Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 CVM Efforts in Wellington, New Zealand John Louie, Nevada Seismological Lab Collaborators: R. Benites, G. McVerry, & W. R. Stephenson, GNS Science S.

Similar presentations


Presentation on theme: "1 CVM Efforts in Wellington, New Zealand John Louie, Nevada Seismological Lab Collaborators: R. Benites, G. McVerry, & W. R. Stephenson, GNS Science S."— Presentation transcript:

1 1 CVM Efforts in Wellington, New Zealand John Louie, Nevada Seismological Lab Collaborators: R. Benites, G. McVerry, & W. R. Stephenson, GNS Science S. Pullammanappallil & B. Honjas, Optim Inc. VUW student: Anna Kaiser, MSc Geophysics 2006 GNS Science summer student employee Support from Fulbright New Zealand and: M. Henderson, M. Savage, T. Stern, Victoria Univ. SES S. Harder & G. Kaip, Univ. of Texas El Paso John Louie, Nevada Seismological Lab Collaborators: R. Benites, G. McVerry, & W. R. Stephenson, GNS Science S. Pullammanappallil & B. Honjas, Optim Inc. VUW student: Anna Kaiser, MSc Geophysics 2006 GNS Science summer student employee Support from Fulbright New Zealand and: M. Henderson, M. Savage, T. Stern, Victoria Univ. SES S. Harder & G. Kaip, Univ. of Texas El Paso

2 2 Benites & Olsen Model Benites & Olsen created a 3-d grid at 40-m spacing to cover the Wellington fault and nearby basins. Problem of many small basins similar to the many basins surrounding Reno and Las Vegas.  Bull. Seismological Society of America, v. 95, no. 6, pp. 2180–2196, Dec. 2005.  I hope to promote and extend their model, and make the computations easier and more flexible.

3 3 Benites & Olsen Model Assembled from “all available geological and geophysical (borehole, bathymetry, gravity, and seismic) data, down to about 800-m depth.” Ground motions computed with Olsen’s staggered-grid viscoelastic finite- difference code up to 1.5 Hz.  Bull. Seismological Society of America, v. 95, no. 6, pp. 2180–2196, Dec. 2005.

4 4 Benites & Olsen Model The low-frequency computed motions are mainly dependent on: Fault geometry Rupture model Basin-floor geometry Site velocities  Bull. Seismological Society of America, v. 95, no. 6, pp. 2180–2196, Dec. 2005.

5 5 Benites & Olsen Model  Adapt into MA-CME for Larsen’s E3D f-d code  Create a Wellington Community Seismic Modeling Environment for extensions to models, addition of scenarios, and grids at multiple scales. Easily generate grids for local (Parkway) and regional (Wairarapa) scenarios. Easily generate grids for local (Parkway) and regional (Wairarapa) scenarios. Topography Basin Topography Free-surface topography is NOT in computations here.

6 6 Benites & Olsen Model  ModelAssembler’s rule-based gridding combines into the CME: Benites & Olsen’s bedrock depth map (Z1.5, depth to Vs=1.5 km/s), and Benites & Olsen’s bedrock depth map (Z1.5, depth to Vs=1.5 km/s), and Benites & Olsen’s Vs30 map, 1.5 km/s in bedrock, 0.175 km/s in basins. Benites & Olsen’s Vs30 map, 1.5 km/s in bedrock, 0.175 km/s in basins. Vs30 Map 1-D Predicted Amplification Section

7 7 Benites & Olsen Model  A trial E3D computation for a shallow event in Wellington, 2-sec Waves Preserving the 175 m/s shallow Vs in basins causes energy trapping Preserving the 175 m/s shallow Vs in basins causes energy trapping GM Snapshot at 8 sec

8 8 Benites & Olsen Model  Computed ground motions at 0.5 Hz depend more on Vs30 than Z1.5 1-D Predicted Amplification Max. Horiz. Ground Vel.

9 9 Wellington – Lower Hutt Measurements  Most Vs30 measurements not as low or as high as Benites & Olsen model predicts Vs30 Map, Measured vs. Benites & Olsen Predictions 1-D Predicted Amplif. from Vs30 & Z 0.6

10 10 Assembly of Model Refinements  ModelAssembler allows changes to the rules that govern modeling and gridding.  Alter default Vs30 values for basin and rock sites to better fit measurements. Use arithmetic averages of slownesses to preserve travel time. Use arithmetic averages of slownesses to preserve travel time. Assume Rock Vs30=660 m/s.  V = 114 m/s Assume Rock Vs30=660 m/s.  V = 114 m/s Assume Basin Vs30=250 m/s.  V = 73 m/s Assume Basin Vs30=250 m/s.  V = 73 m/s Vs30 Measurements and New Predictions

11 11 Assembly of Model Refinements  Predicted amplification versus computed maximum GM: 1-D Predicted Amplif. From Vs30 & Z 0.6 3-D Max. Ground Velocity

12 12 Assembly of Model Refinements  75% greater max. horiz. ground velocity– despite higher Vs30 in basins: Ratio of Max. Ground Velocities– New over Benites Max. Ground Velocity, New Vs30 Assumptions

13 13 Regional Extension of the CME  Detailed Benites & Olsen (orig. or modified) model set into basin- thickness map from gravity or geologic maps.  Small event scenario, 10-sec waves.

14 14 ConclusionsConclusions  Benites & Olsen’s WnLH model has been adapted into a flexible CME for Wellington.  Characterization of 21 GM recording sites in Wn-LH: Average basin Vs30=250 m/s Average basin Vs30=250 m/s Average rock Vs30=660 m/s Average rock Vs30=660 m/s  Including a slow surface layer on bedrock increases shaking in basins.  Gravity and geology allow CME extension to larger regions.  Benites & Olsen’s WnLH model has been adapted into a flexible CME for Wellington.  Characterization of 21 GM recording sites in Wn-LH: Average basin Vs30=250 m/s Average basin Vs30=250 m/s Average rock Vs30=660 m/s Average rock Vs30=660 m/s  Including a slow surface layer on bedrock increases shaking in basins.  Gravity and geology allow CME extension to larger regions.


Download ppt "1 CVM Efforts in Wellington, New Zealand John Louie, Nevada Seismological Lab Collaborators: R. Benites, G. McVerry, & W. R. Stephenson, GNS Science S."

Similar presentations


Ads by Google