Quantifying Risk Requires More than Hazard Data

The insurance industry primarily utilizes earthquake data through the use of probabilistic risk models.

Ultimately, the goal of using a probabilistic risk model is to become more resilient.  With catastrophe risk models and analytics to help quantify damage and losses from an earthquake, as well as identify the location and extent of impacts, risk managers can become more prepared and more resilient.

The first step in assessing risk is understanding hazard. National seismic hazard maps (Figure 1), like this example from the U.S. Geological Survey, are a common use of hazard data, outlining the severity and probability of potential earthquake occurrence.

But there’s more to risk than just hazard. Assessing risk takes it a step further by incorporating proximity to exposure and vulnerability, or damageability, of structures.

There are four steps in probabilistic earthquake risk modeling (Figure 2).

The first step defines the hazard. This starts by identifying all sources of a potential earthquake, such as known faults and areas of seismicity, and then assigning a range of magnitudes to each source and determining the probability of each magnitude actually occurring.

The second step takes each possible event and calculates the ground motion, or level of shaking, to estimate a footprint of the impacted area.

The third step is the vulnerability component, which relates hazard to damage. In this step we take into consideration things like building practices and codes to compute damage for different structure types.

The final step is a sophisticated financial model that computes the financial loss for each possible event.

The input to a probabilistic risk model is generally a portfolio of assets.

Important questions a risk manager would have include: Where are the risks? How concentrated are they? How vulnerable are they based on the building’s age and construction type?

For the insurance industry, it is important to evaluate the full spectrum of risk. The more frequent, yet generally less damaging, events can have an impact on cash flow, while extreme events can impact the solvency of a company (Figure 3).

The advantage of using a probabilistic risk model, rather than just a hazard map, is that you can capture the range of potential events to gain insight into the realm of possibilities and the resulting impact.

Even though probabilistic risk models can estimate loss of life or casualty impacts, they are most commonly used for estimating financial loss.

In addition to estimating financial loss, probabilistic risk models can provide added insights to help insurers assess where risk comes from. For example, insurers might want to know which fault sources or magnitudes are driving the risk.

Ultimately, the goal of using a probabilistic risk model is to become more resilient. With catastrophe risk models and analytics to help quantify damage and losses from an earthquake, as well as identify the location and extent of impacts, risk managers can become more prepared and more resilient.

CoreLogic.com