Master Course

Basics of Seismology and Seismic Hazard

Assessment

[Link]@[Link]

3. Seismometers and seismograms.

Epicentral location

April 2026
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

SEISMOMETRY..................................................................................................................... 3
Early instruments 3
LINEAR OSCILLATORS .......................................................................................................... 4
SDOF damped forced vibrations........................................................ 4
Displacement and acceleration response ratios................................. 5
SEISMOGRAPHS .................................................................................................................. 6
Old seismographs.............................................................................. 6
Optical and electromagnetic seismographs ....................................... 7
Digital seismographs ......................................................................... 8
SEISMOMETRIC NETWORKS .................................................................................................. 9
World-Wide Standard Seismograph Network- WWSSN .................... 9
Global Seismographic Network (GSN)............................................. 10
Seismometric network of Dominican Republic ................................. 10
Real time seismicity on WEB ........................................................... 11
Distribution of Earthquakes in Dominican Republic (1907 — 2025) . 12
Software Seismic eruption ............................................................... 13
SEISMOGRAM INTERPRETATION ...........................................................................................13
TRAVEL TIMES ....................................................................................................................16
LOCATING EARTHQUAKES ..................................................................................................18
Graphical method of earthquake location ........................................ 19
Rigorous determination of origin parameters ................................... 19

F. Sabetta 3.2
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Seismometry
Quantitative analysis of seismic waves requires that the vibrations be instrumentally recorded.
The instrumentation must:
• be able to detect transient vibrations within a moving reference frame;
• operate continuously with absolute UTC (GMT) time so that there is a common reference for
stations throughout the world;
• have a fully known response to ground motion (instrument calibration).
Such a recording system is called a seismograph and the sensor converting ground motions in
some form of signal is called seismometer or geophone.
The design and development of seismic recording systems is called seismometry

Early instruments
The first known instrument for earthquakes measurement is the Chang seismoscope built in
China in 132 B.C. Balls were held in the dragons’ mouths by lever devices connected to an
internal pendulum. The direction of the epicenter was
reputed to be indicated by the first ball released.

The first true seismograph, which recorded the relative

motion of a pendulum and the Earth as a function of
time, was built by Filippo Cecchi in Italy in 1875. The
oldest known record produced with this system is
dated February 23 1887.

A period of rapid instrument development began after

1880 due to a group of British seismologists working in
Japan (John Milne, James Ewing, and Thomas Gray)
In these early seismographs mechanical or optical systems amplified the mass motion and friction
provided the only damping of the pendulum oscillators. By 1900 the first array of 40
photographically recording seismographs built by John Milne provided the initial data base for
studying elastic waves, earthquake sources, and Earth structure.

F. Sabetta 3.3
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Linear oscillators

The simplest type of seismograph can be

illustrated by a mass-spring-damper single
degree-of-freedom (SDOF) system whose
response to shaking is given by the equation
of motion:

  cu  ku  mu
mu g

Where u is the seismograph displacement

(relative motion between base and oscillator)
and ug is the ground displacement.

Dividing by m and introducing the undamped natural

circular frequency 0 = k/m and the damping ratio  c c c c0
   
(ratio of the damping coefficient c to the critical damping cc 2 km 2m0 2k
coefficient cc)

We get the differential equation of motion of a damped SDOF oscillator

in terms of relative displacement u:
  20u  02u  u
u g
SDOF damped forced vibrations
If the ground acceleration is simply harmonic at a circular frequency g

ug (t )  A sin( g t ) u g (t )   A /  g2 sin( g t )

u  e  0t (C1 sin  d t  C 2 cos  d t ) 

A 1
k (1   )  ( 2 )
2 2 2

(1   2 ) sin  g t  2 cos  g t 
A sum of a damped transient response, that rapidly dies out, and of a steady-state response at
the frequency of the applied harmonic load but out of phase with the loading

A
u H sin( g t   )
k

1  2 
H   tan 1  
2 
1  
(1   )  (2 )
2 2 2

F. Sabetta 3.4
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

A u(t) 2g
u  H sin(gt  ) k=0m D(g) 
k ug(t) (20  2g)2  (20g)2
u(t) g
V(g) 
1  2  u g(t) (20  2g)2  (20g)2
H   tan1 
2
2 2
(1   )  (2) 2
 1   u(t) 1
A(g) 
g(t) (2  2 )2  (2  )2
u 0 g 0 g

Max when g = 0(1-22)1/2

Displacement and acceleration response ratios

Displacement response ratio The displacement response ratio (the ratio of
trace displacement amplitude to ground
displacement amplitude) is represented by:

For ground motion frequencies well above the

natural freq. of the seismograph (i.e. large values of
) the trace amplitude correspond to the ground
motion amplitude.

The acceleration response ratio (the ratio of trace

Acceleration response ratio (0 =1 displ. amplitude to ground acceleration amplitude)
is given by:

The same physical system can act as both a

displacement seismograph (0<<g) and an
accelerograph (0>>g). It measures
displacements at frequencies well above, and
accelerations at frequencies well below its
 =g/0 natural frequency.

F. Sabetta 3.5
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Seismographs
Old seismographs

1906 San
Francisco
earthquake

F. Sabetta 3.6
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Optical and electromagnetic seismographs

from Longwell et al. 1969

In 1914 B. Galitzin introduced an electromagnetic moving coil transducer to convert

pendulum mass motion into an electric current, proportional to the mass velocity and used to
rotate a galvanometer coil. Light reflected from a mirror on the galvanometer coil was
recorded on photographic paper and a long optical lever arm was used to produce large
magnification. This type of electromagnetic system has dominated instrumentation in the past
century with the optical recording eliminating friction. The product of the pendulum, transducer,
and galvanometer frequency responses (each one having different damping and resonant
frequency), controls the shape of the overall response allowing to emphasize a particular
frequency pass-band.

The Wood Anderson seismograph (1922), used by C. Richter in the early 1930s to develop the
first magnitude scale, used a small mass suspended eccentrically on a thin tungsten torsion wire.
A mirror attached to the wire allowed optical recording with a ground motion magnification of
2800. Damping was provided electromagnetically at 80% of critical. The damped natural period
was about 0.8 sec.

F. Sabetta 3.7
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

from Lay and Wallace- 1995

Digital seismographs

UTC time
signal

Data acquisition
and A/D converter

Sensor

In the early 1970s seismic recording systems began to forgo the analogue recording, in favour of
digital recording on magnetic tape or solid state memory. This is generally achieved trough
an A/D converter, sampling the output current from the sensor. The most recent seismometers
are generally characterised by broadband (flat response over wide ranges of period 0.1-100
sec.) and high dynamic range (up to 140 dB) due to the force-feedback strategy that greatly
extends the bandwidth and linearity of a seismometer because the mass cannot make large
excursions that bend the springs or levers.

F. Sabetta 3.8
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Seismographs are generally located on hard rock sites (to avoid modification of the seismic waves
by soil layers) and remote from sources of noise such as traffic, machinery and sea waves.

Seismometric networks
World-Wide Standard Seismograph Network- WWSSN
Understanding of earthquakes, Earth structure, and tectonic processes improved dramatically with
the establishment of the World-Wide Standard Seismograph Network (WWSSN) in 1961. Before
that time data were obtained from a wide variety of different instruments and organizations.

The WWSSN used standardized instrumentation that consisted of 120 stations in 60 countries,
each equipped with accurate chronometry, a triaxial electromagnetic short-period
seismograph and a triaxial long-period seismograph. The WWSSN recordings have been
extensively used because the original photographic records were filmed on microfiches and
provided to major international seismic data centers. The establishment of the WWSSN had much
to do with monitoring atomic test ban treaties but the benefits for seismology were enormous,
coming at the time of plate tectonics revolution, when accurate recordings were critical for
determining faulting patterns.

During 1980’s analogue seismographs were replaced by digital ones.

F. Sabetta 3.9
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Global Seismographic Network (GSN)

During 1980’s analogue seismographs were replaced by digital ones.
Another great imrovement in data collecttion has been made with the birth of the Incorporated Research
Institutions for Seismology (IRIS), a consortium of United States Universities that have research programs in
seismology. IRIS, jointly with USGS, is operating the Global Seismographic Network (GSN)
[Link]
The goal of the GSN is to deploy digital recording stations uniformly over the earth's surface. As of 2003, the
GSN was made up of over 128 stations. Eight new stations are planned for completion in 2003-2005.
.
These stations continuously
record seismic data from very
broad band seismometers at 20
samples per second.
It is also the goal of the GSN to
record data with a dynamic range
of 140 db (24 bit digitizers).
These instruments can record
both Earth tides and high
frequency body waves, ranging
from minimum Earth noise up
to the accelerations expected
for a magnitude 9.5 earthquake
3300 km away.

Seismometric network of Dominican Republic

F. Sabetta 3.10
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Real time seismicity on WEB

F. Sabetta 3.11
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Distribution of Earthquakes in Dominican Republic (1907 — 2025)

F. Sabetta 3.12
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Software Seismic eruption

Seismogram interpretation
Our knowledge of the Earth velocity structure and of the seismic sources is the result of
interpreting seismograms that are a complicated mixture of source radiation effects,
propagation phenomena and frequency band-limiting effects of the recording instrument.

The main task of the observatory seismologists is to identify the arrivals of different waves and
phases on the seismograms and note the exact time of their arrivals. This is the result of
nearly a century of experience in reading millions of seismograms (I.S.C. data base has more
than 7 million arrival times provided by 3000 worldwide stations since 1964. The dimension of the
IRIS or ORFEUS archives is of the order of tera-bytes).

Main International Agencies managing seismic networks and collecting and disseminating
seismometric data:
• IRIS - Incorporated Research Institutions for Seismology - [Link]
• USGS - U.S. Geological Survey - [Link]
• ISC - International Seismological Centre - [Link]
• NEIC - National Earthquake Information Centre - www. [Link]

F. Sabetta 3.13
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

• ORFEUS-Observatories and Research Facilities for European [Link]

• EMSC - European Mediterranean Seismological Centre - [Link]

The timing of various

horiz. arrivals is a predictable
comp. function of the distance and
depth of the seismic
source. It’s an “interactive”
and “loop process”
because, to locate phases,
you need an Earth velocity
vertical model or travel time
comp. tables and the epicentral
location that are, on their
turn, derived from
seismograms.

Help in identifying the different wave arrivals and

P, R implementing travel times curves can be provided by
seismic reflection and refraction surveys using
S, L underground explosions (in this case exact origin time
and spatial location of the source are known).Another
help comes from looking at the different (horiz. or vert.)
components of the ground motion.

F. Sabetta 3.14
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Vertical components R1,R2,R3: Rayleigh waves Transverse components G1,G2,G3: Love waves

Note that distances are measured in angular degrees from the source; a simple conversion to
remember is that on the Earth’s surface, 1° latitude is approximately equal to 111 km.

F. Sabetta 3.15
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Travel times

Travel-time curves for

teleseismic distances derived
empirically from recordings on
seismographs (Willmore,
1979).

At short epicentral distances seismograms are dominated by the details of the highly
heterogeneous crustal structure.
At large distances by the relatively simpler velocity structure of the deep mantle and core.
There are 4 classification of seism. based on epic. distance:

1. Local distances < 1° R<100 km

2. Regional distances 1°<< 13° 100 < R< 1400 km

3. Upper mantle distances 13°<<30° 1400 < R< 3300 km

4. Teleseismic distances > 30° R > 3300 km

F. Sabetta 3.16
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Magnitude 9.0 OFF W COAST OF NORTHERN SUMATRA

Sunday, December 26, 2004 at 00:58:53 UTC
Preliminary Earthquake Report
U.S. Geological Survey, National Earthquake Information Center

The magnitude 9.0 earthquake near

Sumatra on December 26, 2004 was one of
the most significant seismic events on Earth
Magnitude 9.0 OFF W COAST OF during the past 100 years.
NORTHERN SUMATRA
U.S. Geological Survey, National This plot displays worldwide records of
Earthquake Information Center GSN seismometers (note the 1 cm scale
bar at the bottom for scale). The traces are
arranged by distance from the epicenter in
degrees.

The earliest signal is that of (P) wave,

which takes about 22 minutes to reach the
other side of the planet (the antipode). The
largest signals are surface waves which
reach the antipode after about 100 minutes.
The surface waves can be clearly seen to
circle the planet and return to the epicentral
region after about 200 minutes. A major
aftershock (magnitude 7.1) can be seen at
the closest stations starting just after the
200 minute mark (note the relative size of
this aftershock, which would be considered
a major earthquake under ordinary
circumstances, compared to the
mainshock).

F. Sabetta 3.17
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Locating Earthquakes

Focus or hypocenter is the point of initiation of

rupture on the fault plane.
Asperities are discontinuities on the fault plane
where stresses may concentrate; these can rupture
prior to a large earthquake giving rise to
foreshocks.
Other discontinuities, called barriers, can remain
unbroken after an earthquake, subsequently
rupturing as aftershocks (Bolt,1993).

One of the most important tasks in observational

seismology is locating seismic [Link] location
in space is defined by the geographical co-
ordinates of the epicentre (N°, E°) and the vertical distance between the epicentre and the focus,
called the focal or hypocentral depth h.
An earthquake is fixed in time by the moment at which the rupture began and the first waves
were radiated from the source. This is known as the origin time t0, which is always given in
U.T.C.
Therefore, the complete location of the earthquake is given by 4 parameters: the latitude
and longitude of the epicentre, the focal depth and the origin time.
In general what is required is the identification of seismic phases and measuring their arrival
times, as well as knowing the velocity structure between the hypocenter and the seismic station.
In general the arrival times of various seismic phases at many seismic stations are required to
locate an earthquake
accurately, but even
from a single station
it’s possible to obtain a
rough estimate of the
hypocentral distance.

The rule of thumb

tp ts D= ts-p x 8
is very crude but
nonetheless could be
D D 1 1
tp  ; ts  t s  t p  D(  ) very useful as a first
Vp Vs Vs V p estimate of the dis-
tance in a given seis-
mogram or accelero-
VsV p Vp gram.
D  ts p  ts p  t s  p x8 Knowing the distance
V p  Vs 3 1 and entering the travel
times available for the
region under consid-
D= hypocentral dist. in km eration, one can esti-
ts-p = arrivals difference in sec. mate the P travel time
and thereby determine
the origin time of the
Assuming constant velocities, Poisson solid
earthquake.
(Vs=Vp/√3), and Vp= 5.8 km/s

F. Sabetta 3.18
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Graphical method of earthquake location

If records are available from a minimum of three stations, with reasonably good azimuthal
distribution, the location of the epicentre, for crustal events, can be estimated with the method of
the circles.

modified from Press and Siever (1994)

If the arcs do not intersect at a clearly defined point but overshoot by similar amounts to define a
triangle it may indicate that the focus is at some depth below the Earth’s surface, since D is
actually the focal rather than the epicentral distance.
This method for locating the origin of an earthquake is clearly very crude and would rarely
be used in practice other than to obtain a preliminary estimate of the location.

Rigorous determination of origin parameters

The procedure is simple : guess a location, depth and origin time; compare the predicted arrival
times of the wave from
your guessed location
with the observed times
at each station; then
move the location a little
in the direction that
reduces the difference
between the observed
and calculated times.
Then repeat this
procedure.
[Link]
o/eqlocation/[Link]

Consider several globally

distributed seismic
stations. We need to
determine 4 unknowns: the 3 hypocentral coordinates and the origin time.

F. Sabetta 3.19
Basics of Seismology and Seismic Hazard Assessment 3.-Seismometry

Mathematically, the problem is solved by setting up a system of linear equations, one for each
station. The equations express the difference between the observed arrival times and those
calculated from the initial estimate of the hypocenter.
We must also have a model of the crustal velocities under the seismic network to calculate the
travel times of waves from an earthquake at a given depth to a station at a given distance.

The process begins with an initial guessed hypocenter, performs several hypocentral
adjustments, each found by a least squares solution to the system of linear equations, and
iterates to a hypocenter that best fits the observed set of wave arrival times at the stations of the
seismic network.

Each station reports its own location (Ni, Ei) and the P wave arrival time ti . We assume an initial
estimate of the origin parameters (N0, E0, h0, t0) and then follow these steps:
1. The distance of each station, i°, from the assumed epicentre is calculated and the travel-
time curves are used to find the corresponding travel-time, Ti*.
2. The travel-time is estimated a second time from the difference between the arrival time
and the assumed origin time: Ti = ti-t0
3. The residual, Ri, for each station is calculated as the difference between the two estimates
of the travel-time: Ri = Ti*- Ti
4. It is assumed that the residuals are due to random errors (such the fact that the Earth
model represented by the travel-time curves assumes a laterally homogenous and
spherical planet, and errors in readings of arrival times) and the errors in the initially
assumed origin: dN0, dE0, dh0, dt0.
5. A least squares regression technique is applied to determine the values of these errors
that minimise the sum of the squares of the residuals Ri.
6. The original estimate of the earthquake origin is updated by the error values (dN0, dE0,
dh0, dt0) and the process is repeated from step (1) until convergence is achieved, usually
defined by tolerances on the error values.

The entire process, including reading P waves first arrivals and values from travel-time curves, is
normally performed automatically using computers, although seismologists generally check
the input and output in order to remove spurious and wrongly associated data.
The final output for each event is presented in catalogues giving the date, origin time, epicentral
co-ordinates and focal depth of the events, each together with the estimates of their errors

ISC origin parameters for selected earthquakes in Central America

DATE TIME EPICENTRE DEPTH
Year Month Day (UTC) N° E° (km)
1982 I 09 01:55:11.9  0.64 13.11  0.080 -88.53  0.089 89  7.2
1982 I 12 05:48:19  1.2 13.15  0.031 -87.56  0.028 10  7.9
1982 VI 19 06:21:57.9  0.21 13.29  0.016 -89.39  0.016 83  1.9
1982 VI 23 08:51:39.5  0.59 13.95  0.076 -88.60  0.072 33

The errors are the standard deviations calculated as part of the regression analysis and it can be
appreciated that there are not insignificant, recalling that 1° is equivalent to about 111 km. The
routine epicentral determinations, such as those performed by ISC, generally carry an error of
the order of 10-15 km.
More accurate locations can be attained using dense regional seismograph networks. A
minimum of 4 stations is required but, as more are the stations and as better is their azimuthal
distribution around the epicenter, as better will be the earthquake location (errors in epicentral
location can be reduced to less than 1 km).

F. Sabetta 3.20