ShakeMap Scientific Background
- Maps - General Information
- Map Types
- Global Earthquake ShakeMaps
- ShakeMap Atlas
- Scenario Earthquakes
For additional background information, download the ShakeMap Manual
A ShakeMap is a representation of ground shaking produced by an earthquake. The information it presents is different from the earthquake magnitude and epicenter that are released after an earthquake because ShakeMap focuses on the ground shaking produced by the earthquake, rather than the parameters describing the earthquake source. So, while an earthquake has one magnitude and one epicenter, it produces a range of ground shaking levels at sites throughout the region depending on distance from the earthquake, the rock and soil conditions at sites, and variations in the propagation of seismic waves from the earthquake due to complexities in the structure of the Earth's crust.
Part of the strategy for generating rapid-response ground motion maps is to determine the best format for reliable presentation of the maps given the diverse audience, which includes scientists, businesses, emergency response agencies, media, and the general public. In an effort to simplify and maximize the flow of information to the public, we have developed a means of generating not only peak ground acceleration and velocity maps, but also an instrumentally-derived, estimated Modified Mercalli Intensity map. This map makes it easier to relate the recorded ground motions to the expected felt and damage distribution. The Instrumental Intensity map is based on a combined regression of recorded peak acceleration and velocity amplitudes. (see Intensity Maps)
Even with the current seismic station distribution in Calfornia, data gaps are common, particularly for events outside the densely-instrumented metropolitan regions surrounding Los Angeles and San Francisco. In order to stabilize contouring and minimize the misrepresentation of the ground motion pattern due to data gaps, we augment the data with predicted values in areas without data. Given the epicenter and magnitude (and for larger earthquakes, fault geometry if avaliable), peak motion amplitudes in spare regions are estimated from the ground motion prediction equations.
The instrumental intensity map shows station symbols as open triangles in order to see the underlying intensity value. The legend bar at the bottom explains the colors (and see Intensity Maps below). For the intensity map as with other maps, station locations are the best indicator of where the map is most accurate: Near seismic stations the shaking is well constrained by data; far from such stations, the shaking is estimated using standard seismological inferences and interpolation.
Note: ShakeMaps are generated automatically following moderate and large earthquakes. These are preliminary ground shaking maps, normally posted within several minutes of the earthquake origin time. The acceleration and velocity values are raw and are at least initially, NOT checked by humans. Further, since ground motions and intensities typically can vary significantly over small distances, these maps are only APPROXIMATE. At small scales, they should be considered unreliable. Finally, the input data is raw and unchecked, and may contain errors. (See Disclaimer)
Maps - General Information
- The Red star usually located near the center of the map is the epicenter.
- Small unfilled circles, shown on some maps, are points where strong motion values were estimated and used to fill gaps in the station distribution.
- The colored triangles indicate reporting stations. In California,
- Red triangles are stations from the Caltech/USGS digital telemetered network.
- Blue triangles represent California Geological Survey (CGS) and California Strong Motion Instrumentation Program (CSMIP) dial-up stations.
- Yellow triangles represent stations from the ANZA Regional Network.
- Green triangles represent National Strong Motion Project (NSMP) dial-up stations or, on historical maps, non-digital stations from which strong motion records were digitized.
- On the Intensity maps, station symbols are open, allowing the underlying intensity color to show thru. On the other maps station symbols (circles, triangles) are color coded according to the data type mentioned above.
The station information includes the station code and name, the agency that manages the station, the station location coordinates in degrees latitude and longitude, and the peak acceleration, velocity, and spectral acceleration for each component of ground motion (when available). Spectral acceleration maps are only made for larger earthquakes (normally, magntidue greater than 5.5). When the peak ground motion maps are made, the value from the peak horizontal component of ground motion is used as the value for the station. This value is highlighted in bold in the station information.
Components from many stations are defined by three letter codes. The last letter indicates the orientation (Z = vertical, N = horizontal north, E = horizontal east). The first two letters indicate the instrument class:
|VL||low gain channels on the analog network|
|VH||high gain channels on the analog network|
|AS||FBA's on the analog network|
|HL||FBA's on the digital network|
|HN||FBA's on the digital network|
|BH||broadband data streams|
|HH||broadband data streams|
FBA's (force balance accelerometers) are designed to record extremely large ground motions and can accurately record waves from very large earthquakes. However, ground motions from small and moderate earthquakes are often too small to trigger these instruments or rise above instrument noise. On the other hand, Broadband seismic sensors can record extremely small ground motions and accurately record waves from earthquakes that range from very small up to moderately large. A number of stations have both FBA and broadband sensors. For ShakeMap, the network tends to emphasize FBA recordings for large ground motions and broadband recordings for small ground motions.
Occassionally, station channels will be flagged due to problems with the station or possibly anomalous peak values. In this case, the popup window of station information will indicate the flagging with the following codes:
|G||Glitch (clipped or below noise)|
|I||Incomplete time series|
|N||Not in list of known stations|
Peak Acceleration Maps
Peak horizontal acceleration at each station is contoured in units of percent-g (where g = acceleration due to the force of gravity = 981 cm/s/s). The peak values of the vertical components are not used in the construction of the maps because they are, on average, lower than the horizontal amplitudes and ground motion prediciton equations used to fill in data gaps between stations are based on peak horizontal amplitudes. The contour interval varies greatly and is based on the maximum recorded value over the network for each event.
For moderate to large events, the pattern of peak ground acceleration is typically quite complicated, with extreme variability over distances of a few km. This is attributed to the small scale geological differences near the sites that can significantly change the high-frequency acceleration amplitude and waveform character. Although distance to the causative fault clearly dominates the pattern, there are often exceptions, due to local focussing and amplification. This makes interpolation of ground motions at one site to a nearby neighbor somewhat risky. Peak acceleration pattern usually reflects what is felt from low levels of shaking up to to moderate levels of damage.
Peak Velocity Maps
Peak velocity values are contoured for the maximum horizontal velocity (in cm/sec) at each station. As with the acceleration maps, the vertical component amplitudes are disregarded for consistency with the regression relationships used to estimate values in gaps in the station distribution. Typically, for moderate to large events, the pattern of peak ground velocity reflects the pattern of the earthquake faulting geometry, with largest amplitudes in the near-source region, and in the direction of rupture (directivity). Differences between rock and soil sites are apparent, but the overall pattern is normally simpler than the peak acceleration pattern. Severe damage, and damage to flexible structures is best related to ground velocity. For reference, the largest recorded ground velocity from the 1994 Northridge (Magnitude 6.7) earthquake made at the Rinaldi Receiving station, reached 183 cm/sec.
Spectral Response Maps
Following earthquakes larger than magnitude 5.5, spectral response maps are made. Response spectra portray the response of a damped, single-degree-of-freedom oscillator to the recorded ground motions. This data representation is useful for engineers determining how a structure will react to ground motions. The response is calculated for a range of periods. Within that range, the Uniform Building Code (UBC) refers to particular reference periods that help define the shape of the "design spectra" that reflects the building code.
ShakeMap spectral response maps are made for the response at three UBC reference periods: 0.3, 1.0, and 3.0 seconds. For each station, the value used is the peak horizontal value of 5% critically damped pseudo-acceleration.
Rapid Instrumental Intensity Maps
As an effort to simplify and maximize the flow of information to the public, we have developed a means of generating estimated Modified Mercalli Intensity maps based on instrumental ground motion recordings (Wald et al., 1999). These "Instrumental Intensities" are based on a combined regression of peak acceleration and velocity amplitudes vs. observed intensity for eight significant California earthquakes (1971 San Fernando, 1979 Imperial Valley, 1986 North Palm Springs, 1987 Whittier, 1989 Loma Preita, 1991 Sierra Madre, 1992 Landers, and 1994 Northridge).
From the comparison with observed intensity maps, we find that a regression based on peak velocity for intensity > VII and on peak acceleration for intensity < VII is most suitable. This is consistent with the notion that low intensities are determined by felt accounts (sensitive to acceleration). Moderate damage, at intensity VI-VII, typically occurs in rigid structures (masonry walls, chimneys, etc.) which also are sensitive to high-frequency (acceleration) ground motions. As damage levels increase, damage also occurs in flexible structures, for which damage is proportional to the ground velocity, not acceleration. By relating recorded ground motions to Modified Mercalli intensities, we can now estimate shaking intensities within a few minutes of the event based on the recorded peak motions made at seismic stations.
A descriptive table of Modified Mercalli Intensity is available from ABAG (Association of Bay Area Governments). A table of intensity descriptions with the corresponding peak ground acceleration (PGA) and peak ground velocity (PGV) values used in the ShakeMaps is given below. ShakeMap uses PGA to estimate intensities lower than V, it linearly combines PGA & PGV for intensities greater than V and less than VII, and it uses PGV for intensities greater than VII (See Wald et al., 1999b, for more details).
Global Earthquake ShakeMaps
For regions around the world were there are insufficient near-real time strong motion seismic stations to generate an adequate, strong-ground-motion data-controlled ShakeMap, we can still provide a very useful estimate of the shaking distribution using the ShakeMap software.
Initially, a point source approximation (hypocenter and magnitude) is used to constrain region-specific empirical ground motion estimation. Site amplification is approximated from a relationship developed between topographic gradient and shear-wave velocity. Additional constraints for these predictive maps come primarily from three important sources, the availability of which varies depending on the region in which the earthquake occurred, as well as a function of time after the earthquake occurrence. These constraints include: (1) additional earthquake source information, particularly fault rupture dimensions, (2) observed macroseismic intensities (including via the USGS “Did You Feel It?” system, and (3) observed strong ground motions, where and when available.
When intensity data are used directly, the peak ground motion parameters are inferred from the macroseismic observations using the equations of Wald et al, (1999a). This is the opposite approach normally used in ShakeMaps where numerous seismic recording are available, and from them the intensities are then inferred.
Input data for global ShakeMaps are depicted with circles for intensity reports and triangles for strong motion data. Intensities are further separated by color: Blue circles indicate data collected via the USGS "Did You Feel It?" Community Internet Intensity Map system and Yellow circles indicate traditional Modifed Mercalli Intensity assignments. For the intensity maps, circles are open, allowing the underlying intensity color to show thru. On the other maps observation location symbols (circles, triangles) are color coded according to the data type mentioned above.
An atlas of maps of peak ground motions and intensity “ShakeMaps” has been developed for approximately 1,000 recent and historical global earthquakes (Allen and others, 2008). These maps are produced using established ShakeMap methodology (Wald and others, 1999c, 2005) and constraints from macroseismic intensity data, instrumental ground motions, regional topographically-based site amplifications (Wald and Allen, 2007), and published earthquake rupture models. The Atlas of ShakeMaps provides a consistent and quantitative description of the distribution of shaking intensity for recent global earthquakes (January 1973 – September 2007 for Version 1.0). We anticipate that the Atlas will be regularly updated with more data constraints for historical events and the addition of future significant events as time progresses.
The Atlas was developed specifically for calibrating global earthquake loss estimation methodologies to be used in the USGS Prompt Assessment of Global Earthquakes for Response (PAGER) Project. PAGER will employ these loss models to rapidly estimate the impact of global earthquakes as part of the USGS National Earthquake Information Center’s earthquake response protocol. Though developed primarily for PAGER, we anticipate many other uses for the historical ShakeMap Atlas, including disaster response planning, capacity building and outreach programs, in addition to calibration of other global loss methodology approaches.
The primary sources for instrumental data are:
- Pacific Earthquake Engineering Research Center’s Next Generation Attenuation (NGA) database
- Consortium of Organizations for Strong-Motion Observation Systems (COSMOS)
- European Strong-Motion Database (ESD)
- Kyoshin Network (K-NET), Japan
Macroseismic intensity observations are gathered from several online sources, including:
- National Geophysical Data Center, National Oceanic and Atmospheric Administration (NOAA)
- Centro Regional de Sismología para América del Sur (CERESIS)
- Istituto Nazionale di Geofisica e Vulcanologia (INGV), Italy
Community Internet Intensity Maps data obtained from the USGS Did You Feel It? (DYFI?) system (Wald and others, 1999a; Atkinson and Wald, 2007) are a valuable ground-shaking constraint and are used for U.S. events since 1999 and for global events since 2003. Additional intensity data were gathered or digitized from numerous earthquake reconnaissance reports and peer reviewed publications, or through the generous contribution from colleagues around the world.
An important, consolidated source for finite fault models is provided by Martin Mai of the Swiss Seismological Service, Zurich. Other fault rupture dimensions were digitized from observations of surface displacement, finite-fault source inversions and slip distribution determined from teleseismic observations, or more recently, from InSAR observations. Less well-constrained faults have been estimated from earthquake aftershock distributions.
The ShakeMap Atlas aims to provide the best estimate of the shaking distribution for historical earthquakes. One important feature of the Atlas is that data constraints are not limited to those data produced by the National Earthquake Information Center, and we will openly accept data contributions that help to improve the representation of the shaking distribution for any of the events. References to all data sources in the Atlas of ShakeMaps (as of August 2008) are provided in Allen and others (2008). Please contact Trevor Allen or David Wald with any comments or data contributions.
Below we indicate the default values and predictive equations used in constructing the Atlas of ShakeMaps. Variations in ground-motion prediction equations (GMPEs) and other configurations may change on an event-by-event basis. Please see the event info.xml on individual ShakeMap download pages or Allen and others (2008) for more information.
Date Range: January 1973 – September 2007
Active crust GMPE: Boore and others (1997)
Subduction zone GMPE (intraslab and interface): Youngs and others (1997)
Stable continent GMPE: Atkinson and Boore (2006)
Seismic site-conditions: Wald and Allen (2007) <http://earthquake.usgs.gov/vs30/>
Peak ground motion to macroseismic intensity conversions: Wald and others (1999b)
- Allen, T.I., Wald, D.J., Hotovec, A.J., Lin, K., Earle, P.S., and Marano, K.D., 2008, An Atlas of ShakeMaps for selected global earthquakes: U.S. Geological Survey Open-File Report, 2008-1236, 34 p.
- Atkinson, G.M., and Boore, D.M., 2006, Earthquake ground-motion predictions for eastern North America: Bull. Seism. Soc. Am., v. 96, p. 2181-2205.
- Atkinson, G.M., and Wald, D.J., 2007, "Did You Feel It?" intensity data: A surprisingly good measure of earthquake ground motion: Seism. Res. Lett., v. 78, no. 3, p. 362-368.
- Boore, D.M., Joyner, W.B., and Fumal, T.E., 1997, Equations for estimating horizontal response spectra and peak acceleration from Western North American earthquakes: A summary of recent work: Seismological Research Letters, v. 68, no. 1, p. 128-153.
- Wald, D.J., and Allen, T.I., 2007, Topographic slope as a proxy for seismic site conditions and amplification: Bull. Seism. Soc. Am., v. 97, no. 5, p. 1379-1395.
- Wald, D.J., Quitoriano, V., Dengler, L., and Dewey, J.W., 1999a, Utilization of the Internet for rapid Community Intensity Maps: Seism. Res. Lett., v. 70, p. 680-697.
- Wald, D.J., Quitoriano, V., Heaton, T.H., and Kanamori, H., 1999b, Relationship between Peak Ground Acceleration, Peak Ground Velocity, and Modified Mercalli Intensity in California: Earthquake Spectra, v. 15, no. 3, p. 557-564.
- Wald, D.J., Quitoriano, V., Heaton, T.H., Kanamori, H., Scrivner, C.W., and Worden, B.C., 1999c, TriNet "ShakeMaps": Rapid generation of peak ground-motion and intensity maps for earthquakes in southern California: Earthquake Spectra, v. 15, no. 3, p. 537-556.
- Wald, D.J., Worden, B.C., Quitoriano, V., and Pankow, K.L., 2005, ShakeMap manual: technical manual, user's guide, and software guide: U.S. Geological Survey, 132 p.
- Youngs, R.R., Chiou, S.-J., Silva, W.J., and Humphrey, J.R., 1997, Strong ground motion attenuation relationships for subduction zone earthquakes: Seism. Res. Lett., v. 68, no. 1, p. 58-73.
Earthquake Scenarios describe the expected ground motions and effects of specific hypothetical large earthquakes. In planning and coordinating emergency response, utilities, emergency responders, and other agencies are best served by conducting training exercises based on realistic earthquake situations, ones that they are most likely to face. Scenario earthquakes can fill this role; they can be generated for any potential hypothetical future or past historic earthquake by the following steps.
First, assume a particular fault or fault segment will rupture over a certain length relying on consensus-based information about the potential behavior of the fault. For historic events, the actual rupture dimensions may be constrained based on existing observations or models. Second, estimate ground motions at all locations in a chosen region surrounding the causative fault.
These earthquake scenarios are not earthquake predictions. That is, no one knows in advance when or how large a future earthquake will be. However, if we make assumptions about the size and location of a hypothetical future earthquake, we can make a reasonable prediction of the effects of the assumed earthquake, particularly the way in which the ground will shake. This knowledge of the potential shaking effects is the main benefit of the earthquake scenario for planning and preparedness purposes.
In California, the California Geological Survey (CGS) and the U.S. Geological Survey (USGS) have evaluated the probabilistic hazard from active faults in the state as part of the Probabilistic Seismic Hazard Assessment For The State of California described by Peterson et al. (1996) and the National Seismic Hazard Mapping Project described by Frankel et al. (1996). From these maps it is possible to prioritize the best scenario earthquakes to be used in planning exercises by considering the most likely candidate earthquake fault first, followed by the next likely, and so on. Such an analysis is easily accomplished by hazard deaggregation, in which the contributions of individual earthquakes to the total seismic hazard their probability of occurrence and the severity of the ground motions are ranked in order. U sing the individual components ("deaggregations") of these hazard maps, a user can properly select the appropriate scenarios given their location, regional extent, and specific planning requirements. Currently, the ShakeMap scenario events come directly out of the CGS catalog of fault source parameters that make up the statewide probabilistic seismic hazard assessment.
Details of the choice of scenarios for the northern California Bay Region can be found here.
Users interested in specific scenarios for planning purposes are encouraged to make such a request by filling out a Comment Form.
Estimating Ground Motions for Scenario Earthquake ShakeMaps
At present, ground motions are estimated using an empirical attenuation relationship, which is a predictive relationship that allows the estimation of the peak ground motions at a given distance and for an assumed magnitude. Thus, ground motions are estimated for a given magnitude earthquake, and at a particular distance from the assumed fault, in a manner consistent with recordings of past earthquakes under similar conditions. For ShakeMap, we use the relationship of Boore et al. (1997) for peak and spectral acceleration, and we use Joyner and Boore's (1988) relationship for peak velocity. We use these predictive relationships to estimate peak ground motions on rock sites, and then correct the amplitude at that location based on the site soil conditions as we do in the general ShakeMap interpolation scheme. Site conditions come directly from the Statewide Site Conditions Map for California (Wills et al., 2000) and we correct for site amplification with the amplitude and frequency-dependent factors determined by Borcherdt (1994).
Attributes and Limitations of Current Maps
Our approach is simple and approximate. We account for fault finiteness by measuring the distance to the surface projection of the fault location (Joyner and Boore's distance definition), but we do not consider the direction of rupture nor do we modify the peak motions by a directivity term. With this approach, the location of the earthquake epicenter does not have any effect on the resulting ground motions; only the location and dimensions of the fault matter. If we were to add directivity to the calculations, than different choices of epicentral location would result in significantly different motions for the same magnitude earthquake and fault segment. Rather, our approach here is to show the average effect since it is difficult to show results for every possible epicentral location.
Our empirical predictive approach also only gives average peak ground motions values so it does not account for all the expected variability in motions, other than the aforementioned site amplification variations. Actual ground motions show significant variability for a given distance, magnitude, and site condition and, hence, the scenario ground motions are more uniform than would be expected for an actual earthquake. The true variations are partially attributable to 2D and 3D wave propagation, path effects (such as basin edge amplification and focusing), differences in motions among earthquakes of the same magnitude, and complex site effects not accounted for by our method.
Earthquake scenarios are used heavily in emergency response planning. Primary users for response planning include city, county, state and federal government agencies (e.g., the California Office of Emergency Services, FEMA, the Army Corp of Engineers), emergency response planners and managers for utilities, businesses, and other large organizations. Scenarios are also used for loss-estimation by utilities, governments, and industry.
Scenarios are of fundamental interest to the community and scientific audiences interested in the nature of the ground shaking likely experienced in past earthquakes as well as the possible effects due to rupture on known faults in the future.
In addition, more detailed and careful analysis of the ground motion time histories (seismograms) produced by such scenario earthquakes is highly beneficial for earthquake engineering considerations. Engineers require site-specific ground motions for detailed structural response analysis of existing structures and future structures designed around specified performance levels. In the future, with these scenarios we will also provide synthetic time histories of strong ground motions that include rupture directivity effects.
Future Scientific Advances
While current earthquake modeling techniques are sufficient for providing useful motion time histories and scenario ShakeMaps based on empirical means (e.g., ground motion attenuation relations), substantial improvement will require developing cost-effective numerical tools for proper theoretical inclusion of known complex ground motion effects. These efforts are underway and must continue in order to obtain site, basin and deeper crustal structure, to characterize and test 3D earth models (including attenuation and nonlinearity), and to improve numerical wave propagation methods to obtain useful, site-specific, ground motion time histories.