Return to Hazards Center Home Page
The session focused on developments in the field of loss estimation, with examples of some specific projects.
Jim Davis discussed the role of modeling and risk assessment in the public policy sector. Loss estimation modeling is useful in identifying, assessing, and managing risk. This information is used by government officials to determine the acceptable level of risk and to come to consensus on how to manage it. Davis stated that public policies should be based on the finding of public models. No black boxes should be permitted so that all assumptions can be openly tested and examined publicly. Risk analysis using models is useful for helping the public to understand the risks. This often leads to using earthquake resistant building practices to reduce losses. In California, earthquake risk assessment techniques utilize models and historical data on the frequency and magnitude of earthquakes on individual faults to predict the probability and random distribution of occurrences on other faults.
Ron Eguchi discussed how the Northridge earthquake may be viewed in terms of capital losses. EQE compared estimated direct economic losses to actual losses from that event. He stated that gathering actual loss data is important for calibrating loss estimation models following major events. The Northridge earthquake was a moderate event hitting a suburban area. A similar event in a more densely populated region would likely result in significantly higher losses. EQE summed various loss data including insured losses; losses accounted through FEMA's Public Assistance, Hazard Mitigation, and Individual Assistance grant programs; and loans from the Small Business Administration. They also used EPEDAT--an EQE product--to estimate the losses. The study concluded that 1) the estimated undocumented losses, which are significant, may be low by a factor of two; 2) documented costs may not be known for at least two years, when they begin to level off; and 3) the cost of the Northridge earthquake could reach $40 billion.
Patricia Grossi, a graduate student at the University of Pennsylvania, is researching ways that the public and private sectors can manage risks from catastrophic natural hazards. The study, currently underway, examines the impact of alternative mitigation strategies on reducing losses from future natural disasters. Calculations are performed using various software methodologies, including HAZUS, developed by the Federal Emergency Management Agency and the National Institute of Building Sciences.
Chris Tucker discussed risk assessment and natural hazards in Canada, where losses from disasters, particularly floods and ice storms, have increased dramatically in recent years. Tucker is using NHEMATIS, an ArcView based loss model. It is currently used for research, planning, and education and can run real-time scenarios for earthquakes, floods, and landslides. He is tracking the costs of disasters by teaming with insurance companies and officials in the various provinces. One major problem in Canada is collecting data, which is primarily not in the public domain.
Greta Ljung described the Institute for Business and Home Safety (IBHS) insured losses database. It has information on earthquakes, hurricanes, tropical storms, wind events with hail, and other wind events. The data is broken down by state and may also be mapped.
Return to Hazards Center Home Page
Return
to Index of 1998 Session Summaries