Earthquake Probability Mapping Help

This web site will generate a map of earthquake probabilities for the region surrounding a site that you specify. We call these EPMs, for "earthquake probability maps." Besides the site location, you control two other input parameters. First, you enter the minimum magnitude for sources to be considered. Second, you enter the time span in years. The website will then make a map for you.

There are two source categories in these maps. These are subduction sources and "everything else." Subduction sources are of specific interest because of their potential to produce tsunamis in many (but not all) instances. Whether you get a map of probability of subduction earthquake depends on the minimum magnitude you enter. Many regions do not have subduction events, and for these, you may get a gray map (no data) if you enter a minimum magnitude that flags the computer to look for subduction only.

Time Span

Any positive time interval, T (units: years) may be entered. However, the time interval you enter should be related to the relative rate of seismicity of the region you are looking at. Typically, for the central and eastern U.S. (CEUS), T on the order of several hundred or even a few thousand years make sense to get a significant probability in your EPM. However, in much of the western U.S., time intervals of a few tens of years often make sense. The most seismically active places in the U.S. are southern Alaska, eastern Hawaii (i.e., the Big Island), and the western third of California. Time intervals like 10 to 30 years often work well when making EPMs for these regions. For the mid-west time intervals of a few hundred years often work well. T = 30 to 50 years is often a good choice for Washington and Oregon EPMs. These are only suggestions. Feel free to experiment. Generally the smaller you choose M, the smaller you can choose T and get a satisfactory map.

Minimum Magnitude, M

Only certain minimum magnitudes may be entered on the drop-down menu. These are 5.0, 6.0, 6.5, 7.0, 7.5, 8.0, and 9.0. However, not all of them are appropriate for any given region. At this time, you should only enter M 8 or 9 for sites in southern Alaska and the Aleutian Islands, in the Pacific Northwest, or in northern California. Entering M 8 or M 9 will produce a map of just subduction-earthquake probability.

M is defined as moment magnitude in these EPMs. For the CEUS, the minimum magnitude that the USGS PSHA uses is m_b 5.0 (body-wave magnitude). This m_b corresponds to a moment magnitude of about 4.5 to 4.8. Therefore, if you enter minimum magnitude 5 for sites in the CEUS, your EPM will be labeled with a lower (moment) magnitude M of 4.75.

Maximum Distance

If a subduction-source map is produced, the maximum distance used for including the source in the probability calculations is automatically set to 200 km. For all other maps, the maximum distance is automatically set to 50 km. These distance limits may or may not be suitable for your needs. We compute a site-specific earthquake probability table that gives detailed analysis for sources within any maximum distance that you specify. This table should be available at this web site at a future time.

The following is region-specific detail that may be helpful.

Conterminous U.S. (48 states)

For most sites in the conterminous U.S., you should generally ask for EPMs of non-subduction sources. Cascadia subduction sources are not included when you enter minimum magnitude, M, less than 8.0. Generally, you should only ask for M 8.0 or M 9.0 EPMs when you are looking at sites in the Pacific Northwest or northern California. In the USGS model, any Cascadia source with magnitude less than 8 is treated as a crustal source (onshore or offshore). Valid M choices for preparing EPMs of crustal and Benioff sources in this region are 5.0, 6.0, 6.5, 7.0, and 7.5.

To obtain EPMs for large Cascadia subduction events, enter M 8.0 or 9.0. The code is not set up at this time to make EPMs for crustal-event sources with M greater than or equal to 8, even though such a map might be of interest to some (it would show positive probability for the San Andreas fault region, under certain rupture scenarios).  The information used to compute earthquake probabilities is contained in the 2002 update to the USGS probabilistic seismic hazard mapping project, documented in Open-File Report 02-420 by A. Frankel and others.

If you enter M = 9.0 on the drop-down menu, you will get a map showing positive probability at sites that are within 200 km of one or more of the M9 Cascadia megathrust scenarios (there are 4 such scenarios, discussed in OFR 02-420). The maximum probability is derived from an estimated annual rate of M9 megathrust events equal to 0.002, i.e., 1 such event per 500 years on average, which is the rate used in the USGS PSHA for this source. The M9 source occurs on an epistemic branch with weight 0.5, the alternate branch being Cascadia M8.3 earthquakes. In general, when computing seismic hazard (and earthquake probabilities) we multiply rates times epistemic weights; however, in this special instance (M9) we compute probability with the source rate unreduced by the branch weight. Selecting M9 on the drop-down menu implies that you choose the M9 branch or branches of the model (i.e., eliminate magnitude uncertainty from Cascadia megathrust source description), which implicitly gives these scenarios a combined weight of 1. This is the only instance of EPM probability estimates that alters the weight for a source from that which appears in the USGS 2002 PSHA.

Alaska

For Alaska, subduction sources will be included in the EPM if they are within 50 km of the site. Cities like Anchorage, Seward, and Valdez are within 50 km of subduction sources. M7.5+ subduction sources occur with relatively high frequency in this region. If you want to see EPMs for subduction sources in southern Alaska, and if you wish to omit crustal sources, enter M of 7.5, 8, or 9 on the drop-down menu. These choices will accumulate rates of subduction events in the model that fit the criteria and that are within 200 km of each site on the map. When you select M less than 7.5 on the drop-down menu, the Alaska EPMs do not exclude subduction sources from the probability calculations, unlike the conterminous US EPMs.

Alaska earthquake data used in these EPMs are those of the USGS probabilistic seismic hazard maps, documented in Open-file Report 99-36, by R. Wesson and others.  The maximum magnitude of Alaska's PSHA sources is M 9.2, approximately the size of the great Prince William Sound earthquake of March 27, 1964. For Alaska EPMs, the M9+ scenario rates are multiplied by their respective epistemic weights when estimating probability of source (c.f., Cascadia megathrust treatment, discussed above).

Hawaii

For Hawaii, there are no subduction sources within 200 km of any onshore site. Therefore, you should not enter minimum magnitude greater than 7.0 for Hawaii-region EPMs. Hawaii earthquakes are related to volcanoes, with the most frequent damaging earthquakes spatially associated with the most active volcanoes, which are in the southern part of the Big Island, and offshore at the Loihi seamount. Good background information on the earthquake hazard of Hawaii is available at the Hawaii Volcano Observatory web site. The Hawaii earthquake data used in these EPMs is that of the USGS probabilistic seismic hazard maps.

Puerto Rico and U.S. Virgin Islands

For Puerto Rico, you should enter M 7.5 if you want to see only subduction source EPMs. Atlantic plate subduction zones are more than 50 km from land (or onshore) sites in Puerto Rico, and they will automatically be excluded if you enter M less than 7.5. You should not enter M8 or M9 for Puerto Rico as these magnitudes are greater than those of the PSHA model. Tsunami-producing earthquakes are of interest to many Puerto Ricans. In addition to subduction sources, there are several offshore earthquake zones that may produce tsunamis, especially the Mona Passage zone northwest and west of the island. Such sources contribute to the probabilities that you will see if you enter M6.5 or M7 on the drop-down menu.

The Puerto Rico and Virgin Island regional earthquake data used in these EPMs is that of the USGS probabilistic seismic hazard maps. See Documentation for 2003 USGS Seismic Hazard Maps for Puerto Rico and the U. S. Virgin Islands by C. S. Mueller, A. D. Frankel, M. D. Petersen, and E. V. Leyendecker. These source calculations are not meant to extend beyond the U.S. Virgin Islands.

Historical Seismicity

Earthquakes that have been recorded instrumentally or whose epicenters have been determined from newspaper accounts and similar archive material provide important information for estimating future earthquake probabilities. Earthquake catalogs have been searched and earthquakes with magnitude greater than or equal to 6.0 in the western U.S. or greater than or equal to 5.0 (body-wave magnitude) in the CEUS are plotted in these EPMs as small black circles. (Small blue circles are used for deep earthquakes in Puerto Rico, black for shallow earthquakes.) These epicenters should be current through about 2004. There is no attempt to achieve completeness or best epicenter estimates in data shown with the EPMs, just a general idea of known seismicity in different regions.

Earthquake Probability

The earthquake probability that is plotted in the EPM is associated with the mean rate estimate for each seismic source. A companion article at this web site briefly discusses uncertainty in mean rate and in other earthquake parameters (size and location).

The color scale that is shown in the legend on each EPM is scaled to the maximum probability that was found in the interior of the map. These color scales change from map to map. If you are going to compare two maps side by side, it might make sense to use a common color scale. This feature may be added to this web site at a future time if we learn of a demand.

Comments and Questions

This web site was initially developed in 2004 and released to public in early 2005. We encourage comments and suggestions about how to improve the presentation of earthquake probability data. We will do our best to answer your questions as well.