June 28, 1997

Prepared by:

Douglas B. Bausch and David S. Brumbaugh and

Arizona Earthquake Information Center

Northern Arizona University

P.O. Box 4099

Flagstaff, Arizona 86011


The following report was prepared by the Arizona Earthquake Information Center (AEIC) as a continuation of studies for the Arizona Division of Emergency Management's Earthquake Program. The accompanying maps were prepared and presented to the Prescott Community during 1993 with funding provided by the Federal Emergency Management Agency (FEMA) under cooperative agreement number AZ102EPSA. These seismic hazard evaluations began in 1988, and included several products at the state level, as well as more detailed mapping provided for seven key communities within Arizona. The summary below describes the products prepared by the AEIC under these programs.

Two products were prepared at the state level at a scale of 1:1,000,000. The first product was a State of Arizona Maximum Intensity Ground Shaking Map (1887-1987) (Morrison and others, 1991) that contours the maximum levels of historical ground shaking experienced within the state. The second product prepared for the state was the Arizona 100-Year Accelerations contour map (Bausch and others, 1993). The acceleration mapping required a detailed analysis of seismic sources affecting Arizona, and their probability for recurrence. These interpretations were input into the U.S. Geological Survey computer program SEISRISK III (Bender and Perkins, 1987). The 100-year contour map illustrates the force of gravity expressed as a percentage of 1.0 g that has a 90-percent chance of not being exceeded during the next 100 years. The output consisted of a grid of about 4,000 data points that were contoured for Arizona and outlying regions. The preparation of these two map products, conclusions and recommendations are explained in the State Report (Bausch and Brumbaugh, 1994). More detailed seismic hazard mapping was provided for seven key communities in Arizona. The communities were selected based on their proximity to faulting, historical seismicity, and population. The communities of Grand Canyon Village, Flagstaff, Winslow, Prescott, Phoenix, Tucson and Yuma were selected for these studies. The community-based seismic hazard evaluation provides mapping of key geologic units that are expected to exhibit different intensities of ground shaking, as well as neotectonic faults, selected critical facilities, and the 50, 100 and 250 year acceleration data points and values from the state-wide project. The mapping is prepared utilizing 7.5-minute U.S. Geological Survey base maps that cover the urbanized area of each community. These maps are accompanied by this report for Yavapai County that analyzes the vulnerabilities of the County and provides conclusions and recommendations.


Maximum Intensity Ground Shaking Map (1887-1987)

Arizona 100-Year Probabilistic Acceleration Contour Map


Seismic Hazard Mapping of Local Geology and Accelerations for Seven Key Arizona Communities

1) Grand Canyon Village 2) Prescott 3) Flagstaff

4) Phoenix 5) Winslow 6) Tucson

7) Yuma

Communities were selected using a combination of three criteria: 1) historic seismicity; 2) proximity of faulting; and, 3) population. 

These products are prepared for the application of earthquake hazard planning and preparedness, and should not be considered for site-specific construction design. Copies of this report can be obtained through the Arizona Division of Emergency Management, 5636 East McDowell Road, Phoenix, Arizona 85008, Attention: Mr. Al Franco, Earthquake Program Manager (602) 392-7510. Comments concerning these publications are welcome.



The Prescott community of Yavapai County was selected as a key community based upon its relatively large and rapidly expanding population, proximity to seismic sources, and damaging historical earthquakes. This report addresses the earthquake risk to all of Yavapai County and includes the more detailed maps prepared during previous studies for the Prescott community. The risk of ground shaking in the Yavapai County area is considered moderate. However, the overall seismic risk to Yavapai County is increased by the presence of neotectonic faulting, the growing population, and unreinforced masonry buildings. Because of these factors, this report contains a detailed analysis of the seismic hazard to the Yavapai County region.

Yavapai County is subject to ground shaking from earthquakes originating on neotectonic faults within the County, as well as from other nearby earthquake sources, such as the Hurricane or Toroweap faults and the Northern Arizona Seismic Belt (NASB). Historically, earthquakes originating in this belt have resulted in ground shaking to the Yavapai County region in 1906 (M 6.2), 1910 (M 6.0), and 1912 (M 6.2). The ML 5.1 Chino Valley earthquake of February 1976 resulted in minor damage to several Yavapai County communities. Other historical accounts describe earthquake shaking in the Yavapai County area (DuBois and others, 1982). 

Portions of Yavapai County are underlain by a northwest trending system of faults, including the Aubrey, Big Chino, Verde and Horseshoe faults. These faults bisect the County from the northwest to the southeast. Paleoseismological studies by Euge and others (1992) indicate movement within the past 100,000 years and the potential to produce a magnitude 7.25 earthquake. A large ground rupturing earthquake on either of these faults is considered a worst-case scenario for the Yavapai County community.

Because of these risks, Arizona is designated by the Federal Emergency Management Agency National Earthquake Hazards Reduction Program as a "High Risk" state for earthquakes.





1.1 What to do Before, During and After an Earthquake  


2.1 Federal Legislation

2.2 Earthquake Insurance

2.2.1 Federal Earthquake Insurance Proposals


    1. Ground Shaking
      1. Predicting Ground Motion
    2. Ground Failure

3.2.1 Slope Stability

3.2.2 Liquefaction

      1. Ground Rupture

3.3 Neotectonic Faulting of Yavapai County

      1. Aubrey Fault
      2. Aubrey West Fault
      3. Big Chino Fault
      4. North, Central and Southern Verde Faults
      5. Horseshoe Fault


4.1 Peak Ground Acceleration Mapping for Yavapai County

4.2 Effects of Local Geology

4.2.1 Alluvium and Young Alluvium

4.2.2 Sedimentary and Volcanic Rock

4.2.3 Basement Rock: Igneous and Metamorphic Rock

4.2.4 Shallow Ground Water


    1. Yavapai County Historical Earthquakes
      1. The ML 5.1 February, 1976 Chino Valley Earthquake

5.2 Shaking Effects in Yavapai County from the 1906, 1910 and 1912 Northern Arizona Earthquakes


6.1 Grand Wash Fault System

6.2 North, Central, and Southern Hurricane Faults

6.3 North and South Toroweap Fault


7.1 Impact of the Design Earthquake on Yavapai County


8.1 Ground Shaking Parameters

8.2 Hazardous Buildings and Structures

8.3 Critical Facilities

8.4 Lifelines



10.1 Special Development Regulations

10.2 Hazard Reduction

10.3 Recovery and Reconstruction

10.4 Recommended Goals and Policies

10.4.1 Retrofit and Strengthening of Existing Facilities

10.4.2 Strengthening and Enforcing Seismic Codes

10.4.3 Public Education







































Earthquakes occur when stresses within the earth's crust are relieved by slippage along rupture surfaces known as faults. The rupture process generates waves that radiate from the fault source, affecting people and structures on the surface of the earth. Although the process is conceptually simple, the factors controlling the precise nature of an earthquake are not completely understood. Further geological and seismological research is needed to assess where and when earthquakes will occur in the future, as well as how large they are likely to be, and to anticipate the probable effects on various types of man made structures.

The focus, or hypocenter, of the earthquake is the point within the earth's crust where the initial rupture of the rocks occurs and where the elastic waves from the earthquake are first released. The majority of earthquakes recorded in the United States have had shallow focal depths; 15 km or less; and have occurred in regions containing faults outcropping at the surface. In other regions, however, earthquakes occur at deeper locations within the earth's crust, so that a surface rupture is not often observable in the field. The later process is perhaps the most common within the study area. Relatively deep earthquakes often exceeding 15 km are recorded for Arizona (AEIC Catalog of Earthquakes). In addition, no surface rupturing events associated with earthquakes within Arizona have been observed during historic time.

The epicenter of an earthquake is the projection of the focus up onto the earth's surface. In the absence of instrumental data, epicenters have often been established on the basis of felt reports and the damage that is observed. However, epicenters are now typically located by the relative arrival times of seismic wave components received at various instruments operating within a seismograph network. In Arizona, there are twelve seismic stations located throughout the state at Tucson, Yuma, Phoenix, Flagstaff, Williams, Jerome, Sunset Crater, Blue Ridge Reservoir, Pipe Springs National Monument and the south and north rims of the Grand Canyon (Figure 1). However, in view of Arizona's earthquake risk and size, this is considered sparse seismic station coverage. Historically, seismic station coverage for Arizona is considered very sparse. The earthquake risk to the state can best be determined by adequate seismic station coverage that collects and processes accurate earthquake data. The accuracy of epicenter and hypocenter locations depends upon: (1) The number of reliable recording stations; (2) geologic interpretations of crustal structures; and (3) knowledge of local earthquake wave propagation velocities in various areas.


Figure 1 - Locations of seismograph stations within Arizona. Stations maintained by the AEIC are shown by open triangles, and include remote analog stations PSNM and SCN operated by the National Park Service, and JRAR operated by the State Parks service. Other operators in Arizona include: Arizona State University (ASU), Caltech (YMA), and University of Arizona (TUC).

Earthquakes are normally classified as to severity according to their magnitude (usually using the Richter scale), or their seismic intensity. Richter magnitude is a logarithmic measure of the maximum motions of the seismic waves as recorded by a seismograph. Because this size classification is based on a logarithmic scale, a magnitude 8 earthquake is not twice as big as a magnitude 4 earthquake, but rather, 10,000 (i.e., 104 or 10x10x10x10) times larger. More recently, seismologists have shown that magnitude is also proportional to the energy released during an earthquake, but at a level 32 times greater between earthquake magnitudes (e.g., a magnitude 6 earthquake releases 32 times the energy as a magnitude 5 earthquake).

 The magnitude of an earthquake is intended to be a measurement of its size, independent of the place of observation. It is calculated from measurements on seismographs. Physically, the magnitude can be correlated with the energy released by an earthquake, as well as with the fault rupture length and the maximum fault displacement. At present, at least four different magnitudes are in common use for classifying earthquakes: (1) local magnitude (ML), the classic Richter magnitude based on peak response of a calibrated instrument; (2) body-wave magnitude (mb), based on the response amplitude of the primary (P-wave) body-wave; (3) surface wave magnitude (MS), based on the response amplitudes of long-period surface waves; and (4) the moment magnitude (MW), which is the most complete measure of earthquake size. Moment magnitude (MW) is directly based on the amount of energy released during an earthquake and can be measured by a geologist in the field examining the fault geometry, as well as by a seismologist studying the digital waveforms. Each of these magnitudes are used in this report, and are derived from a well-calibrated instrument, knowledge of the characteristics of the rock through which the seismic waves must travel and the local conditions at the seismograph station.

In the absence of instrumental recordings of ground motion, seismologists have described the ground movement by assigning intensity numbers according to subjective intensity scales. Following an earthquake, the assignment of an intensity to a given location is based on interviews with inhabitants of the area and on observations of damage in the area. Assigned intensity values from different locations are then combined to formulate a map containing a series of isoseismals, contours that separate regions of successive intensity rating. The shape and extent of the isoseismals are influenced by the tectonic features of the area, indicating predominant directions along which seismic waves are transmitted and the manner in which the earthquake originates (NUREG, 1975). In addition, several other factors influence the felt intensity of an earthquake, including: population density, local geology, shallow ground water, and building type.

The destructiveness of an earthquake at a particular location is commonly reported using the Modified Mercalli Scale of seismic intensity. Seismic intensities are subjective classifications based on reports of ground shaking and damage caused by past earthquakes. There are several seismic intensity scales; the one used most often is the Modified Mercalli Intensity (MMI) scale. The MMI scale was modified in the 1930's to address construction practices and affects on new inventions such as automobiles, and the scale is undergoing modification during the writing of this report to address modern construction practices, such as steel frame buildings. This scale has 12 levels of intensity; the higher the number, the greater the ground shaking intensity and/or damage.

Earthquakes have only one magnitude, but they have variable intensities that generally decrease with increasing distance away from the source. However, other factors such as local geology, shallow ground water and building type affect the intensities of earthquakes at a site. For example, greater intensities are associated with poorly consolidated alluvial soils, high ground water levels, poor construction practices and unreinforced masonry structures. Certain soils greatly amplify the shaking in an earthquake. Seismic waves travel at different speeds in different types of rock, and when seismic waves pass from rock to soil they generally slow down and get bigger. The looser and thicker the soil is, the greater the amplification will be. For example, ground motion that damaged regions underlain by poorly consolidated sediment in the Loma Prieta earthquake were 10 times greater than neighboring regions. In addition, earthquakes such as Northridge 1994 and Kobe 1995 have demonstrated the influence of fault rupture directivity on intensity distribution. When the earthquake rupture moves along the fault, it focuses energy in the direction it is moving so that a site in that direction will receive more shaking than a site the same distance away but in the opposite direction.

1.1 What to do Before During and After an Earthquake

The following is a list of tasks that individuals at the home or office should undertake to lessen the overall impact of a major earthquake.

Before an Earthquake:

Remove or correct interior nonstructural hazards, such as top-heavy bookcases and storage cabinets, water heaters and other appliances. Anchor furniture and water heaters against the wall and provide gas-fired appliances with flexible connections.

Set aside a supply of emergency food and water, and obtain first aid materials, a gas shut-off wrench, fire extinguisher, and battery-powered radio. Identify neighbors with first aid training and check for an emergency supply of medication for all members of the family, especially children, disabled, and elderly.

Practice taking cover. This exercise will make people aware of the safest places during an earthquake, such as under a desk, table, bed or strong doorway. The maximum duration of shaking from an earthquake impacting Yavapai County is expected to be roughly 20 to 30 seconds.

Practice exiting. Walk the possible escape routes from your house or office and plan to avoid light fixtures, masonry chimneys, unsupported walls and other overhead hazards. Power for elevators and escalators may fail in high-occupancy facilities, so be aware of alternate exits. Do not panic or run; crowded exits should be evacuated in an orderly manner to avoid additional injuries in a rush for the door, emergency loud speaker systems may give instructions.

Practice turning off electricity and water and know how to turn off gas at the main. Replace rigid inlet gas connection lines to water heaters with flexible line. For safety reasons, do not practice gas shut-off; only the utility company should turn the gas back on. Be sure anyone in the household can locate main switches and valves.

Review the responsibilities of each family member after an earthquake. Plans for picking up children from schools, day-care centers, or other facilities with dependents should be regularly checked and reviewed. Have the phone number available of the person outside of the area for management of family messages.

Contact the County and neighbors about forming a neighborhood co-op self-help group.

During an Earthquake:

Stay calm.

If you are indoors remain indoors.

If in your place of residence, crouch under a desk or table, or brace yourself in a doorway (Be aware that it is possible for doors to swing shut during a quake). Try to protect your head with a coat, cushions, etc. Stay away from windows or brick masonry (fireplaces), china cabinets, hanging cabinets, or anything else that might possibly fall on you.

If you are in a high-rise building stay away from outside walls and windows. Do not use the elevator.

If outdoors remain outdoors. Try to move away from buildings, powerlines, trees, or anything that might fall on you.

If you are in a car try to move away from overpasses. Stop slowly and remain in your car. If possible try not to park where building material may fall on your car.

After an Earthquake:

Check for injuries in your family and neighborhood.


Extinguish small fires and check for additional fire hazards, such as cracked walls, roof lines and attics, and other physical signs of structural damage that can cause a malfunction in the electrical wiring.

Check for the smell of leaking gas, and if detected, shut off gas at the gas meter. Unanchored gas heaters or gas-fired hot water heaters may experience damage to valves and service connections, especially those without flexible line connections.

Shut off electrical power if there is damage to the wiring or there is a gas leak. The main switch is usually located in or next to the main fuse or circuit breaker box.

Clean up flammable liquids, medicines, and other harmful substances.

Check for structural and nonstructural damage, such as cracked chimneys, fallen power lines, and other objects that may become unstable and fall during an aftershock.

Try not to use water, it may result in a drop in water pressure for firefighting purposes (fire flow). Toilets should not be flushed until both incoming water lines and outgoing sewer lines have been checked to see if they are open.

Try not to use the phone unless it is a genuine emergency. Emergency and damage report alerts, and other information may be obtained by turning on your radio.

Report serious injuries and significant damage to a nearby city or county emergency reception center.



2.1 Federal Legislation

 At the federal level, there are two important pieces of legislation relating to local seismic hazard assessment. These are Public Law 93-288, amended in 1988 as the Stafford Act that establishes basic rules for federal disaster assistance and relief, and the Earthquake Hazards Reduction Act of 1977, amended in 1990, which establishes the National Earthquake Hazards Reduction Program (NEHRP).

The Stafford Act briefly mentions "construction and land use" as possible mitigation measures to be used after a disaster to forestall repetition of damage and destruction in subsequent events. However, the final rules promulgated by the Federal Emergency Management Agency (FEMA) to implement the Stafford Act (44 CFR Part 206, Subparts M and N) require post-disaster state-local hazard mitigation plans to be prepared as a prerequisite for local governments to receive disaster assistance funds to repair and restore damaged or destroyed public facilities. Under the regulations implementing Section 409 of the Stafford Act, a city or county must adopt a hazard mitigation plan acceptable to FEMA if it is to receive facilities restoration assistance authorized under Section 406.

The overall purpose of the National Earthquake Hazards Reduction Act is to reduce risks to life and property from earthquakes. This is to be carried out through activities such as: hazard identification and vulnerability studies; development and dissemination of seismic design and construction standards; development of an earthquake prediction capability; preparation of national, state and local plans for mitigation, preparedness and response; conduct basic and applied research into causes and implications of earthquake hazards; and, education of the public about earthquakes. While this bears less directly on earthquake hazards for a particular local government, much of the growing body of earthquake-related scientific and engineering knowledge has been developed through NEHRP funded research, including this study.

2.2 Earthquake Insurance

 After every major earthquake, the problem of financing recovery and reconstruction reemerges. As urban settlement has expanded worldwide, disasters have been experienced with increasing frequency and publicity. So it would seem reasonable for private and public sector organizations to plan in advance to provide more adequately for such contingencies.

Yet disaster relief and recovery resources are not consistently adequate or timely. Federal and state disaster assistance covers only a portion of the loss encountered in major earthquakes. Sometimes it is not received until long after the critical needs for such assistance are experienced. Consequently, many households, businesses and industries are significantly disrupted and many smaller enterprises go out of business after a major disaster.

The optimum solution would be to build cities strong enough and located so as to withstand the worst damage likely to be caused by natural disasters. Gradually, as older cities are renewed with each cycle of rebuilding and reinvestment and as new cities are built with better codes and land use practices, this goal will come closer to being achieved. But the experience of hazard mitigation to date, together with the evolving state of scientific knowledge and incomplete coverage of federal and state disaster assistance, suggest that additional sources of financial support for post-disaster recovery and reconstruction are needed. One possible source is earthquake insurance.

Historically, insurance coverage for earthquake damage has been either unavailable or prohibitively expensive. This was because there were previously no established actuarial methods for accurately estimating earthquake losses in advance of such disasters. Because scientific methods of earthquake prediction and loss estimation were in their infancy, insurance companies have found it difficult to reasonably estimate what probable maximum loss they might incur by insuring against for a catastrophic disaster. Therefore, they could not be sure whether or not they could remain in business after such an event if payment of claims plus operating costs were to exceed premiums collected.

Improvements in loss estimation have proceeded sufficiently in the past decade, however, so that more insurance companies have begun to provide earthquake insurance, although still costly, with high deductibles. Some insurance companies are today better able to distinguish more clearly the level of risk by geographic area. Computerized methods, including Geographic Information Systems (GIS), have made it possible for some of the more sophisticated companies to begin to model and forecast potential losses, based upon information gathered and maintained about localized areas and the structures now being insured.

Although this growing trend in sophistication has helped insurance companies improve their available coverage to some degree, less than a third of all property owners in earthquake prone areas are estimated to be participating in earthquake insurance. Moreover, sophisticated new industry technology cannot overcome the problem of a nationwide impact likely to be created by a catastrophic event. Destruction anticipated in various catastrophic earthquake scenarios is so large and difficult to estimate that a national program to cover seismically induced losses has been slow to evolve and faces serious difficulties in enactment. Direct losses of such an event have been estimated varyingly in the tens of billions of dollars and indirect losses at several magnitudes more.

2.2.1 Federal Earthquake Insurance Proposals

In recent years, the insurance industry has approached the federal government to enact legislation which would require 100% mandatory coverage in all homeowner and commercial risk insurance, backed up by a federally sponsored reinsurance pool to which loans could be made by the federal government to offset losses incurred in a catastrophic event (Earthquake Project, 1989). Such loans would be paid back through future premium receipts.

Passage of such legislation has been stalled by disagreement over the issue of whether or not federally required earthquake insurance should be accompanied by hazard mitigation in high risk areas to reduce the potential magnitude of losses over time. The argument for hazard mitigation as part of a national earthquake insurance program is predicated on principles similar to those underlying the federal flood insurance program which has been in place for nearly two decades. That program has identified high risk areas by issuance of Flood Insurance Rate Maps which cities and counties are obligated to observe through requiring hazard mitigation measures in their local planning and zoning practices in high flood hazard areas.

The argument for mandatory mitigation goes, in short, why penalize those who are not in high risk areas by requiring them to absorb costs of losses which might otherwise be avoided through proper hazard mitigation? The counter argument is that mandatory mitigation would increase local development costs in many communities where earthquake losses may not be experienced. Although involved interests are far from agreement on the role and level of required mitigation, there is a reasonable expectation that a compromise will be worked out as the probabilities of a catastrophic earthquake disaster increase each year.



The seismic hazard evaluation for Yavapai County is determined by the analysis of the following factors:

1.) Preparation of community ground shaking maps utilizing five U.S. Geological Survey 7.5-minute quadrangles as a base (Plates 1-5, In Pocket). The community maps are prepared to illustrate: the differences in ground shaking intensity based on geologic type; 50, 100 and 250 year acceleration data, and reference facilities and landmarks.

2.) Geographical factors such as the pattern, type, and movement of a nearby potentially active fault or fault system, and the distance of the fault to the area under investigation. An evaluation of the previous work performed in the area provides information for this segment of the earthquake hazard evaluation.

3.) The spatial and temporal distribution of historic earthquake epicenters. Historic records are utilized for this portion of the analysis.

4.) Evaluation of isoseismal maps (based on the Modified Mercalli Intensity scale that uses felt reports to map the extent and magnitude of earthshaking), to provide information on the expected intensity, type of ground motion, and the distribution of future earthquakes.

5.) Geologic criteria such as slope stability, ground rupture, liquefaction and other seismically induced geologic hazards. The geometry of the underlying fracture system, the profile of the overlying surficial deposits and the basement/soil interaction are important in the evaluation of seismic risk. Seismic amplification and dampening are controlled by the soil/basement profile and topographic effects, while mass movement may result from the applied seismic force (Seed and others, 1969).

6.) Vulnerability of critical facilities and lifelines based on structural type and seismic building code conformance. For this information the Uniform Building Code (U.B.C.) should be consulted. The U.B.C. for a given area varies with seismic zonation and by the importance of the structure to the community. Yavapai County lies within U.B.C. Zone 2b, as shown on the current U.B.C. zonation mapping (Figure 2).

Given the generalized nature of this study, the seismic history of the area is significant. The frequency, ground acceleration, magnitude and intensity of past earthquakes are essential data (Haley and Hunt, 1974). The maps prepared for this study are recommended for planning purposes only; site specific investigation, especially for critical facilities are warranted.


Figure 2: Seismic zone map of the United States from the 1994 Uniform Building Code (UBC). Designations range from 0 to 4, with 4 representing the greatest ground shaking potential. Yavapai County lies within zone 2b of the national mapping. States and local communities are allowed to exceed the UBC requirements based on the local knowledge of their engineering communities.

3.1 Ground Shaking

Several faults have the potential of generating earthquakes that will cause strong ground motions in Arizona, including Yavapai County. The Prescott region is located within an area of moderate risk of earthquake ground shaking, however, northern portions of Yavapai County are underlain by high-risk areas associated with major neotectonic faults (Figure 3). Each of these potential earthquakes will affect Yavapai County differently, depending on the distance between the earthquake-generating fault and Yavapai County, the size and rupture mechanism of the earthquake, and the local geologic conditions. Some faults are also more likely to cause an earthquake than others. The Yavapai County area is affected by earthquakes occurring on seismogenic sources outside the County, such as within the Northern Arizona Seismic Belt that defines the tectonic boundary of the Colorado Plateau (Brumbaugh, 1987). In addition, large earthquakes on the "Boundary Faults", such as the Hurricane and Toroweap with relatively rapid rates of displacement could result in significant ground shaking in Yavapai County.

Seismic waves propagating through the earth's crust are responsible for the ground vibrations normally felt during an earthquake. Seismic waves vibrate up and down and side to side at different frequencies, depending on the frequency content of the earthquake rupture mechanism, the distance from the earthquake source to a particular site, and the path and material through which the waves are propagating. As seismic waves travel through the earth's crust, their energy is lost due to the inelastic behavior of the ground motion, and due to scattering, diffraction and deflection of the waves as they cross materials of different physical properties. The overall effect, known as attenuation, alters the form and frequency content of the seismic waves with distance away from the earthquake's source.

Near-field earthquakes, which occur within approximately 10 miles of the site of reference, generate rough, jerky, high-frequency seismic waves that are generally more efficient in causing short buildings, such as single-family residential structures, to vibrate. Longer period wave forms, characteristic of far-field earthquakes, are felt at greater distances from the earthquake source. These longer-period waves, manifested as a slow rolling motion, are more likely to cause high-rise buildings and buildings with large floor areas to vibrate vigorously. An earthquake on the southern Hurricane or Toroweap faults within the Arizona Strip would be an example of a far-field earthquake affecting Yavapai County, as opposed to a relatively near-field events on the Big Chino or Verde faults.


Figure 3: General ground shaking risk map for Arizona, based on 100 year probabilistic acceleration mapping of Bausch and Brumbaugh (1994). Overall earthquake risk not only includes the risk of ground shaking, but is increased by factors such as population density, building-type and age, and local geologic conditions that are not illustrated by this general map.









Approximate duration in millions of years


Millions of years ago








Approximately the last 10,000 years













































































































Faults in Arizona have formed over millions of years as a response to various tectonic stress regimes. Some of these faults are generally considered inactive under the present geologic conditions, that is, they are unlikely to generate future earthquakes. Other faults are known to be accumulating strain as a result of current shifting of the earth's plates. Such faults have either generated earthquakes in historical times, or show geologic and geomorphic characteristics that suggest they might move in the relatively recent future, within a time span of concern to the residents of the area, or for long-term consideration in building design.

In general, the probability of an earthquake occurring on a given fault decreases with age of the latest proven faulting. That is, geologically young faults (Quaternary) are more likely to move than pre-Quaternary faults. However, it is at times difficult to determine with a certain degree of confidence, which faults are capable of moving in the future, and which ones are not likely to move under the present stress regime. Geologic evidence suggests that some faults may remain dormant for hundreds to thousands of years between major displacements. The geologic time scale (Table 1) is often used as a yardstick of latest proven faulting to evaluate the risk a fault may pose to development. For Arizona, existing studies (Scarborough and others, 1983; Menges and Pearthree, 1983; Pearthree and others, 1983; and Scarborough and others, 1986) define neotectonic faults as those that exhibit signs of surface displacement within the last about 4 million years (Late Pliocene-Quaternary).

Faults with infrequent recurrence, however, should be considered in the design phase and seismic analyses of many types of projects, such as nuclear facilities, dams and emergency operation centers. When their risk cannot be established, these faults may also be treated in the same manner as active faults, including designating building setbacks if necessary. The activity classification of faults may also change as geologic field studies along the trace of the fault are conducted, or if an earthquake occurs on a fault previously considered inactive. Some historical earthquakes have occurred along previously unknown faults.

Maximum Probable Earthquake: A maximum probable earthquake is the largest earthquake a fault is predicted capable of generating within a specified time period of concern, say 30 or 100 years. Maximum probable earthquakes are most likely to occur within the time span of most developments, and therefore, are commonly used in assessing seismic risk.  



Intensity Value and Description


Average peak velocity (centimeters per second)


Average peak acceleration (g is gravity = 9.80 meters per second squared)


I. Not felt except by a very few under especially favorable circumstances (I Rossi-Forel scale)




II. Felt only by a few persons at rest, especially on upper floors of high-rise buildings. Delicately suspended objects may swing. (I to II Rossi-Forel scale)




III. Felt quite noticeably indoors, especially on upper floors of buildings, but many people do not recognize it as an earthquake. Standing automobiles may rock slightly. Vibration like passing of truck. Duration estimated. (III Rossi-Forel scale)




IV. During the day felt indoors by many, outdoors by few. At night some awakened. Dishes, windows, doors disturbed; walls make creaking sound. Sensation like a heavy truck striking building. Standing automobiles rocked noticeably. (IV to V Rossi-Forel scale)






V. Felt by nearly everyone, many awakened. Some dishes, windows, and so on broken; cracked plaster in a few places; unstable objects overturned. Disturbances of trees, poles, and other tall objects sometimes noticed. Pendulum clocks may stop. (V to VI Rossi-Forel scale)






VI. Felt by all, many frightened and run outdoors. Some heavy furniture moved, a few instances of fallen plaster and damaged chimneys. Damage slight. (VI to VII Rossi-Forel scale)






VII. Everybody runs outdoors. Damage negligible in buildings of good design and construction; slight to moderate in well-built ordinary structures; considerable in poorly built or badly designed structures; some chimneys broken. Noticed by persons driving cars. (VIII Rossi-Forel scale)






VIII. Damage slight in specially designed structures; considerable in ordinary substantial buildings with partial collapse; great in poorly built structures. Panel walls thrown out of frame structures. Fall of chimneys, factory stacks, columns, monuments, and walls. Heavy furniture overturned. Sand and mud ejected in small amounts. Changes in well water. Persons driving cars disturbed. (VIII+ to IX Rossi-Forel scale)






IX. Damage considerable in specially designed structures; well-designed frame structures thrown out of plumb; great in substantial buildings with partial collapse. Buildings shifted off foundations. Ground cracked conspicuously. Underground pipes broken. (IX+ Rossi-Forel scale)






X. Some well-built wooden structures destroyed; most masonry and frame structures destroyed with foundations; ground badly cracked. Rails bent. Landslides considerable from river banks and steep slopes. Shifted sand and mud. Water splashed, slopped over banks (X Rossi-Forel scale)


More than 60


More than 0.60g


XI. Few, if any, (masonry) structures remain standing. Bridges destroyed. Broad fissures in ground. Underground pipelines completely out of service. Earth slumps and land slips in soft ground. Rails bent greatly.




XII. Damage total. Waves seen on ground surface. Lines of sight and level distorted. Objects thrown into air.



Primary Source: Bolt (1993)

Maximum Credible Earthquake: The maximum credible earthquake, i.e. the largest earthquake a fault is believed capable of generating, is nevertheless, often considered in a number of planning and engineering decisions. For example, maximum credible earthquakes are considered in the design of critical facilities such as dams, nuclear power plants, and emergency operation centers. They are also used in urban and emergency planning to identify and mitigate the risk of worst-case scenarios.

Earthquakes are normally classified as to severity according to their magnitude (usually using the Richter scale), or their seismic intensity. Richter magnitude is a logarithmic measure of the maximum motions of the seismic waves as recorded by a seismograph. Because this size classification is based on a logarithmic scale, a magnitude 8 earthquake is not twice as big as a magnitude 4 earthquake, but rather, 10,000 (i.e. 104 or 10x10x10x10) times larger. The destructiveness of an earthquake at a particular location is commonly reported using a seismic intensity scale. Seismic intensities are subjective classifications based on reports of ground shaking and damage caused by past earthquakes. There are several seismic intensity scales; the one used most often is the Modified Mercalli Intensity (MMI) scale (Table 2). This scale has 12 levels of intensity; the higher the number, the greater the ground shaking intensity and/or damage. Earthquakes have only one magnitude, but they have variable intensities that generally decrease with increasing distance away from the source. However, other factors such as local geology, shallow ground water and building type affect the intensities of earthquakes at a site.

3.1.1 Predicting Ground Motion

Ground motion caused by earthquakes is generally characterized using the parameters of ground displacement, velocity, and acceleration (Figure 4). Engineers traditionally work with ground acceleration, rather than with velocity or displacement, since acceleration is directly related to the dynamic forces that earthquakes induce on structures. The most often used measure of the strength of ground motion is peak ground acceleration. Peak ground accelerations are generally calculated using empirical attenuation equations that describe the behavior of the ground motions as a function of the magnitude of the earthquake, and the distance between the site and the seismic source (the causative fault). The increasingly larger pool of seismic data recorded in the world, and particularly in the western United States, has allowed researchers to develop reliable empirical attenuation equations that are used to model the ground motions generated during an earthquake.

Although computer models are now routinely used to predict the ground motions expected at a given site as a result of an earthquake, it is still difficult to anticipate the damage sustained by different types of structures during an earthquake. This is so because the response of structures to ground shaking depends on many parameters, including the amplitude and frequency content of the seismic waves, and the duration of shaking. The frequency content of the ground motion, in turn, depends on the rupture mechanism of the earthquake, the properties of the materials that attenuate the seismic energy, and the regional and local site conditions that may amplify, focus, or defocus the seismic waves arriving at the site of interest. In addition, different structures, because of differences in their natural frequencies and modes of vibration, respond differently to a given ground motion. For planning of critical facilities, therefore, it is often best to study the effects of the worst-case scenario, using as standard the maximum credible earthquake of the fault nearest to the site, such as an M 7+ event on the Big Chino or Verde faults. Structures are then designed accordingly, assuming that earthquakes of lesser magnitude and intensity will effect the study area to a lessor degree.


Figure 4: Probabilistic acceleration contour mapping for Arizona. Peak horizontal ground accelerations at bedrock expressed as a percentage of gravity and based on a 90% probability of not being exceeded in a 100 year time-frame (Bausch and Brumbaugh, 1994).

An acceleration contour map prepared for the State of Arizona (Figure 4) by Bausch and others (1993) (see Executive Summary), provides the acceleration data presented on the Prescott community maps (Plates 1-5, in-pocket). The data for the state acceleration map comprise a grid of approximately 4,000 specific data points for Arizona and the outlying regions. The data that lie within the areal extent of the Prescott community maps and the data point number (Bausch and Brumbaugh, 1994) are plotted on Plates 1 through 5. Values are expressed as the force of gravity, and represent the anticipated horizontal accelerations at bedrock in the area. Values for 50, 100 and 250 years are presented based on the generally accepted probability of 90-percent non-exceedance (or 10-percent chance of occurring) within the stated time frame.

The ground shaking maps presented in this report, if used together with inventories of potentially hazardous buildings, can help identify areas most likely to be damaged during an earthquake. The maps can also be used to identify areas where response capability operations, such as heavy rescue operations, will be vital in the case of an earthquake. The ground shaking maps should be used only for general planning purposes, and should not be used for specific building design requirements. Boundaries between the geologic units are compiled from small scale regional geologic publications and should be checked in the field if the site in question falls close to a geologic contact. In addition, the difference in ground shaking characteristics across these contacts should be taken as gradual, rather than an abrupt change in ground shaking intensity. Site-specific studies are required to adequately characterize the seismic parameters used in the design of a structure.

Modified Mercalli Intensity levels for Yavapai County may be calculated from the acceleration values presented from a combination of Richter's (1958) and Evernden and Thompson's (1988) empirical relationships for predicting intensity in terms of peak and root-mean-square (RMS) accelerations. Evernden and Thompson (1988) made a comprehensive study on the correlation between intensity and different seismic parameters. It can be assumed that RMS acceleration amplitudes for alluvial sites are about 70% of the corresponding peak ground acceleration values. The ratios of the RMS to peak amplitude for alluvial sites usually are higher than the ratios of the RMS to peak amplitude for stiff soil or rock sites. This is also consistent with Evernden and Thompson's (1988) recommendation of negative corrections for intensity scales for sites that are not underlain by alluvial materials. Richter (1958) reported an acceleration-intensity relationship based on the Modified Mercalli Intensity scale. Evernden and Thompson (1988) stated that Modified Mercalli Intensities are not a linear scale in terms of the level of ground shaking. They argued that the Rossi-Forel scale provides a better linear relationship. The relation between the two scales are qualitative.

Predicted intensity may have advantages over probabilistic bedrock accelerations for sites underlain by thick sequences of alluvium, such as the Yavapai County community. Alluvium is known to amplify the effects of ground shaking in both intensity and duration of shaking. Other parameters that increase reported earthquake intensities such as shallow ground water, building type and soil properties may be included in the conversion of the predicted bedrock acceleration values to predicted intensity.

3.2 Ground Failure

 In areas where ground failure might occur, seismic intensity values may increase by one or two levels. Areas of suspected ground failure in the event of a large magnitude near-field earthquake, including liquefaction and slope instability, are discussed further in this report. Geologic effects caused by earthquakes are divided into two principal categories: primary effects and secondary effects. Primary effects are those caused by deep-seated forces in the earth and include fault rupture, tectonic uplift and subsidence. Secondary geologic effects are those caused by ground shaking and include liquefaction, compaction of sediment and various forms of mass movement (Youd, 1986).

3.2.1 Slope Stability

Earthquake induced landslides and rockfalls often result in a considerable portion of the damage associated with historical earthquakes. Falls of precarious rocks may be triggered by small earthquakes in steep terrain. Slopes in their natural condition are generally far less susceptible to instability than those that are altered by activities of man. Mining activities within the County, such as at Bagdad, can greatly enhance slope instability. Therefore, design of man-made slopes should include parameters for seismically induced ground shaking based on acceleration mapping.

The rapid population growth in Yavapai County will result in more development within vulnerable hillside regions. The current risk to the Yavapai County community as a result of earthquake-induced slope instability is expected to be low.

3.2.2 Liquefaction

Liquefaction occurs primarily in saturated, loose, fine to medium-grained soils in areas where the ground water table is 50 feet or less below the ground surface. When these sediments are shaken, such as during an earthquake, a sudden increase in pore water pressure causes the soils to lose strength and behave as a liquid. The resulting features are called sand boils, sand blows or "sand volcanoes". Liquefaction-related effects include loss of bearing strength, ground oscillations, lateral spreading, and flow failures or slumping (Yerkes, 1985). Ground failure caused by liquefaction is a major cause of earthquake damage. For example, most of the extensive damage caused by the 1964 Alaska and the 1989 Loma Prieta earthquake was a consequence of liquefaction.

Only very localized regions of Yavapai County meet the criteria for liquefaction to occur. Limited areas near the Verde River Valley, Big Chino Wash and smaller stream valleys are underlain by relatively unconsolidated soil and shallow ground water.

      1. Ground Rupture

Within the greater Yavapai County area there are several mapped neotectonic faults (Scarborough and others, 1983; Menges and Pearthree, 1983; Pearthree and others, 1983; and, Scarborough and others, 1986). An analysis of fault rupture hazard for a particular fault requires that the subject fault be located exactly, and that its potential for fault rupture be known, if only approximately. The historical record, is too short a time period to characterize the earthquake recurrence of most faults. Geologists use repeatedly offset stratigraphic or physiographic features along a fault to reconstruct the earthquake history of the fault. Approximate recurrence intervals for major earthquakes on that particular fault can often be obtained from the analysis of field data. These recurrence intervals are then used in engineering design and planning decisions. Unfortunately, these kinds of data are available for only a very few faults affecting Arizona. However, data have been collected towards the assessment of seismogenic characteristics of Yavapai County faults.

The analysis of fault rupture potential also assumes that a fault will slip along the same or nearly the same surface on which the fault last slipped. This assumption is generally true, based on observations from past surface-rupture events that show most ground ruptures do follow closely pre-existing fault traces. However, during an earthquake some sections of a fault surface may rupture, while others may not. In conducting a fault-rupture hazard analysis the worst-case scenario is assumed, that is, that during a moderate to major earthquake the subject fault surface will rupture in the area of study. An earthquake producing surface rupture along the Big Chino or Verde faults in Yavapai County could be associated with an earthquake of magnitude 7+. The nearby Aubrey and Horseshoe faults could also produce damaging ground accelerations should they rupture in a M 7+ event. Surface rupture occurs when part of the stress released during an earthquake ruptures the fault plane at the earth's surface. In general terms, if the displacement is more than a few inches, structures that straddle the fault trace will be damaged. It is very costly to design structures to withstand large vertical or horizontal displacements.

Reconnaissance mapping of neotectonic faults in Yavapai County has been completed by previous researchers (Figure 5). The major neotectonic faults affecting Yavapai County include the Aubrey, Big Chino, Verde and Horseshoe. A generally northwesterly trending system of faults bisecting the County from the northwest to the southeast. Additional data concerning the state-of-activity of the Aubrey, Big Chino and Verde faults were compiled by Euge and others (1992). Of these neotectonic faults, Euge and others (1992) summarized that the Big Chino was likely to have ruptured last and they collected paleoseismic data from three trench sites along the Big Chino fault. The Horseshoe fault was extensively studied by Piety and Anderson (1991) prior to retrofit of the Stewart Mountain Dam. These mapped neotectonic faults within or near Yavapai County are summarized in Table 3, below:





Age of Youngest Event


Fault Length (miles)


Fault Orientation


Maximum Credible Earthquake






15,000 years








Located in southeasternmost Yavapai County. Was extensively studied by the U.S. Bureau of Reclamation (Piety and Anderson, 1991) prior to retrofit of the Stewart Mountain Dam.




50,000 years

 18 + each splay





 Consists of a northern, central and southern segments. Characterized by high-angle normal faulting. Responsible for formation of the Verde Valley.


Big Chino










May be youngest fault in Yavapai County. Trenching indicates 80 feet of vertical displacement in last 200,000 years. Forms narrow asymmetric graben. May also be responsible for the ML 5.1 Chino Valley earthquake. Paleoseismic studies indicate 2-3 events in last 100,000 years.




30,000 years








Accommodates transition in Arizona from northward trending Toroweap and Hurricane faults to northwesterly trending Big Chino and Verde faults. Located in northwesternmost Yavapai County.



Figure 5: Reconnaissance neotectonic fault mapping in the Yavapai County region (from Euge and others, 1992).

3.3 Neotectonic Faults of Yavapai County

Existing data for known neotectonic faults within Yavapai County may be summarized as follows:

3.3.1 Aubrey Fault: The Aubrey fault is a north-south trending normal fault that lies within the Arizona Mountains source zone. The Aubrey fault may be divided into three segments based on changes in its trend (Menges and Pearthree, 1983). Subsurface data (via trenching) has been obtained across the southernmost segment, and the results indicate recurrent displacement on the Aubrey fault with the latest rupture about 30,000 years ago (Euge and others, 1992). The trenching data suggests slip rates ranging from 0.01 mm/yr to 0.03 mm/yr. No events in historic time have been specifically linked to the Aubrey fault. Nevertheless, the presence of degraded scarps, and the analysis of trenching data indicates that this fault belongs with the most active in Arizona. The maximum earthquake estimated for this fault with complete fault rupture is M 7.25 (Euge and others, 1992).

3.3.2 Aubrey West Fault: The Aubrey West is a rather subtle feature, but is observed on aerial photographs. Its surface expression is poor, and is included here as a distinct source because of its structural/spatial relation with the Aubrey fault to the east. The latest displacement is estimated to be Quaternary by Euge and others (1992), and the total fault length is approximately 10 km, and is the segment shown just west of the main Aubrey fault on Figure 5. Based on fault length, the MCE is estimated to be M 6.4.

3.3.3 Big Chino Fault: The Big Chino fault strikes northwesterly along the northeastern boundary of the Big Chino Valley for a distance of 56 km. It is marked at the surface by an escarpment stretching along the base of the southwestern flank of Big Black Mesa (Figure 6). The fault controlled escarpment was trenched in three places (Euge and others, 1992) (Figure 7). One trench showed a fault surface with 26 feet of total displacement. Fault controlled depositional wedges of colluvium suggested this represented two or three events. These events occurred within the last 100,000 years and average slip rates are estimated to be 0.06-0.12 mm/yr. The Maximum Credible Event, from fault rupture/magnitude relations, would be M 7.25 (Euge and others, 1992).

Although there has been no historic surface rupture associated with the Big Chino fault, at least one moderate earthquake may have been associated with subsurface rupture and slip. The February, 1976 Chino Valley event of ML 5.1 may well have occurred on the Big Chino fault (Eberhart-Phillips and others, 1981).


Figure 6: Location map of the Big Chino fault in Yavapai County. Study area shown in the following Figure 7 is outlined (from Euge and others, 1992).


Figure 7: Location of trenches along Big Chino fault in the Sheep Camp area. The Big Chino fault is one of only a few faults in Arizona for which paleoseismic data has been collected (from Euge and others, 1992).

3.3.4 North, Central, and Southern Verde Faults: The Verde fault is a northwest striking, high-angle normal fault, largely responsible for the geomorphic expression of the Verde Valley that forms the hanging wall block of this fault. Although the Verde fault has had a complex movement history, much of its movement has been documented to have occurred along the Central Verde fault between 7-8 million years before present (Carr, 1986). The Verde Valley was characterized by internal drainage until about 2 million years ago when the Verde River began draining the valley to the southeast. Therefore, displacement of these dissected deposits indicate movement of the Verde fault within the last 2 million years.

The Northern and Central Verde faults may be considered as one segment about 30 km long and offset from the overlapping southern segment near Copper Canyon. The southern segment, also about 30 km long, appears to have been the most recently active. Fault scarps in Allen Canyon as much as 27 feet high suggest 3 to 4 movements on this part of the Verde fault with recurrence time possibly as short as 50,000 years and slip rates as high as 0.03 mm/yr (Euge and others, 1992). The MCE event on the Verde fault would have a magnitude of M 7.25.

3.3.5 Horseshoe Fault: The Horseshoe fault is a north to northwest trending fault that consists of two segments, both of which have been active during the last 300,000 years (Piety and Anderson, 1991).

The northernmost of the two segments is about 12 km long and bounds Horseshoe basin on its western side. Perhaps as many as two surface rupturing events have occurred on this segment in about the last 150,000 years. The most recent event may have occurred during the last 15,000-30,000 years. The southernmost segment of the Horseshoe fault cuts across the basin for a distance of about 10 km. Trenches cut across this segment indicate that at least two surface rupturing events have occurred, one about 15,000 years ago, and another about 100,000 years ago (Piety and Anderson, 1991). The MCE for the Horseshoe fault has been estimated at M 7.



Plates 1 through 5 were prepared to illustrate the probabilistic peak ground accelerations at bedrock, and ground shaking units based on geologic-type.

4.1 Peak Ground Acceleration Mapping for the Prescott Community

The Peak Ground Acceleration (PGA) mapping represents peak horizontal acceleration of the ground at bedrock. The approach of representing peak horizontal ground acceleration on bedrock is a common and widely used method of showing ground accelerations. Indeed it has been utilized in national reports (Algermissen and others, 1982; 1990) and on one other report in Arizona (Euge and others, 1992). In fact, such an approach is often the only feasible one because of a lack of adequate data on spectral accelerations of different rock types. Over the last decade more rock type acceleration numbers have become available through the installation of accelerometers in the western U.S. At present, however, Arizona is sadly lacking in such data. The National Hazard Maps and subsequently the Uniform Building Code, which is based upon the national maps prepared periodically by the U.S. Geological Survey, are a result of probabilistic acceleration mapping. The construction of probabilistic acceleration maps are a result of three types of basic input parameters:

1) Attenuation of ground shaking with distance from the earthquake source;

  1. Frequency of earthquakes within an area or region, termed recurrence; and

3) The character and extent of regions and faults that generate earthquakes.

Several probabilistic assessments have been performed for the study area (Table 4).

Previous Studies

Values Expressed for "Downtown" Prescott





100 YEAR


250 YEAR

 Arizona Earthquake Information Center (Bausch and Brumbaugh, 1994)







 Arizona Department of Transportation (Euge and others, 1992)







 Building Seismic Safety Council (Algermissen and others, 1990)




 1996 National Seismic Hazard Maps (Frankel and others, 1996)







 Accelerations are expressed as a percent of gravity at bedrock, based on 90% non-exceedance.

 The studies listed above utilized several different methods in determining the probabilistic accelerations for northern Arizona. Bausch and Brumbaugh (1994), Euge and others (1992), and Pearthree and others (1996) all represent regional reports, while Algermissen and others (1990) and Frankel and others (1996) represent the national mapping of the U.S. Geological Survey. The regional reports all utilized the computer program SEISRISK III (Bender and Perkins, 1987), and include line sources (faults), as well as historic seismicity.

A new method of determining probabilistic accelerations was utilized during preparation of the most recent national maps by Frankel and others (1996). Rather than defining seismogenic source zones, this method moves a one-square kilometers grid across the historic seismicity database, thereby providing a running average of the occurrence values for the region. The primary advantage of this method is in eliminating the uncertainties in defining source zone boundaries. For example, the location of the ML 5.1 Chino Valley earthquake north of Prescott locally increases the acceleration values for the region.

Values of horizontal accelerations exceeding 0.10 g, or 10-percent of the force of gravity, are generally accepted as being destructive to weakly constructed structures (Richter, 1958). Figure 8 was prepared to compare the acceleration values with other Arizona communities that are considered high (Yuma) to low (Phoenix). The anticipated accelerations for the San Francisco region, which has the highest earthquake risk in the U.S., are also illustrated for comparison.

4.2 Effects of Local Geology

 An analysis of ground shaking intensity reported during historic earthquakes affecting Arizona indicated differences in ground shaking based on the underlying geology (Morrison and others, 1991). For these studies regarding Arizona key communities, geologic earth units that occur throughout the state were categorized into three groups: 1) alluvium (map symbol-Q); 2) sedimentary and volcanic bedrock (map symbol-S/V); and, 3) granitic bedrock (map symbol-gr). The analysis of historic intensities indicated higher reported intensities for alluvial sediments compared to bedrock areas, as well as slightly higher intensities noted for sedimentary and volcanic rock as compared to areas underlain by granitic rock. For example, during the ML 6.2 1906 earthquake the Prescott newspaper, The Arizona Weekly Miner on 1/31/1906 reported that, "One peculiar freak of the temblor was that it affected only the eastern part of town, the shock being quite marked in that section east of Cortez street". The later comment may reflect the earthquake effects to structures underlain by young sediments and shallow ground water associated with Granite Creek. Plates 1 through 5 (In-Pocket) illustrate the areal extent of the basic geologic ground shaking types transferred from regional published geologic maps.


Figure 8: Acceleration probability expressed as a percentage of the force of gravity against time. The graph illustrates the relatively high values anticipated for the Yuma community in comparison with moderate (Prescott) and low (Phoenix) Arizona communities. San Francisco, the U.S. city with the highest earthquake risk, is shown for comparison. Ten-percent of the force of gravity is generally accepted as the onset of damage to weakly constructed structures (Richter, 1958).

      1. Alluvium (Q): Alluvial materials associated with Granite Creek, as well as older dissected alluvial fans are shown on the plates prepared for this study. The alluvium (Q) consists of moderately-well to poorly consolidated sand, silt and gravel. Alluvium is Holocene to middle Pleistocene in age (Reynolds, 1988).
      2. Sedimentary and Volcanic Rock (S/V): Volcanic rock consists of basalt and locally includes tuff and agglomerate. The basaltic rock is Tertiary in age.
      3. Granitic Basement Rock (gr): The granitic basement consists of Precambrian granite and schist. Local granitic intrusions are mapped southwest of Prescott that are late Tertiary to early Cretaceous.
      4. Shallow Ground Water: Shallow ground water can increase the expected seismic intensity values at a site. For most earthquake scenarios, seismic intensity values increase by one level on the Modified Mercalli Intensity scale (see Table 2) in those areas of Yavapai County where shallow ground water is present. The accuracy of these interpretations are dependent on the accuracy of the ground water data available for Yavapai County. An increase of one level in the expected seismic intensities for a given scenario earthquake may be applied to any area of the county where shallow ground water (less than 30 feet) may be reported in the future.



Yavapai County has experienced several strong earthquakes from seismogenic sources within northern Arizona.

The ML 5.1 Chino Valley earthquake of February 4, 1976 occurred a few miles north of Prescott. Observations associated with this earthquake include:

The most significant shaking effects were noted within Chino Valley. MMI VI effects were reported at Chino Valley, Cottonwood, Miller Valley and Paulden.

Arizona's only strong motion reading was obtained at the Prescott VA Hospital at 0.04 g.

Within Yavapai County, MMI V shaking effects were noted for the following largest northern Arizona earthquakes: a) 1906, January 25, MS 6.2; b) 1910, September 24, MS 6.0; and c) 1912, August 18, MS 6.2

Most of Yavapai County is located within a geomorphic province termed the Arizona Transition Zone, an area characterized by the transition between the Basin and Range of Arizona's deserts to the Colorado Plateau. The Transition Zone includes neotectonic faulting, as well as earthquakes.


Figure 9: Seismicity of Arizona 1830 to 1993 showing the Northern Arizona Seismic Belt (NASB) (Arizona Earthquake Information Center Archives).

5.1 Yavapai County Historical Earthquakes

The following provides a listing of the historical earthquakes located and felt in Yavapai County. The community that felt the strongest effects is listed.

March 11, 1870, MMI V, Prescott: "The first shock of earthquake ever felt by whites in this part. The wave, all agreed came from the southwest" (DuBois and others, 1982).

August 12, 1870, MMI IV, Prescott: "Late last week, the pioneer earthquake shook men and things in every settlement around Prescott" (DuBois and others, 1982).

(note: the 1906, 1910 and 1912 northern Arizona earthquakes were all strongly felt in Yavapai County, see below)

July 16, 1930, MMI V, Constellation: "Shocks were accompanied by heavy rumbling. Two shocks. Aftershock recorded two hours later" (DuBois and others, 1982).

July 28, 1931, MMI VI, Cottonwood: This event was recorded and located by at least six regional seismograph stations at 34.7N x 112.0W. "Several housewives stated that their chinaware in pantries rattled and in some instances broke" (DuBois and others, 1982).

February 8, 1932, MMI II, Perkinsville: "Slight" (DuBois and others, 1982).

November 27, 1933, MMI V, Hillside: "Accompanied by subterranean sounds. Trembling motion felt by 75% of the population. Buildings swayed, rattled and creaked" (DuBois and others, 1982).

May 1, 1967, MS 3.8, 34.46N x 112.86W: No felt reports were found to document this event. Recorded at five regional seismograph stations with a focal depth of 26 km reported (DuBois and others, 1982).

5.1.1 The 1976 Chino Valley Earthquake

The Chino Valley earthquake occurred on February 4, 1976, and was assigned a magnitude of ML 5.1, and location of 34.66N x 112.50W. Yavapai County's largest historic earthquake resulted in MMI VI shaking effects. Minor damage was reported in the epicentral region, however, the shaking effects in Prescott were moderate. Minor damage was reported at four Chino Valley area communities (Chino Valley, Cottonwood, Miller Valley and Paulden). Figure 10 illustrates the areal extent of its shaking effects in Arizona, and Figure 11 shows the epicenter in relation to Prescott and regional neotectonic faults. Forty-two seismograph stations recorded the event with an origin time of 00:04:58 GMT (5:04 pm local time) . In addition, one strong motion reading of 0.04 g was obtained from an accelerometer located at the Prescott VA Hospital (DuBois and others, 1982). This remains the only strong motion reading of an Arizona earthquake. The preferred focal plane indicates a northwesterly trending fault, and a focal depth of 10-15 km was determined. Eberhart-Phillips and others (1981) attribute the 1976 event to the Big Chino fault. Five magnitude 2.0-3.0 aftershocks were recorded during the remainder of February 1976. Additional micro-seismicity studies in the epicentral region made by the University of Arizona in 1978 and 1979 indicated a rate of one event every three days (DuBois and others, 1982).


Figure 10: Isoseismal map of the ML 5.1 Chino Valley earthquake of February 4, 1976. Slight damage occurred in the epicentral region (from DuBois and others, 1982).


Figure 11: Local seismicity and faulting in Yavapai County. The location of the February 1976 Chino Valley earthquake is illustrated. Fault trace locations are adapted from Euge and others (1992), and epicenter data are from the Arizona Earthquake Information Center archives.

5.1.2 Shaking Effects in Yavapai County from the 1906, 1910 and 1912 Northern Arizona Earthquakes

The strongest earthquakes to strike northern Arizona occurred generally north of Flagstaff, however, significant shaking effects were recorded for communities within the relatively sparsely populated Yavapai County (Figures 12, 13, and 14).

Effects in Yavapai County as a result of the MS 6.2, 1906 Earthquake

Prescott MMI V

Genuine earthquake shocks this city (Arizona Weekly Miner, Prescott).

One peculiar freak of the temblor was that it affected only the eastern part of town, the shock being quite marked in that section east of Cortez street (Arizona Weekly Miner, Prescott).

Many persons felt the shock yesterday. Allen Hill stated, "A little after 1:30 p.m. I was sitting in the chair in front of my desk talking to a friend when I felt a rocking sensation. I saw the pictures on the walls move, in fact quite noticeably. The building trembled, the windows rattled, and I believe that the shock lasted from 15 to 20 seconds. I thought when I felt the first trembling movement that a boiler had exploded in the neighborhood or that perhaps the chimney had fallen from the building." (Arizona Weekly Miner, Prescott)

H. Wm. Stevens, who was visiting in the offices of R.E. Morrison (located in the second story of the Electric Light building), said, "I have felt earthquakes in Los Angeles and San Francisco, but I have never experienced one of such duration as this one. The chandeliers in the Morrison offices moved. I rocked in the chair that I was seated in, the pictures moved on the walls, and the windows rattled perceptibly. I later called into the offices occupied by Mr. Job in the same building, and the stenographer there stated that she plainly felt the earthquake, but did not realize what was the cause of the shaking." (Arizona Weekly Miner, Prescott).

Ash Fork MMI III

  Slightly felt here (Williams News, 1/27/06).

Jerome MMI V

  On Thursday Jerome experienced an earthquake, two distinct shocks being plainly felt by many persons (Jerome Mining News).

The oscillation was from east to west. It occurred at 1:29 in the afternoon, and shook heavy buildings in a manner which caused some alarm (Jerome Mining News).


Effects in Yavapai County as a result of the MS 6.0, 1910 Earthquake

Cedar Glade MMI V

  An earthquake caused some alarm here tonight at 9:06 o'clock, but no damage was done. The depot building rolled and windows rattled, alarming several in the waiting room. The jar was distinctly felt for several seconds (Arizona Weekly Miner, Prescott, 9/28/10).

Skull Valley MMI VI

  At 9:05 o'clock tonight an earthquake shook the buildings here, alarming many of the occupants. The railroad depot rolled as if falling from its foundation. Women residents were particularly alarmed, many calling to their husbands. Many in the depot report that the building rumbled and rattled as if hit by a locomotive running at high speed, and several rushed outside to inquire into the cause of the strange disturbance (Arizona Weekly Miner, Prescott, 9/28/10).

Jerome MMI V

  Inhabitants of this place were alarmed tonight, at 9:06 o'clock by the rumblings of an earthquake. The shock was distinctly felt for several seconds. Many buildings were shaken badly (Arizona Weekly Miner, Prescott, 9/28/10).

At the Boyd Hotel the plaster of the walls and ceilings of the second and third floors was cracked in many places, some falling to the floor. No damage is reported at the United Verde smelters or other mines in this vicinity, although the shock was general throughout the Verde district (Arizona Weekly Miner, Prescott, 9/28/10).

Felt south to Jerome (Winslow Mail, 10/1/10).

A slight earthquake shock was felt here, at Jerome, Williams, Flagstaff, and Kingman at 9:06 o'clock tonight. No damage was reported (The Arizona Republican, 9/24/10).

Prescott MMI V

Data pertinent to mapping ground shaking differences throughout the city are presented within the felt reports below. Similar to felt report differences in the 1906 earthquake, these data can be useful in forecasting ground shaking hazards for the Prescott community.

A slight earthquake shock was felt here, at Jerome, Williams, Flagstaff, and Kingman at 9:06 o'clock tonight. No damage was reported (The Arizona Republican, 9/24/10).

At six minutes past 9 o'clock, last night an earthquake was felt in this city and many places throughout northern Arizona (Arizona Weekly Miner, Prescott, 9/28/10).

By far the most violent shock in this city was reported from the S.F.,P. & P. storehouse, near the railroad depot. The building is reported to have been badly shaken, jarring supplies from the walls. Peculiar as it may seem, no shock was felt in the depot, only a short distance away. This it is believed is due to the fact that the depot is of reinforced concrete construction, the walls being reputed strong enough to support the heaviest engine on the Santa Fe lines. Not even the slightest tremble was felt in the train dispatcher's office, in the second story (Arizona Weekly Miner, Prescott, 9/28/10).

Residents of Nob Hill report that they were attracted by a vibration in their homes, that dishes rattled in their pantries, and windows in particular, jarred perceptibly. No chimneys are reported to have fallen, although the plaster of walls and ceilings cracked in many places (Arizona Weekly Miner, Prescott, 9/28/10).

Residents of the Thumb Butte section, arriving late last night, report that they felt no tremblor in that district and that the shock did not disturb that Locality (Arizona Weekly Miner, Prescott, 9/28/10).

The 'quake was not felt in any of the large brick or stone buildings of the city. None of the inmates of the Hotel St. Michael, Prescott, Head or Schuerman hotels were cognizant of the temblor until informed later by the occupants of other buildings (Arizona Weekly Miner, Prescott, 9/28/10).

A slight flicker of the electric lights was noticeable all over the city the time of the 'quake, but this did not attract any unusual attention in places where the shock was not felt (Arizona Weekly Miner, Prescott, 9/28/10).

Walker MMI IV

Telephone advices received from Walker state that buildings of the town were perceptibly shaken by the earthquake, and that many sitting at their porches were alarmed at the rumbling sound of the tremblor. No damage is reported (Arizona Weekly Miner, Prescott, 9/28/10).

Poland Junction MMI IV

The 'quake is reported to have awakened several residents of Poland Junction who had retired for the night. The jar was felt several seconds (Arizona Weekly Miner, Prescott, 9/28/10).

Effects in Yavapai County as a result of the MS 6.2, 1912 Earthquake

 Ash Fork MMI VI

  W.E. Henning of Detroit, who arrived Monday afternoon from Ash Fork, stated that he was in the Escalante and two distinct shocks were felt. The glass windows rattled, and between the first and second vibrations but a few seconds intervened. Two shocks are commonly reported for single earthquakes, which indicates that the observors may be reporting the effects of the two primary earthquake waves, the P- and the S-wave. In the railroad offices employees had their attention directed to the occurrence, but the slight disturbance did not occasion any alarm (The Weekly Miner, Prescott, 8/28/12). (Note on `Escalante': A geographic names search did not indicate an `Escalante' near Ash Fork, however, the interpretation by DuBois and others (1980) that the Escalante referred to is in Utah is likely incorrect. In addition, the article states that Mr. Henning arrived from Ash Fork, not Utah. Travelling from Utah to Prescott in less than 24 hours was probably not practical in 1912.)

Seligman MMI V

  J.W. Sullivan, who arrived yesterday morning from Seligman, say's that when he arrived in that town on Sunday afternoon from his range, ten miles to the south, Mike McBride and other residents were somewhat excited over the seismic disturbance, and all were absorbed in discussing the temblor and the general shakeup that followed. The glass windows gave the first warning in being almost shaken from their casing, and the buildings also rocked to and fro. Every building in the town was affected but not damaged (The Weekly Miner, Prescott, 8/28/12).

Kirkland Valley MMI IV

  L.J. Hasefield, who returned yesterday to Kirkland valley, stated that at 2 o'clock on Sunday afternoon his merchandise store and hotel building at that place were shaken up like dice in a box, and the rumbling of the glass in the windows, with the buildings moving perceptibly, was conclusive evidence of the earthquake reaching that place (Arizona Weekly Miner, Prescott, 8/28/12). (Note: It appears that Mr. Hasefield was not actually in Kirkland Valley at the time of the earthquake on Sunday, the effects to his buildings may have been reported to him by another party.)


Figure 12: Isoseismal Map of M 6.2 1906 Northern Arizona Earthquake.


Figure 13 - Isoseismal Map of the M 6.0 1910 Northern Arizona Earthquake


Figure 14 - Isoseismal Map of M 6.2 1912 Northern Arizona Earthquake



As mentioned previously, most northern Arizona seismicity falls within the northwest trending Northern Arizona Seismic Belt (NASB) (see Figure 9). Most well-located earthquakes and well-defined focal solutions indicate that this region of Arizona is extending in a northeast-southwest direction along northwesterly trending normal faults. Subsequently, the northwesterly trending neotectonic fault systems, such as the Big Chino and Verde, appear to control much of the earthquake hazard to Yavapai County. However, earthquakes that originate from seismogenic sources outside the County, including the Hurricane and Toroweap faults, as well as the NASB, add to the earthquake risk of Yavapai County.

More distant earthquake sources, such as the Hurricane and Toroweap faults along the western margin of the Colorado Plateau, may produce localized damage in Yavapai County and because of their relatively rapid rates of displacement, they are considered the likely sources for the design Maximum Probable Earthquake (MPE). Other possible source(s) for the MPE include moderate (M 6.0) earthquakes occurring on local fault systems that may not rupture the ground surface. This MPE would be significantly larger that the 1976 earthquake and result in local damage. The source(s) for the Maximum Credible Earthquake (MCE) for Yavapai County is the Aubrey-Big Chino-Verde-Horseshoe fault system that includes several segments capable of producing M 7+ earthquakes. The later fault systems are described in more detail above within the Ground Failure section, while the more distant sources for the MPE are outlined below:

6.1 Grand Wash Fault System

The Grand Wash system is located in northwesternmost Arizona, southern Utah and southern Nevada. These "boundary faults" comprise northward trending high angle normal faults that provide the physiographic transition of the Colorado Plateau to the east and the Basin and Range to the west. The longest segment of the Grand Wash system is about 60 miles in length. Included within the Grand Wash system are the Virgin Mountains and Washington faults. The Grand Wash and Washington faults do not have documented Holocene movement, while the Virgin Mountains system in southern Nevada displaces Holocene strata. A Maximum Credible Earthquake magnitude of 7.5 may be assigned to the Grand Wash system.

6.2 North, Central and Southern Hurricane Faults

The Hurricane fault is considered part of the southerly extension of the Wasatch fault system of central Utah. Based on neotectonic evidence, and length, the Hurricane fault is one of the greater seismic hazards for Arizona. The northerly trending Hurricane fault is one of a series of normal faults in the zone forming a series of easterly tilted blocks between the Colorado Plateau and Basin and Range. The fault plane solution for the ML 5.9 St. George, Utah earthquake of September 2, 1992 projects to the surface near the Hurricane fault (Pearthree and Wallace, 1992). Hamblin and others (1981) document vertical displacement of a 290,000 year old basalt flow approximately 300 feet along the Hurricane fault near Hurricane, Utah. In Arizona, Pearthree and others (1983) and Scarborough and others (1986) document displacements of up to 2-3 feet within young alluvial fan deposits (10,000 to 20,000 years old). The late-Quaternary recurrence interval for the Hurricane is estimated between 5,000 and 20,000 years (Hamblin and others, 1981).

Euge and others (1992) assign a slip rate to the Hurricane fault of 0.3 to 0.5 mm/yr. A MCE of 7.75 is assigned the Hurricane fault, which is the largest MCE of any Arizona fault.

6.3 North and South Toroweap Fault

The Toroweap fault consists of north-northeast trending southern and northern primary segments. In actuality the fault is comprised of many segments, the entire fault zone is about 265 miles in length (Euge and others, 1992). As with other faults in this region, displacement and seismic activity decreases southward (Hamblin and others, 1981). The Toroweap fault is the easternmost of the Boundary Faults that form the transition between the Colorado Plateau and the Basin and Range. The Toroweap is a normal fault that forms eastward tilted fault blocks of Colorado Plateau Paleozoic and Mesozoic sedimentary rocks. Although the Toroweap fault is segmented in nature, we do not believe that each of the short segments are uniquely seismogenic. Therefore, the Maximum Credible Earthquake (MCE) is expected to rupture multiple segments. The MCE assigned the Toroweap for this study is 7.75. Euge and others (1992) assign the latest movement of the Toroweap as late to mid Holocene (5,000 years ago). The late-Quaternary recurrence interval for the Toroweap is estimated between 15,000 and 20,000 years (Hamblin and others, 1981). Comparison of the Toroweap with the Hurricane fault to the west indicates equivalent MCEs, however, the Hurricane fault has a smaller recurrence time and a higher slip rate (0.3-0.5 mm/yr vs. 0.06-0.4 mm/yr).



The design earthquakes for Yavapai County are the maximum credible and probable earthquakes presented in Table 5. Because most earthquakes are believed to originate as a result of fault breaks, the rock motion at any particular site will depend on: (1) the amount of energy released along the fault during the earthquake; and (2) the distance of the site from the zone of energy release. In general the amplitude of these motions decrease with increasing distance from the zone of energy release, although other factors, such as geologic structure and orientation, will also have some effects (Seed and others, 1969). The magnitude of an earthquake is a convenient indication of the amount of energy released when a fault breaks. Because the magnitude, M, of an earthquake and the amount of energy released, E, are related by:

log E = 11.4 + 1.5 M

an increase of 1 unit on the magnitude scale corresponds approximately to a 30-fold increase in the amount of energy released. Another factor that needs to be considered is the depth of focus or hypocenter. Depths of earthquakes in California and northern Mexico are usually 6 to 15 miles (10 to 25 km) and are considered to be shallow. For events this shallow the hypocentral and epicentral distances are not appreciably different when the distance between the site and causative fault exceeds about 40 miles (60 km) (Seed and others, 1969). In the case of the maximum probable design earthquake the causative fault would be at least 40 miles (60 km) from the study area. The length of the fault break must also be considered, as the site to fault distance and duration of shaking are dependent on the length and direction of rupturing. In the case of the maximum probable earthquake for Yavapai County the causative faults are short in length relative to the distance of the site from the fault. Therefore, the distance of the site from the fault can be expressed by the epicentral distance for the Yavapai County area.

In relating the design earthquake to engineering seismology the following parameters must be considered: (1) magnitude of the event; (2) maximum amplitude of the horizontal acceleration; (3) duration of strong motion; and (4) the predominant frequency or period of motion of the site (Haley and Hunt, 1974). These parameters are given in Table 5 for MMI of VIII-IX and X-XI events, the MPE and MCE, respectively. The design parameters are calculated from the acceleration curves and from the predominant period calculations of Seed and others (1969). These equations result in an increase in period of the earthquake with increasing distance from the epicenter.

7.1 Impact of the Design Earthquake to Yavapai County

The following compilation of potential earthquake hazards are based on the response of the materials underlying Yavapai County to the design earthquake parameters listed in Table 5.

  1. Within local regions of Yavapai County that are underlain by alluvial deposits and areas of shallow ground water, the potential exists for greater damage than normal to structures. The earthquake damage index to structures increases with increasing thickness of the alluvial layer and softness (compaction) of the subsoil (Kanai, 1983).
  2. The natural period of one story wooden structures ranges (i.e. most single-family residences) from 0.2 to 0.3 seconds (Kanai, 1983). The predominant period of the MPE has a value of 0.8 which inhibits the development of resonance (period of structure = period of earthquake) and the accompanying increase in vibration and destruction. Multi-story buildings will display higher periods of vibration making resonance more probable. It is important to note that the MCE (Table 5) displays a period value conducive to the development of resonance in single story wooden structures. However, the predominate period of individual buildings may vary depending on design and construction materials.
  3. Documented earthquake damage studies indicate brick buildings experience an overall increase in damage on soft ground with an increase in the number of stories. The thickness of the sediment package is a controlling factor in the expected damage to concrete structures.
  4. However, thickness can be offset by variations in the natural density of the material underlying a specific reinforced concrete structure. An increase in the softness of the ground causes a concomitant increase in the extent of damage to reinforced concrete buildings (Kanai, 1983). The demonstration of low values for penetration tests during geotechnical investigations are a suitable measure of ground softness.
  5. By using the empirical equation of the natural period for reinforced concrete buildings a range of periods for different heights of structures can be developed:

1) 1 story = .06 - .12 seconds

2) 2 story = .12 - .21 seconds

3) 3 story = .21 - .30 seconds

4) 4 story = .30 - .40 seconds

5) 5 story = .40 - .60 seconds

6) 8 story - 10 story = .60 - .90 seconds

The duration of motion and the period of .55 seconds resulting from earthquakes in Tokyo provided suitable conditions for the development of resonance in 5 story structures (Kanai, 1983). The MPE has a predominate period of 0.8 sec., thereby, resulting in resonance to 5-story structures. However, the MCE predominant period of 0.15-0.25 seconds may result in resonance in concrete structures of about 2 to 3-stories. This table should be used as a guide, a more precise value of predominant period should be determined by a structural engineer taking into account the specific building design and construction.





ML = 7.75




A. Predominant period of vibratory motion:

0.80 sec

B. Maximum horizontal accelerations:

.40 g (40%)

C. Duration of motion:

20 - 30 sec

Predicted characteristics of ground motion for Yavapai County, Arizona. Caused by a large magnitude (M 7.75) earthquake on any of the large faults that comprise the Boundary Fault System (Grand Wash, Hurricane, and Toroweap) of northwestern Arizona about 100 miles northwest of Yavapai County. No primary ground rupture in Yavapai County would be expected with the MPE.


ML = 7.25




A. Predominant period of vibratory motion:

.15-.25 sec

B. Maximum horizontal accelerations:

1.2 g (120%)

C. Duration of motion:

20 - 30 sec

Predicted characteristics of ground motion for Yavapai County, Arizona produced by a surface rupturing event along a segment of the Aubrey-Big Chino-Verde-Horseshoe systems (calculated from Seed and others, 1969).



This section assesses the earthquake vulnerability in Yavapai County to structures and facilities of the United States, and are based upon past earthquake performance in the U.S. Beyond the scope of this study are the effects of the MCE and MPE on particular structures within Yavapai County.

Although it is not possible to prevent earthquakes from occurring, their destructive effects can be minimized. Comprehensive hazard mitigation programs that include the identification and mapping of hazards, prudent planning and enforcement of building codes, and expedient retrofitting and rehabilitation of weak structures can reduce significantly the scope of an earthquake disaster.

Various geologic phenomena that can cause property damage and loss of life are triggered by earthquakes. These hazards include ground shaking, fault rupture, landslides, and foundation failures caused by liquefaction or subsidence. Earthquakes can also cause a variety of localized, but not less destructive hazards such as urban fires, dam failures, and toxic chemical releases.

This section identifies and discusses the opportunities available for seismic upgrading of existing development and capital facilities, including potentially hazardous buildings, utilities, transportation infrastructure, and other critical facilities. Many of the issues and opportunities available to the County apply to new development, redevelopment and infilling. Issues involving rehabilitation and strengthening of existing development are decidedly more complex given the economic and societal impacts inherent to these issues.

Prioritizing rehabilitation and strengthening projects requires that the County consider where its resources would be better spent to reduce earthquake hazards in the existing development, and how the proposed mitigation programs can be implemented so as not to cause undue hardship on the community. The hazard evaluation is intended to define the scope of the problem. Rehabilitation programs should target, on a priority basis, potentially hazardous buildings, critical facilities, and high-risk lifeline utilities. Rehabilitation issues can be best addressed by the County.

8.1 Ground Shaking Parameters

The level of seismicity within the Yavapai County area is moderate, however, this may be a result of historically poor seismograph station coverage for the region and does not take into account damaging earthquakes outside of the region, such as along the Hurricane or Toroweap faults along the western margin of the Colorado Plateau.

If the maximum credible event were to occur on the Big Chino or Verde faults, the effects to Yavapai County could be extensive. These effects include extensive failure of unreinforced masonry construction. Resonance could develop in reinforced concrete structures that are 2-3 stories in height. The duration of strong motion (20-30 seconds) and the maximum horizontal accelerations (1.2 g) will be great enough to cause damage to other structures.

8.2 Hazardous Buildings and Structures

Most of the loss of life and injuries that occur during an earthquake are related to the collapse of hazardous buildings and structures. FEMA (1985) defines a hazardous building as "any inadequately earthquake resistant building, located in a seismically active area, that presents a potential for life loss or serious injury when a damaging earthquake occurs." Building codes have generally been made more stringent following a damaging earthquake. However, a large percentage of structures in Yavapai County built prior to improved building codes, have not been upgraded to current building code standards, and may, therefore, be potentially hazardous during an earthquake.

Damage to buildings is commonly classified as either structural or non-structural. Structural damage means the building's structural support has been impaired. Structural support includes any vertical and lateral force resisting systems, such as the building frames, walls, and columns. Non-structural damage does not affect the integrity of the structural support system. Examples of non-structural damage include broken windows, collapsed or rotated chimneys, and fallen ceilings. During an earthquake, buildings get thrown from side to side, and up and down. Heavier buildings are subjected to higher forces than lightweight buildings, given the same acceleration. Damage occurs when structural members are overloaded, or differential movements between different parts of the structure strain the structural components.


Figure 15: Unreinforced Masonry Building (URM). Prepared by the Applied Technology Council for the Federal Emergency Management Agency (1988), Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook: Earthquake Hazards Reduction Series 41, FEMA 154.

Larger earthquakes and longer shaking durations tend to damage structures more. The level of damage resulting from a major earthquake can be predicted only in general terms, since no two buildings undergo the exact same motions during a seismic event. Past earthquakes have shown us, however, that some buildings are more likely to perform more poorly than others.

Unreinforced masonry buildings (URMs) (see Figure 15), based on observations from past earthquakes, are prone to failure due to inadequate anchorage of the masonry walls to the roof and floor diaphragms, due to the limited strength and ductility of the building materials, and sometimes due to poor construction workmanship. These buildings are generally old, some dating to even the 19th century. Unless the buildings have been appropriately reinforced and strengthened, an earthquake may cause irreparable damage, and even collapse of some of these URMs, with the resultant threat to human life and property. Deterioration of the mortar (often of lime and sand with little or no cement, having very little shear strength) and of the wood framing as a result of weather exposure may also contribute to the weakening and poor performance of these structures during an earthquake. Parapets and cornices that are not positively anchored to the roofs may fall out.

Wall diaphragms are generally made of wood. These diaphragms are therefore very flexible, allowing large out-of-plane deflections at the wall transverse. This large drift can cause the masonry walls to collapse. Some tall URM buildings have thin walls that may buckle out-of-plane under severe lateral loads. If the wall is a non-load bearing wall, it may fail; collapse of a load-bearing wall will result in partial or total collapse of the structure.

Although URMs have been assumed to be the most hazardous buildings in a community, the collapse of URM buildings during an earthquake, in general, does not pose as much a hazard to loss of life as the collapse of other types of structures. Several other more numerous building types are also known to perform poorly during moderate to strong earthquakes, although they have not been targeted for upgrading and strengthening. Of these, soft-story (Figure 16) buildings (those with a story, generally the first floor, lacking adequate strength or toughness, due to few shear walls, i.e., buildings where the first floor is the garage) are of particular concern. The Northridge Meadows apartment collapse that killed 16 people during the January 17, 1994 Northridge earthquake was a building that included a mixture of residential units and parking garage on the first story (EERC, 1994).


Figure 16: Soft story-timber pole construction. Prepared by the Applied Technology Council for the Federal Emergency Management Agency (1988), Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook: Earthquake Hazards Reduction Series 41, FEMA 154.

The walls with large door openings common to garages have almost no resistance to lateral forces. If a second or more stories sit on top of the garage, the building may sustain significant amounts of damage. A soft-story building can also be one that has large window openings for display purposes on the first floor. Some soft-story buildings have a timber pole construction. The poles are often subject to wood deterioration with exposure to the elements. This deterioration, if unchecked, may also contribute to unsatisfactory seismic performance.

Structural damage to wood frame structures often results from a lack of, or inadequate connection, between the superstructure and the foundation; these buildings may slide off their foundations, with consequent damage to plumbing and electrical connections. Unreinforced masonry chimneys may also collapse. These types of damage are generally not life threatening, although they may be costly to repair. Wood frame buildings with stud walls generally perform well in an earthquake, unless they have no foundation or have a weak foundation constructed of unreinforced masonry or poorly reinforced concrete. Damage to wood frame buildings is generally limited to cracking of the stucco, which in fact, dissipates much of the earthquake's induced energy. The collapse of wood frame structures, if it happens, generally does not generate heavy debris; but rather, the wood and plaster debris can be cut or broken into smaller pieces by hand-held equipment and removed by hand in order to reach victims (FEMA, 1988).

Partial or total collapse of buildings where the floors, walls and roofs fail as large intact units, such as large precast concrete panels, cause the greatest concern in terms of life loss and difficulties in victim rescue and extrication (FEMA, 1988). Thousands of people have died as a result of collapse of these kinds of buildings during earthquakes, such as in Mexico City (1985), Armenia (1988), Nicaragua (1972), El Salvador (1986), and Philippines (1990). Many of the parking structures that failed spectacularly in Northridge (1994) consisted of pre-cast components (EERC, 1994).

The heavy debris that results from collapse of these types of buildings requires heavy mechanical equipment to remove the rubble. Location and extrication of victims trapped under the rubble is generally a dangerous process that requires equipment to tunnel and lift heavy debris Extrication of trapped victims within the first 24 hours after the earthquake becomes critical for victim survivability. In most instances, however, the planning resources available to procure the necessary equipment for victim rescue are ill-defined to non-existent. However, the implementation of the establishment of Heavy Urban Search and Rescue teams as recommended by FEMA (1988) has improved victim extrication. Buildings that are more likely to generate heavy debris in case of failure need to be identified, so that appropriate mitigation and planning procedures are defined prior to an earthquake.

Precast concrete frame buildings vary in their performance during earthquakes, dependent in part on the strength and toughness of the details connecting the structural elements (Figure 17). If poorly designed connections between prefabricated elements fail, the damage and loss of life can be significant. Precast frames are often weakened due to a combination of the accumulated stresses that may result from shrinkage and creep, and due to stresses incurred during transportation. Vertical support may fail if the building was designed with an inadequate bearing area and/or with insufficient connections between floor elements and columns. Corrosion of the metal connectors between prefabricated elements may also occur weakening the structure. Multi-story concrete and reinforced masonry buildings with concrete floor slabs may collapse with the floor slabs falling, nearly intact, one on top of the other, becoming closely stacked ("pancake" style of failure). The floor slabs prevent access to, and extrication of victims. Given the large size and weight of these slabs, some weighing up to 250 tons, the slabs generally need to be cut into smaller segments prior to removal by heavy cranes. The cutting of these slabs is a time-consuming process that requires breaking the concrete and cutting the reinforcing steel.

Tilt-up buildings have concrete wall panels, often cast on the ground, or fabricated off-site and trucked in, that are tilted upward into their final position. The weak connections and anchors have been observed to pull out of the walls during an earthquake, causing the floors or roofs to collapse. A high rate of failure was observed for this type of construction in the 1971 Sylmar, California earthquake. Tilt up buildings may generate heavy debris if concrete wall panels fail.


Reinforced masonry buildings often perform well in moderate earthquakes if they are adequately reinforced and grouted, and if sufficient diaphragm anchorage exists. Poor construction workmanship, resulting in ungrouted and unreinforced walls may lead to failure of the building during a large earthquake. Insufficient reinforcement can also result in heavy damage to walls, while lack of positive connections of the floor and roof diaphragms can also result in structural damage.


Figure 17: Precast concrete frame construction. Prepared by the Applied Technology Council for the Federal Emergency Management Agency (1988), Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook: Earthquake Hazards Reduction Series 41, FEMA 154.

Reinforced concrete frame buildings with or without reinforced infill walls display low ductility. Typical problems with this kind of building during an earthquake include shear failure if there are large tie spacings in columns, or if columns have insufficient shear strength, column failure if inadequate rebar splices are placed at the same location, shear failure of columns with insufficient tie anchorage, hinge deformation due to lack of continuous beam reinforcement, column failure due to inadequate reinforcing of beam-column joints, and non-structural damage due to the relatively low stiffness of the frame. A common type of failure observed following the MW 6.7 January 17, 1994 Northridge earthquake was confined column collapse (EERC, 1994), where infilling between columns confined the length of the columns that were allowed to move laterally in the earthquake.

Multi-story steel buildings generally also have concrete floor slabs. However, these buildings are less likely to collapse than concrete structures. Common damage to these types of buildings is generally non-structural, including collapsed exterior curtain wall (cladding), and damage to interior partitions and equipment. Older, pre-1945 steel frame structures may have unreinforced masonry such as bricks, clay tiles and terra cotta tiles as cladding or infilling. Cladding in newer buildings may be glass, infill panels or pre-cast panels that may fail and generate a band of debris around the building exterior (with considerable threat to pedestrians in the streets below). Structural damage may occur if the structural members are subject to plastic deformation which can cause permanent displacements; if some walls fail while others remain intact, torsion or soft-story problems may result. Pounding with adjacent buildings can also occur.

Buildings are often, however, a combination of steel, concrete, reinforced masonry and wood, consisting of different structural systems on different floors or different sections of the building. Of these, those types considered to be potentially hazardous (and that have not been discussed above) include: concrete frame buildings without special reinforcing, precast concrete and precast-composite buildings, steel frame or concrete frame buildings with unreinforced masonry walls, reinforced concrete wall buildings with no special detailing or reinforcement, large capacity buildings with long-span roof structures (such as theaters and auditoriums), large unengineered wood-frame buildings, buildings with inadequately anchored exterior cladding and glazing, and buildings with poorly anchored parapets and appendages (FEMA, 1985). Additional types of potentially hazardous buildings may be identified in the future as a result of observations of damaged structures after an earthquake.

The configuration of the building, that is, its vertical and/or horizontal shape can also be an important factor in earthquake performance. Simple, regular and symmetric buildings in general perform better than non-symmetric buildings. Non-symmetric buildings tend to twist in addition to shake laterally, while wings on a building tend to act independently during an earthquake, resulting in differential movements, and cracking. The geometry of the lateral load resisting systems is another component of a building's configuration (for example buildings with one or two walls made mostly of glass, while the remaining walls are made of concrete or brick). Asymmetry in the placement of bracing systems that provide a building with earthquake resistance can result in twisting or differential motions, with resultant damage.

Site-related seismic hazards may include the potential for neighboring buildings to "pound" the structure, or for the neighboring buildings to collapse onto the adjacent building. Pounding occurs when there is little clearance between adjacent buildings, and the buildings "pound" against each other as they deflect during an earthquake. The effects of pounding can be specially damaging if the floors of the buildings impacting each other are at different elevations, so that, for example, the floor of one building hits a supporting column of the adjacent building. Damage to the column can result in partial or total collapse of the impacted building.

8.3 Critical Facilities

Critical facilities are considered parts of a community's infrastructure that must remain operational after an earthquake, or facilities that pose unacceptable risks to public safety if severely damaged. Essential facilities are needed during an emergency, such as hospitals, fire and police stations, emergency operation centers and communication centers. High-risk facilities, if severely damaged, may result in a disaster far beyond the facilities themselves. Examples include nuclear power plants, dams and flood control structures, freeway interchanges and bridges, industrial plants that use or store explosives, toxic materials or petroleum products. High-occupancy facilities have the potential of resulting in a large number of casualties or crowd control problems. This category includes high-rise buildings, large assembly facilities, and large multifamily residential complexes. Dependent care facilities house populations with special evacuation considerations, such as preschools and schools, rehabilitation centers, prisons, group care homes, and nursing and convalescent homes. Economic facilities are those facilities that should remain operational to avoid severe economic impacts, such as banks, archiving and vital record keeping facilities, airports and ports, and large industrial and commercial centers.

It is essential that critical facilities designed for human occupancy have no structural weaknesses that can lead to collapse. The Federal Emergency Management Agency (FEMA, 1988) has suggested the following seismic performance goals for health care facilities:

The damage to the facilities should be limited to what might be reasonably expected after a destructive earthquake and should be repairable and not life-threatening.

Patients, visitors, and medical, nursing, technical and support staff within and immediately outside the facility should be protected during an earthquake.

Emergency utility systems in the facility should remain operational after an earthquake.

Occupants should be able to evacuate the facility safely after an earthquake.

Rescue and emergency workers should be able to enter the facility immediately after an earthquake and should encounter only minimum interference and danger.

The facility should be available for its planned disaster response role after an earthquake.

8.4 Lifelines

Certain critical facilities designed to remain functional during and immediately after an earthquake may be able to provide the public with only limited services if the lifelines that they depend on are disrupted. Our understanding of the seismic hazards to new and existing lifeline facilities relies in great part on the several workshops and publications dedicated to the subject, including research completed as a result of the 1989 Loma Prieta earthquake. The issue of seismic hazard mitigation for lifelines is very complex, given the diversity of lifeline facilities. The general comments on the effect of strong ground motion to buildings apply to structures from where a given lifeline service is provided, such as the control tower in an airport, or the buildings housing the computers and telephone circuits central to the communication lifelines of a district. Properly designed, manufactured and laid out buried pipelines, in general, are not damaged by strong ground motions, but can be severely disrupted in areas of surface rupture, liquefaction, or landsliding. Some freeway interchanges and bridges have been damaged by strong ground motions in past earthquakes; certain bridge designs have been prioritized in retrofitting programs because of their poor past performance in regions of seismic activity.

A hazard analysis should focus on four lifeline categories: (1) water and sewer facilities, (2) transportation facilities, (3) electric power facilities, and (4) gas and liquid fuel lines. Retrofit and upgrading programs for lifelines generally require careful planning to ensure that the public is not inconvenienced by irregular or discontinued service. The extensive systems of cable and pipe used to distribute electrical energy, gas, telephone communications, and water, or to collect sewer and storm drain water also require that potential problem spots in the system be identified and prioritized for an effective mitigation program to be implemented.



Federal Programs:

The Stafford Act requires post-disaster state-local hazard mitigation plans to be prepared as a prerequisite for local governments to receive disaster assistance funds to repair and restore damaged or destroyed public facilities. This report may meet the requirement for a hazard mitigation plan for Yavapai County. 

Ground Failure:

The major neotectonic faults of Yavapai County include the Horseshoe, Verde, Big Chino and Aubrey. These fault systems are all capable of generating M 7+ earthquakes should a major surface rupturing earthquake occur. These faults are summarized below:

  Aubrey Fault: Accommodates transition in Arizona from northward trending Toroweap and Hurricane faults to northwesterly trending Big Chino and Verde faults. Located in northwesternmost Yavapai County.

Big Chino Fault: May be youngest fault in Yavapai County. Trenching indicates 80 feet of vertical displacement in last 200,000 years. Forms narrow asymmetric graben. May also be responsible for the ML 5.1 Chino Valley earthquake. Paleoseismic studies indicate 2-3 events in last 100,000 years.

Verde Fault: Consists of northern, central and southern segments. Characterized by high-angle normal faulting. The Verde fault last broke the ground surface about 50,000 years ago. Responsible for formation of the Verde Valley.

Horseshoe Fault: Located in southeasternmost Yavapai County. Last ruptured the surface about 15,000 years ago. The Horseshoe fault was extensively studied by the U.S. Bureau of Reclamation (Piety and Anderson, 1991) prior to retrofit of the Stewart Mountain Dam.

Data collection opportunities for Yavapai County area neotectonic faults, such as determining the recurrence interval and recency of ground rupturing earthquakes through paleoseismological studies should be pursued. Threat determination and necessary mitigation should be performed prior to locating critical facilities on or near all neotectonic faults.

In the mountainous terrains of portions of Yavapai County, falls of precarious rocks may be triggered by small earthquakes. However, slopes in their natural condition are generally far less susceptible to instability than those that are altered by activities of man. Mining activities within the County, such as at Bagdad, can greatly enhance slope instability. Therefore, design of man-made slopes should include parameters for seismically induced ground shaking based on acceleration mapping. The rapid population growth in Yavapai County will result in more development within vulnerable hillside regions. The current risk to the Yavapai County community as a result of earthquake-induced slope instability is expected to be low.

Only very localized regions of Yavapai County meet the criteria for liquefaction to occur. Limited areas near the Verde River Valley, Big Chino Wash and smaller stream valleys that are underlain by relatively unconsolidated soil and shallow ground water may be susceptible to liquefaction induced ground failure in an earthquake.

Historic Seismicity:

Yavapai County has experienced several strong earthquakes from seismogenic sources within northern Arizona.

The ML 5.1 Chino Valley earthquake of February 4, 1976 occurred a few miles north of Prescott. The most significant shaking effects were noted within Chino Valley. MMI VI effects were reported at Chino Valley, Cottonwood, Miller Valley and Paulden. Arizona's only strong motion reading was obtained from this earthquake at the Prescott VA Hospital at 0.04 g.

Most of Yavapai County is located within a geomorphic province termed the Arizona Transition Zone, an area characterized by the transition between the Basin and Range of Arizona's deserts to the Colorado Plateau. The Transition Zone includes neotectonic faulting, as well as earthquakes.

The Yavapai County area is affected by earthquakes occurring in the Northern Arizona Seismic Belt. Historically, earthquakes originating in this belt have resulted in MMI VI shaking within Yavapai County in 1906 (M 6.2), 1910 (M 6.0) and 1912 (M 6.2).

Earthquake Hazard and Risk:

  Yavapai County is a rapidly growing community, and while the County has enforced Uniform Building Code (UBC) Seismic Zone 2b construction practices for the last seven years, a number of older unreinforced masonry buildings (URMs) are located within downtown Prescott and other communities. In addition, many remodels have produced a "soft-story" on the ground floor, and the frost-wedging cycle, common to environments where the temperature fluctuates about the freezing point, has undoubtedly contributed to a weakening of the URMs during their 100+ year lifetimes.

The seismic hazard for Yavapai County is considered moderate. However, the risk of earthquake ground shaking in Yavapai County is significantly higher than more populous regions of the state such as Phoenix or Tucson.

Yavapai County was selected for seismic hazard analysis based upon its close proximity to seismic sources, historical damaging earthquakes, large and growing population, and unreinforced masonry buildings.

Community planning maps for the Prescott area are provided as part of this project that outline soil versus bedrock locations, and the force that may be expected by an earthquake, should one occur, based on 50, 100 and 250 year time frames.



This report should be reviewed by members of ACES, and comments should be incorporated into updated seismic hazard assessments for Yavapai County. Although many earthquake related hazards exist for the community, effective mitigation requires prioritization of mitigation opportunities. Therefore, the primary mitigation opportunity would be to investigate the hazards associated with the aging unreinforced masonry buildings of downtown Prescott, while a secondary opportunity would be to further determine the state-of-activity of nearby neotectonic faults through paleoseismological and seismological studies. Probabilistic bedrock accelerations included in this report provide a useful guide for assessing ground shaking potential.

10.1 Special Development Regulations

Special development regulations reinforce and augment existing code standards by raising the level of hazard conscious project design and hazard mitigation engineering. Special development regulations include additional geologic/geotechnical investigation and construction standards. Foundation investigations are required in the Yavapai County Building Code, however, it may be in Yavapai County's best interest to emphasize the level of investigation and protection. Some standards may apply only to critical facilities, such as detailed seismic analyses in high risk areas. Special construction standards may include additional reinforcement of foundations in areas of potential ground failure. Avoidance of the hazard may be appropriate in some cases where engineering methods cannot mitigate the hazards, such as is the case where ground rupture along active or potentially active fault traces are identified during project investigation. Special minimum setbacks away from active faults can be defined for structures, lifelines, or critical facilities planned on or traversing the project site. In the case of critical facilities, setbacks from potentially active faults should be mandatory.

10.2 Hazard Reduction

Hazard reduction programs are designed to improve the safety of existing development. For example, some older structures, having been built to now-outmoded code standards, could benefit from seismic upgrading. Owners of older residences or commercial structures may voluntarily upgrade, or if a commercial facility is undergoing significant reconstruction, newer safety standards can be incorporated. Examples of hazard reduction programs include:

Strengthening of pipelines and development of emergency back-up capability by public utilities serving Yavapai County.

Regular fire safety inspections and fire flow tests to identify areas with cracked or damaged water lines.

Encouraging the construction of auxiliary water systems to supplement existing water lines. This will help ensure that adequate water flow for fire suppression will be available regardless whether the main water lines are damaged.

Planning for emergency response at the government and individual level to reduce the risk to the public from these hazards.

Identification of unsafe structures and posting public notices.


10.3 Recovery and Reconstruction

After major disasters around the world, local governments have been beset with pressures to act quickly to put their communities back together and restore normalcy. Businesses and governments alike are generally unprepared to deal efficiently and wisely with the flood of decisions which must be made in such situations regarding how best to recover and reconstruct. Poor timing of disaster assistance leads to failure of businesses which might otherwise survive. Buildings are rebuilt in an unsafe way or in hazardous locations so that similar destruction is likely to reoccur.

Local government has a major responsibility to prepare in advance to handle community recovery efficiently and wisely. Yet relatively little attention has so far been given to the need for practical guidance and training by which cities and counties can actively plan ahead for managing recovery and reconstruction in the best possible way. Delays and confusion caused by competing priorities for action and resources following a major disaster can be diminished by designing management systems and policies to address pre-event hazard mitigation as well as post-event short-term recovery and long-term reconstruction actions.

Hard experience from California earthquakes in the past decade has demonstrated the importance to cities of preparing in advance of disasters for actions required to be taken after. General Plan Guidelines encourage local governments to consider inclusion of contingency plans for immediate post-disaster response and longer term reconstruction activities in areas potentially subject to significant damage. Moreover, incorporation of disaster recovery and reconstruction policies by local governments can help communities to see more clearly the critical importance of timely hazard mitigation to the challenge of rebuilding a community. Pre-event mitigation of structural and natural hazards from earthquakes and their effects directly reduces post-event expenditure of human and financial resources needed to respond to the emergency, as well as to recover and reconstruct.

The benefits of this "ounce of prevention" principal have been demonstrated repeatedly both in flood control engineering and in flood plain management, two highly differing approaches to mitigating flood hazards. Flood control engineering leads to the creation of levees that reduces the hazards of building in or near natural riverbeds. Flood plain management precludes building in riverbeds and related overflow areas and thus limits potential damage. Both approaches reduce loss of life and property from flooding and concurrent disaster response, recovery and reconstruction costs. The combination of engineering mitigation and hazard avoidance may be directly applied toward the reduction of Yavapai County's earthquake risk.

Beyond this basic principal, it is necessary to face the probability that all known hazards will not be fully mitigated by the time disaster strikes. To the extent that a community has prepared itself to deal with the sudden onslaught of recovery and reconstruction issues as well as emergency response requirements, it will be in a better position to facilitate restoration of normalcy. To do this effectively, however, all levels and departments of local government must be willing to look beyond the more familiar emergency response functions to the less well known recovery and rebuilding issues. A wealth of literature chronicling real disaster recovery experiences has begun to emerge. From these new materials, local communities can draw parallels to their own circumstances. Recovery activity in cities such as Whittier and Santa Cruz can be instructive in shaping a community's own local recovery and reconstruction plan.

Adoption of a reconstruction and recovery ordinance, a pre-event ordinance authorizing post-event recovery and reconstruction organization and duties, is recommended for Yavapai County. If adopted before a disaster, such an ordinance can automatically go into effect with the declaration of a local emergency. This has the advantage not only of saving time it might otherwise take to get organized after a disaster, but also of providing a stronger legal basis for potentially controversial post-event recovery and reconstruction decisions.

Among the recommendations: is the pre-event adoption of an ordinance, to be activated with the declaration of emergency, which clearly spells out the recovery organization and the duties and responsibilities of its member departments and officials. Such an ordinance can be extremely useful in organizing recovery efforts which in many communities are an afterthought, if recognized at all, up to that point. The ordinance can deal with such matters as authorities and powers of different officials, distribution of responsibility and accountability, authorization of extraordinary procedures for streamlining repair permit issuance and simplifying public notice, interim joint powers and contract procedures, criteria for establishing building moratoria, standards for expedited repair permit processing, criteria for restoration of standard operating procedures and organization for planning and implementation of long-term reconstruction and redevelopment programs.


10.4 Recommended Goals and Policies

Goal: Minimize the economic impact of strong ground motion, liquefaction and fault rupture on public and private property, and protect the public from earthquake hazards.

10.4.1 Retrofit and Strengthening of Existing Facilities

The retrofit of unreinforced masonry buildings by adopting an ordinance that requires mandatory upgrading of this type of structures is encouraged for Yavapai County. In addition, Yavapai County should consider encouraging structural and nonstructural assessment and mitigation of other types of potentially hazardous buildings, including soft-stories, older residential structures, and tilt-up concrete buildings. This policy could be enforced when a potentially hazardous building is undergoing substantial repair or improvements resulting in more than half the assessed property value. Potential implementation measures should include:

Conducting an inventory and structural assessment of potentially hazardous buildings based on screening methods such as those developed by the Federal Emergency Management Agency;

the use of variances, tax rebates or credits, or public recognition as incentives; and

development of a mandatory upgrading program for high-occupancy, essential, dependent or high-risk facilities.

Yavapai County should advocate Federal legislation requiring that hospital buildings and structures be upgraded to comply to current building and fire code standards. Yavapai County should identify and mitigate nonstructural and structural hazards in all public owned buildings, especially critical facilities to ensure the safety of its employees and the survival of the structure. This should include:

Providing a nonstructural hazard mitigation public education system to reach all employees, businesses and the non-English speaking community.

Adopting an ordinance requiring correction of nonstructural hazards in commercial and industrial facilities.

Enforcing this ordinance through Yavapai County's established inspection process conducted by the various departments.

Yavapai County should coordinate with the public utility companies and oil companies, to strengthen, relocate or take other appropriate measures to safeguard natural gas and crude oil pipelines that extend through areas of high liquefaction potential, or areas that may settle differentially during an earthquake.

Yavapai County should coordinate and support efforts by the Arizona Department of Transportation to promote the expeditious strengthening of single-column bridges and other potentially hazardous freeway structures that do not meet seismic safety standards, or that are seated on potentially active faults.


10.4.2 Strengthening and Enforcing Seismic Codes

Yavapai County should strengthen the project permit and review process to ensure that proper actions are taken to mitigate the impact of seismic hazards, to encourage structural and nonstructural seismic design and construction practices that minimize earthquake damage in critical facilities, and to prevent the total collapse of any structure designed for human occupancy.

Yavapai County should provide review and enforce seismic design provisions and to identify and prevent structural and nonstructural design flaws in projects involving dependent, essential, high-risk, high-occupancy, or major commercial projects requiring approval. This provision should include training programs for plan checkers and building inspectors, or the retention of a State-certified structural engineer.

Yavapai County should coordinate with building owners, architects and structural engineers early in the review process to identify unacceptable nonstructural irregularities such as overhangs or parapets. Through the environmental review process, Yavapai County shall encourage special development standards, designs, and construction practices that reduce seismic risks to acceptable levels for projects involving critical facilities, large scale residential developments, and major commercial or industrial developments.

Planned lifeline utilities should be designed, located, structurally upgraded, fit with safety shut-off valves, designed for easy maintenance, and provided with redundant back-up systems as a condition of project approval if areas of liquefaction cannot be avoided.

Yavapai County should require geological and geotechnical investigations in areas of potential seismic or geologic hazards as part of the environmental and developmental review process.

Yavapai County should establish "Hazard Management Zones" that identify areas susceptible to faulting, liquefaction, settlement and slope instability. These zones may be established based on mapping provided in this report.

The County should enforce provisions for seismic analyses contained in the latest edition of the Uniform Building Code (UBC) for Seismic Zone 2b, including requirements for site amplification studies and dynamic analyses for critical facilities proposed in alluvial areas.

Geotechnical and engineering geological consultant recommendations should be peer reviewed by Yavapai County in-house or retained geotechnical engineer and/or geologist, State-registered in the corresponding discipline. Yavapai County should encourage alternative project designs or low intensity land uses during the environmental and developmental review process in areas determined to have significant seismic or geologic constraints.

      1. Public Education

Yavapai County should improve the public reduction of risks from seismic and/or geologic hazards during property transfers, including encouraging structural strengthening and site maintenance to reduce the risk to tolerable levels. Yavapai County should consider the following:

Providing and/or improving the visibility of hazard declaration statements on subdivision tract, parcel and zoning maps that includes the most current data on the location of potentially active faults; and

require property owners to sign a notice confirming their awareness and waiver of unmitigated risk identified in engineering, geologic or geotechnical investigation reports. Although the legal value of the notices are uncertain, they do provide educational value.

Yavapai County should make available pamphlets, brochures and in-house expertise to educate homeowners on earthquake preparedness, including the identification of nonstructural hazards.

Yavapai County should consider the preparation of community-specific hazard education tapes for broadcast over the local cable network, and establish educational "walking tours" to illustrate the greatest hazard issues affecting Yavapai County.

Yavapai County should provide "hands-on" training as a community resource for emergency preparation and response, that includes involving members of the community in practice drills.



Algermissen, S.T., Perkins, D.M., Thenhaus, P.C., Hanson, S.L., and Bender, B.L., 1990, Probabilistic Earthquake Acceleration and Velocity Maps for the United States and Puerto Rico: U.S. Geological Survey Miscellaneous Field Studies Map, MF-2120.

Algermissen, S.T., Perkins, D.M., Thenhaus, P.C., Hanson, S.L., and Bender, B.L., 1982, Probabilistic Estimates of Maximum Acceleration and Velocities in Rock in the Contiguous United States: U.S. Geological Survey, Open-File Report 82-1033, 107 pp.

Anderson, J. G., and Luco, J.E., 1983, Parametric Study of Near-Field Ground Motions for Oblique-Slip and Dip-Slip Dislocation Models: Seismological Society of America Bulletin, Vol. 73, p. 45-57.

Bausch, D.B., and Brumbaugh, D.S., 1997, Relocation Study of Early Arizona Earthquakes: Events of 1906, 1910 and 1912: Arizona Earthquake Information Center, Prepared for the Arizona Division of Emergency Management, Federal Emergency Management Agency Cooperative Agreement, ARS 35-148(A), dated March 1, 1997, 66 pp., and appendices

Bausch, D.B., and Brumbaugh, D.S., 1994, The Cataract Creek Earthquake Sequence of May-April 1993: Preliminary Findings: Seismological Research Letters, Vol. 65, No. 1, p. 31.

Bausch, D.B., and Brumbaugh, D.S., 1994, Earthquake Hazards in Arizona: Arizona Earthquake Information Center, Prepared for the Arizona Division of Emergency Services, Federal Emergency Management Agency Cooperative Agreement, AZ102EPSA, dated May 23, 1994, 49 pp., 2 plates and appendices.

Bausch, D.B., Brumbaugh, D.S., Morrison, S.J., and Daughton, T., 1993, Arizona 100-Year Acceleration Contour Map: Arizona Earthquake Information Center Contribution 93-2, Federal Emergency Management Agency Cooperative Agreement, AZ102EPSA, 1:1,000,000.

Bay Area Regional Earthquake Preparedness Project, BAREPP, 1986, Local Incentive Program: Case Studies, Federal Emergency Management Agency and Seismic Safety Commission Cooperative Agreement EMF-86-K-0253.

-------, 1986a, Local Incentive Program: Case Studies, Federal Emergency Management Agency and Seismic Safety Commission Cooperative Agreement EMF-86-K-0253.

-------, 1986b, Hazardous Buildings: Case Studies, Federal Emergency Management Agency Cooperative Agreement, EMF-86-K-0253.

-------, 1985, Earthquake Hazard Mitigation at the Community Level, Workshop Curriculum Materials: Spangle and Associates and the Reitherman Company, Cooperative Agreement No. EMG-85-K-0227, 54 p.

Beatley, T. and Burke, P., 1990, Seismic Safety through Public Incentives: The Palo Alto Seismic Hazard Identification Program, in Earthquake Spectra, Vol. 6, No. 1, p. 57-80.

Bender, B. and Perkins, D.M., 1987, SEISRISK III: A Computer Program for Seismic Hazard Estimation: U.S. Geological Survey Bulletin 1772, 48 pp.

Benuska, Lee, (editor), 1990, Loma Prieta Earthquake Reconnaissance Report: Earthquake Spectra, Supplement to Volume 6, 448 p.

Bolt, B.A., 1993, Earthquakes: W.H. Freeman and Company, New York, 331 pp.

Bonilla, M.G., Mark, R.K., and Lienkaemper, J.J., 1984, Statistical Relations Among Earthquake Magnitude, Surface Rupture Length, and Surface Fault Displacement: Bulletin of the Seismological Society of America, v. 74, p. 2379-2411.

Brumbaugh, D.S., 1991, Instrumental Magnitudes of Early Arizona Earthquakes: Seismological Research Letters, v. 62, p. 51.

Brumbaugh, D.S., 1987, A Tectonic Boundary for the Southern Colorado Plateau: Tectonophysics, v. 136, pp. 125-136.

Brumbaugh, D.S., Morrison, S.J., and Bausch, D.B., 1993, Arizona Earthquakes (1830-1992) Map and Catalog: Arizona Earthquake Information Center Contribution 93-1, 1:1,000,000, 12 pp.

Campbell, K.W., 1981, Near-source attenuation of peak horizontal acceleration: Bulletin of the Seismological Society of America, Vol. 71, pp. 2039-2070.

------, 1988, Preliminary Report on Empirical Studies of Vertical Strong Ground Motion for the Diablo Canyon Site, California: Report prepared for the U.S. Geological Survey, and to the U.S. Nuclear Regulatory Commission, dated October.

------, 1989, Preliminary Report on Empirical Studies of Vertical Strong Motion for Diablo Canyon Site, California, Prepared for the U.S. Nuclear Regulatory Commission.

Corley, W. G., 1986, BSSC/NEHRP seismic design program: Proceedings of the 1986 Annual Conference Western States Seismic Policy Council, p. 172-178.

Deppe, K., 1988, The Whittier Narrows, California Earthquake of October 1987 -Evaluation of Strengthened and Unstrengthened Unreinforced Masonry in Los Angeles City: EERI - Earthquake Spectra, Vol. 4, No. 1, p. 157-180.

DuBois, S.M., Sbar, M.L., and Nowak, T.A., 1982, Historical Seismicity in Arizona: Arizona Bureau of Geology and Mineral Technology; University of Arizona, Open-File Report 82-21, 199 pp.

DuBois, S.M., and Smith, A.W., 1980, The 1887 Earthquake in the San Bernardino Valley, Sonora: Historical Accounts and Intensity Patterns in Arizona: State of Arizona Bureau of Geology and Mineral Technology; University of Arizona, Special Paper No. 3, 112 pp.

Earthquake Engineering Research Center (EERC), 1994, Preliminary Report on the Seismological and Engineering Aspects of the January 17, 1994 Northridge Earthquake: College of Engineering, University of California at Berkeley, Report No. UCB/EERC-94/01, January 24, 1994.

Earthquake Engineering Research Institute (EERI), 1989, Loma Prieta Earthquake, October 17, 1989, Preliminary Reconnaissance Report: National Science Foundation, Grant No. CES8822367, Publication No. 89-03, 51 p.

-------, 1986, Reducing Earthquake Hazards: Lessons Learned from Earthquakes: Earthquake Engineering Research Institute, Publication No. 86-02, 208 p.

Earthquake Project, 1989, Catastrophic Earthquake, The Need to Insure Against Economic Disaster, Boston, Mass: National Committee on Property Insurance.

Eberhart-Phillips, D., Richardson, R.M., Sbar, M.L., and Herrmann, R.B., 1981, Analysis of the 4 February 1976 Chino Valley, Arizona Earthquake: Bulletin of the Seismological Society of America, v. 71, no. 3, p. 787-801.

Euge, K.M., Schell, B.A., and Lam, I.P., 1992, Development of Seismic Acceleration Maps for Arizona: Arizona Department of Transportation Report No. AZ92-344, 327 pp.

Evernden, J.F. and Thompson, J.M., 1988, Predictive model for important ground motion parameters associated with large and great earthquakes: U.S. Geological Survey Bulletin 1838, U.S. Government Printing Office, 27 p.

Federal Emergency Management Agency (FEMA), 1985, Comprehensive Earthquake Preparedness Planning Guidelines: Earthquake Hazards Reduction Series 2, FEMA 73, May, 81 p. plus references.

------, 1985, An action plan for reducing earthquake hazards of existing buildings: Earthquake Hazards Reduction Series 16, FEMA 90, December, 75p.

------, 1987, Abatement of Seismic Hazards to Lifelines: Water and Sewer Lifelines and Special Work Shop Presentations: Earthquake Hazard Reduction Series 2, FEMA 135, 181 p.

------, 1988, Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook: Earthquake Hazards Reduction Series 41, FEMA 154, July, 185 p.

------, 1988, Rapid Visual Screening of Buildings for Potential Seismic Hazards: Supporting Documentation: Earthquake Hazards Reduction Series 42, FEMA 155, September, 137 p.

------, 1988, Earthquake Damaged Buildings: An Overview of Heavy Debris and Victim Extrication: Earthquake Hazards Reduction Series 43, FEMA 158, September.

------, 1988, Seismic considerations: Health Care Facilities: Earthquake Hazards Reduction Series 35, FEMA 150, April, 105 p.

------, 1989, Establishing Programs and Priorities for the Seismic Rehabilitation of Buildings, A Handbook: Earthquake Hazards Reduction Series 45, FEMA 174, May, 122 p.

------, 1989, Establishing Programs and Priorities for the Seismic Rehabilitation of Buildings: Supporting Report: Earthquake Hazards Reduction Series 46, FEMA 173, May, 190 p.

------, 1990, Seismic Safety of Existing Buildings, Societal Implications: L 172, August.

------, 1990, Seismic Safety of Existing Buildings, Engineering Considerations: L 171, August.

Frankel, A., Mueller, C., Barnhard, T., Perkins, D., Leyendecker, E.V., Dickman, N., Hanson, S., Hopper, M., 1996, Interim National Seismic Hazard Maps: Documentation, 2 Plates, Fault Table; draft dated January 18, 1996, 31 pp.

Haley, S., and Hunt, G., 1974, Determination of design earthquake characteristics: American Society of Civil Engineers, Los Angeles, 19 pp.

Hamblin, W.K., Damon, P.E., and Bull, W.B., 1981, Estimates of Vertical Crustal Strain Rates Along the Western Margin of the Colorado Plateau: Geology, v. 9., p. 293-298.

Hayes, E., 1988, A Review of Information on Seismic Hazards Needed for the Earthquake-Resistant Design of Lifeline Systems in the United States: U.S. Geological Survey, Open File Report ICSSCTR-10, 71 p.

Jaffe, M., Butler, J., and Thurow, C., 1981, Reducing Earthquake Risks: A Planners Guide: American Planning Association (Planning Advisory Service), Report No. 364, 82 p.

Jaumé, S.C., and Sykes, L.R., 1992, Changes in State-of-Stress on the Southern San Andreas Fault Resulting from the California Earthquake Sequence of April to June 1992: Science, November 20, v. 258, p. 1325-1328.

Jones, L.M., 1992, Landers Aftershocks and Earthquake Probabilities for the San Andreas Fault in Southern California: EOS Abstract, v. 73, p. 357.

Joyner, W. B. and Boore, D. M., 1982, Estimation of Response--Spectral Values as a Function of Magnitude, Distance and Site Conditions: U.S. Geological Survey, Open File Report B2-881, 14 p.

-------, 1981, Peak Horizontal Acceleration and Velocity from Strong Motion Records Including Records from the 1979 Imperial Valley, California Earthquake: Seismology Society of America Bulletin, Vol. 71, pp. 2011 - 2038.

Joyner, W. B., and Fumal, T. E., 1985, Predictive Mapping of Earthquake Ground Motion: in Ziony, J. I., ed., Evaluating Earthquake Hazards in the Los Angeles Region, An Earth-Science Perspective: U.S. Geological Survey Professional Paper 1360, p. 203-217.

Joyner, W.B, and Boore, D.M, 1988, Measurement, characterization, and prediction of strong ground motion: in Von Thun, J.L., (editor), Earthquake Engineering and Soil Dynamics II - Recent Advances in Ground-Motion Evaluation, Proceedings of the specialty conference sponsored by the Geotechnical Engineering Division of the American Society of Civil Engineers, June 27-30, Park City, Utah, pp. 43-102.

Kanai, K., 1983, Engineering seismology: University of Tokyo Press, Tokyo, 251 pp.

Menges, C.M., and Pearthree, P.A., 1983, Map of Neotectonic (Latest Pliocene-Quaternary) Deformation in Arizona: Arizona Bureau of Geology and Mineral Technology, Open-File Report 83-22.

Mokhtar, T. A., 1979, The relationship between the seismicity and late Cenozoic tectonics in Arizona: University of Arizona Master Thesis, 53 pp.

Morrison, S.J., Brumbaugh, D.S., and Daughton, T., 1991, State of Arizona Historical Intensity Map (1887-1987): Arizona Earthquake Information Center Contribution 91-1, Federal Emergency Management Agency Cooperative Agreement, AZ102EPSA, 1:1,000,000.

NUREG, 1975, Procedures for evaluation of vibratory ground motions of soil deposits at nuclear power plant sites: U. S. Department of Commerce, NUREG-75/072, 64 pp.

Pearthree, P.A., and Wallace, T.C., 1992, The St. George Earthquake of September 2, 1992: Arizona Geology, Arizona Geological Survey, v. 22, no. 4, p. 7-8.

Pearthree, P.A., Menges, C.M., and Mayer, L., 1983, Distribution, Recurrence and Possible Tectonic Implications of Late Quaternary Faulting on Arizona: Arizona Bureau of Geology and Mineral Technology Open-File Report 83-20, 51 pp.

Piety, L.A., and Anderson, L.W., 1991, The Horseshoe Fault Evidence for Prehistoric Surface-Rupturing Earthquakes in Central Arizona: Arizona Geology, Arizona Geological Survey, v. 21, no.3, p. 1-4.

Reynolds, S.J., 1988, Geologic Map of Arizona: Arizona Geological Survey, Map 26, scale 1:1,000,000.

Richter, C.F., 1958, Elementary Seismology: W.H. Freeman and Company, San Francisco, p. 768.

Rogers, A. M., Tinsely, J. C., and Borcherdt, R. D., 1985, Predicting Relative Ground Response: in Ziony, ed., Evaluating Earthquake Hazards in the Los Angeles Region, An Earth Science Perspective: U.S. Geological Survey, Professional Paper 1360, p. 221-248.


Scarborough, R.B., Menges, C.M., and Pearthree, P.A., 1986, Map of Late Pliocene-Quaternary (post-4-m.y.) Faults, Folds, and Volcanic Outcrops in Arizona: Arizona Bureau of Geology and Mineral Technology Map 22, scale 1:1,000,000.

Seed, H. B.,1981, Earthquake Resistant Design of Earth and Rock Fill Dams: Geotechnique, Vol. 29, No. 3, pp. 215-263.

-------, 1979, Soil Liquefaction and Cyclic Mobility Evaluation for Level Ground During Earthquakes: Journal of the Geotechnical Engineering Division, ASCE, Vol. 105, No. GT2, p. 201-255.

Seed, H. B., Idriss, I. M., and Kieffer, F. W., 1969, Characteristics of rock motions during earthquakes: Journal of the Soil Mechanics and Foundations Division, Proceedings of the American Society of Civil Engineers, p. 1199-1218.

Seed, H. B., Idriss, I. M., and Arango, I., 1983, Evaluation of Liquefaction Potential Using Field Performance Data: American Society of Civil Engineers Journal of Geotechnical Engineering Division, Vol. 109, No. 3, p. 458-482.

Seed, H. B., Tokimatsu, K., Harder, L. F., and Chung, R. M., 1985, Influence of SPT Procedures in Soil Liquefaction Resistance Evaluations: Journal of the Geotechnical Engineering Division, ASCE Vol. III, No. 12, December.

Sharpe, R., 1982, An Investigation of the Correlation between Earthquake Ground Motion and Building Performance: Applied Technology Council, U.S. Contract Survey, No. 14-08-0001-19892, 113 pp.

Shoemaker, E.M., Squires, R.L., and Abrams, M.J., 1978, Bright Angel and Mesa Butte Fault Systems of Northern Arizona: in Smith, R.B., and Eaton, G.P., eds., Cenozoic Tectonics and Regional Geophysics of the Western Cordillera: Geological Society of America Memoir 152, p. 341-367.

Steinbrugge, K.V, and Algermissen, S.T., 1990, Earthquake Losses to single-family dwellings: California Experience: U.S. Geological Survey Bulletin 1939-A, U.S. Government Printing Office, 65 pp.


Sudduth, C., 1986, Evaluating Earthquake Ground Failure Potential for Development Decisions, Summary of Working Group V and Audience Discussion: in Future Directions in Evaluating Earthquake Hazards in Los Angeles Region, An Earth-Science Perspective: U.S. Geological Survey, Professional Paper 1360, p. 320.

Tinsley, J. C., Youd, T. L., Perkins, D. M., and Chen A. T. F., 1985, Evaluating Liquefaction Potential: in Ziony, ed., Evaluating Earthquake Hazards in the Los Angeles Region, An Earth Science Perspective: U.S. Geological Survey, Professional Paper 1360, p. 263-315.

Trifurnac, M. D., and Brady, A. G., 1975, On the correlation of seismic intensity scales with the peaks of recorded strong ground motion: Bulletin of the Seismological Society of America, v. 65, no. 1, p. 139-163.

Wallace, T.C., 1992, The 1992 Landers Earthquake Sequence: Arizona Geology, Arizona Geological Survey, v. 22, no. 3, p. 1-4.

Wesnousky, S.G., 1986, Earthquakes, Quaternary Faults, and Seismic Hazard in California: Journal of Geophysical Research, v. 91, no. B12, p. 12,587-12,631.

Wolfe, M. R., Bolton, P. A., Heikkala, Greene M. M., May, P. J., 1986, Land-Use Planning for Earthquake Hazard Mitigation: A Handbook for Planners: Natural Hazards Research and Applications Information Center, Special Publication 14, 122 p.

Wong, I.G., and Chapman, D.S., 1986, Deep intraplate earthquakes in the intermountain U.S.: Implications to thermal and stress conditions in the lower crust and upper mantle: Earthquake Notes, v. 47, p. 6.

Working Group on the Probabilities of Future Large Earthquakes in Southern California, 1992, Future Seismic Hazards in Southern California, Phase I: Implications of the Landers Earthquake Sequence: National Earthquake Prediction Council, California Earthquake Prediction Evaluation Council, and Southern California Earthquake Center, 42 pp.

Wyss, M., 1979, Estimating Maximum Expectable Magnitude of Earthquakes from Fault Dimensions: Geology, v. 7, p. 336-340.

Youd, T. L., 1978, Major Cause of Earthquake Damage is Ground Failure: Civil Engineering, Vol. 48, No. 4, p. 47-51.

Youd, L. T., 1986, Geologic effects-liquefaction and associated ground failure: Proceedings of the 1986 Annual Conference Western Seismic Policy Council, p. 8-30.

Youd, T.L., and Perkins, D.M., 1978, Mapping liquefaction-induced ground failure potential: Proceedings of the American Society of Civil Engineers, Journal of the Geotechnical Engineering Division, Vol. 104, No. GT4, pp. 433-446.

Youd, T. L., and Keefer, D. K., 1981, Earthquake-induced ground failures, in Hays, W. W., ed., Facing Geologic and Hydrologic Hazards: U. S. Geologic Survey Professional Paper 1240-B, p. 23-31.


ACES - Arizona Council on Earthquake Safety.

AEIC - Arizona Earthquake Information Center.

AZEPP - Arizona Earthquake Preparedness Program.

Abatement - The reduction or elimination of a hazardous condition, including but not limited to strengthening, occupancy restrictions, or demolition.

Accelerograph - An instrument used to record very large earthquakes.

Aftershocks - Earthquakes that follow a main larger earthquake, most commonly smaller than the main shock.

Alluvium - Surficial sediments of poorly consolidated, gravels, sands, silts, and clays deposited by flowing water.

Bedrock - Designates hard rock that is in its natural intact position and underlies soil or other unconsolidated surficial material.

Catastrophic Earthquake - A great earthquake, generally near Magnitude 8.0 or greater, capable of widespread tremendous damage in the epicentral area accompanied by localized severe damage hundreds of miles from earthquake source.

Coseismic Rupture - Rupture occurring during an earthquake.

Critical Facilities - The following are categories of critical facilities:

Any lifeline system facility whose operations are essential at times of emergencies, such as energy transmission systems, major utility lines, evacuation and disaster routes, and warehouses storing heavy rescue and debris cleanup equipment;

essential facilities that are needed in times of emergencies, such as medical facilities, fire and police stations, emergency operation centers, and communication centers;

high-risk facilities that, if severely damaged, could cause catastrophic casualties or disasters far beyond the facility itself, such as nuclear power plants, dams and flood control infrastructure, freeway-to-freeway interchanges and bridges, and industrial plants which manufacture or store explosives, toxic materials, or petroleum products;

high occupancy facilities having potential for catastrophic fatalities and crowd control problems such as high-rise buildings, large assembly public and private facilities and large multifamily residential complexes;

dependent care facilities that house populations with special evacuation considerations, such as educational facilities for preschool, public and private schools, rehabilitation centers, prisons, and major social care facilities, including group care homes and elderly care facilities; and

economic facilities whose continued operation is paramount to avoiding severe economic impact, such as banking, vital record keeping, archiving activities, airport and port facilities, and large industrial/commercial complexes.

Dekameters (Dam3) - Unit of dam storage capacity where 1 Dam3 is equal to .8 acre-feet. An acre-foot is the volume of an acre of area that is one foot in thickness.

Deterministic - Synonymous with designating a maximum credible event for the seismic design of a structure.

Differential Settlement - Nonuniform settlement; the uneven lowering of different parts of an engineering structure, often resulting in damage to the structure. Sometimes included with liquefaction as ground failure phenomenon.

Dynamic Analyses - A complex earthquake-resistant engineering design technique (UBC - used for critical facilities) capable of modeling the entire frequency spectra, or composition, of ground motion. The method is used to evaluate the stability of a site or structure by considering the motion from any source on mass, such as that dynamic motion produced by machinery or a seismic event.

EERC - Earthquake Engineering Research Center.

EERI - Earthquake Engineering Research Institute.

Earthquake - Vibratory motion propagating within the earth or along its surface caused by the abrupt release of strain from elastically deformed rock by displacement along a fault.

Emergency Operations Center - A room or building equipped and staffed to monitor and direct the response to an emergency situation.

FEMA - Federal Emergency Management Agency.

Fault - A fracture (rupture) or a zone of fractures along which there has been displacement of adjacent earth material.

Fault Mechanism - A model of the sense and orientation of the generating fault displacement.

Fault Slip Rate - The average long-term movement of a fault as determined from geologic evidence (measured in cm/year).

Frequency Composition - The range of frequencies as measured at a site, including high frequency earthquake ground motion (greater than 1 Hz) and lower frequency energy (less than 1 Hz).

Ground Failure - Permanent ground displacement produced by fault rupture, differential settlement, liquefaction, or slope failure.

Ground Rupture - Displacement of the earth's surface as a result of fault movement associated with an earthquake.

High Occupancy - High occupancy can be defined as exceeding 100, 200, and 500 person capacity, depending on importance factors of the facility, i.e., dependent care and other building uses.

Intensity - A subjective numerical measure of the effects of an earthquake at a particular place. Intensity depends on the earthquake magnitude, distance from the epicenter, and on the local geology, and the subjective and/or empirical observations of damage in particular cultural setting, i.e., construction standards.

Isoseismal Area - An area composed of points of equal earthquake intensity on the earth's surface.

Lateral Force - The force of the horizontal side-to-side motion on the earth surface as measured on a particular mass, either a building or structure.

Lateral Spread - Shallow angle slope failure caused by liquefaction of a subsurface layer; very large displacement.

Levee - An artificial embankment of earth fill, built along the bank of a water course or an arm of the sea and designed to protect land from inundation.

Lifeline System - Linear conduits or corridors for the delivery of services or movement of people and information (e.g., pipelines, telephones, freeways, railroads).

Liquefaction - Changing of soils (unconsolidated alluvium) from a solid state to weaker state unable to support structures; where the material behaves similar to a liquid as a consequence of earthquake shaking.

Magnitude - A measure of the size of an earthquake, as determined by measurements from seismographic records.

Major Earthquake - Capable of widespread, heavy damage up to 50+ miles from epicenter; generally near Magnitude Range 6.5 to 7.0 or greater, but can be less, depending on rupture mechanism, depth of earthquake, and location relative to urban center, etc.

Moderate Earthquake - Capable of causing considerable to severe damage, generally in the range of Magnitude 5.0 to 6.0 (Modified Mercalli Intensity <VI), but highly dependent on rupture mechanism, depth of earthquake, and location relative to urban center, etc.

Near-Field Earthquake - Used to describe a local earthquake within approximately a few fault zone widths of fault which is characterized by high frequency waveforms that are destructive to above-ground utilities and short-period structures (less than 2- or 3-stories high).

Neotectonic Fault - A fault showing evidence of movement within recent geologic time, generally considered as the last 5 million years, and capable of generating damaging earthquakes.

Nonductile - Refers to concrete construction using columns and beams with insufficient steel reinforcing, making a structure brittle and prone to collapse during earthquakes.

Nonstructural Hazard - Fallen or broken interior or exterior building components such as stains, piping, windows, wall partitions, electrical conduits, lights, brick veneer, equipment, etc.

Parapet - The portion of any wall which extends above the roof line, or a wall that serves as a guard at the edge of a balcony or roof.

Piling - Used to pin a structure to stable bedrock below using long piles emplaced through an unstable zone of potential ground failure; commonly utilized for landslide and liquefaction mitigation.

Peak Acceleration - The greatest amplitude of acceleration measured across a record left on an accelerogram.

Professional Geologist - A geologist who is certified by the State as qualified to apply geologic data, principles, and interpretation to naturally-occurring earth materials so that geologic factors affecting planning, design, construction, and maintenance of civil engineering works are properly recognized and used. A professional geologist is particularly needed to conduct investigations, often with civil engineers, of sites with potential ground failure hazards.

Project - A development application involving zone changes, variances, conditional use permits, tentative parcel maps, tentative tract maps, and plan amendments.

Power Intertie - Terminal facilities in a lifeline network system, such as water and sewage pumping stations, electrical power substations, and power generating facilities.

Poisson Distribution - A probability distribution that characterizes discrete events occurring independently of one another in time.

Probabilistic Assessment - Seismic risk assessment that accounts for how often and how strongly the ground will shake from surrounding seismic sources. The designation of a maximum probable earthquake is probabilistic.

Pseudostatic - A simplified engineering analysis of the stability of a site or structure that translates earthquake ground motion into a static force in performing stability calculations.

Reinforced Masonry - Masonry construction with steel reinforcement.

Resonance - Amplification of ground motion frequencies within bands matching the natural frequency of a structure and often causing partial or complete structural collapse; effects may demonstrate minor damage to single-story residential structures while adjacent 3- or 4-story buildings may collapse because of corresponding frequencies, or vice-versa.

Response Spectra - The range of potentially damaging frequencies of a given earthquake applied to a specific site and for a particular building or structure.

SASO - Southern Arizona Seismic Observatory.

Seismic Capacity - The measure of resistance of a structure to strong earthquake shaking, or seismic loading as commonly used when applying lateral forces to a structure.

Seismogenic - Capable of producing earthquake activity.

Seismology - The study of earthquakes.

Soft-Story Construction - A structure that has at least one story, often the ground level, with significantly less rigidity or strength than other floors in the structure during seismic loading, e.g., apartment dwellings with garages on the lower floor, etc. A soft-story is considered to have less than 70% of the stiffness of the story above, while a weak-story possesses less than 80% of the strength of the story above.

Site Amplification - An increase in the seismic signal amplitude within some frequency resulting from propagation through crust, topography, earth materials at the site, bedrock and alluvium contact, etc.

Standard Cone Penetrometer - A engineering borehole measurement, which among other applications, is able to translate penetration resistance (as gauged from a narrow cylindrical instrument of standard weight dropped from a standard height) of unconsolidated earth materials into a measure of liquefaction susceptibility.

Storage Capacity - Is measured in acre-feet and decameters, including dead storage.

Strike-Slip Fault - A fault with a vertical to subvertical fault surface in three dimensions that displays evidence of horizontal and opposite displacement.

Structural Engineer - A licensed Civil Engineer certified by the State as qualified to design and supervise the construction of engineered structures.

Tectonics - A mechanical model relating the movement of several rigid crustal plates over an underlying plastic layer. The juncture of adjacent plates can be one of convergence, spreading, or sliding along a transform tectonic plate boundary.

Tilt-up Wall - A poured concrete-and-steel-reinforced wall raised or lifted into place and tied to other walls, and to the roof and floor.

USBR - United States Bureau of Reclamation.

Veneer - A covering layer of material for walls that is attached to the wall, but not bounded adequately so as to act with it during earthquake shaking.

Water Table - The upper surface of ground water that saturates pores and fractures in rock or surficial earth materials.