20 September 2000 10:00 AM PDT

New FEMA Study Estimates U.S. Losses From Earthquakes At $4.4 Billion Per Year

• Statement by the Director of the Nevada Seismological Laboratory
• Statement by the Nevada State Geologist
• FEMA Press Release
• FEMA Report
  • Local copy of report
  • Adobe Acrobat Reader version 4 is required to view the report.
• Nevada Seismological Laboratory
• Nevada Bureau of Mines and Geology
• Statement by Thomas Henyey, Director of the Southern California Earthquake Center

The Federal Emergency Management Agency (FEMA) has just released a document entitled HAZUS99 Estimated Annualized Earthquake Losses for the United States. The document presents estimates, using the HAZUS computer program written for FEMA, of annualized losses due to earthquakes in all 50 states plus the District of Columbia, and in 40 metropolitan areas of the United States. The report also estimates the ratio of average annualized loss to replacement value for the states and metropolitan areas.

The Nevada Seismological Laboratory has only just received a copy of the report and has not had a chance to analyze it thoroughly. The following discussion is based on our first impression of the contents of the document. We will follow this with additional comments as appropriate.

In Nevada this document is likely to attract considerable attention for several reasons:

1. In terms of annualized losses, Nevada ranks fifth of all the states. Among metropolitan areas outside California, Las Vegas ranks 7th and Reno ranks 11th. (The report uses Washoe County to represent Reno; and combines Clark County, Nye County, and Arizona's Mohave County into the Las Vegas metro area.)

2. In the loss ratio, Nevada again ranks fifth among all states. Among metropolitan regions outside California, Reno ranks third highest after Hilo and Anchorage, and Las Vegas ranks tenth.

3. The average annualized losses are high. For Nevada, the model finds an average loss of about $55 million dollars per year. This breaks into about $28 million dollars per year for Las Vegas and $18 million dollars per year for Reno, with the remainder presumably spread among the more rural counties. One way to think of these numbers is as the consequence of a Northridge earthquake sized loss ($28 billion) about once every 1000 years for Las Vegas and somewhat smaller sized but more frequent losses for the Reno-Carson City urban corridor (e.g., about $2 billion every 100 years). The complete model includes a combination of more frequent losses of smaller size and enormous losses that occur less frequently.

The following are my interpretations of the numbers in this report.

1. Considering the rankings, the states and metropolitan areas with the highest seismic hazards generally rank highest. In average loss, the top five states are California, Washington, Oregon, New York, and Nevada. In loss ratio, the top five are California, Alaska, Oregon, Washington, and Nevada. New York ranks higher than Nevada in average loss per year, due to the risk to New York City: its enormous size, its lack of seismic resistance in its buildings, and the consequent potential for extremely extensive damage if struck by even a very rare, moderate sized earthquake. In loss ratio, New York drops to number twenty. California is number one in both average loss per year and loss ratio by a wide margin.

2. The Nevada Seismological Laboratory has reported often that Nevada is seismically the 3rd most active state. The FEMA results are consistent with this, and from that perspective, they are not a surprise. Our ranking behind Washington and Oregon in average annualized loss is the result of the more extensive urban development in those states, and our ranking behind them in the loss ratio by state is because the greatest part of the population in those states lives in an area more hazardous than Las Vegas.

3. The fact that loss ratios approximately follow the seismic hazard indicates that in the HAZUS model, building to higher building code levels does not fully compensate for the higher seismic hazard that is present.

4. There are major uncertainties in the input to FEMA's loss estimates, but Nevada was treated the same way as other states, so therefore in terms of relative hazard the report is probably about right.

Specific comments for Nevada:

1. The absolute values may be off by a wide margin. Furthermore, the report uses 1990 census data for population, very rudimentary estimates for effects of the type of soil, and similarly rudimentary estimates for the characteristics of the building stock. The uncertainties in the estimates of the loss from individual earthquakes is at least a factor of 3-4, and may be as high as a factor of ten. In spite of this, it should be emphasized that the report uses a national methodology that can be regarded as the best available. In a discussion of the results among members of the Nevada Seismological Laboratory, although there are a variety of ideas about how the report can be improved; one thing that we all agreed on is that we cannot demonstrate that FEMA's estimates are wrong in its main conclusions.

2. Even though the HAZUS results have very large uncertainties, Nevada still has an enormous average annual loss.

3. The uncertainties can be reduced by more research. Specifically, more research can help to characterize the earthquake hazards that face Nevada much more precisely than what is presently known. Research can include mapping faults, characterizing the ground motion that will affect structures including complications caused by local soil conditions, and improved understanding of the locations and rates of earthquakes.

4. Some research is ongoing. The Reno-Carson City urban corridor, and Las Vegas to a lesser extent, have benefited from research supported through the National Earthquake Hazard Reduction Program (NEHRP) that is administered by the US Geological Survey. The seismic network operated by the Nevada Seismological Laboratory in this part of Nevada is supported by NEHRP in partnership with the State of Nevada. The FEMA report confirms that it has been an appropriate decision on the part of the USGS to support this network and other research projects. It also confirms the importance of continuing and hopefully expanding that support. Everyone in Nevada will benefit if Congress approves and fully funds continuation of the National Earthquake Hazard Reduction Program.

5. Las Vegas has never been a priority for seismic monitoring, either with Nevada or Federal support. It is our hope that this will change in the near future. The Nevada Seismological Laboratory has installed a minimal monitoring network around Las Vegas, but without substantial Federal support an adequate network will not be possible. It is our hope that Congress will approve and fully fund the Advanced National Seismic System (ANSS). The ANSS initiative, led by the US Geological Survey with input from the seismic network operators nationwide (including the Nevada Seismological Laboratory) explicitly designates Las Vegas as a priority area for seismic monitoring.

6. There are actions Nevadans can take to reduce their losses. The estimates include non-structural losses (i.e. loss of contents) and business interruption losses. Loss to contents can be mitigated. Business interruption losses can be minimized by following an aggressive program of preparing our major industries (especially gaming) to deal with the problems. Long-term activities include developing a thorough understanding of the earthquake hazard in Nevada, and the implementation and enforcement of high standards in building codes.

For more information about earthquakes in Nevada and how to be prepared, please contact John G. Anderson, Director, Nevada Seismological Laboratory (775-784-4265), Jonathan G. Price, Director and State Geologist, Nevada Bureau of Mines and Geology (775-784-6691 extension 126), or Frank Siracusa, Chief, Nevada Division of Emergency Management (775-687-4240) and see the Web sites of the Nevada Seismological Laboratory (www.seismo.unr.edu), the Nevada Earthquake Safety Council (www.nbmg.unr.edu/nesc/index.html), the Nevada Bureau of Mines and Geology (www.nbmg.unr.edu), and the Nevada Division of Emergency Management (www.state.nv.us/dem/index.htm). The column "Opinion: Exceeding Seismic Code Pays in the Long Run" (which is accessed from the page on "Preparedness Information from the NESC page of the Seismological Laboratory web site: http://www.seismo.unr.edu/nesc/rgj-jga980125.html) may be of particular interest.

Nevada has significant earthquake hazards and risks. On Wednesday, September 20, 2000, the Federal Emergency Management Agency (FEMA) released a report titled "HAZUS 99 Estimated Annualized Earthquake Losses for the United States." In this report FEMA estimates that the loss from earthquakes for residential and business buildings in Nevada is likely to be, on the average, approximately $55 million per year. In any given year, the actual loss is likely to be either considerably lower or higher, because earthquakes are infrequent events. One way to think of these numbers is as the result of approximately a Northridge earthquake-sized loss once every 300 years in Las Vegas, and also once every 300 years in Reno. This model includes direct damage to buildings, non-structural damage, and part of the economic loss from not being able to use many buildings immediately after an earthquake. The figure is a minimum estimate of potential loss, because it does not include damage to infrastructure (such as roads, bridges, and electrical, water, sewage, and communication systems) or building contents, and it does not consider indirect economic losses, such as effects on tourism.

Nevada has experienced many historical earthquakes. Although earthquakes don’t occur at regular intervals, in historical times we have experienced a magnitude 7 or greater earthquake somewhere in the state about once every 30 years. The last major earthquake was a magnitude 7.2 in 1954 at Fairview Peak 45 miles (70 kilometers) southeast of Fallon. The largest historical earthquake in Nevada was in 1915 in Pleasant Valley 35 miles (60 kilometers) south of Winnemucca; this event has been estimated to have been a magnitude 7.3 to 7.8. In the geological past, large earthquakes have occurred near what are now urban areas. For example, within the last 3,000 years two earthquakes have occurred along the Genoa fault, which extends from south of Woodfords, California, through the town of Genoa, into Carson City. The most recent event was between 500 and 600 years ago, and the penultimate event was between 2,000 and 2,200 years ago. In each event, the fault moved as much as 18 feet (5.5 meters), and the earthquakes were probably magnitude 7.2 to 7.5.

Risk from earthquakes is a function of the geological hazard and the structures and number of people at risk. Among the states, California ranks first (with an estimated average annual loss of $3.3 billion) because it not only has a high hazard from many active faults, such as the San Andreas fault, but it also has many buildings and a large population at risk. The FEMA loss estimation computer model uses available national data sets and makes assumptions about certain parameters in the data, such as soil and rock responses to earthquakes and types of buildings at risk. At a high level of confidence, the FEMA numbers can be viewed as order-of-magnitude estimates. (That is, they could be off by a factor of ten, although calibrations with the Northridge 1994 and Loma Prieta 1989 earthquakes would suggest that the calculations may be accurate within a factor of two.) It is therefore unwise to make too much of small differences in the rankings from state to state or from city to city. Nonetheless, FEMA’s calculations place Nevada fifth among the states in terms of estimated average annual loss from earthquakes.

Among the metropolitan areas that FEMA evaluated in its loss estimation model, Las Vegas and Reno ranked high on the list (at $28 million and approximately $18 million per year, respectively) in terms of estimated loss. Although the hazard is higher in Reno than in Las Vegas (that is, the geological and seismic evidence suggest to us that there is a higher probability of having a major earthquake in the Reno-Sparks-Carson City area than in the Las Vegas area), the calculated loss is higher in Las Vegas because the greater size of Las Vegas more than offsets the greater number of earthquakes in the vicinity of Reno.

The FEMA report illustrates earthquake risk in two ways — average annual dollar loss and the ratio of dollar loss to the value of the building stock. The latter way of portraying the numbers closely reflects the earthquake hazards as presented in the probabilistic seismic hazard maps of the U.S. Geological Survey, which provide some of the input for FEMA's computer model. In comparing estimated earthquake losses for metropolitan areas, Reno ranks higher than Las Vegas in terms of the ratio of dollar loss to the value of the building stock. That is, there is a higher percentage of the building stock at risk from earthquakes in Reno than in Las Vegas. By any measure, the Reno-Sparks-Carson City and Las Vegas metropolitan areas have high earthquake risks.

There is a good chance that you will experience a major earthquake if you live in western Nevada. There are approximately 30 faults that are likely to cause damage in the Reno-Carson City urban corridor. The probability of at least one magnitude 6 or greater event in the next fifty years is between 34 and 98%. The probability of at least one magnitude 7 or greater event in the next fifty years is between 4 and 50%. Hazards include intense ground shaking, ruptures of the ground, liquefaction, landslides, and ancillary problems, such as fires and hazardous waste spills.

An urban earthquake today could cause billions of dollars in damage. FEMA's analysis uses a nationally consistent approach that looks at long-term probabilities rather than worst-case or likely scenarios. Although the numbers are reasonable estimates of average annual loss, the actual loss may be considerably higher in any given event.

We don’t know everything that needs to be known. More research can accomplish two things: it can reduce the uncertainties in the estimates of the seismic risk, and it can lead to a more precise description of the hazard for planning, earthquake-resistant construction, and hazard mitigation. For instance, only 15% of the state is adequately mapped geologically, and much more needs to be learned about the geological history of faults. Seismometers adequately cover only small portions of the state, and only a few strong-motion instruments have been installed. The effects of soils on ground shaking have not been adequately determined anyplace in Nevada. There are not enough continuous global positioning system (GPS) sites recording strain in Nevada.

We can’t predict when earthquakes will occur, but we can be better prepared. Earthquake risks can be reduced through non-structural mitigation; structural hardening or rehabilitation of buildings and infrastructure; better understanding and mapping of the hazards; zoning restrictions where appropriate; strengthening and enforcing building codes; insurance and related incentives for mitigation; engineering, seismological, and geological research; and education of the public.

For more information about earthquakes in Nevada and how to be prepared, please contact John G. Anderson, Director, Nevada Seismological Laboratory (775-784-4265), Jonathan G. Price, Director and State Geologist, Nevada Bureau of Mines and Geology (775-784-6691 extension 126), or Frank Siracusa, Chief, Nevada Division of Emergency Management (775-687-4240) and see the Web sites of the Nevada Earthquake Safety Council (www.nbmg.unr.edu/nesc/index.html), the Nevada Bureau of Mines and Geology (www.nbmg.unr.edu), the Nevada Seismological Laboratory (www.seismo.unr.edu), and the Nevada Division of Emergency Management (www.state.nv.us/dem/index.htm).