Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Re: Mistreatment of Seismology Issues

[Subject Prev][Subject Next][Thread Prev][Thread Next]
On 10-23-1998 Ted Smith of  California Department of Conservation wrote:

> Richter magnitude doesn't tell you anything about distance to the fault or
> epicenter, subsurface conditions, or your structural design parameters.
> Thus, were someone to attempt a simple declaration that the potential for a
> 7.5 M earthquake is seismically unacceptable, you'd have to ask "At what
> distance, for what ground conditions, and for what type of structure,
> occupancy, and use?" This essentially is the reason that the state codes
> (Seismic Hazards Mapping Act, for example) use the term "acceptable level of
> risk" < 

To put it in a very simple way, Ted Smith says that there are four (4)
parameters which must be considered to determine the seismicity issues which
should be taken into account for structures built in California. 

1. distance from the fault or epicenter
2. subsurface conditions
3. structural design parameters (type of structure)
4. occupancy and use for the structure.

I am looking at the "Fault Activity Map of California and Adjacent Areas"
published in 1994 by the Department of Conservation, Division of Mines and
Geology, compiled by Charles W. Jennings. 

To begin with, from the de-facto seismology point of view, California is
divided into two parts. Part One: All areas of California west from the San
Andreas fault, and Part Two: The remaining areas of California east from the
San Andreas fault. The reason for this division is: about 95% of all active or
inactive seismic faults in California find themselves "west from the San
Andreas fault". The exceptions are: The Greater San Francisco Bay Area East,
White Wolf Fault, Garlock Fault, and numerous historically active and inactive
faults north-east from Yucaipa-Cabazon (near Riverside). 

The "Fault Activity Map of California" distinguishes the following types
(kinds) of faults:

1. Late Quaternary Period / Historic; Years Before Present (YBP) = 200
    Evidence of displacements during historic time exists
2. Late Quaternary / Holocene; YBP = 10,000
    Evidence of displacements during this time exists
3. Late Quaternary / Pleistocene; YBP = 700,000
    Evidence of displacements exists
4. Early Quatemary / Pleistocenel; YBP = 1,600,000
    Evidence of faults within Sierra Nevada exists; but is not restricted to
Foothills fault system
5. Pre-Quaternary; YBP > 1,600,000
    There is evidence of recognized displacements. Faults are not necessarily

Looking at the "Fault Activity Map of California" with a magnifying glass or
without (the map is about 5'-6" ft tall and 4 ft wide), one can clearly see
that there is not one point on the surface of the "west of the San Andreas
part of California" that is located further than 10 miles away FROM ANY MAP
IDENTIFIED FAULT. It means that there is not one-100 square miles (10 miles by
10 miles) land surface in "western" California which is "fault-free" (immune
to seismic faults, immune to a possibility of an earthquakes occuring at any
time, anywhere in "western" California). One does not have to have an
overblown imagination to see "western" California as an irregularly formed
spider web consisting of hundreds of seismic faults stretching from the
Mexican border to Mendocino and beyond. All these faults can be designated as
"non-San-Andreas-faults". They can generate earthquakes of magnitude about 6.6
/ 6.8 anywhere in "western" California. 

Just to make things more interesting, to the best of my recollection (which
can be varified very easily), the City of Northridge was not positioned on an
active or inactive fault, or on any KNOWN active or non-active fault, as per
the U.S. Geological Survey or Department of Conservation maps. It means the
Northridge fault was a newly "discovered" fault that should not have been
there to begin with, and which could not be detected by current methods used
by modern seismologists. The lesson of this event teaches us that regardless
of the yet undetermined type of structure which is to be built in a certain
yet undetermined locality, the magnitude of future earthquake on map
identified or unidentified fault anywhere in California could follow a similar
traumatic experience. The only exception is the San Andreas "tunnel"
(corridor) where the expected magnitude is much higher.

Now, let us look at earthquakes occuring ON the San Andreas fault.  Since the
San Andreas fault is a major fault historically capable of generating
earthquaes up to 8.6 (or more), this fault is prominantly displayed and
identified on the Fault Activity Map of California. The magnitude of shaking
at any given locality, some distance away from the fault but caused by seismic
energy release of the San Andreas is inversely proportional to the distance
from its epicenter (either lineary, parabolically, elliptically, or anywhere
in between). Example: The investigated location (say, Pasadena) is 35 miles
away from San Andreas M7.4 quake at its epicenter. The distance between the
epicenter perpendicularly to the "long axis of the fault" and the point (on
the undefined shape line/curve) where the magnitude of the quake equals
zero/near zero is (say) 280 miles. Mathematically expressed: (M7.4 x
(280-35))/280 = M6.5 using linear magnitude dissipation (it is an unreasonably
rosy picture). When parabollic / elliptical magnitude dissipation is assumed
(much more realistic assumption), the resulting magnitude in Pasadena would be
M6.9 / M7.2. (see Properties of Parabolla / Ellipse in any "steel book"). Thus
we can establish the magnitude of shaking at any location in California due to
the San Andreas fault seismic energy release mathematically.

The Tunnel -  By comparing numbers of earth shakings generated by the San
Andreas fault anywhere within say, a 70 mile wide tunnel running on top of the
San Andreas (35 miles on each side of the San Andreas longitudinal axis)
against expected earthquakes magnitude in the "western" part of California, it
becomes obvious that California, from the structural engineering point of view
should be divided into three parts.

Part 1.  The land areas within say, a 70 mile wide corridor running on top of
the San Andreas longitudinal axis. Minimum design magnitude should be greater
than 7.3. (Historical earthquakes that occured within this corridor area are:
San Francisco, 1906 - M8.2; Kern County, 1952 - M7.7; Loma Prieta, 1989 -
M7.1; Landers, 1992 - M7.3). 

Part 2.  All areas west from the San Andreas fault (reasonable design
magnitude should be somewhere around M6.8) except for the corridor (tunnel)
area described under Part 1. 
Part 3.  Relatively small areas designated as "exceptions" in the fifth
paragraph of this message. The "exceptions" require individual treatment as
far as the magnitude of the earthquakes is concerned.

What is interesting about the myth of "acceptable level of risk" concept is
that after you took a walk, after the Northridge disaster, along Ventura Blvd
in Sherman Oaks or Encino which are located about 10 miles south from
Northridge (all three communities lie within the San Fernando Valley, near Los
Angeles), and observed hundreds of badly damaged or partially collapsed
commercial structures, then walked along the side streets where many single
story wood construction residences and almost EVERY 2, 3 or 4 story apartment
structure (and there were hundreds of them) were red-tagged and every tenant
was forced to leave the area, you knew that something very basic is wrong with
the concept of "acceptable level of risk". In my judgement, the codes under
which all these structures were designed were wrong, and in many cases, they
were deadly wrong. For all practical purposes, many of these areas became
ghost towns until they were rebuilt which took on the average 9-12 months or
longer. I do not want to talk here about collapsed bridges and other "heavy
duty" damage all over the Los Angeles county as everyone knows what happened
by having watched television after the event. One characteristic of the damage
was clearly visible: the uneven distribution of damage was very much related
to soil characteristics which differed from area to area (GEOLOGY part of the

The above comments are related to Ted Smith's positions 1, 3 and 4. With due
respect to his position and experience, my opinion still stands that the
"acceptable level of risk" concept is misleading, harmful and meaningless. 

What remains to be determined is the "subsurface conditions" (Ted Smith's
position # 2). Since I am not a geologist I cannot say anything about the
subject. However, it seems that under the worst but realistic and proven
seismic conditions as described above, the "subsurface conditions" is the only
variable that has to be individually established for every structure built in