​​​​​​​

Update of Hazus Annualized Earthquake Loss Estimates for California*


Rui Chen and Chris J. Wills
California Geological Survey

September 2016

*This is a brief summary of loss modeling for California documented in Chen, R., K.S. Jaiswal, D. Bausch, H. Seligson, and C.J. Wills (2016). Annualized Earthquake Loss Estimates for California and Their Sensitivity to Site Amplification, Seismological Research Letters, 87 (8), Pre-Issue Publication Article, doi: 10.1785/0220160099.


Earthquake risk estimation requires an integration of long-term earthquake hazards, an inventory of the human and building stock subjected to the underlying hazards, and their relative vulnerability/susceptibility to damage when exposed to such hazards. Annualized earthquake loss (AEL) provides a long-term average loss per year in a specified geographic area (such as state, county, or census tract). AEL is a reasonable indicator of relative regional earthquake risk and, therefore, facilitates understanding and comparison of earthquake risk among different communities. Earthquake loss estimates support stakeholders in preparing emergency response plans, developing earthquake-hazard mitigation strategies, and establishing earthquake insurance policies.

AEL estimation is dependent upon a number of input datasets and models that characterize earthquake ground-motion hazards and their effects on the built environment. Updates to these datasets and models are necessary to account for continuing development of the built environment and evolving methods in characterizing earthquake hazards; for example, development of  new earthquake ground-motion prediction equations (GMPEs), seismic-hazard assessment techniques, and site-effect considerations.

Nearly all inputs used in the 2010 CGS loss analysis are outdated.  Since that study, the Federal Emergency Management Agency completed enhancement of the inventory data within their loss estimation tool, Hazus, relying upon the 2010 census data and incorporating 2014 building and content exposure cost valuations. Similarly, the U.S. Geological Survey updated National Seismic Hazard Model (NSHM) in 2014 (Petersen et al., 2014, 2015). In addition, a new semi-empirical non-linear site amplification model was developed by Seyhan and Stewart (2014) and a revised map of time-averaged shear-wave velocity in the upper 30 m (VS30) for California was developed by Wills et al. (2015). Both affect ground-motion calculations in seismic-hazard assessment. ​

This investigation updates the 2010 CGS loss analysis for California using aforementioned latest input datasets and models. In addition to AELs, estimated losses are presented in terms of annualized earthquake loss ratio (AELR) that are calculated as the ratio of AEL for a specified geographic area to the total building replacement value of that area. To facilitate presentation and discussion, AELR is multiplied by 100 and termed annualized percent earthquake loss (APEL). Because APEL is normalized by the building replacement value, it is an alternative indicator of relative seismic hazard and building vulnerability in different geographic areas. Estimated AELs are presented at multiple resolutions, starting with the state level assessment and followed by assessments for counties, metropolitan statistical areas (MSAs), and incorporated cities.

Economic losses reported in this study are from shaking-related building damage. They do not include losses due to ground-failure effects such as landslide, liquefaction, and surface fault rupture, or due to other secondary losses such as fires following earthquakes. Building economic losses are direct economic losses, including structural damage, nonstructural damage, and content damage; as well as building damage-related economic losses, such as inventory loss, relocation cost, loss of proprietors’ income, and rental income loss. They do not include losses associated with business interruption. Losses from individual earthquakes can be very different from AEL estimates and may vary significantly depending on earthquake location, magnitude, and many other factors. Also, typically losses would occur in major events in which several billions of dollars would be lost interspersed with periods in which few losses would be incurred.

The updated statewide AEL estimate from this study is $3.7 billion for California. This estimate is about 11% higher than the 2010 CGS estimate. The difference is attributed to the combined effects of increased building inventory value (17%), differences between the 2008 and 2014 NSHMs, and to a lesser degree, differences in VS30 maps (Wills and Clahan, 2006 vs. Wills et al., 2015). Some difference in statewide AEL estimates for California between this study and that of the 2010 CGS study is also due to different approaches used to incorporate site amplification. Ground-motion scaling for site effects in this study is based on the Seyhan and Stewart (2014) model, whereas in the 2010 CGS study, ground motions at each grid point were calculated and scaled for site condition using GMPEs in the 2008 update of the NSHM (i.e., GMPE scaling). 

Figure 1 depicts building AEL and APEL at the county level; the top 10 counties are noted on each map. Table 1 lists 10 counties with the highest estimated AEL, county AEL as a percent of the statewide AEL, and 10 counties with the highest APEL. Los Angeles County, with both high exposure (over 25% of the state total exposure) and high hazards, tops the list with almost one-third of the state AEL. More than 80% of the state AEL occurs in the top 10 counties. Counties with high APEL are generally in areas with high-seismic hazards, such as along the San Andreas fault system, near the Cascadia subduction zone, and in the Transverse Ranges. 

Table 2 lists ten MSAs and ten cities with the highest estimated AEL. More than 70% of the state AEL occurs in the top three MSAs, namely Los Angeles-Long Beach-Santa Ana, San Francisco-Oakland-Fremont, and Riverside-San Bernardino-Ontario. Nearly 25% of state total AEL is contributed by the top five cities: Los Angeles, San Jose, San Francisco, Oakland, and San Diego. A complete listing of AELs and APELs for all California counties, MSAs, and incorporated cities with estimated AEL greater than $1000 can be downloaded in the Supplement Materials appended to the end of this summary (Tables S1, S2, and S3, respectively). Maps of building AEL and APEL at census tract level are also provided in the Supplemental Materials (Figures S1 and S2, respectively).  Because boundaries of MSAs and cities are not standard, we provide boundary polygon ArcGIS shapefiles as zipped folders in the Supplemental Materials.



​​
Table 1. Ten California counties with the highest estimated annualized earthquake loss (AEL) and ten counties with the highest estimated annualized percent earthquake loss (APEL)
​ ​R​ank​​ County AEL
($ Million)
% ofState Total Rank County APEL (%)
1​Los Angeles1139.430.61San Benito0.217
2Santa Clara331.88.92Humboldt0.200
3Alameda297.28.03Imperial0.175
4Orange259.07.04Alameda0.164
5San Bernardino225.26.05Santa Clara0.158
6Riverside208.45.66Del Norte0.147
7Contra Costa171.44.67San Mateo0.139
8San Francisco142.13.88Contra Costa0.138
9San Mateo131.73.59Santa Cruz0.135
10San Diego123.83.310Napa0.130
 Top 10 county sub-total​ 3030.0  81.3 


Table 2. Ten California metropolitan statistical areas and ten cities with the highest estimated annualized building loss

Rank Metropolitan Statistical Area AEL
($ Million)
% of State Total City AEL
($ Million)
% of State Total
1Los Angeles-Long Beach-Santa Ana1398.437.54Los Angeles483.513.0
2San Francisco-Oakland-Fremont776.720.85San Jose148.14.0
3Riverside-San Bernardino-Ontario433.711.64San Francisco138.03.7
4San Jose-Sunnyvale-Santa Clara343.09.21Oakland75.82.0
5San Diego-Carlsbad-San Marcos123.83.32San Diego64.21.7
6Oxnard-Thousand Oaks-Ventura89.12.39Fremont50.21.3
7Santa Rosa-Petaluma77.32.07Long Beach36.61.0
8Sacramento--Arden-Arcade-Roseville51.01.37San Bernardino31.20.8
9Vallejo-Fairfield43.21.16Hayward29.90.8
10Santa Cruz-Watsonville42.91.15Santa Clarita28.90.8


Conclusions

1. The updated total AEL for California is ∼3:7 billion, which is about 0.16% of the 2014 California gross state product. It is about seven times the projected direct economic losses from the 2014 M 6.0 South Napa earthquake, which are reported to be on the order of $0.5 billion by Earthquake Engineering Research Institute. 

2. More than 70% of state AEL occurs in three MSAs: Los Angeles–Long Beach–Santa Ana, San Francisco–Oakland–Fremont, and Riverside–San Bernardino–Ontario.

3. More than 80% of the state AEL occurs in 10 out of 58 California counties: Los Angeles, Santa Clara, Alameda, Orange, San Bernardino, Riverside, Contra Costa, San Francisco, San Mateo, and San Diego. 

4. Given the very high economic exposure and population in addition to its proximity to many of the most seismogenic faults in the country, Los Angeles County contributes a significant portion (over 30%) of the estimated state AEL.

References

Chen, R., K.S. Jaiswal, D. Bausch, H. Seligson, and C.J. Wills (2016). Annualized Earthquake Loss Estimates for California and Their Sensitivity to Site Amplification, Seismological Research Letters, v 87 (8), Pre-Issue Publication Article, doi: 10.1785/0220160099.

Petersen, M.D., M.P. Moschetti, P.M. Powers, C.S. Mueller, K.M. Haller, A.D. Frankel, Y. Zeng, S. Rezaeian, S.C. Harmsen, O.S. Boyd, N. Field, R. Chen, K.S. Rukstales, N. Luco, R.L. Wheeler, R.S. Williams, and A.H. Olsen (2015). The 2014 United States national seismic hazard model, Earthquake Spectra, v 31 (S1), S1-S30, doi: 10.1193/120814EQS210M. 

Petersen, M.D., M.P. Moschetti, P.M. Powers, C.S. Mueller, K.M. Haller, A.D. Frankel, Y. Zeng, S. Rezaeian, S.C. Harmsen, O.S. Boyd, N. Field, R. Chen, K.S. Rukstales, N. Luco, R.L. Wheeler, R.S. Williams, and A.H. Olsen (2014). Documentation for the 2014 Update of the United States National Seismic Hazard Maps, USGS Open-File Report 2014-1091​, 243 pp. 

Seyhan, E., and J.P. Stewart (2014). Semi-empirical nonlinear site amplification from NGA-West2 data and simulations, Earthquake Spectra, v 30 (3), 1241-1256, doi: 10.1193/063013EQS181M.

Wills, C.J., C.I. Gutierrez, F.G. Perez, and D.M. Branum (2015). A next-generation VS30 map for California based on geology and topography, Bulletin of the Seismological Society of America, v 105 (6), 3083-3091, doi: 10.1785/0120150105.

Wills, C. J., and K. B. Clahan (2006). Developing a map of geologically defined site-condition categories for California, Bulletin Seismological Society of America, v 96 (4A), 1483–1501, doi: 10.1785/0120050179.


Sup​plemental Materials

Tables

Table S1. Estimated AELs for all California counties (in alphabetic order), county AEL as a percentage of the statewide AEL, and APELs.
Table S2. Estimated AELs for MSAs, MSA AEL as a percentage of statewide AEL, and MSA APELs.
Table S3. Estimated AELs for California incorporated cities with estimated AEL greater than $1000 (in alphabetic order), city AEL as a percentage of the statewide AEL, and city APELs.
________________________________________

Figures

Figure S1. Distribution of building AEL in California by census tract.
Figure S2. Distribution of APEL in California by census tract.
________________________________________

Other

Download: Metropolitan Statistical Areas.zip [Zipped ArcGIS File; ∼1751 KB] ArcGIS shapefile for boundary polygons.
Download: Incorporated Cities.zip​ [Zipped ArcGIS File; ~7735 KB] ArcGIS shapefile for boundary polygons.