About earthquake engineering

This is about science produced by the California Institute of Technology

and originally broadcast by station KPCC Pasadena California. The

programs are made available to this station by national educational radio.

This program is about earthquake engineering with host Dr. Robert

McGregor Leon and his guest Dr. Paul Jennings professor of applied mechanics.

Here now is Dr. McGrath at.

The start I would call the first thing you better do is tell me just what you mean by earthquake

engineering.

Well by earthquake engineering what we mean is a study of the special problems that are related to

the protection of structures against earthquakes.

Well certainly in this area of California Southern California we're very

conscious of earthquakes. It seems to be a long history of

such experiences. How frequent are earth quakes throughout the world.

Well they are very frequent. It's been estimated that maybe a million of them occur a year

but fortunately very few of these perhaps 100000 of them are strong

enough to be felt by anyone. And of the ones that are felt very few

of those perhaps 100 of them are such that they might cause damage.

The 100 or so damage which occurs

is measurable in terms of structural damage building

destruction loss of human life. Is this what you

mean in terms of highly populated areas.

Well one of the problems is that the people are.

Increasing and as a result more and more of the seismically active

areas of the world are becoming populated so that the number of

earthquakes that could be damaging increases because people are living in more

places even if the number of earthquakes that are happening is relatively constant.

As the years go by we can expect somewhat more more problems in this area because

people are now living in places they didn't live before.

Just moving out to these more active areas. What are some of the active areas throughout the world.

Well most of the strong earthquakes are about 80 percent of them occur around the

Pacific Ocean and this is often called the ring of

fire ring of fire because an earthquake activity is also

often associated with volcanism active volcanoes and there are many of these

around.

Perimeter of the Pacific Ocean so extending all the way from white Australia

through the eastern coast of Asia up through Alaska back

oh yes Australia not so much but New Zealand is very active Eisley

has of course is Japan.

They can check out an insular Russia illusion islands are

extremely active all along the coast Pacific coast all the way

down to Mexico and beyond into South America through the western coast

of the United States and down into South America right about other areas into

Europe and Africa. Yes there's another belt of earthquake.

Occurrence called the Alpine belt that runs through Turkey

India northern India Russia Tashkent is on the northern

portion of this earthquake belt and into Outer Mongolia.

Do we have any idea how recent this activity is is it something for example the Pacific

Ring of Fire as we call it. Is this something that's been going on for millions of years.

Is it more recent than the one through the Near East.

Well you have to talk to a geologist to say for sure we have reason to think

it's been going on for a long time in Japan where our historic records are the longest.

There's been lots of earthquake activity at roughly a constant rate for

several centuries. The hymn records back to.

Oh I think maybe 3 or 500 B.C.

So it's been going on for him for a long time.

So certainly we have a good deal of recorded history. Certainly in more

recent times. But you say that you don't know personally the

long geologic history of how these chains have developed here.

No this is tectonics outside of my area

so I'd rather not comment on that. My turn to the damaging

earthquakes even this is more the problem of the earthquake engineer.

We're very interested in what type of motions cause damage and how they reach

doctors respond to these motions.

For one think how do you measure the magnitude of an earthquake. You mention the

frequency but the question of the magnitude.

Well the common magnitude measurement is it on the

Richter scale which measures the maximum

trace on it. That aside I want the sized model just

instrument gives him this is not. Too great a value

for the earthquake engineer because what we're interested in is a slightly different concept we're interested

in the strong motion right at the site of the structure. Now the

seismologist instruments will not give you this. These are very sensitive instruments that are

set to major earthquakes that have occurred very far away and what

they're trying to do is find out something about the construction of the earth the origin of

the quakes and so on.

What is the physical quantity that's measured is that an acceleration as it isn't it's an acceleration

because this by Newton's Law is most directly related to the forces or structure

would experience during an earthquake so that the instruments that the

engineer uses in earthquake design think. I believe this must be the feeling which

are immediately concerned here. Our device is much like that of the seismologist but

specifically designed and located.

Well yes we we have somewhat different breed of cat in our strong ocean instrument in

that it's designed to record a motion much stronger first in the.

Size seismologist instrument and secondly it what we

want to get a little better resolution so we have to run the paper faster it records on

paper the instruments we have now. What you would mean by resolution is you want more detail we want more.

Yes we want more detail in the traces of the motion because we make rather extensive

analyses of these records to find out how they would shake different simple

structures to get an idea of what happened at the site and what would happen if this

earthquake occurred somewhere else to tell us what kind of effect it would have on certain

structures.

Well you mention it in that you measure accelerations what might be the magnitude of a

large earthquake in terms of acceleration me magnitude might be

as high as.

50 percent gravity in fact we recorded that Coast and Geodetic surveys

instruments recorded that during the Parkfield earthquake. This was a smaller earthquake Richter

magnitude about five. It was just one one or two pulses in

one of those pulses an acceleration of half. It has a gravity which meant

that if your shoes are nailed to the floor that you just receive a force

sideways or roughly half your weight.

So that would be considered an extremely large.

Well this is a large acceleration. Fortunately this earthquake didn't last very long

and there is very very little structural damage associated with this or earthquake so that the

time period and as innocent as the duration of the exit ation is important.

We have reason to believe that even in a large earthquake such as the Alaskan

earthquake that the accelerations didn't even at the fault didn't exceed

50 percent G perhaps or in that magnitude but the duration was quite long

maybe 45 50 seconds for the strongest portion and the perceptible motion

last for 5 minutes. If you have the strong separation for a long time well

then this can be very damaging to structures and of course it was in the last minute.

And you mentioned earlier that the.

Half acceleration I think and gravitational acceleration

field that had been experienced in terms of short pulses that might be what he would have seconds

then.

Well yes a second at the instrument There was right close to the fault.

The acceptation was just a couple three seconds long the strong part of it so that

something that's approaching a minute or many minutes down is the one that we're very much concerned with right

structures in what way are structures affected by these

accelerations.

Well basically two things can happen there. One of the problems is with the foundation

in the soil itself. For example during the Alaskan earthquake there were many

slides larger slides you may remember the pictures in the papers. You know you can arm

slide for instance a very large slide in a nice residential district

and there were submarine slides on the waterfront areas and some of the small towns around that means

submarine. Under the Earth immediately under the

water surface near the shore of some of the small towns as I was on a

very steep slope and slid into the ocean and taking some of the waterfront structures with

it. And as a result also I generated a local wave which

was anywhere from 10 to 30 feet high and this way then came back

over the land and caused great damage to the thing.

So here you are talking about the displacement of the material of the earth.

Yes and what it does in disrupting to mandate it and you have a building on top of us of a

major slide. There's just not much you can do right

now.

Is there any other aspect of this. Oh yes assuming now that the foundation

behaved itself the soil doesn't fail in that matter.

Then you can talk about the shaking of the building and if the shaking is strong

enough some buildings receive damage it's very hard to relate

the damage you get or the damage that you see to the

acceleration unless you have an instrument because it depends so much on the quality of construction

the structures.

Especially someone they have in foreign countries such as say Agadir Morocco where there's a

disastrous earthquake while back consisted mainly of boulders piled on

top of each other and during the earthquake even a smaller earthquake is sufficient to

set some of these down.

I would making any. Large scale attempt to gain

experience and records the nature of earthquakes magnitude frequency of some of

these vibrations which occur.

Yes the frequency in occurrence of earthquakes is a study that has been

taken care of pretty well by a seismologist there very interested of course in the

results of earthquakes as well as what they can learn from them about why they occur and what the

character of the earth is. But the recording of strong ground motion

has been the province of the earthquake engineer and at Caltech

and elsewhere were very active in encouraging people to put in instruments Coast and

Geodetic Survey has an instrument program and we received a set of very

interesting records from the recent Parkfield earthquake.

How many local governments have they been participating in this investigation.

Yes the city of Los Angeles recently passed an ordinance whereby

buildings over I believe 10 stories have to have three accelerometers in

them. Next an accelerometer is the name we use for our instrument and this instrument

records the acceleration and at the rough and middle and in the basement building.

Now in the event of an earthquake and we and these instruments work well we have

records that will help us in this and determining the.

The accuracy of the methods of analysis and design that we use. I should mention too that these

instruments have been developed over a period of several years and are quite reliable.

These are checked periodically you talk about these as accelerometers and such. I

presume that these are the essential sensing element of what we

call the seismologist instrument.

Well it's it's a similar sort of thing it has a different technical characteristics

because we're looking at something that is a little bit different in terms of our motion

our instruments that sit there and wait for the earthquake to set them off. I see so there's a sort of

triggering. Yeah see it's a little pendulum there that any acceleration reaches a certain

set level that the instruments start and then it will run for a while until it senses that there is

no more strong shaking you know will shut itself off and shut itself off so that you have a continuous

graphic display that.

Right. You mentioned Los Angeles is that has this become widespread throughout the

United States.

No I believe that Los Angeles is unique in requiring that more tall

buildings at these accelerometers be in strong because we certainly have that for

motivation.

Yes yes the San Andreas Fault is here to remind us that

earthquakes have occurred in this area.

What sort of problems does the

earthquake engineering counter. This matter of the

foundation design what aspects of the problem can he contend with.

Well he has a very formidable problem there because soil is a very

difficult thing to work with and that from a analytical point of view in trying

to decide what its properties are this in itself is very difficult. Some of the things that

can happen to the foundation in addition to the slides is the

lip liquefaction. And this is especially prominent in recent

earthquake in Japan.

What do you mean by liquefaction.

Well what happens is that the sand is considered now sand thats

saturated with water the water table is maybe three or four feet below the ground you have a couple

hundred feet of sand.

Well if this sand is not quite dense during the shaking the sand grains can come apart from

each other so that the.

The whole soil then assumes much of the properties of a fluid and it can

flow. For example in the recent got earthquake in Japan

buildings tilted several degrees hundreds of buildings tilted some

buildings and tilted to the extent that they had to be taken down one

of them just layed slowly over on its side.

Yes I think I remember seeing some editors of that one woman was hanging or washing up on the

roof of this building and was able to walk down the side of the building after it slowly

rotated and walked off into the building and promptly fainted.

How about with regard to the structure then you talk about the foundation problems and

the displacement of the soil ready earth beneath the building and the

liquefaction problem. How about the design of the building itself to withstand a vibration.

What kind of forces for example were the nature of the forces that are induced by this acceleration.

Why was this the earth shaking is three dimensional in that

it shakes north south east west and up and down. So which are the

predominant forces in the lateral forces. I think what you could call on are south and

east west are the stronger ones.

About two to three times usually the vertical acceleration going down sways back and

forwards it so that an up and down.

Yes it's a swaying motion back and forth that we're worried about. Buildings generally

have a.

Sufficient factor of safety for vertical vertical loads anyway so

that this is not considered a critical problem. These are lateral

forces that are distributed over the building. They are proportional to

the weight of the structure because it isn't what we call the inertial loading and it's

proportional to how strong an earthquake you have in attempting to design

for this kind of a condition.

The engineers in present times were in terms of what we call dynamic on

houses literally.

No no you have to realize that in civil engineering

structures like buildings and dams are not unique and so the engineer has a limited amount of time to

devote to the design of each one it's not like an aircraft we going to design build

hundreds of them he can analyze one thing in detail. The

design is based on building codes which represent a standard

rules. Yes these are standard rules and Compendium of experience with

Pastor earthquakes. However special structures where the

risk might be very high and you want to be extremely sure such as say nuclear

reactors in a dynamic analysis is me so that they literally do a computation of the

motion of the building under certain types of forces or perhaps the weight of them ASA fee it is it would

be lateral forces are based on.

The same sort of configuration that you'd expect in a dynamic response. So you apply a

lateral load is proportional the weight of the structure and whose magnitude is specified by the code.

Are there any factors such as the shape or the general configuration of the building

which enters into such question.

Yes from a dynamic point of view here thinking of your building

a swing back and forth. It's important that the resistance of the

building piece symmetric in that you want to have resistance where you have

weight of the structure so that the building will not tend to vibrate

torsional your twist twisting motion when it occurs can be very

damaging and Hastert quakes we've seen cases where this is cause that your

buildings. So you like to keep the structure.

Dynamically symmetrical by that I mean that these differences are distributed in the same way that the way it is

so that regards the which direction the earthquake came with me motion came the

building it would have uniforms and direction.

Another property of interest to me talking about building design is to separate buildings especially

those of different heights because when the earthquake comes along the building will tend to

sway at a period determined by its properties and the building next

door will have different properties and sway at a different period.

If these aren't separated by enough room they can hammer together and cause quite a bit of damage I see so that

literally the proximity then leads to interaction or pounding on the engine

of a hammering.

You want they want to generally are separated by a few inches to a foot

depending on the situation.

I don't in addition to these codes which you mention which sort of rules of

design or certain principles in terms of the structural

sense now that I applied in deciding whether or not a building is well

configured for earthquake resistance.

Well we're it's way has been called the great inspector and that

it seems to seek out any weak point in your structure.

So what you want to have of course then is that the structure is an enterable unit and

to have a good structure for earthquake purposes it's very important to design

the joint so that the structure does act together and doesn't come apart because of a weak join or a

lack of attention to some detail. So this is very important in

the Alaskan earthquake. Some of the failures could be laid to

inattention to detail with a particular elements were joined together poorly and came apart during the

earthquake and then want to starts going apart at a much weaker structure and if

the earthquake continues for a long time as it did in Anchorage Well then it would shake the structure down.

What is the.

The level to which you design you attempt to designed to resist

earthquakes are 99 percent Well no I was not kind of economically

feasible on the present philosophy is that for the smaller earthquakes

which you might have many of in the lifetime of a building that he mentions for example a

hundred thousand with my family's home number that one would encounter.

Well not in a particular that's 100000 in the world doing well but

particular building depending on where it's located and you might be able to shake it three or four times during his

lifetime with smaller earthquakes. It's expected that a well-designed building would not

exhibit any damage to something of that magnitude so it would be completely resilient to a

smaller earthquake a small rise. However it's just not economically feasible in art

and in fact desirable to design it to remain completely elastic during the

strongest earthquake because. Two reasons the cost of

it would be prohibitive. Since you are designing against something that may or may not happen

we don't know that much about the occurrence of earthquakes within any reason you know time span like

what 50 years or 100 years we don't know if an ice quit going to come.

The other thing to consider of this is the deflections. If you have a

completely elastic structure and a strong earthquake it may shake quite

a bit in the amplitudes may build up.

So the idea is used that in the strong earthquakes we rely upon

the yielding of some members and the plastic action

is called to dissipate energy even in Lebanon bending.

Yes the structures can be designed to be very ductile so that they will

retain their strength even after they're bending just like a Cuesta coat hanger.

Yes you can twist it many times before it breaks and it takes energy to do that well this is the sort of thing

that where it is considered and design of buildings with a strong earthquake comes along

willing to accept some damage no loss of life or or a

hazard due to the persons involved but some damage can be

accepted in view of the rarity of the event.

So for these more severe situations then you have designed a building so that it may

suffer some permanent affirmation and literally be twisted out of shape but would not collapse.

You know it would not collapse and it would be repairable.

What is the frequency of earthquakes in this area and time.

Yes I think I'm just quote done from studies that some other people have made.

Clarence Dr. Allen turns on the psychological lab and

Richter I'm adding the scales name and some other people have done some

studies on this and the data are very scant

and you had to just kind of take a rough average over what you've experienced in the past

but it appears that somewhere in southern California area including

certain parts of northern Mexico you can expect a greater earthquake

Richter magnitude seven and three quarters or greater.

About once every 50 years 2050 would this be of the magnitude and the cause permanent

defamation and buildings.

Yes if they were close to the fault you'd expect. Some

signs of distress do not collapse and even in a well-designed building.

How about these much lighter earthquakes and we experience them every so often. I've

never read it. I tend to cultivate keep track of them. How many of those do we really experience in this area.

I want to plans on a size you're thinking of something that might be noticeable.

Well you might say oh and I'm going to 3 earthquake might be noticeable

bending on the local area.

Some places are even and so I go for near much more quiet than others. You might get three or four of

those a year and be strong enough to so that you'd notice and readily see

what sort of damage is caused by these major earthquakes.

What are we talking about in terms of the economics.

Well I think you turned out a very significant economic loss

in the Alaskan earthquake the damage it's very hard to spin it down to

somewhere between 300 and 500 million dollars. The

earthquake of the same year in the God of Japan the damages been estimated about 1 billion

dollars. I see.

How about some of the more recent earthquakes in a southern California area.

They have been damaged in an extremely minor I think

of course we haven't had a real strong earthquake either since a Long Beach earthquake in 1933

was that tens of millions on earth was about 40 million dollars dollars and at

that time.

What is the nature of the work that's going on today Paul and in connection with the research

associated with the design of the structures and so on.

Well I should point out that at Caltech we don't actually design structures this is the job

of the practicing engineer what we're working on is the research problems associated

with Earthquake Engineering attempt to understand a problem yes we're trying to understand the problem we're

trying to understand what kind of motions can occur during earthquakes.

So we try to encourage people putting out

accelerometers so they can measure the motion we perform

analyses on this motion. We try to understand some of the soil

effects that can happen. And we spend a lot of time working on the structural

dynamics to try to find out why given certain motion what can this do

to special structures try to understand just what it is it goes on during during

the earthquake.

And you mentioned. Building structures are unique and the

response of these buildings will be different one from the other so that you don't really have a standard

approach to the problem.

When I feel not quite fair we have a standard approach but it has to be a rather general approach

because it has to encompass for the practicing engineer has to compass encompass

a variety of buildings that have come up again so that you can translate into the set of code.

Well yes there are dogs and set of principles to go by where they

decide depending on the type of construction they use what kind of forces should be

designed for.

Is all of this work that's done in the research area and all the tickler

mathematical.

There's been no knowing one of the recent things it's been done at Cal

Tech is the development of machines to shake buildings this was done for

the State Department of Architecture and these building the building shakers

can be placed in a building and they will then put a very

small force on this building and shake it back and forth.

And this helps us determine what the period of the building as the oscillation character is yes the

oscillation characteristics and what kind of shape it assumes when it sways back and forth.

As I remember you did something like this would a dam. Yes Rita yes the.

I can't recall a good book a Canyon Dam was shaken this way to see what kind of

things would happen there to see if it were something that we could learn by shaking dams. It

appears as if you could pick up some of the dynamic properties of a dam by shaking it so that there is

literally field testing if you like.

Yes this is very true and we learn a lot about structures when we can.

But these machines on and shake them doesn't make sense to talk about the dynamical properties

in terms of modeling. It can work with small scale devices. I don't have to go

to full size building.

You can work with small scale devices the problem is getting the scales

correct because they are the dynamical characteristics and he has one of the problems as I mentioned before was with the

joints of the structure and to scale the joints satisfactorily is very

difficult to get a little to duplicate the joints of a steel building

would be very difficult on a very small scale so that a lot of problems and so this business of having it

in or go design which is uniform in it's well balanced in its

characteristics throughout so that there are not weak points in it to to simulate this in a

model is a difficult piece difficult there's some significant work being done now

in this area. One of the things we like to do too and I think

well the Japanese are starting this problem is to build a big platform about 50 foot square so

that I draw likely shaking the base of it you could build it to a full size structure

on and shake it.

This was about science with host Dr. Robert McGregor Leon and his guest

Dr. Paul Jennings professor of applied mechanics at the California Institute of Technology.

Join us again for our next program when two more members of the Cal Tech faculty will

discuss a subject of interest about science is produced by the

California Institute of Technology and is originally broadcast by station

KPP C. Pasadena California. The programs are made available to

this station by national educational radio. This is the national

educational radio network.