Earthquakes and earthquake faults are an important fact of life in Southern California. Though using GIS in conjunction with GPS and seismological data to study earthquake activity is not new, using ArcView GIS with 3D Analyst extension can give these data a new twist. The author used publicly available information obtained off the Internet and published maps to create an interesting, informative project for a GIS class she took at San Bernardino Valley College in San Bernardino, California.

Sources of Seismic Data

California has more than 200 known faults capable of producing earthquakes of magnitude 6.0 or greater. Some of these faults are hidden beneath the sediments in valleys, some are visible at the surface. The biggest concentration of faults is in the valley and mountainous region surrounding the San Andreas Fault zone and the San Jacinto Fault zone. There are hundreds of earthquakes every year along these faults though most are too small to feel.

The biggest concentration of faults is in the valley and mouatainous region surrounding the San Andreas fault zone and the San Jacinto fault zone.

Because Southern California is such a seismically active area, the United States Geological Survey (USGS), the California Division of Mines and Geology, and California Institute of Technology (Caltech) in cooperation with other agencies, corporate partners, and institutions have developed the Southern California Earthquake Center (SCEC) and TriNet Seismic Network. This organization has a number of seismographs strategically placed around Southern California to monitor earthquake activity.

These seismographs are linked to a computer system using a combination of GPS technology and landlines. Seismography records are sent to central computers within a few seconds or minutes of an event. The seismic information is provided to the public via the Internet. Visit the ArcUser Jump Station at the ArcUser Web site for a list of these sites. Data at these sites, which are updated frequently during the day, are saved to catalogs that are searchable by various parameters.

Defining the Scope of the Project

After deciding to create a project using ArcView GIS and 3D Analyst to study earthquakes and faults in Southern California and identifying data sources, the next step was to determine the project's objective and scope. Once the pertinent data was obtained and converted to a usable form, maps could be created and analysis performed.

The project's objective was to ascertain whether there is a pattern to the earthquakes in the inland valleys and mountains of Southern California. The project hoped to provide answers for the following questions. Are there earthquakes occurring in areas where no faults are mapped? If so, do these earthquakes create a pattern suggesting more faulting not yet mapped? Is there a pattern to the depths of earthquakes?

The study area needed to be small enough to make the project manageable. The area chosen stretches from 116.70 degrees to 117.83 degrees west longitude, and 33.80 degrees to 34.56 degrees north latitude which is an area approximately 67 miles by 50 miles. The time frame selected was January 1, 1995, to June 30, 1998. The project would deal with data on earthquakes of magnitude 2.0 and greater.

ArcView GIS 3.0a, 3D Analyst, and the beta version of ArcView GIS 3.1 were used to create the project. USGS maps and information downloaded from the Caltech and SCEC Internet sites provided the fault data. The searchable Internet sites of SCEC were used for the basic earthquake data. The geographic data that comes with ArcView GIS was used to create the base map.

Collecting and Converting Data

The SCEC Data Center Earthquake and Hypocenter and Phase Database was queried for data on the study area. Available search parameters available included start and end dates, minimum and maximum magnitudes, minimum and maximum depths in kilometers (km), and latitude and longitude ranges. The results were printed out. A second site, ftp/ca.earthquakes, a cooperative effort of Caltech and the USGS that is maintained by Kate Hutton and Lucy Jones, was used for additional data.

Resolving Data Issues

These two Internet sites used different units of measurement to store data. The SCEC site uses Universal Time for time measurements, has latitude and longitude expressed as a decimal degree to two decimal places, and includes depth information not available from the Caltech-USGS site.

The Caltech-USGS site uses local time (either Pacific Standard or Daylight Time) and has latitude and longitude data expressed in degrees and decimal minutes. Though this site did not have depth information, it did have other descriptive information as to location (i.e. 3 miles NW of San Bernardino) and whether the earthquake was felt by residents.

Time measurements from the SCEC site were converted to Pacific Standard Time for use with the Caltech-USGS data. Latitude and longitude measurements in degrees and decimal minutes from the Caltech-USGS site were converted and expressed as measurements in decimal degree to three decimal places. Additional data, whether depth or descriptive data, were added to the project database.

Microsoft Excel was used to create an earthquake data table. The tabular data was saved as a dBASE IV file usable by ArcView GIS. The table was added to an ArcView GIS project and then used to create an event theme. The theme legend editor was used to classify the earthquakes by magnitude using graduated symbols. Each size symbol was then given a different color for easier identification. The location of earthquakes was double-checked against the descriptive location for accuracy and any errors were corrected.

Constructing the Base Map

ArcView GIS comes with a large amount of geographic data that can be used to construct a base map. The highway and street data (highways and dtl_st files) were used. Since the data covers the entire United States, data were limited to the project study area. The select button was used to select Riverside, San Bernardino, Los Angeles and Orange Counties. These selections were converted to a shapefile. Additional editing to the shapefile was performed using the polygon splitting function. A combination of the select button, conversion to shapefile, and line splitting functions was used to edit the highway theme.

FInding Faults

There was not single source for the fault data. A combination of data from several Internet sites, a California Division of Mines map, and a USGS geologic map of California were used to determine the locations of faults. Using the line drawing tool, new a new line theme was created. As each line was drawn, the name of the fault was added to the table created. Once drawn, the fault lines were fine-tuned with the vertex editing function. Highways and county boundaries were used as reference points for correct line placement. None of the maps completely agreed with each other, so an average was used where there were discrepancies. Faults were not included if they were very short and found only on one map. Some of the minor San Bernardino Mountain faults were not drawn.

A triangulated irregular network (TIN) created from the earthquake point theme. Faults and earthquake epicenters are visible on the surface. Areas in yellows and reds indicate shallower epicenter depths. Blue areas indicate deeper epicenters.

Creating a Contour, TIN, and Three-Dimensional Scene

Once the map was completed, a three-dimensional theme and scene could be created. The earthquake theme was used to create a three-dimensional point theme with depth as the z value. This was used with the fault theme to create a three-dimensional scene. Since the depths range from zero to 20.8 km, the resultant scene looked like a string of beads hanging down from the surface. A vertical exaggeration of 0.1, set by adjusting the z factor in the three-dimensional theme properties dialog, compressed the points into a more intelligible form.

In order to make a triangular irregular network (TIN) theme, the themes from the first view were copied into a new second view. A breakline theme, created using the make new theme choice, was created parallel to and on each side of an observed pattern of earthquakes perpendicular to the San Jacinto fault. Using the earthquake point theme and breakline themes, an inverse distance weighting (IDW) contour theme was created. The contour theme was used to create a TIN theme. This TIN was added to the three-dimensional scene and compressed to 0.1 vertical exaggeration. The addition of this TIN caused the scene to rotate very slowly. As the three-dimensional scene was spinning, a pattern became apparent.

Analyzing the Map and Data

The combination of map, TIN and three-dimensional scene revealed a pattern of earthquakes in the San Bernardino Valley in an area where faults have not been mapped. There appears to be a pattern of quakes parallel to the San Jose and Cucamonga faults and perpendicular to the San Jacinto and San Andreas faults.

One map shows a barrier near here. There is a pattern of northwest trending quakes near the San Jacinto and San Andreas faults and another pattern of earthquakes perpendicular to these faults. At the juncture of the San Jacinto fault and the barrier earthquake pattern, there is a large cluster of earthquakes ranging in depth from just below the surface to 20 km deep.

From this data there appears to be at least one fault not shown on maps. There is also a pattern to the depths of earthquakes with the deepest ones occur in the Banning Pass area where several faults come together and the shallowest ones located in the mountains.

Sources of Additional Information

Go to the ArcUser Jump Site for the URLs of many excellent Web sites on related topics. For more information, contact Margaret Gooding at mgooding@gistech.com.

About the Author

Margaret Gooding graduated from California State University at Fullerton with a bachelor's degree in Geology and a minor in Art. She earned a GIS certificate at San Bernardino Valley College and completed an internship at Esri in July of 1998.