Sismologia na era da comunicação digital
111
Sismologia na era da comunicação digital Susana Custódio [email protected] ESB, 23/Fev/2017 Bogota, Colombia, M6.2, 2015 Virginia, USA, M5.8, 2011
-
Upload
rosa-duarte -
Category
Science
-
view
69 -
download
0
Transcript of Sismologia na era da comunicação digital
Sismologia na era da comunicação digital
Susana Custódio
ESB, 23/Fev/2017
Bogota, Colombia, M6.2, 2015 Virginia, USA, M5.8, 2011
A nossa Terra
ETHZ-HPCSED
Placas litosféricas que se movem
NASA
O que é um sismo
ISC, 1960 – present
O que é um sismo?
Geração de ondas sísmicas
L. Matias, FCUL
[Hayes, Wald, et al.]
Destruição causada por sismos
Sichuan, China, M7.9, 2008
Haiti, M7, 2010
Christchurch, New Zealand, M6.3, 2011 Lorca, Espanha, M5.1, 2011
Mega-sismos: Sumatra, 2004, M9.1
Mega-sismos: Sumatra, 2004, M9.1
Perdas humanas e económicas
[Roger Bilham, Univ Colorado]
Como reduzir as perdas causadas pelos sismos?
REAKT (From an original figure of T. Jordan redrawn by S. Wiemer)
Long-term Hazard Mapping
Long-term Forecasting
Short-term Forecasting
Early Warning ShakeMaps &
Rapid Loss
Aftershock Forecasting
Decades Years Days to Minutes Seconds Minutes to Hours
Minutes to Days
Earthquake
Estimar os impactos dos sismos
decades years days
Early
W arning
Long- Term
Forecast ing
Short - Term
Forecast ing
& Predict ion
Long- term
Hazard
Mapping
ShakeMaps
&Rapid Loss
Assessm ent
0 m inutes
Ear thquake
seconds days/
m onths
Aftershock
Hazard
The different time scales of the
earthquake process
From an original figure of
Tom Jordan redrawn by
Stefan Wiemer
REAL-TIME SEISMOLOGY 1. Fonte
2. Propagação
3. Sítio
4. Vulnerabilidade & Inventário
REAKT
Sismologia em tempo real
Long-term Hazard Mapping
Long-term Forecasting
Short-term Forecasting
Early Warning ShakeMaps &
Rapid Loss
Aftershock Forecasting
Decades Years Days to Minutes Seconds Minutes to Hours
Minutes to Days
Earthquake
REAKT (From an original figure of T. Jordan redrawn by S. Wiemer)
Sismologia em tempo real com dados sísmicos
Kanamori, Nature 1997
Qual a informação que temos disponível em tempo real?
Localização do Hipocentro (< 1 min)
SeisComP3 Hanka et al, ORFEUS Newsletter 2008 Guralp
Shakemaps (5-7 min)
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
• ‘Ground Motion
Prediction Equations’
(propagação das
ondas sísmicas)
• Interpolação entre
estações
Antecipar o movimento do solo
decades years days
Early
W arning
Long- Term
Forecast ing
Short - Term
Forecast ing
& Predict ion
Long- term
Hazard
Mapping
ShakeMaps
&Rapid Loss
Assessm ent
0 m inutes
Ear thquake
seconds days/
m onths
Aftershock
Hazard
The different time scales of the
earthquake process
From an original figure of
Tom Jordan redrawn by
Stefan Wiemer
REAL-TIME SEISMOLOGY 1. Fonte
2. Propagação
3. Sítio
4. Vulnerabilidade & Inventário
REAKT
Estimativa de danos
PAGER
M 7.8, IRAN-PAKISTAN BORDER REGIONOrigin Time: Tue 2013-04-16 10:44:20 UTC (15:14:20 local)
Location: 28.11oN 62.05
oE Depth: 82 km
PAGERVersion 6
Estimated Fatalities Estimated Economic Losses
Created: 7 hours, 31 minutes after earthquake
Estimated Population Exposed to Earthquake ShakingESTIMATED POPULATION
EXPOSURE (k = x1000)- -* - -* 135k* 2,600k* 1,682k 377k 2k 0 0
ESTIMATED MODIFIEDMERCALLI INTENSITY
PERCEIVED SHAKING Not felt Weak Light Moderate Strong Very Strong Severe Violent Extreme
POTENTIALDAMAGE
ResistantStructures
VulnerableStructures
none
none
none
none
none
none
V. Light
Light
Light
Moderate
Moderate
Moderate/Heavy
Moderate/Heavy
Heavy
Heavy
V. Heavy
V. Heavy
V. Heavy
*Estimated exposure only includes population within the map area.
Population Exposure population per ~1 sq. km from Landscan
Orange alert level for shaking-related fatalities.Significant casualties are likely and the disasteris potentially widespread. Past events with thisalert level have required a regional or nationallevel response.
Yellow alert level for economic losses. Somedamage is possible. Estimated economic lossesare less than 1% of GDP of the Islamic Republicof Iran.
Structures:Overall, the population in this region residesin structures that are highly vulnerable toearthquake shaking, though some resistantstructures exist. The predominant vulnerablebuilding types are adobe block andunreinforced brick with timber floorconstruction.
Historical Earthquakes (with MMI levels):
Date
(UTC)
Dist.
(km)
Mag. Max
MMI(#)
Shaking
Deaths
1989-12-07 389 5.9 VIII(127) 0
1994-02-23 339 6.1 VII(508) 6
2003-12-26 386 6.6 IX(70k) 26k
Recent earthquakes in this area have causedsecondary hazards such as landslides thatmight have contributed to losses.
PAGER content is automatically generated, and only considers losses due to structural damage.Limitations of input data, shaking estimates, and loss models may add uncertainty.
http://earthquake.usgs.gov/pager Event ID: usb000g7x7
Selected City Exposurefrom GeoNames.org
MMI City Population
VII Khash 70k
VI Zahedan 552k
VI Iranshahr 131k
VI Rudbar 11k
VI `Alaqahdari Dishu 9k
VI Mirabad 14k
V Zaranj < 1k
V Turbat 76k
V Zabol 122k
V Chah Bahar 47k
IV Gwadar 52kbold cities appear on map (k = x1000)
Sismologia em tempo real
Long-term Hazard Mapping
Long-term Forecasting
Short-term Forecasting
Early Warning ShakeMaps &
Rapid Loss
Aftershock Forecasting
Decades Years Days to Minutes Seconds Minutes to Hours
Minutes to Days
Earthquake
REAKT (From an original figure of T. Jordan redrawn by S. Wiemer)
Aviso Antecipado
Aviso Antecipado
1.
As primeiras ondas (P)
têm menor amplitude.
2.
As primeiras ondas (P)
contêm informação sobre
a magnitude dos sismos.
Allen et al, 2003
Aviso Antecipado
JMA
Aviso Antecipado
Allen et al, SRL 2009
REDES DE CIDADÃOS
“CITIZEN EMPOWERED SEISMOLOGY”
CROWDSOURCING
Did You Feel It?
IPMA
Did You Feel It?
Wald et al, AG 2012
M 4.2, VA M 4.4, CA
Detecção de sismos por acessos à página do EMSC
Detecção de sismos por acessos à página do EMSC
Bossu et al, AG 2012
Detecção de sismos por Tweets
Earle et al, AG 2012
“earthquake”, “gempa”, “temblor”, “terremoto”, “sismo”
Redes de MEMS (Micro Electro Mechanical Systems)
Quake Catcher Network
O que é que vamos ver com redes muito densas?
Clayton et al
O que é que vamos ver com redes muito densas?
Clayton et al
O que é que vamos ver com redes muito densas?
Clayton et al, AG 2012
E em Portugal?
As magnitudes dos sismos
Zoback and Gorelick, 2012
Fast Earthquake W arnings
Email SMS
Radio Fax
Telefone
Civil Protection
W arnings [ operator validat ion]
Delay after Origin Time: 3m40s Delay after 1º record: 3m31s Delay after minimum set of records (10): 2m43s
SeiscomP 2.6, SEISAN 8.1 (modified) and in-house software
Portugal: Hipocentro e Magnitude
Imagem cedida por F. Carrilho, IPMA
Shakemap: Portugal
Sentiu um sismo?
Alguns links…
• IPMA: – https://www.ipma.pt/pt/geofisica/sismologia/
• EMSC: – http://www.emsc-csem.org
– http://www.citizenseismology.eu/index.html#materials
• USGS: – http://earthquake.usgs.gov
Webpages
• PAGER: http://earthquake.usgs.gov/earthquakes/pager/
Aviso Antecipado
Aviso Antecipado
Ondas sísmicas
Um século de sismos
Ammon et al., SRL 2010
Detecção de sismos por Tweets
Earle et al, AG 2012
W phase: CMT (Centroid Moment Tensor)
Duputel et al, preliminary results for the 2011 Tohoku earthquake
W-phase
Lockwood and Kanamori, G3 2006
W-phase
W phase source inversion group http://wphase.unistra.fr/
Onda de longo período (100 – 1000 s)
W-phase
Seth and Wyssession, Blackwell Publishing 2003
Modelação de Fontes Sísmicas: Inversões
Source
Source
Modelação de Fontes Sísmicas: Inversões
Source
Propagação das ondas
Modelação de Fontes Sísmicas: Inversões
Source
Propagação das ondas
Modelado
Modelação de Fontes Sísmicas: Inversões
Source
Propagação das ondas
Modelado
Dados
Modelação de Fontes Sísmicas: Inversões
Source
Propagação das ondas
Modelado
Dados
Modelação de Fontes Sísmicas: Inversões
W phase: CMT (Centroid Moment Tensor)
Duputel et al, preliminary results for the 2011 Tohoku earthquake
W phase: CMT (Centroid Moment Tensor)
Duputel et al, preliminary results for the 2011 Tohoku earthquake
Redes de MEMS
QCN
Mwp
Tsuboi, GRL 2000
W phase: CMT (Centroid Moment Tensor)
Actualização dos Shakemaps e dos danos previstos
Mapas de velocidade de pico cumulativa Terashake
Human and economic losses
[Roger Bilham, Univ Colorado]
This is how much
we borrowed
from Troika
Qual a informação que temos disponível em tempo real?
Portugal: Factores de Amplificação
Teves-Costa et al, 2010
Determinação de Magnitudes
Bobby Motwani
Determinação de Magnitudes
SeisComp3 Hanka et al, ORFEUS Newsletter 2008
Mecanismo Focal: Inversões do Tensor Momento
Source
Propagação das ondas
Sismo TO Tempo
Redes Globais
Estimativa de danos
Sismo TO
Redes Regionais
Portugal 3 min 5 min
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
W-phase
20-50 min
5-7 min
5-20 min
Shakemaps
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
Modelos de Fonte Finita
Hayes et al., SRL 2011
Sismo TO Tempo
Redes Globais
Mecanismo Focal Estimativa de danos
Sismo TO
Redes Regionais
Portugal 3 min 5 min
Falha finita
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
>20 min
> 2 hr
5-10 min
20 min – 10 hr
W-phase
20-50 min
5-7 min
5-20 min
Shakemaps
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
Aviso Antecipado
Project REAKT, FP7
Aviso Antecipado
http://www.youtube.com/watch?v=33JI929aCS8 Courtesy of Aldo Zollo
T0
PRESTo PLUS Eqk
Epicenter
Vulnerable
Site
S-Waves
Station
Alert Level
Magnitude
Accelerograms P- and S-waves Windows
Depth
Courtesy of Aldo Zollo
Aviso Antecipado
Aldo Zollo
Did You Feel It?
Wald et al, AG 2012
Did You Feel It 1999 – 2011 1.6 million responses
USA Hazard Map 10%, PGA, 50 yrs
Wald et al, AG 2012
Did You Feel It?
Wald et al, AG 2012
A Terra por dentro
Wiley
Energy Released in Earthquakes
IRIS
Tectónica de placas
USGS
J. Renner
How much energy does a mega-earthquake release?
Porquê fazer sismologia em tempo real?
Did You Feel It? Wald et al., AG 2012
3. Porque avança a ciência
Porquê fazer sismologia em tempo real?
4. Porque contribui para o avanço da ciência.
Global CMT
Localização do Hipocentro
Guralp
STA
LTA
• Ondas P
• Picks automáticos
Localização do Hipocentro
• Ondas P
• Picks automáticos
• Agregação de picks
• Actualização
automática
SeisComP3 Hanka et al, ORFEUS Newsletter 2008
Shakemaps
Input
Shakemaps
GMPE
(Groud Motion Prediction Earthquakes)
Shakemaps
Factores de Amplificação
Shakemaps
• Mapas de: – PGA
– PGV
– PGD
– SA
– Intensidade
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
Mecanismo Focal: Inversões do Tensor Momento
Domingues et al., GJI 2012
Detecção de sismos por Tweets
Earle et al, AG 2012
Redes de MEMS
Micro Electro Mechanical Systems
Sismologia na era da comunicação digital
Susana Custódio
Kan
amo
ri, Natu
re 19
97
ESB, 23/Fev/2017
Porquê fazer sismologia em tempo real?
1. Porque reduz os danos materiais e humanos.
Communicating the warning Japan
TV and radio announcements
• 124 of 127 TV stations (98%) • 41 AM, 35 FM radio (75%)
J-Alert messages
• 226 municipalities receive the warnings • 102 announce them with public address systems
Cell phones
• 3 companies (Docomo, AU, Softbank) • 52 million can receive them (47%)
Dedicated providers serve
• power plants • factories • schools • hospitals
• shopping malls
earthquake location and
hazard
estimated shaking in your
area
Aviso antecipado, Japão Tohoku 2011 Courtesy of Richard Allen, UC Berkeley
Shakemap, M4.3 LA Worden et al., BSSA 2010
Shakemap Aviso Antecipado
Porquê fazer sismologia em tempo real?
Bogota, Colombia, M6.2, 2015 Virginia, USA, M5.8, 2011
2. Porque queremos (e esperamos) saber rapidamente o que se está a passar.
Porquê fazer sismologia em tempo real?
Ellsworth and Beroza Science, 1995
3. Porque contribui para o avanço da ciência.
Qual a informação que temos disponível em tempo real?
Ammon et al., SRL 2010
Sismo TO Tempo
Redes Globais
Sismo TO
Redes Regionais
Portugal 3 min 5 min
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
30 min 1 hr 15 min
Sismo TO Tempo
Redes Globais
Sismo TO
Redes Regionais
Portugal 3 min 5 min
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
Sismo TO Tempo
Redes Globais
Sismo TO
Redes Regionais
Portugal 3 min 5 min
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
5-7 min
5-20 min
Shakemaps
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
Sismo TO Tempo
Redes Globais
Estimativa de danos
Sismo TO
Redes Regionais
Portugal 3 min 5 min
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
5-7 min
5-20 min
Shakemaps
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
Sismo TO Tempo
Redes Globais
Mecanismo Focal Estimativa de danos
Sismo TO
Redes Regionais
Portugal 3 min 5 min
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
5-10 min
20 min – 10 hr
W-phase
20-50 min
5-7 min
5-20 min
Shakemaps
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
Sismo TO Tempo
Redes Globais
Mecanismo Focal Estimativa de danos
Sismo TO
Redes Regionais
Portugal 3 min 5 min
Falha finita
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
>20 min
> 2 hr
5-10 min
20 min – 10 hr
W-phase
20-50 min
5-7 min
5-20 min
Shakemaps
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
Mecanismo Focal: Inversões do Tensor Momento
Domingues et al., GJI 2012
Sismo TO Tempo
Redes Globais
Mecanismo Focal Estimativa de danos
Sismo TO
Redes Regionais
Portugal 3 min 5 min
Falha finita
2 hr 1 hr 30 min 15 min 5 min
Hipocentro
5-15 min
< 1 min
Magnitude
5-20 min
2-3 min
30 min 1 hr 15 min
5-7 min
5-20 min
Shakemaps
ShakeMap Manual DRAFT: Version 1.0 6/19/06
shows as a topographic low on the ShakeMap intensity Map (Figure 2.8) and second, the
focusing of strong motion to the southeast, the direction the earthquake rupture appears to have
propagated. The offset of the strongest shaking to the southeast from the epicenter, and the
amplification within the basin of sediments underlying Napa and along the northern shore of San
Pablo Bay are also clear on the map of instrumental intensity.
ShakeMap quality strong motion instrumentation coverage in the San Francisco Bay area has
also substantially improved because the 2000 Napa earthquake, so future earthquakes will have
substantially better station control.
TECHNICAL MANUAL 64 Recent Example ShakeMaps
W-phase
20-50 min
>20 min
> 2 hr
5-10 min
20 min – 10 hr
Aviso Antecipado
1.
As primeiras ondas (P)
têm menor amplitude.
2.
As primeiras ondas (P)
contêm informação sobre
a magnitude dos sismos.
Allen et al, 2003
Qual a informação que temos disponível em tempo real?
