Upload
others
View
8
Download
0
Embed Size (px)
Citation preview
The 2015 Gorkha (Nepal) earthquake:unfinished business
Large earthquakes in the Himalaya and India
James Jackson, Bullard Laboratories, University of Cambridge
8 metres
1857 California: San Andreas Fault
Earthquakes happen on faults
The Earthquake Mechanism
epicentre
300 million years ago
U.S. Geological Survey
Contrasts within the continents (old vs. young)
Earthquake depth distributions
Moho(base ofthe crust)
Earthquakes either in upper crust or in whole crust…but not significantly in continental mantle
Maggi et al 2000
Teleseismic depths:well-determinedfrom P and SH body-wavemodelling
T cutoff ~350oC
Mantle has earthquakes if colder than 600oC
T. Craig et al, GJI, 2014
CRUST
CRUST
MOR
CRUST:
ISENTROPIC INTERIOR:Tp = 1315oC
MANTLEconstantheat flux9.8 mw/m2
surface heat flux 58.6 mW/m2
Moho T = ~600 oC
‘LITHOSPHERE’thickness= 242 km
50 km
SIBERIANSHIELD
earthquakes
Moho
McKenzie et al, EPSL, 2005
…needs DRY lower crust(granulite composition)
Moho
Depth to baselithospherekm
9000 paths
Priestley & McKenzie, EPSL, 2006
resolution: X-Y ~150-200 kmZ ~ 30 km
Africa rifting
Archean
TeleseismicCraig et al, GJI, 2011
Lithosphere thicknessGeology
black: Proterozoicor older
grey: Pan-African
from local microearthquake
surveys
Craig et al, GJI, 2011
North South
Craig et al, GJI, 2011
26
30
15
40
40
Central Asia
rifting
Sloan et al, GJI, 2011
Siberia
Mongolia
India
forelands
Jackson, GSA Today, 2002; Sloan et al GJI, 2011
Moho depthEarthquake depth
Alps
forelands
Kastrup, Deichmann et al, JGR, 2004
Deichmann, PEPI, 1992
“...temperatures in the lower crustal seismic zone are well above 450oC.”
Monsalve et al, JGR, 2006
CRUST
MANTLE
Himalaya and south Tibet
Moho ~500oC
600oC
57
Priestley et al 2008 Monsalve et al 2006
CRUST
MANTLE
Earthquakes M>5.5 and deeper than 60 km 1965-2006
Copley & McKenzie, GJI, 2007topographic front
India Kazakhstan
Sloan et al, GJI, 2011
The lower continental crust can have earthquakes up to 600oC ….. usually dry granulite composition associated with shields
Usually only the upper continental crust has earthquakes (up to about 350oC)
Strong lower crust supports mountain fronts, not the mantle
What’s special about ancient shields?
Implications
…earthquake size depends on fault length and area
Copley et al, JGR, 2011
INDIA 18972001
Vicky Stevens
The 2015 Gorkha (Nepal) earthquake
Tibet
India
2005 Muzaffrabad 2008 Sichuan
2015 Nepal
TIBET
NEPAL
J P Avouac, Caltech
Moho
locked
1934
TIBET
INDIA
Bollinger et al., 2014, JGR, Fig8; also Sapkota et al, 2013, Nature Geoscience
2015
2015
Before the earthquake…. monitoring with GPS
GPS in Nepal
Tibet
India
Nepal
velocities relative to India
J-P Avouac
GPS velocities relative to India
J-P Avouac
locked fault width 110 km
creeping zone~17 mm/yr
TIBETINDIA
J-P Avouac
Kathmandu
Earthquakes 1995-1999
J-P Avouac
up to 3500m3000-5000m>5000m
Larger earthquakes (M>5.5) 1965-2005
Base of the locked zone follows the 3500m contour
The Mw 7.8 GorkhaEarthquake, Nepal
Preliminary Observations Jean-Philippe Avouac
Bullard LaboratoriesUniversity of Cambridge
CollaboratorsVicky StevensKristel ChanardThomas AderMarion ThomasPierre BettinelliLaurent BollingerFrancois AyoubRodolphe CattinJerome Lave
National Seismic Center (DMG, Nepal):Madhab PandeySudhir RajaureSom Sapkota
Institute for Tibet Research (China)Jing Liu
IISER Kolkata (India) Supriyo Mitra
John GaleztkaJeff GenrichMireille FlouzatDiego MelgarShengji WeiJean-Paul AmpueroLingsen MengTeng Wang
+ John Elliott (Oxford)NOT James JacksonKeith Priestley (Cambridge)
J P Avouac John Elliott
John Elliott
Mw 7.8 25th April 2015: aftershocks in first two weeks John Elliott
Mw 7.8 25TH April 2015 Mw 7.3 12TH May 2015
25 April
12 May
John Elliott
Displacement GPS seismogram at KKN4north of Kathmandu
Processed by SOPAC(5 samples/second)
J-P Avouac
AFTERMainshock
(M 7.8) of 25 April
BEFOREAftershock
(M7.3)of 12 May
Aftershocks M>4
J-P Avouac
AFTERAftershock
(M7.3)of 12 May
Aftershocks M>4
Diego Melgar
Interseismic coupling from Ader et al (2012)Yellow patch, slip >1m, joint inversion of teleseismicwaveforms an SAR amplitude offsets.
J P Avouac
Elliott et al, Nature Geoscience, in press
Pass 2: post earthquake
phase (distance) shiftdue to ground motion
780
kmRadar interferometry
= 5.6 cm
Pass 1: pre earthquake
Tim Wright
DOWN
UP
John Elliott
Changes in line-of-sight distance to satellite(each fringe = 10cm)
90 km
John Elliott
4.5m
0m
Kathmandu
faultslip
90 km
John Elliott
Galetzka et al, Science, 2015
Static stress drop after the Gorkha mainshock.Aftershocks M>4Aftershocks M>6
John Elliott
??
Aftershocks M>4 recorded until 15 September 2015
J-P Avouac, Supriyo Mitra, Keith Priestley
3 33 moreTo west
Post-mainshock GPS displacements 25 April to 1 July
J-P Avouac
Kathmandu
Post-mainshock GPS displacements 25 April to 15 September
J-P Avouac
50 mm
(landslide)
Kathmandu
Kathmandu valley
… a former lake basin
Kathmandu pre-earthquake
Kathmandu pre-earthquake
Kathmandupre-earthquake
Kathmandupre-earthquake
Bedrock Kathmandu basinseconds
Tomijung village
Majgaun (upper village)
Thulo Sabro Roger Bilham
Roger BilhamGogne
John Elliott
??It’s not over yet
Conclusions (Gorkha EQ)
• Blind rupture, unzipped down dip edge ofLocked Main Himalayan Thrust
• Repetition of similar past events (1833)recurring every about 100yrs
• Clear pulse like rupture with 5 6s rise time,3km/s propagation.
• Put Kathmandu basin in resonance at 4 5s, butnot much shaking at higher frequencies(16%g).
What can be done?
Amod Dixit
Roger BilhamBarpak reconstruction
NSET: National Society for Earthquake Technology
Retrofitted school, Lalitpur
Retrofitted school, Lalitpur
NSET
Women’s group, Kirtipur
Women’s group, Kirtipur
Earthquakes 1962-2003
The earthquake belt of the Mediterranean – Middle East - Asia
East-west trade routes: the ‘Silk Road’
Trade routes follow the geological structures:the edges of deserts, plateaus and mountains
are all formed by earthquake faulting
>10,000 dead>100,000 dead
Earthquakes that killed more than 10,000 people: 1000-2008 AD
36 of them in the last 100 years120 in total (worldwide):
1887, 1889, 1911destroyed in 855, 958, 1137, 1830
Bishkek
faults
Also: Ashkhabad (1948), Tashkent (1966), Wenchuan (2008)....
1885
Tabriz, Iran
Daily problems of:congestion, pollution, air quality, water supply (and health, poverty) etc.…
Complex Physical SystemEarthquakes without frontiers
At risk population density
earthquakes1964 2007
M>5
1900 2000M>6
cities>500,000
inhabitants
people per 5km2 inearthquake zones
ItalyGreeceTurkey
IranArmeniaAzerbaijan
NepalIndiaPakistan
China
KazakhstanKyrgyzstanUzbekistanTurkmenistan
Earthquakes without frontiers
Bam 2003
Vision
Better knowledge of the earthquakehazards and their context
Find effective pathwaysto improved resilience
Training and exchange of knowledge
‘For years I fought a losing battle to keep away from involvement with the notion of earthquake prediction. The press and public will go toward the suggestion of prediction like hogs to the trough. Meanwhile, other objects of investigation are neglected or distorted; and aid is given to the people who would like to forget the fact that for public safety we don’t need prediction –-that earthquake risk could be removed, almost completely, by proper building construction and regulation.’ (Charles Richter, retirement speech 1970)
Earthquake prediction is NOT the answer