Probabilistic evaluation of liquefaction-induced settlement mapping through multiscale random field models

Understanding and assessing the spatial extent of liquefaction hazard requires the spatial dependence of geotechnical properties to be taken into account. In this work, novel multiscale random field models will be developed to characterize and quantify spatial variability of geotechnical parameters across scales. Such models will then be integrated with a simplified procedure for the probabilistic evaluation of liquefaction-induced settlements based on in-situ field tests such as the cone penetration test (CPT). Joint distribution of peak ground acceleration and moment magnitude of earthquake will be derived from the Seismic Hazard Maps using a simplified procedure. A unique feature of the proposed work is its ability to refine and obtain higher resolution random fields for geotechnical properties in critical areas, such as adjacent to important infrastructure, or in areas with detailed small-scale field data. The proposed framework is then applied to evaluate the liquefactioninduced settlement of an earthquake-prone area to demonstrate its applicability.