The National Seismic Hazard Maps and Eastern Washington Seismic Hazard Assessment

1
The National Seismic Hazard Maps and Eastern
Washington Seismic Hazard Assessment

• Art Frankel
• U.S. Geological Survey
• Golden, CO

for Eastern Washington Hazards and Risks Workshop
Yakima, WA October 16, 2007
2
This map is used in building codes in 47 states
3 states will enact this in 2008
3
The national seismic hazard maps are the basis of
seismic design maps in the International Building
Code (used in 47 states) and International
Residential Code (used in 45 states).The maps
have a variety of other applications, including

• new AASHTO design guidelines for bridges
• EPA regulations on landfills
• Loss estimation using HAZUS
• Inputs used for determination of earthquake
  insurance premiums
• Inputs used for earthquake scenarios for
  emergency management and assessment of impact on
  built environment

4
National Seismic Hazard Mapping Project Members

• M. Petersen, Project chief
• A. Frankel, S. Harmsen, C. Mueller, K. Haller, R.
  Wheeler, Y. Zeng, N. Luco, R. Wesson, O. Boyd, D.
  Perkins

With help from Craig Weaver, Joan Gomberg, Brian
Sherrod
5

• The national seismic hazard maps represent an
  average estimate of seismic hazard using
  alternative models of fault parameters,
  seismicity, and attenuation relations they are
  not worst-case maps
• Website with hazard maps, lookup by lat-lon,
  input data, fault parameters, deaggregations,
  documentation, Seismic Hazard 101
  eqhazmaps.usgs.gov

6
Process for 2007 Update of the National Maps
Attenuation Oct 2005
National User-Needs Workshop DEC 2006 CA
CA Oct 2006
Comments From Outside Community June-July
Final Prob Maps Oct-Nov 2007 Design maps Dec 2007
PacNW Mar 2006
Draft maps (Project 07) Feb 15, 2007
Draft maps June 2007
InterMtn West June 2006
For 2009 NEHRP Provisions, 2010 ASCE, 2012 IBC
External Review Panel on Maps May 2007
CEUS May 2006
External Review Panel on NGA Sep, 2006

7
Hazard Methodology Example
a
b
Earthquake Sources
a
Ground motion
Hazard curve
r1
d1
d4
annual probability of exceeding pga
peak ground acceleration
d2
r2
M7.6
d3
high seismicity zone
San Andreas fault
0.25g
0.5g
distance
peak ground acceleration (pga)
r3
Specify recurrence rates of earthquakes for each
source that can affect site of interest
Attenuation relations tell you median ground
motions that each potential earthquake will
produce at site, and variability
Hazard curve describes probability of having
ground motions a certain intensity
8
Line shows 2 Prob. of Exceedance in 50
year Approx. 2500 yr return time
IBC for eastern WA Uses 2/3 of 2 in 50
yr Spectral accelerations
9
M4 since 1963 M5 since 1930 M6 since 1850
0.33
M8.0-8.7
0.67
also from GPS
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15

16
Used faults with documented Quaternary
displacement
Saddle Mtns 0.052
Rattlesnake Wallula 0.043
Mill Creek Thrust 0.038
Hite 0.02
(Toppenish Ridge)
Horse Heaven Hills 0.031
Wallowa 0.14
Mean vertical slip rates in mm/yr
values from Geomatrix 1996 These represent rates
averaged over 10-15 m.y.
17
Recurrence Times Estimated from Fault Slip Rates
18
Comparison of hazard estimates for Hanford
(values in g 46.5N, 119.5W)
19
(No Transcript)
20
(No Transcript)
21
Changes in Pacific Northwest for 2007 Update of
National Seismic Hazard Maps

• Cascadia subduction zone treatment of weighting
  of M9 and M8s
• Added 3 newly-studied faults (Stonewall
  Anticline, Boulder Creek fault, Lake
  Creek-Boundary Creek fault)
• Included uncertainty in magnitude determinations
  when calculating seismicity rates from earthquake
  catalog
• Changed attenuation relations for crustal
  earthquakes (NGA) and added two attenuation
  relations for Cascadia subduction zone (now use
  Youngs et al., 1997 Atkinson and Boore, 2003
  Zhao et al., 2006).

22
Working Group on Washington-Oregon Faults for the
National Seismic Hazard Maps

• Provides recommendations to NSHMP about faults to
  add to the hazard maps, parameters to use for
  the added faults, and fault parameters to revise.
  
• Ian Madin, Mark Molinari, Brian Sherrod, Tim
  Walsh
• Joan Gomberg of USGS led the discussions

23
Eastern WA Issues Discussed with Working Group
for 2007 Update

• Should we add other faults in Yakima fold belt,
  even though they dont have documented late
  Quaternary movement (about 100,000 yr), as
  suggested by Bob Yeats? WG recommended that we
  do not, in keeping with policy of NSHMP, because
  of lack of study on these other faults
• WG recommended that we consider giving some
  weight to faster late Quaternary slip rates
  rather than the longer-term rates. However there
  is only one published paper with Quaternary
  rates West et al., 1996 for Saddle Mountains
  anticline, but this uses normal slip on hanging
  wall to infer slip rate on underlying reverse
  fault.

24
Next Generation of Attenuation (NGA) relations
for crustal earthquakes

• Sponsored by Pacific Earthquake Engineering
  Research (PEER) Center
• We had the relations available in Sept., 2006
  reviewed by a team of ground-motion experts R.
  Archuleta, J. Anderson, M. Chapman, C.B. Crouse,
  R. Graves, T. Heaton, W. Holmes, J. Stewart
• Used Boore and Atkinson Campbell and Bozorgnia
  Chiou and Youngs, based on recommendation of team
  of experts
• These new relations make significant changes to
  the national maps and to scenario ground-motion
  maps

25
Provided by Ken Campbell, EQECAT
NGA Project Database

• NGA strong-motion database
• 172 worldwide earthquakes
• 1,400 recording stations
• 3,500 multi-component strong-motion recordings
• Over 100 parameters describing source, path, and
  site conditions

Strong motion dataset prior to NGA database (red)
in NGA database (blue)
26
Figure and caption from Campbell and Bozorgnia
27
f
Figure modified from Campbell and Bozorgnia
Arrows possible choices for 90 Confidence Interv
al for Median for CA events With M 6.9 Black
arrow Increases map values by about 7. Red arrow
Increases map values by about 20. (when hazard
dominated by M 6.9 events)
0.2 ln units 22 change
28
Two Sets of Scenarios for Cascadia Subduction
Zone revised for 2007 Update

• M8.8-9.2 rupturing entire CSZ 500 year average
  recurrence time from paleoevent dates
  (probability of 0.67)
• M8.0-M8.7 floating rupture zones rupture under
  any location along coast with rate of 1/500, on
  average (probability of 0.33)
• equal probabilities assigned to 8.3, 8.4,
  8.5, 8.6, 8.7 scenarios 8.0,8.1,8.2 assigned
  half probability of others

29
Squares data from M8.3 Tokachi-Oki EQ, B and C
sites
Used in 2002 maps
30
Squares data from M8.3 Tokachi-Oki EQ, B and C
sites
Used in 2007 maps
31
Squares data from M8.3 Tokachi-Oki EQ, B and C
sites
32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
Improving Hazard Assessment in Eastern WA

• Do more trenching studies of prehistoric
  earthquakes with LIDAR as guide determine timing
  of late Quaternary earthquakes and late
  Quaternary slip rates (most recent 100,000 yr or
  less)
• Use GPS and regional deformation models
  (Earthscope/Plate Boundary Observatory provides
  denser GPS station configuration)
• Conduct more seismic reflection/refraction
  surveys to characterize buried faults detailed
  gravity and aeromag.
• Include effects of site response (soil over
  basalts) and deep sedimentary basin on ground
  motions

37
Plans for NSHMP Products

• Complete design maps based on instructions from
  Building Seismic Safety Council
• Update all products on our website to the 2007
  maps custom mapping, deaggregations, hazard
  lookup
• Determine uncertainty estimates
• Produce scenario ground-motion maps for all
  faults in the national maps

38
(No Transcript)
39
Figure from Boore and Atkinson
40
Possible configurations for rupture zone of great
Cascadia Earthquakes
From Flueck et al. (1997)
41
0.2 wt
0.2 wt
0.1 wt
0.5 wt
42
Coupling factor Derived from GPS By McCaffrey et
al., 2007
Figure Modified From McCaffrey et al., 2007
Note that our model Has eastern edge Between 20
and 30 km Depth of subducted plate
43
Figure from Boore and Atkinson
44
New velocity field for Pacific Northwest derived
from campaign (blue) and continuous sites (red).
Reference frame is North America and ellipses
are 70 confidence. Vancouver Island data from
Geological Survey of Canada (Herb Dragert et al.)
From McCaffrey et al., 2007
45
Figure from Campbell and Bozorgnia
46
Figure modified from Campbell and Bozorgnia
47
Direct Inputs to Hazard Maps

• Earthquake catalogs (instrumental and historic)
• Fault data (geologic slip rates, dates of past
  events from trenching, fault geometry, etc.)
• Effects of prehistoric earthquakes
  paleoliquefaction (New Madrid, Charleston, Wabash
  Valley), subsidence and uplift (Cascadia,
  Seattle flt), turbidites (Cascadia)
• Geodetic data (NV-CA, Puget Lowland)
• Ground-motion attenuation relations

48
Figure 17 Comparison of new NGA equation (Chiou
and Youngs) and older Equation by Sadigh et al.
(1997) used in 2002 maps.