The focus of our next series of blog posts is seismic risk assessments (SRAs).
We’ll delve into the aspects of SRAs that may be unfamiliar to many, especially given the somewhat recent standardization of these assessments and updated language. This post will tackle the concepts of scenario expected loss (SEL) and scenario upper loss (SUL), as they are far more commonly used in the industry than probable loss.
Redefining Probable Maximum Loss
Before getting into SEL & SUL, it’s important to establish the definitions and terms laid out by the latest ASTM Seismic Standards (E2026-16a). Historically, the term probable maximum loss (PML) has been used to describe building damageability in earthquakes, that is, the likelihood a building will experience damage during a seismic event. Without a standard definition, inconsistencies resulted, and PML now has “a number of very different explicit and implicit definitions.”
So, ASTM cleaned the slate and introduced the concepts of probable loss and scenario loss. Probable loss is more conservative, as it takes into account all of the uncertainties that can impact damage in an earthquake, whereas scenario loss reflects damage losses for specific earthquake scenarios, further defined as SEL & SUL. While the term PML is discouraged from use, it is not wholly unacceptable, given that it is defined in terms of probable or scenario loss.
Commonalities and Differences in SEL and SUL
SEL and SUL have some commonalities, such as the scenarios applied to their formulation. Although other scenarios are possible and sometimes used, the most common scenario earthquake used is the “475-year earthquake,” an earthquake that will typically occur once every 475 years. For any given assessment, the SEL and SUL share the same scenario earthquake event. Furthermore, the SEL and SUL assessments share the same building damageability.
The major difference between SEL and SUL is the confidence level of the loss estimate (the conclusion of the assessment), detailed below:
Scenario Expected Loss (SEL)
50% confidence level on the loss estimate;
A 50% confidence level on the loss estimate, the SEL, means that half of all similar buildings would suffer the same amount of damage assessed, or less.
Scenario Upper Loss (SUL)
90% confidence level on the loss estimate;
A 90% confidence level on the loss estimate, the SUL, means that nine out of ten similar buildings would suffer the same amount of damage assessed, or less.
Rolling the Dice on Damageability
For example, imagine rolling a six-sided die. A single roll would be the scenario, the values on each side the building damageability, and the result of the dice roll the actual building damage. Remember, the scenario and the damageability are the same for both the SEL and SUL. However, the confidence level—and hence the conclusions—of SEL and SUL for the assessment are different.
The SEL for a single roll of a six-sided dice is three. Half of all dice rolls will have a value of three or less. The SEL for a single roll of two six-sided dice would be seven.
The SUL for a single roll of a six-sided dice is six. Nine out of ten dice rolls will have a value of six or less. (Indeed, all dice rolls will have a value of six or less but an assessment of five would only have an 83% confidence level. So the SUL has to be higher and the only higher value in this assessment is six.) The SUL for a single roll of two six-sided dice would be ten (10).
Probability, Not Prediction
This example demonstrates how scenario loss is a probabilistic assessment, not a prediction of earthquake damage. There are too many unknown factors to make such a prediction for both the hazard and building damageability.
Despite considerable research into estimating scenario earthquakes, specifics such as magnitude, duration, frequency content, pulse effects, direction, etc., can vary considerably. Even some aspects of site and building stability cannot be completely or perfectly known, which compounds the uncertainty, making predictions of earthquake damage highly impractical.
Lenders, insurers, property owners, and investors can gain tremendous value in understanding these integral concepts of SRAs. An outcome of 20% for SEL versus 20% for SUL indicate very different conditions. Also, lenders often have different requirements, either based on SEL or SUL. Determining whether to use SEL or SUL will largely depend on the client’s risk aversion. As such, no seismic assessment is (or should be) created equal; EBI Consulting’s team of experts follow the highest ASTM standards to provide quality, custom seismic risk assessments and analyses to fit your specific business needs. For more information on seismic risk assessments, read more on the EBI website or contact us.