing it to be more easily
overtopped.

4. Seepage and Piping
Increasing levels of water in the river will cause the water table in the
levee to rise. This will also increase fluid pressure and may result
in seepage (water being pushed through the levee to rise as springs
on the surrounding flood plains). If a high rate of flow is
developed due to the increased fluid pressures, then a high velocity
pathway to the flood plain may develop piping may occur. Piping
will erode the material under the levee, undermining it and causing
its collapse and failure.
Dam Failures
Failure of natural dams or human made dams results in flooding
downstream from the dam. Natural dams result from natural events that
block streams, such as landslides, lava flows, or pyroclastic flows into
streams. Humans build dams for flood control, water storage, and the for
the generation of electricity.
Hazards Associated with Flooding
Hazards associated with flooding can be divided into primary hazards that occur
due to contact with water, secondary effects that occur because of the flooding,
such as disruption of services, health impacts such as famine and disease, and
tertiary effects such as changes in the position of river channels. hroughout the
last century flooding has been one of the most costly disasters in terms of both
property damage and human casualties. !a"or floods in #hina, for e$ample, %illed
about & million people in '((), nearly * million in '+,', and about ' million in
'+,( he '++, flood on the upper !ississippi -iver and !idwest %illed only *)
people, but the ../. Army #orps of 0ngineers estimates the total economic loss at
between '1 and &2 billion dollars .Again, the primary effects of floods are those
due to direct contact with the flood waters. As seen in the video last lecture, water
velocities tend to be high in floods. As discharge increases velocity increases.
3ith higher velocities, streams are able to transport larger particles as suspended
load. /uch large particles include not only roc%s and sediment, but, during a flood,
could include such large ob"ects as automobiles, houses and bridges.
!assive amounts of erosion can be accomplished by flood waters. /uch erosion
can undermine bridge structures, levees, and buildings causing their collapse.
3ater entering human built structures cause water damage. 0ven with minor
flooding of homes, furniture is ruined, floors and walls are damaged, and anything
that comes in contact with the water is li%ely to be damaged or lost. Flooding of
automobiles usually results in damage that cannot easily be repaired.
he high velocity of flood waters allows the water to carry more sediment as
suspended load. 3hen the flood waters retreat, velocity is generally much lower
and sediment is deposited. After retreat of the floodwaters everything is usually
covered with a thic% layer of stream deposited mud, including the interior of
buildings.
Flooding of farmland usually results in crop loss. 4ivestoc%, pets, and other
animals are often carried away and drown.
Humans that get caught in the high velocity flood waters are often drowned by the
water.
Floodwaters can concentrate garbage, debris, and to$ic pollutants that can cause
the secondary effects of health hazards.

/econdary and ertiary 0ffects
-emember that secondary effects are those that occur because of the primary
effects and tertiary effects are the long term changes that ta%e place. Among the
secondary effects of a flood are5
6isruption of services 7
6rin%ing water supplies may become polluted, especially if sewerage treatment
plants are flooded. his may result in disease and other health effects, especially
in under developed countries.
8as and electrical service may be disrupted.
ransportation systems may be disrupted, resulting in shortages of food and clean7
up supplies. 9n under developed countries food shortages often lead to starvation.
4ong 7 term effects :tertiary effects;7
4ocation of river channels may change as the result of flooding, new channels
develop, leaving the old channels dry.
/ediment deposited by flooding may destroy farm land :although silt deposited by
floodwaters could also help to increase agricultural productivity;.
<obs may be lost due to the disruption of services, destruction of business, etc.
:although "obs may be gained in the construction industry to help rebuild or repair
flood damage;.
9nsurance rates may increase.
#orruption may result from misuse of relief funds.
6estruction of wildlife habitat.
=redicting -iver Flooding
Floods can be such devastating disasters that anyone can be affected at almost
anytime. As we have seen, when water falls on the surface of the 0arth, it has to
go somewhere. 9n order to reduce the ris% due to floods, three main approaches are
ta%en to flood prediction. /tatistical studies can be underta%en to attempt to
determine the probability and fre>uency of high discharges of streams that cause
flooding. Floods can be modeled and maps can be made to determine the e$tent
of possible flooding when it occurs in the future. And, since the main causes of
flooding are abnormal amounts of rainfall and sudden thawing of snow or ice,
storms and snow levels can be monitored to provide short7term flood prediction.
Fre>uency of Flooding
9n your homewor% e$ercise you will see how flood fre>uencies can be determined
for any given stream if data is available for discharge of the stream over an
e$tended period of time. /uch data allows statistical analysis to determine how
often a given discharge or stage of a river is e$pected. From this analysis a
recurrence interval can be determined and a probability calculated for the
li%elihood of a given discharge in the stream for any year. he data needed to
perform this analysis are the yearly ma$imum discharge of a stream from one
gaging station over a long enough period of time.
9n order to determine the recurrence interval, the yearly discharge values are first
ran%ed. 0ach discharge is associated with a ran%, m, with m ? ' given to the
ma$imum discharge over the years of record, m ? & given to the second highest
discharge, m ? , given to the third highest discharge, etc.
he smallest discharge will receive a ran% e>ual to the number of years over which
there is a record, n. hus, the discharge with the smallest value will have m ? n.
he number of years of record, n, and the ran% for each pea% discharge are then
used to calculate recurrence interval, - by the following e>uation, called the
3eibull e>uation5
- ? :n@';Am
A graph is then made plotting discharge for each year of the record versus
recurrence interval. he graph usually plots recurrence interval on a logarithmic
scale. An e$ample of such a plot is shown here for the -ed -iver of the Borth
gaging station at Fargo, Borth 6a%ota.
A best7fit line is then drawn through the data points. From the best7fit line, one
can determine the discharge associated with the a flood with a recurrence interval
of say '2 years. his would be called the '27year flood.
For the data on the -ed -iver, above, the discharge associated with the '27year
flood is about '&,222 cubic feet per second. /imilarly the discharge associated with
a flood with a recurrence interval of 12 years :the 127year flood; would have a
discharge of about &',222 cubic feet per second. he '227year flood would have a
discharge of about &1,222 cubic feet per second.
Bote that for the -ed -iver data, shown above, the April '(, '++) flood had a
discharge of ,2,222 ft,Asec, which is e>uivalent to a &127year flood. Also note
that a flood that reached a similar stage occurred on the -ed -iver in Fargo in the
year '((), only ''2 years before. Furthermore, the -ed -iver reached a bit more
than ,2,222 ft,Asec in &22+. 6oes this ma%e the statistical analysis unreliableC
he answer is no. As we shall see, it is possible to have two '227year floods
occurring '22 years apart, 12 years apart, or even & in the same year.
he probability, =e, of a certain discharge can be calculated using the inverse of
the 3eibull e>uation5
=e ? mA:n@';
he value, =e, is called the annual e$ceedence probability. For e$ample, a
discharge e>ual to that of a '27year flood would have an annual e$ceedence
probability of 'A'2 ? 2.' or '2D. his would say that in any given year, the
probability that a flood with a discharge e>ual to or greater than that of a '2 year
flood would be 2.' or '2D. /imilarly, the probability of a flood with discharge
e$ceeding the '22 year flood in any given year would be 'A'22 ? 2.2', or 'D.
Bote that such probabilities are the same for every year. /o, for e$ample, the
probability that discharge of the -ed -iver at Fargo, Borth 6a%ota will e$ceed
&1,222 ft,Asec :the discharge of the '227year flood; this year or any other year
would 'D. Eou can thin% of this in the same way you would thin% about rolling
dice. he probability on any roll that you will end up with a si$, rolling only on
die, is ' in F or 'F.F)D. 0ach time you roll that one die the probability is the same,
although you %now that it is possible to roll two or three si$es in a row.
hus, it is important to remember that even though a &127year flood occurred in
Fargo in '++) and &22+, there is still a 2.*D probability that such a flood, or one of
even greater magnitude, will occur this year.

6espite the fact that the '22 year flood has only a 'D chance of occurring each
year, the probabilities do accumulate over time.
he probability of a certain7size flood occurring during any period can be
calculated using the following e>uation5
=t ? '7:'7=e;n
where =t is the probability of occurrence over the entire time period, n, and =e is
the probability of occurrence in any year.
3e can use this e>uation to calculate how the probabilities change over time. he
result is depicted in the graph below for =e ? 2.2' :'22 year flood;
wo important points emerge5
he probability of a '22 year flood occurring in '22 years is BG '22DH :/ee
below;
he probability of a '22 year flood occurring in ,2 years :the lifetime of the
average home mortgage; is &F.2DH
Flood Hazard !apping
Food hazard mapping is used to determine the areas susceptible to flooding when
discharge of a stream e$ceeds the ban%7full stage. .sing historical data on river
stages and discharge of previous floods, along with topographic data, maps can be
constructed to show areas e$pected to be covered with floodwaters for various
discharges or stages.

9n constructing such maps aerial photographs and satellite images of prior floods
are studied to help to determine the areas that would be covered. he illustration
above shows a possible hazard map based on estimated discharges or river stages
for a hypothetical '27year flood, 127year flood, and '227year flood.
9n addition, scale models are often constructed of areas prone to flooding. /uch
models only wor% if they are %ept up to date with current flood prevention
structures and drainage modifications.
!onitoring the =rogress of /torms
9f factors such as amount of rainfall, degree of ground saturation, degree of
permeable soil, and amount of vegetation can be determined, then these can be
correlated to give short7term prediction, in this case called a forecast, of possible
floods. 9f a forecast is issued, then a flood warning can be communicated to warn
the public about the possible e$tent of the flood, and to give people time to move
out of the area. /uch forecasts are very useful for flooding that has a long lag time
between the storm and the pea% discharge. Flash floods, which characteristically
have short lag times, are more problematical. hus, in some areas %nown to be
susceptible to flash floods, a flash flood warning is often issued any time heavy
rainfall is e$pected because there is always the chance of a flash food
accompanying heavy rainfall.
9n con"unction with the Bational 3eather /ervice, various agencies in the ../.
provide information on the internet that forecast potential floods. For e$ample 7
http5AAwater.weather.govAahpsA
Human 9ntervention
Humans can modify the landscape in many ways. /ometimes humans attempt to
modify drainage systems to prevent flooding, but sometimes these efforts have
adverse effects and actually help to cause flooding in other areas. Any
modification of the landscape has the potential to cause changes in the drainage
system, and such changes can have severe conse>uences.
#hannel !odifications
Humans often decide that a stream should flow along a specified path for such
reasons as flood control, enhancement of drainage, control of erosion, increasing
access to the floodplain for development, or improvement of the appearance of the
channel. /uch channel modifications involve measures such as the straightening
the channel, deepening or widening the channel, clearing vegetation from the
ban%s, or lining the channel with concrete. hese modifications are referred to as
channelization.
9n order to control floods, channel modification should involve increasing the
channel cross7sectional area, so that higher discharge will not increase the stage of
the river. /traighter channels also allow higher velocity flow and, enable the
stream to drain faster when discharge increases. 4ining the channel with concrete
provides a smoother surface over which the water can flow, thereby reducing
friction and also increasing the velocity of the stream.
3hile channelization for flood control may reduce the incidence of flooding in the
channelized area, it often results in more severe flooding both upstream and
downstream from the channelized area.

#hannelization can also interfere with the natural habitat of the stream system and
decrease the aesthetic value of the stream.
#hannelization, or any other modification of a stream system, changes the validity
of all historic data collected over the years on that stream. 6uring flooding of the
!ississippi -iver in '+), and '++, water levels rose to higher levels than e$pected
from the statistical data, because modification of the stream had made the data
invalid. he '+), flood caused damage and rose to levels that might be e$pected
from a &227year flood, even though discharge was only at a level predicted for a
,27year flood. hus, it appears that modifications of the drainage system had an
adverse effect.
0ffects of 6evelopment on Flood Hazard
3henever humans modify the landscape in any way changes are to be e$pected in
the way water drains from the land. .nless careful consideration is given to the
possible drainage conse>uences, such landscape modifications can result in higher
incidence of flooding. 6evelopment on floodplains should therefore be underta%en
only with great care. 0$isting developments that have enhanced flooding problems
are often costly to fi$. Among the factors that enhance the flood potential are5
#hannelization 7 As noted above, most channelization is underta%en to reduce
flood hazards. Iut, channelization is also underta%en to allow development on the
floodplain.
9f the channelization results in decreasing the cross7sectional area of the stream, as
in the e$ample above, then the same discharge that may not have produced
flooding prior to channelization, may overflow the ban%s and cause e$tensive
flooding after channelization.
/ubsidence 7 As will be discussed in a later lecture, subsidence often results in
developed areas due to compaction of the sediment, both due to the increasing
weight of structures and hydrocompaction associated with the lowering of the
water table. Any time the elevation of an area is lowered, it becomes sub"ect to
collection of more water, and in severe cases, could drastically change the drainage
pattern.
/torm /ewers 7 9n order to collect run off from streets, par%ing lots, and buildings,
all of which bloc% the infiltration of water into the soil, storm sewers are installed
to provide underground drainage of the surface. 3hile this may prevent local
flooding of streets, it moves water more rapidly to the ma"or stream systems and
thus decreases the lag time and increases the pea% discharge of the streams
collecting the runoff from the storm sewers.
-eduction of infiltration 7 Any time the surface materials of the 0arth are covered
with impermeable materials li%e concrete, asphalt, or buildings, the infiltration of
water into the soil is prevented. .rbanization tends to reduce infiltration, and thus
water must collect in storm sewers and eventually in the main drainage systems.
hus, e$tensive urbanization also decreases the lag time and increases the pea%
discharge even further. .rbanization can therefore lead to a higher incidence of
flash floods.
!itigation of Flood Hazards
!itigation of flood hazards can be attempted in two main ways5 An engineering
approach, to control flooding, and a regulatory approach designed to decrease
vulnerability to flooding.
0ngineering Approaches
#hannel modifications 7 As discussed above, channelization. can enlarge cross7
sectional area and thus create a situation where a higher stage is necessary before
flooding. 9n other words by enlarging the cross7sectional area, higher discharge
can be held within the channel. #hannelization also increases water velocity, and
thus reduces drainage time.

6ams 7 6ams can be used to hold water bac% so that discharge downstream can be
regulated at a desired rate. Human constructed dams have spillways that can be
opened to reduce the level of water in the reservoir behind the dam. hus, the
water level can be lowered prior to a heavy rain, and more water can be trapped in
the reservoir and released later at a controlled discharge.
-etention ponds 7 -etention ponds serve a similar purpose to dams. 3ater can be
trapped in a retention pond and then released at a controlled discharge to prevent
flooding downstream.
4evees, 6i%es, and Floodwalls 7 hese are structures built along side the channel
to increase the stage at which the stream floods. /ome controversy has developed
concerning the use of such structures. For e$ample, during the '++, floods on the
upper !ississippi -iver, the city of /t. 4ouis was prevented from flooding by
closing the floodwalls. his essentially narrowed the river channel as it passed /t.
4ouis and caused slowing of the -iver. Iecause of the restricted channel, flood
waters were forced to flow into areas both upstream and downstream from /t.
4ouis, perhaps increasing the damage in the these areas.
Floodways 7 Floodways are areas that can be built to provide an outlet to a stream
and allow it flood into an area that has been designated as a floodway. Floodways
are areas where no construction is allowed, and where the land is used for
agricultural or recreational purposes when there is no threat of a flood, but which
provide an outlet for flood waters during periods of high discharge. he Ionnet
#arrie /pillway west of Bew Grleans is such a floodway. 6uring low stages of the
!ississippi -iver the land between the -iver and 4a%e =ontchartrain is used for
recreational purposes 7 hunting, fishing, and dirt bi%e riding for e$ample. 6uring
high stages of the -iver when there is a potential for the -iver to rise to flood stage
in Bew Grleans, the spillway is opened so that water drains into 4a%e
=ontchartrain. his lowers the level of water in the !ississippi and reduces the
possibility of a levee brea% or water overtopping the levee.

-egulatory Approaches to -educe Julnerability

3ith a better understanding of the behavior of streams, the probability of flooding,
and areas li%ely to be flooded during high discharge, humans can underta%e
measures to reduce vulnerability to flooding. Among these non7structural
measures are5

Floodplain zoning 7 4aws can be passed that restrict construction and habitation of
floodplains. 9nstead floodplains can be zoned for agricultural use, recreation, or
other uses wherein lives and property are not endangered when :note that 9 did not
use the word if; flood waters re7occupy the floodplain.
Floodplain building codes 7 /tructures that are allowed within the floodplain could
be restricted those that can withstand the high velocity of flood waters and are high
enough off the ground to reduce ris% of contact with water.
Floodplain buyout programs 7 9n areas that have been recently flooded, it may be
more cost effective for the government, which usually pays for flood damage either
through subsidized flood insurance or direct disaster relief, to buy the rights to the
land rather than pay the cost of reconstruction and then have to pay again the ne$t
time the river floods.

!ortgage limitations 7 4ending institutions could refuse to give loans to buy or
construct dwellings or businesses in flood prone areas.