Towards Seismic Resilience: A Comprehensive Approach to Modeling Earthquake Consequences from Regional Hazard to Household Well-Being

Understanding disaster risk is an integral aspect of disaster risk management and it is essential for
efficient risk reduction investments. Regional disaster risk assessment frameworks, including seismic
risk analysis, are commonly used to quantify disaster risk and inform risk reduction strategies. However, few of the frameworks are able to quantify the full extent of disaster consequences especially as
related to household recovery and the disproportionate effect on the socioeconomically vulnerable
population. This dissertation proposes a set of new computational methods along various steps of
the seismic risk assessment process and lays out the overall methodology for quantifying disaster
impacts on the society considering the entire recovery process. This is achieved by integrating and
leveraging analytical tools from other fields, namely, geostatistics, real estate investment, macroeconomics, and welfare economics.

The dissertation first proposes a framework for simulating spatially distributed ground motions
at multiple intensity measures. The framework allows modelers to rapidly simulate multiple realizations of earthquake ground motion intensity, which are required for regional risk assessments.
Furthermore, the research investigates post-earthquake decisions on damaged commercial buildings
considering the influence of real estate market conditions. An investment decision model for individual buildings that unifies seismic performance estimation and real estate investment analysis is
proposed and applied to four commercial buildings in the Los Angeles area. Results suggest that
the state of the real estate market has a significant influence on decisions and recovery trajectories
of commercial buildings.

Finally, this dissertation integrates regional probabilistic seismic risk analysis, which quantifies
asset losses, with an economic recovery model to estimate the overall economic impact of a large
earthquake on the San Francisco Bay Area. In addition to quantifying changes in economic output, this approach gives insight into changes in income and employment across different economic
sectors. The results of this analysis are combined with information on income, employment sector,
housing values, savings, mortgage, rent, and insurance of households across the Bay Area, to calculate consumption changes throughout the recovery. Furthermore, well-being losses (i.e., changes in the utility of consumption over the recovery period) are calculated and analyzed for different
socioeconomic groups. Well-being losses, unlike asset losses which use absolute monetary values, are
able to reflect the differential impact that a loss in consumption of $1 has on a low income versus a
high income household. The analysis reveals that poorer households suffer only 19% of the overall
physical asset losses, but experience 41% of the well-being losses. Well-being losses also extend over
a larger region than that of severe asset losses, requiring design of policies to help people recover, in
addition to reducing asset losses. We evaluate several pre- and post-disaster risk reduction strategies
in terms of their effect on both asset and well-being losses and examine their efficacy across different
socioeconomic groups.

The tools developed in this dissertation are meant to be used by researchers, practitioners, and
government agencies to understand the extent of the disaster risk. They can help policymakers
identify vulnerable groups within their community who might require pre- and post-disaster assistance, and design efficient investment strategies for disaster resilience considering socioeconomic
characteristics and vulnerabilities of the population at risk.

Markhvida, M and Baker, JW and and . (2020). Towards Seismic Resilience:
A Comprehensive Approach to Modeling Earthquake Consequences from Regional Hazard
to Household Well-Being. Blume Earthquake Engineering Center Technical Report 205.
Stanford Digital Repository. Available at: https://purl.stanford.edu/yg981dy5678