evolution of the mexican seismic alert system...

694 Seismological Research Letters Volume 80, Number 5 September/October 2009 doi: 10.1785/gssrl.80.5.694

BACKGROUND

American plate, represents the source of most of the strong

-

arrival of the Spaniards (Figure 1), who referred to an earth-quake that occurred in the year (One Flint), cor-

is part of the so called !re belt and has been regarded as one of the regions with the highest seismic activity in the world. From

M

INTRODUCTION

-

M

e"ects of that earthquake caused at least 10,000 deaths and

away from the epicenter. #e analysis of seismic records from that event allowed scientists to determine that the e"ect of the

structure of several buildings, which caused their collapse and

nor previous training for a quick emergency response in case of

-mended initiatives to learn from the damages so as to prevent

Evolution of the Mexican Seismic Alert System (SASMEX)J. M. Espinosa-Aranda, A. Cuellar, A. Garcia, G. Ibarrola, R. Islas, S. Maldonado, and F. H. Rodriguez

J. M. Espinosa-Aranda, A. Cuellar, A. Garcia, G. Ibarrola, R. Islas, S. Maldonado, and F. H. RodriguezCentro de Instrumentación y Registro Sísmico, A. C. (CIRES), Mexico

! Figure 1. Time “one flint” (1480), night earthquake reference, prehispanic civilization.

Seismological Research Letters Volume 80, Number 5 September/October 2009 695

research and technological developments, as an engineering approach to ward o" possible future seismic disasters.

-mic risk and where early warning signals from SAS have been useful. #e implementation of SAS made it possible to anticipate the arrival of the e"ects of waves in an average of 60 seconds, a period that allows timely opera-

protect equipment or systems susceptible to damage, such as power plants, computer systems, and telecommunications net-

Shortly a$er the city’s mayor announced the alert system’s availability, a patent trial diverted a huge amount of money to legal matters and delayed implementation of the system. But dur-ing the litigation it was demonstrated that the idea of warning of quake risk in advance, by the use of electrical communications, was proposed long before the 1906 San Francisco earthquake in

prior knowledge invalidated the patentability of the idea.

of an M

(Espinosa-Aranda were received, people successfully responded and evacuated the

At present the SAS disseminates public seismic alert

buildings, and emergency organizations.

installation of alternate emitters for warnings issued by the

Guerrero. #e EASAS of Guerrero provide automatic alert services and control the transmissions of the local commercial broadcasting stations, and they also cover private seismic alert services o"ered to 102 institutional users.

! Figure 2. Large Mexican earthquakes between 1875 and 2009.

696 Seismological Research Letters Volume 80, Number 5 September/October 2009

In order to improve the dissemination of SAS warning

installation of three powerful VHF emitters. #is will allow the SAS to take advantage of the low-cost commercial receiv-ers carrying the Public Alert™, which meet technical standards of U. S. National Oceanic and Atmospheric Administration (NOAA) National Weather Radio (NWR) Speci!c Area Message Encoding (SAME) protocols and the Emergency Alert System (EAS) event codes, such as those used to warn against diverse natural hazards in the United States.

M

--

tant earthquake events as well as !ve preventive alerts for mod-erate ones. In comparison with the original algorithm used in

1992), technological evolution incorporated in the SASO has resulted in better time-e&ciency criteria to de!ne the range of the warnings.

#anks to the interest shown by the governmental author-

functions of SASO and SAS to constitute a single entity, the

number of seismic sensors must be increased to provide e"ec-tive warning of impending regional seismic risk and to dissemi-nate seismic alert notices in vulnerable cities. With the service rendered by EASAS in each region, it will be possible to con-sider seismic e"ects in advance and to de!ne the range of the most suitable alert warning for each particular risk condition, and to review these factors systematically.

DESCRIPTION

It has been a big challenge to have an e"ective earthquake early warning system operating continuously against a known dan-

gerous risk. To ensure that the system is available and reliable over the long run, the principal elements of the SAS and SASO have been kept as basic as possible. To guarantee operation, their principal systems have been programmed to report regu-larly about their service condition. #ey have been designed for low-impact damage due to adverse climate and lightning. All systems operate in an inert atmosphere to avoid corrosion fail-

#e SASMEX also aims to achieve the best availability

e"ective public earthquake alert service.

Seismic Detector SystemSensing !eld stations (FS) are deployed across the seismic-prone regions spaced according to the typical depth of their seismic focus. #e FS are provided with accelerometers and digital elec-tronics for recording and analyzing an earthquake’s evolution to determine in real time the parameters to be transmitted to the central control and registry system in advance of dangerous earthquakes.

Communications SystemEach FS, via !ber optics, links with its own radio transmitter to report regular status messages or eventual seismic warning data, generated by an ongoing earthquake through a VHF and UHF redundant radio network (RRN), to reach the central control system. #e RRN transmitters are normally o", and they only turn on, opening high-power communication paths for two to four seconds, to securely relay a brief coded message about regular service conditions or the risk indices of the ongo-ing earthquake. #e low use factor and low standby power con-sumption make it possible to operate with solar energy.

Central Control and Registry System#e data sent by the FS and RRN arrive at a computerized

! Figure 3. The Seismic Alert System of Mexico.


whose main function is performed when digital messages are received from the FS nearest to the epicenter. #ese data are used to calculate the seismic range warning about the impend-ing occurrence of a dangerous earthquake detected by the !rst

-ing range, either strong or moderate, and with the second one, triggers the warning process, either a public alert or a preven-tive alert. With the scheduled messages from FS and RRN, the

of any anomalies.

Warning Dissemination SystemDuring the initial process of an eventual moderate or

for the automatic generation of signals of either public or pre-ventive alerts, which are sent to the population. #e “preven-tion time” is de!ned as the time elapsed between the beginning of the alert signal and the beginning of the phase, related to the region where it is intended to mitigate the risk.

EVOLUTION OF SAS AND SASO; CURRENT RESULTS

Seismic Alert System of Mexico City (SAS)

-

-

aim of mitigating possible seismic catastrophes. #e SAS has been regarded as the !rst system in the world for earthquake

is because the major seismic e"ects threatening the Valley of

whereas notices broadcasted by radio from the epicenter area can be transmitted instantaneously in anticipation of the seis-mic e"ects.

#e SAS identi!es strong earthquakes along the coast of the State of Guerrero with a linear layout of 12 seismic sensing

it to be able to perform an e&cient survey because many seis-mic foci of this region occur at similar depths.

#e parameters measured from the FS are transmitted to

redundant dedicated communications system connected by

M 6.0 earth-

of the signal was achieved due to support from some television networks and most commercial radio stations, members of the

-tribute as a social service for their audience.

from M M

earthquake detection algorithm are shown in Table 1. Each event in Table 1 has its preliminary information as given by the National Seismologic Service (SSN) of the Instituto de

the epicenter to the nearest SAS FS. #e SAS performance is indicated by the number of FS that registered each warned earthquake, the forecast of magnitude, the type of alert emit-ted, and its prevention time.

To be able to identify the development of dangerous earth-quakes at each FS, an analysis of the ground acceleration is continuously performed by means of an algorithm capable of recognizing the P and seismic phases with a short-term

average square input (ASI) from three channels as a typical function (Espinosa-Aranda P, the proce-dure analyzes the evolution of the short time average whereas for phase , it waits for a second threshold. #e FS algorithm integrates the seismic energy during a variable time period last-ing = 2( P) and determines its growth rate at instant

= 2( P). #ese parameters are transmitted by radio to the

range assigned by SAS to its alert warnings, the relayed param-eters are located within a family of calibration curves built with equivalent parameters obtained by processing historic accelero-

#e pair of parameters, seismic energy reached and its deriva-tive at the instant , de!ne the calibration curve locus to determine the hierarchy of the SAS warnings, i.e., public alert for strong earthquakes or preventive alert for moderate earth-quakes.

SAS public alert warnings are broadcast to the population of

--

vision channels. In the capital city of Toluca the warnings are disseminated

channel and via three radio broadcasting stations a&liated

#ere are also private users that agree to provide services via dedicated radio links and bene!t from the public and


preventive alerts. #ese consist mainly of primary and sec-ondary schools, universities, emergency and safety agencies, government buildings, and civil protection organizations, as well as the subway system operated by Sistema de Transporte Metropolitano (Metro) since 1992.

To promote awareness of seismic prevention e"orts in -

http://www.cires.org.mx) shows the information contained in the alert warnings and the seismic

is automatically updated a few instants a$er the occurrence of an earthquake. SAS bulletins are issued when at least two FS detect the same earthquake, and they are relayed via e-mail

information disclosed by the SSN and by the US Geological Survey (USGS) is also included. To enhance the e&ciency of dissemination of the SAS alert warnings and to make it pos-sible to warn the public in advance about other contingencies

a digital code relay system such as that used by NOAA was

time), when the SAS emitted a signal of preventive alert to warn

of the risk of an M

of the technology derived from NOAA/SAME receivers. #at event validated the e"ective function of this advanced resource, which, with a minor technical adjustment, has been proposed as a means of more e&ciently mitigating seismic vulnerability

Although they were aware of the limited e&cacy of the early warning function because of the shorter distances from their

-

A central registry and control system was installed at each city to collect the information relayed by the SAS sensors and to perform the speci!c adjustment on the warning ranges in terms of the parameters relayed by the FS. #is alternate emitting unit of the Seismic Alert System (EASAS), although capable of adjusting the range of the local seismic alert warnings and of taking into account the particular vulnerability in each region,

! Figure 4. SAS network and earthquakes warned between 1991 and 2009.


TABL

E 1

SAS

Hist

oric

al R

esul

ts a

nd E

volu

tion

Serv

icio

Sis

mol

ogic

o N

acio

nal (

SSN

), UN

AM

SAS

Fore

cast

Res

ult

#Re

gion

Loca

l Dat

e-Ti

me

Latit

ude

Long

itude

Mag

nitu

deDe

pth

(km

)Di

stan

ce

(km

)#

Sens

. Op.

Mag

nitu

deAl

ert R

ange

Prev

entio

n Ti

me

(s)

66Gu

erre

ro27

-04-

2009

11:4

6:45

16.9

0–9

9.58

5.7

718

9M

< 6

Prev

entiv

e54

65Gu

erre

ro27

-03-

2009

02:

48:1

617

.35

–100

.82

5.3

3090

7M

< 6

Prev

entiv

e58

64Gu

erre

ro11

-11-

2008

05:

02:0

616

.62

–100

.80

4.7

1543

4M

> 5

Prev

entiv

e63

Guer

rero

06-1

1-20

07 0

0:35

:42

17.0

8–1

00.14

5.6

99

9M

> 6

Publ

ic68

62Gu

erre

ro28

-04-

2007

08:

56:5

416

.94

–99.

825.

09

156

M <

6Pr

even

tive

61Gu

erre

ro13

-04-

2007

03:

43:4

817

.21

–100

.33

5.4

259

8M

< 6

Prev

entiv

e56

60Gu

erre

ro13

-04-

2007

00:

42:2

217

.09

–100

.44

6.3

4110

6M

> 6

Publ

ic58

59Gu

erre

ro31

-03-

2007

00:

18:5

616

.90

–99.

914.

744

45

M <

5Pr

even

tive

58Gu

erre

ro18

-09-

2005

06:

25:5

317

.05

–100

.02

4.4

918

3M

< 5

Prev

entiv

e57

Guer

rero

Coa

st09

-01-

2003

20:

08:4

216

.89

–100

.36

5.3

2528

7M

< 6

Prev

entiv

e58

56Gu

erre

ro27

-09-

2002

02:

04:5

817

.16–1

00.5

94.

936

54

M <

6Pr

even

tive

5855

Guer

rero

Coa

st25

-09-

2002

13:

14:4

816

.87

–100

.115.

35

144

M <

6Pr

even

tive

5754

Guer

rero

Coa

st16

-02-

2002

22:

10:1

916

.94

–99.

934.

737

76

M <

5Pr

even

tive

53Gu

erre

ro C

oast

07-1

0-20

01 2

1:39

:20

16.9

3–1

00.16

6.1

108

8M

> 6

Publ

ic65

52Gu

erre

ro06

-03-

2001

15:

57:5

617

.20

–100

.105.

37

239

M <

6Pr

even

tive

6351

Guer

rero

Coa

st05

-03-

2001

04:

17:3

016

.70

–100

.01

5.4

9620

8M

< 6

Prev

entiv

e66

50Gu

erre

ro C

oast

14-0

4-20

00 2

0:45

:05

16.8

8–1

00.3

54.

69

285

M <

5Pr

even

tive

6649

Guer

rero

17-0

3-20

00 1

8:50

:58

17.12

–99.

354.

850

417

M <

5Pr

even

tive

6848

Pueb

la-O

axac

a Bo

rder

15-0

6-19

99 1

5:42

:05

18.2

5–9

7.49

6.7

9921

19

M <

5Pr

even

tive

1547

Guer

rero

30-0

5-19

99 0

4:58

:42

17.3

3–1

00.7

94.

412

125

M <

5Pr

even

tive

6846

Guer

rero

24-0

4-19

99 2

2:08

:57

17.2

9–1

00.8

04.

627

74

M <

5Pr

even

tive

6845

Guer

rero

Coa

st07

-09-

1998

01:

53:1

716

.77

–99.

674.

111

103

M <

5Pr

even

tive

44Gu

erre

ro C

oast

09-0

8-19

98 11

:18:

0716

.90

–100

.23

4.5

517

3M

< 6

Prev

entiv

e43

Guer

rero

Coa

st17

-07-

1998

06:

18:0

316

.98

–100

.165.

226

47

M >

6Pu

blic

7442

Guer

rero

Coa

st05

-07-

1998

14:

55:0

716

.83

–100

.125.

25

186

M >

6Pu

blic

6641

Guer

rero

Coa

st09

-05-

1998

12:

03:1

317

.37

–101

.41

5.2

2439

4M

< 5

Prev

entiv

e60

40Gu

erre

ro11

-03-

1998

08:

13:1

217

.04

–100

.124.

236

54

M <

5Pr

even

tive

39Gu

erre

ro C

oast

21-1

2-19

97 2

3:22

:00

17.2

5–1

00.9

05.

610

125

M >

6Pu

blic

6938

Guer

rero

Coa

st26

-08-

1997

19:

13:1

716

.34

–100

.53

4.7

1784

6M

< 5

Prev

entiv

e45

37Gu

erre

ro19

-07-

1997

02:

34:3

717

.22

–100

.43

4.9

5614

6M

< 5

Prev

entiv

e56

36Oa

xaca

-Gue

rrer

o Co

ast

14-0

7-19

97 2

0:26

:18

16.16

–98.

764.

325

453

M <

5Pr

even

tive

35Gu

erre

ro C

oast

11-0

7-19

97 1

7:23

:31

16.5

6–9

9.65

4.3

1029

3M

< 5

Prev

entiv

e34

Low

Bal

sas

Rive

r22

-05-

1997

02:

50:5

518

.43

–101

.79

5.9

6114

74

M <

6Pr

even

tive

33Gu

erre

ro08

-05-

1997

10:

58:3

017

.26

–100

.38

5.1

1415

5M

< 6

Prev

entiv

e55

32Gu

erre

ro23

-03-

1997

14:

23:1

617

.43

–100

.85

4.7

1524

4M

< 6

Prev

entiv

e


TABL

E 1

(Con

tinue

d)SA

S Hi

stor

ical

Res

ults

and

Evo

lutio

n

Serv

icio

Sis

mol

ogic

o N

acio

nal (

SSN

), UN

AM

SAS

Fore

cast

Res

ult

#Re

gion

Loca

l Dat

e-Ti

me

Latit

ude

Long

itude

Mag

nitu

deDe

pth

(km

)Di

stan

ce

(km

)#

Sens

. Op.

Mag

nitu

deAl

ert R

ange

Prev

entio

n Ti

me

(s)

31Gu

erre

ro21

-03-

1997

21:

49:1

617

.03

–99.

824.

827

217

M <

5Pr

even

tive

5530

Mic

hoac

an C

oast

11-0

1-19

97 1

4:28

:29

18.0

9–1

02.8

67.

317

212

7M

< 6

Prev

entiv

e42

29Gu

erre

ro27

-10-

1996

03:

15:4

017

.25

–100

.87

4.4

9210

3M

< 5

Prev

entiv

e28

Guer

rero

19-0

7-19

96 0

4:00

:54

17.3

7–1

00.3

04.

87

276

M <

6Pr

even

tive

27Gu

erre

ro C

oast

15-0

7-19

96 1

6:23

:39

17.4

5–1

00.9

26.

518

216

M <

6Pr

even

tive

6526

Guer

rero

Coa

st13

-03-

1996

15:

04:1

916

.49

–99.

195.

335

246

M >

6Pu

blic

7425

Guer

rero

Coa

st16

-09-

1995

10:

09:0

016

.59

–98.

634.

315

82

M <

6Pr

even

tive

24Gu

erre

ro C

oast

15-0

9-19

95 2

1:20

:05

16.3

2–9

8.62

5.0

344

M >

6Pu

blic

23Oa

xaca

-Gue

rrer

o Co

ast

14-0

9-19

95 0

8:09

:27

17.0

0–9

9.00

375

M <

5Pr

even

tive

22Oa

xaca

-Gue

rrer

o Co

ast

14-0

9-19

95 0

8:04

:35

17.0

0–9

9.00

7.3

4537

9M

> 6

Publ

ic72

21Oa

xaca

Coa

st31

-05-

1995

06:

49:2

815

.97

–98.

774.

617

702

M <

5Pr

even

tive

20Gu

erre

ro C

oast

14-0

4-19

95 0

0:01

:08

16.4

3–9

9.09

4.8

2523

6M

< 6

Prev

entiv

e19

Low

Bal

sas

Rive

r10

-12-

1994

10:

17:4

018

.02

–101

.56

6.3

2097

6M

< 6

Prev

entiv

e34

18Gu

erre

ro C

oast

29-1

0-19

94 1

0:44

:08

16.9

7–9

9.89

5.1

2412

9M

< 6

Prev

entiv

e58

17Hi

gh B

alsa

s Ri

ver

22-0

5-19

94 1

9:41

:46

18.0

3–1

00.5

75.

623

938

M <

6Pr

even

tive

30B

16-1

1-19

93 1

9:11

:00

1M

> 6

Publ

ic (F

ail)

AGu

erre

ro C

oast

24-1

0-19

93 0

1:52

:17

16.5

0–9

9.00

6.7

149

M >

6(F

ail)

16Gu

erre

ro C

oast

10-0

9-19

93 0

4:50

:24

16.6

1–9

9.09

4.9

488

4M

< 6

Prev

entiv

e70

15Gu

erre

ro29

-07-

1993

14:

17:0

117

.39

–100

.66

5.0

2223

4M

< 5

Prev

entiv

e14

Oaxa

ca-G

uerr

ero

Coas

t15

-05-

1993

02:

26:3

116

.46

–98.

634.

820

184

M >

6Pu

blic

13Oa

xaca

-Gue

rrer

o Co

ast

14-0

5-19

93 2

1:12:

0016

.04

–98.

706.

020

606

M >

6Pu

blic

7312

Oaxa

ca-G

uerr

ero

Coas

t14

-05-

1993

21:

09:3

916

.43

–98.

715.

820

206

M >

6Pu

blic

6511

Guer

rero

Coa

st31

-03-

1993

04:

18:1

517

.30

–100

.79

5.1

87

4M

< 6

Prev

entiv

e10

Guer

rero

Coa

st09

-11-

1992

20:

13:2

516

.74

–100

.00

4.8

2118

3M

> 6

Publ

ic9

Guer

rero

Coa

st04

-11-

1992

22:

33:3

716

.72

–99.

754.

37

132

M <

5Pr

even

tive

8Gu

erre

ro C

oast

30-1

0-19

92 0

2:16

:00

17.11

–100

.78

5.0

133

M <

6Pr

even

tive

7Gu

erre

ro C

oast

16-1

0-19

92 11

:28:

0016

.44

–99.

224.

828

5M

< 6

Prev

entiv

e6

Guer

rero

02-0

8-19

92 0

6:54

:00

17.12

–100

.39

4.8

56

M <

5Pr

even

tive

5Oa

xaca

-Gue

rrer

o Co

ast

07-0

6-19

92 11

:41:

0016

.21

–98.

845.

138

3M

< 5

Prev

entiv

e4

Oaxa

ca-G

uerr

ero

Coas

t07

-06-

1992

03:

01:0

016

.15–9

8.87

5.2

453

M <

6Pr

even

tive

3Gu

erre

ro C

oast

15-0

5-19

92 0

2:35

:00

16.8

7–9

9.95

4.9

43

M <

5Pr

even

tive

2Gu

erre

ro C

oast

26-0

4-19

92 1

4:53

:00

16.7

5–1

00.0

74.

820

4M

< 5

Prev

entiv

e1

Oaxa

ca-G

uerr

ero

Coas

t16

-10-

1991

12:

36:0

016

.70

–98.

704.

316

1M

< 6

Prev

entiv

e


failed to increase the times of opportunity between the arrival of the seismic e"ects and the time of delivery of their warning because it is a known fact that prevention time depends on the eventual distance between the vulnerable region and the site of the earthquake epicenter.

Seismic Alert System of Oaxaca (SASO)

e"ects of large earthquakes in its territory, and its public ser-

a technological development that evolved from SAS. SASO

north, and isthmus regions. Since its commissioning, it has issued three public alerts and !ve preventive alerts based on the

6). #e SASO historical results and the evolution of the detec-tor algorithm of earthquakes are shown in Table 2 in a similar format to Table 1. Its functioning has been intermittent due to the irregular !nancial support for its service.

SASO Earthquake Range Algorithms -

duction zone makes the depth of the foci variable, increasing

in terms of the distance to the Paci!c coast. As a result, to the north of its territory and at the Tehuantepec isthmus region, hazardous earthquakes occur at depths larger than 100 km. On the other hand, many catastrophic intermediate-foci earth-

-stance made it necessary to develop better algorithms for the SASO, to reduce the time needed to determine the hierarchy of the warning to be issued and to provide longer times to miti-gate the earthquake e"ects.

To optimize the opportunity time of the SASO alert warn-ings, along the coastal region where the seismic epicenters occur

original SAS criterion developed for Guerrero so that, within a time period of = 2( P) the energy reached can be de!ned as its growth rate, as indices of seismic risk. With such an objec-tive in mind, two e&cient new algorithms were designed for time handling whose simultaneous functions at the FS located

it possible to determine and relay the hierarchy of the SAS warnings, public alert for strong earthquakes or preventive

three seconds. A$er the beginning of the P-wave phase, the

and Ramos 1999), analyzes and calculates the following fac-

! Figure 5. First SAS service with NOAA receivers activated, 27 March 2009 Guerrero M 5.3 earthquake.


tors: dominant period, peak acceleration, and energy, during the interval T 2 = ( S P), half time of the original T . #e second method determines simultaneously, by means of a clas-

#erefore, the FS of the SASO, installed in regions with deep focus, are capable of relaying their results at the end of this

regions were used (Iglesias -

ted the most recent preventive alarm warning, anticipating in

an M

the e&cient identi!cation of dangerous earthquakes with deep focus to gain more prevention time in the SAS warning process. Also, with new acceleration records, the calibration process for the best range estimation alert issuing will be improved.

by SASO are only transmitted to users with special receiving

equipment located mainly at primary and secondary schools, universities, and emergency and civil protection organizations. #e public alerts are broadcast by the local radio and TV com-

NWR-SAME standard transmitters.

SAS-SASO INTEGRATION

Brecha de Guerrero (Guerrero Gap) as the region with the highest probability of generating a catastrophic earthquake,

-mic sensors subsystem included just a strip along the coast of

-tioning, the e"ects of earthquakes generated at other neighbor-ing regions were missed, despite the fact that they could in'ict

historical results achieved with the SAS justify increasing the number of FS in the network in order to cover other regions where harmful earthquakes are likely to take place (Figure 9).

For this purpose, it is initially considered to take advan-tage of the agreements by which the government authorities

-

! Figure 6. SASO Network and earthquakes warned between 2003 and 2007.


TABL

E 2

SASO

His

toric

al R

esul

ts a

nd E

volu

tion

Serv

icio

Sis

mol

ogic

o N

acio

nal (

SSN

), UN

AM

SAS

Fore

cast

Res

ult

#Re

gion

Loca

l Dat

e-Ti

me

Latit

ude

Long

itude

Mag

nitu

deDe

pth

(km

)Di

stan

ce

(km

)#

Sens

. Op

.M

agni

tude

Aler

t Ran

gePr

even

tion

Tim

e (s

)

8Ch

iapa

s05

-07-

2007

20:

09:2

116

.95

–94.

046.

210

010

55

M <

5Pr

even

tive

407

Oaxa

ca15

-03-

2007

07:

13:0

016

.07

–97.

245.

112

206

M <

5Pr

even

tive

6Oa

xaca

24-0

1-20

07 2

2:23

:25

16.2

1–9

7.14

4.3

62

Prev

entiv

e5

Oaxa

ca19

-08-

2006

00:

41:3

015

.91

–97.

305.

552

408

Publ

ic4

Oaxa

ca15

-12-

2004

02:

05:1

416

.05

–95.

434.

510

84

M <

5Pr

even

tive

BOa

xaca

18-0

8-20

04 0

4:03

:10

16.3

0–9

5.12

5.7

6350

9M

> 5

.5(C

omx

Fail)

AVe

racr

uz-O

axac

a07

-08-

2004

06:

49:1

217

.06

–95.

445.

911

289

10(F

ail)

3Gr

o-Oa

x Co

ast

14-0

6-20

04 1

7:54

:23

16.3

1–9

8.06

5.8

1015

6M

> 5

.5Pu

blic

302

Oaxa

ca C

oast

13-0

1-20

04 1

5:28

:58

16.0

0–9

7.16

5.5

149

10M

> 6

Publ

ic1

Oaxa

ca13

-01-

2004

13:

50:1

916

.01

–97.1

65.

114

256

M <

6Pr

even

tive

#e principal advantage of the proposed SASMEX integration M

SASO and SAS functions had been integrated, the SASMEX

than 60 seconds.

DISCUSSION

Any earthquake early warning system like SAS is e&cient when

epicenter, and if it can rapidly forecast the range of an ongoing earthquake, but principally it is e"ective if the time provided is enough to enact any prede!ned preventive action.

To mitigate the possibility of a new seismic disaster such

SAS has been a pioneer project in the world in the dissemina-tion of public early seismic alert signals.

Presently, the SAS warns vulnerable populations in public elementary schools through special SAS receivers and the pop-ulation at large via commercial radio and television stations. Facsimile, SMS, Web site, e-mail, and now NOAA receivers are used to e&ciently disseminate public alert signals prior to the arrival of strong seismic e"ects.

With the same goals in mind, the government authorities

their own SASO to warn against strong earthquakes occurring in their territory and to improve the e&ciency of the work car-

Protection Agency). However, because of the shorter distance between the seismic foci and the vulnerable places where the alert warnings are disseminated, the prevention time is shorter.

#ere are plans to integrate the respective functions of the SAS and SASO to enhance the utility of their service. However,

-ity and availability of this service, to increase the FS required to cover regions where harmful earthquakes can take place, and to promote a “prevention culture,” so there is incentive to improve natural-hazard early warning technologies.

Since damaging earthquakes are not too frequent, the

reliability and availability of its basic elements to guarantee the e"ective dissemination of alert warnings in the event of dam-aging earthquakes. With this aim the SASMEX has its own self-evaluation procedures to ensure its services and to achieve regular technological enhancements, including better forecast-ing criteria of seismic risk.


! Figure 7. SASO seismic alert forecast, 2nd and 3rd criteria.

! Figure 8. SASO alert service 3, on 5 July 2007 Chiapas, M 6.2 earthquake.


! Figure 9. SASMEX seismic sensors network.

! Figure 10. SASO alert service criterion 1, on 14 June 2004 Oaxaca, M 5.8 earthquake.


Public-oriented disclosure together with e&cient earth-quake recognition, risk range evaluation, and reliability of the early warning communications constitute necessary ele-

training campaigns launched among the vulnerable popula-

funded, any early warning system will risk failure to comply with the main objective for which it was designed (United Nations 2006).

#e technological development of early warning systems applicable to earthquakes is constantly improving, and it is wise to analyze the spectrum as a whole. #e Scienti!c Earthquake

to the chairman of the USGS, recommended that a study of the feasibility of earthquake alerting in the United States must include a comprehensive assessment of how such information

issue when stating: “We must step outside our ivory tower and try to anticipate the larger consequences of our developing tools and make sure that they will actually improve earthquake hazard reduction.”

ACKNOWLEDGMENTS

#e authors deeply acknowledge the support received from the

de Obras y Servicios, who have contributed to the develop-

-sion to take advantage of this technological resource capable

has made it possible to install a system to transmit, via low-cost receivers of the NOAA-SAME system, the warnings issued by the SAS system. We also thank Dr. Gerardo Suarez Reinoso

-tive discussion time and orientation.

We would also like to acknowledge the Secretaría de Educación Pública (Public Education Ministry), which has

broadcast the automatic warnings issued by the SAS to the

SASO design and development. Finally but not least, thanks to

-

and SASO resources.

REFERENCES

strong motion network. 4 (

(EUA).

para una alerta sísmíca. Bachelor’s thesis, Facultad de Ingeniería,

1,

Espinosa-Aranda, J. M, A. Jimenez, G. Ibarrola, F. Alcantar, A. Aguilar,

Alert System. 66 (

New York: Academic Press.

4-

Sísmica. 68

-lously low peak acceleration.

93

,

Kecman, V. (2001).

MA: MIT Press.

79

la ocurrencia de un sismo en el intervalo de tiempo S-P, para ser utilizado en una Alerta Sísmíca, usando Máquinas de Soporte Vectorial. Bachelor’s thesis, Facultad de Ingeniería, Universidad

Suarez, R., and V. Acosta (1996). . Vol.

United Nations (2006).

International Strategy for Disaster Reduction, !nal version.

Centro de Instrumentación y Registro Sísmico, A. C. Anaxagoras Av., No. 814

Narvarte Sec., Benito Juarez, 03020 Mexico, D. F.Mexico

[email protected]

evolution of the mexican seismic alert system...

Documents