SCEC Award Number 11171 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title Trimming of the UCERF 2 Logic Tree: Assessing the Value of Hazard Information with Portfolio Loss Exceedance Curves
Investigator(s)
Name Organization
Keith Porter SPA Risk LLC
Other Participants Potentially 1 graduate student or postdoctoral scholar (possibly Dr Mohammad Javanbarg) may contribute to the research.
SCEC Priorities A2, C, B6 SCEC Groups SHRA
Report Due Date 02/29/2012 Date Report Submitted N/A
Project Abstract
 N/A
Intellectual Merit The research quantifies the importance of the individual branches of the logic tree that comprises the Uniform California Earthquake Rupture Forecast version 2 (UCERF2) in terms of a measure of statewide economic risk. It shows that uncertainty in this risk metric—the expected annualized loss to all woodframe single-family dwellings—is dominated by the ground-motion prediction equation, UCERF2 recurrence model, and in the case of Brownian-passage-time recurrence model, aperiodicity. Risk is much less sensitive to the other uncertainties.

One implication is that a statewide risk calculation employing only 5 branches of the UCERF2 tree and 4 NGA relationships (total 20 branches) captures virtually all the uncertainty produced by the full complement of 480 UCERF2 branches and 4 NGA branches (total 1,920). This allows for a 100x reduction in computational effort with no corresponding underestimation of uncertainty. Another implication is that efforts to reduce these two epistemic uncertainties might be the most cost-effective way to reduce uncertainty in statewide risk resulting from UCERF2 and ground-motion-prediction equations.

Note well: the research says nothing about the scientific importance of the trimmed branches, only that a risk calculation may not need them, and that reducing the uncertainty in the remaining branches can reduce calculated upper-bound risk. By “upper-bound risk” is meant large, rare losses such as the commonly used 250-year probable maximum loss. Since upper-bound risk drives reinsurance requirements for insurers, reducing uncertainty can reduce calculated upper-bound risk and thus reduce reinsurance requirements and cost. The tree-trimming exercise identifies those branches where reduction in uncertainty could most efficiently reduce reinsurance costs.
Broader Impacts The research produced a new OpenSHA/OpenRisk tool: the Portfolio EAL Calculator, which now has been fitted with a nearly exhaustive set of vulnerability functions and an estimate of statewide single-family woodframe dwellings. The calculator is capable of estimating the expected annualized loss (EAL) for a portfolio of assets, given an ERF and GMPE, and thus allows one to explore one of the implications of new knowledge about either ERF or GMPE.

The research also strengthened the relationship between CU and SCEC, particularly between Porter and Milner, now a USC graduate student. Porter provided the domain expertise required to develop the OpenSHA Portfolio EAL Calculator. Milner implemented it, and using SCEC's HPC facilities, performed the calculations required by the study.
Exemplary Figure N/A