Expected Housing Losses in an Earthquake

In 2003, ABAG updated the housing loss estimates in the report Preventing the Nightmare to be consistent with the scenarios used in the USGS Working Group 2002 probability models. The housing loss model is based on building damage data from past California earthquakes. For more information on the assumptions used in this model, see sources.

How large will housing losses be?

The Loma Prieta earthquake caused a total of over 16,000 uninhabitable units throughout the Monterey and San Francisco Bay Areas. Thirteen of the 18 potential Bay Area earthquake scenarios have a far larger impact than the Loma Prieta Earthquake. Many Bay Area faults can generate earthquakes and every Bay Area county is significantly affected by at least two. Several of the larger earthquakes will impact several counties severely.

Source: The following estimates of uninhabitable units for Bay Area counties in selected earthquake scenarios is summarized from the ABAG 2003 report Preventing the Nightmare. Download the full table.

What building types will be affected?

Source: adapted from Shaken Awake!, ABAG 1996.

More information about impacts of past California earthquakes on buildings

How many people will need shelter?

Source: Preventing the Nightmare, ABAG 1999

What Can I Do to Prevent Housing Losses?

Residents and building owners

Local governments

Sources

• Building damage data used to develop uninhabitable housing units model
• Location of housing damage in Loma Prieta and Northridge earthquakes
• Shaken Awake! Estimates of uninhabitable dwelling units and peak shelter populations in future earthquakes affecting the San Francisco Bay Area, ABAG 1996. This report outlines the assumptions used in all of ABAG’s uninhabitable housing modeling. Purchase the full report.
• Preventing the Nightmare: Designing a Model Program to Encourage Owners of Homes and Apartments to Do Earthquake Retrofits, ABAG 2003 update of uninhabitable housing losses. Download updated section or purchase original report. Update based on scenarios used in the USGS Working Group 2002 probability models.