Research

3D Site Effects


Topography effects: Wave diffraction of plane wave
incident on wedge at the material’s critical angle


Basin edge effects generating surface waves (left) vs.
ground surface response effects, frequently referred
to as one-dimensional site response.

Site effects are phenomena that arise from the passage of seismic waves through the near-surface layers of the earth’s crust. These layers are strongly heterogeneous, frequently discontinuous, have irregular surface and subsurface geometry (e.g. hills and valleys), and are prone to permanent deformations, liquefaction and landslides. Site effects change the amplitude, frequency and duration of seismic waves.

Depending on the mechanics of these changes, site effects are broadly classified into ground response effects (impedance and resonance effects), basin edge effects (preferential focusing, and/or trapping of seismic energy), and topography effects (resonance of a topographic feature, energy reverberations in confined space). Combining elements of geotechnical engineering, wave propagation and computational mechanics, our research group develops computational tools for high-fidelity predictions of site effects; synthesizes these tools with field and laboratory data by our collaborators to deepen our understanding of these phenomena; and distills our understanding into simplified predictive models of site effects for engineering practice applications. [See videos of our simulations]

Recent Publications

  • Asimaki D. and Jeong S. (2012). Ground motion amplification observations at Hotel Montana during the M7.0 2010 Haiti Earthquake: Topography, Stratigraphy or both?, Bulletin of the Seismological Society of America, 103(5), 2577-2590
  • Asimaki D., Ledezma C., Gonzalo A. Montalva, Tassara A., Mylonakis G. and Boroschek R. (2012). Site Effects and Damage Patterns, Special Issue on the Maule 8.8 Earthquake, Earthquake Spectra, 28(1), S55-74
  • Asimaki D. and Li W. (2012). Site- and ground motion-dependent nonlinear effects in seismological model predictions, Soil Dynamics and Earthquake Engineering, 32(1), 143-151 (Top 25 Most Downloaded SDEE Papers)

top

Soil-foundation-structure Interaction


Soil-foundation-structure interaction
simulation of large diameter foundations
for offshore wind installations


Domniki participated in an earthquake
recon­naissance trip to Chile in 2010, after
the M8.8 earthquake. Here, a member of
her team is measuring the inclination of
a pier, whose foun­dation moved towards
the sea because of liquefaction
(photo credit: GEER)

Foundation design models in engineering practice are based on the so-called load-transfer approach: the interaction between foundation and soil is quantified by empirical functions (aka. soil springs), which relate the force per unit area or length of foundation to the deflection induced to the soil (and vice-versa). Although soil springs have been used in engineering practice for over 40 years, their accuracy, selection and appropriate use is still hotly debated. In almost all cases, controversies are related to the empirical basis of the soil spring method and to the cost of full-scale load tests, which has severely constrained the availability of data. Our research in this area uses high fidelity, multi-scale computational tools to conduct virtual experiments that bypass empirical assumptions of design methodologies. We use these simulated experiments to study, for example, the effects of liquefaction, slope stability failure, soil stiffness degradation (fatigue) and erosion on the performance of foundations subjected to extreme loading conditions (earthquakes, hurricanes etc).  We then develop simplified, physics-based models based on the simulations to improve the state of practice in engineering design. Our models have found applications in the design of waterfront structures, bridges and offshore wind turbines.

Recent Publications

  • Varun V., Asimaki D. and Shafieezadeh A. (2012). Soil-Pile-Structure Interaction Simulations in Liquefiable Soils via Dynamic Macroelements: Formulation and Validation, Invited paper, Soil Dynamics and Earthquake Engineering, Jose Rœsset Special Issue, (dx.doi.org/10.1016/j.soildyn.2012.03.008)
  • Varun V. and Asimaki D. (2012). A Generalized Hysteresis Model for Biaxial Response of Pile Foundations in Cohesionless Soils, Soil Dynamics and Earthquake Engineering, 32(1), 56-70 (Top 25 Most Downloaded SDEE Papers)

top

Inverse Problems in Site Characterization


Horizontally stratified geologic formation in Utah. Because
of the horizontal stratification, site characterization at this
site is based on one-dimensional wave propagation theory

Ground deformation and failure forecasting during earthquakes is based on mathematical models that describe the idealized material behavior in monotonic (unidirectional) loading, and a set of rules that describe the unloading and reloading of the material by cyclic-type loads such as earthquake shaking. Most existing models, however, have been based on small-scale laboratory experiments, and very few studies exist where the inelastic, dynamic soil behavior is measured from field experiments in-situ. Our research group develops optimization algorithms to extract the inelastic soil parameters from ground motion borehole recordings, surface wave testing and ambient noise. Our algorithms extract information on soil deformation, soil strength, material attenuation and scattering. We then use this information for predictions of ground deformation in future earthquakes, which help guide the design of distributed infrastructure systems (pipelines, roads, bridges); as well as to compute the maximum ground motion that a site can sustain, which helps guide the design of critical infrastructure such as nuclear power plants and dams.

Recent Publications

  • Asimaki D., Kallivokas L.F., Kang J.W., Li W., Kucukcoban S. (2012). Time-domain forward and inverse modeling of lossy soils with frequency-independent Q for near-surface applications, Soil Dynamics and Earthquake Engineering, 43, 129-159
  • Asimaki D., Kalos A. and Li W. (2010). A wavelet-based seismogram inversion algorithm for the in-situ characterization of nonlinear soil behavior, Pure and Applied Geophysics, 168(10), 1669 - 1691

top

Near real time liquefaction and ground deformation on regional scales


Landslide in El Salvador during the
M7.7 Earthquake of 2001 (source: Wikipedia)


Landslide in El Salvador
during the M7.7
Earthquake of 2001
(source: Wikipedia)

Damage and loss assessment of urban infrastructure typically relies on uncoupled sequences of probabilistic ground motion models, soil and structural analyses of lifeline systems, and empirical damage and loss estimations. This approach cannot predict liquefaction or permanent soil deformations from 3D wave propagation phenomena like basin effects, or the interaction between soil deformations and buried pipelines on a city-scale. In turn, the reliability of this uncoupled approach is hard to quantify for distributed systems that are vulnerable to ground deformations and liquefaction and are often built on 3D geologic formations. Recent catastrophic have shown how the planning, design and vulnerabilities of distributed systems dictate a city's capacity to absorb, recover from, and mitigate disasters; the unprecedented extent of liquefaction in Christchurch, for example, caused widespread lifeline failures, leaving residents without water supply for days or weeks. Our research in this area seeks to replace the empirical, uncoupled state-of-practice approach with a physics-based assessment framework for distributed systems. To address this problem, we are developing material models and multi-scale algorithms to enable ground deformation, liquefaction and landslide predictions on regional scales. Our collaborators include seismologists, structural engineers, lifeline and visualization experts.

top