Southern California Earthquake Data Center

About

Data Center FAQ

Data

Faults

Stations

Website

General SCEDC questions



Data Questions

Q: What data is archived by the SCEDC?
A: The Data Center is the source of the 1932-present Caltech/USGS earthquake catalog, the most complete archive of seismic data for any region in the United States. The SCEDC archives seismic waveform data from many sources, but primarily from the SCSN/Trinet network and its Southern California predecessors. Most of the waveform data can be categorized as passive source seismic data collected by broad-band, strong-motion and analog instruments.

 

We also archive and distribute data for both LARSE I and II seismic survey, data sets from the portable deployments following the Landers and Northridge earthquakes, and data from the Anza network and SCEC borehole stations. Our collection of non-seismic data includes "survey-mode" precise GPS measurements made in southern California by various universities. The GPS data archive consists of: raw GPS data, RINEX files, indices to the RINEX files, log sheets, and site descriptions.

 

Q: What is the difference between the information in the database and "The Catalog?

A: The Caltech/USGS catalog contains hypocenter information for 1932 through the present. This catalog is available in many formats and contains information such as time, location, depth, magnitude and quality of the solution; the catalog is a summary of the parameters of the event. The information contained in the Oracle database is the full suite of parametric information available, including phases, amplitudes, all origins and magnitudes and pointers to the waveforms for an event.

 

In our current system, all catalog information is derived from the information contained in the database; if an event's magnitude is changed in the database, it is also changed in the catalog. In the past, the supporting parametric data and the catalog existed as separate entities. The problem with this system occurred when the catalog or the supporting data was changed in one place and not in the other -- the underlying data becomes separated from the catalog version of the event. This problem was compounded by the existence of multiple instances of the catalog.

 

Q: How do I tell what data is available from the SCEDC?

A: The best way to check on waveform data availability is to use STP. STP offers several query modes about waveform data availability as well as data about stations, channels and events. An IRIS on-line database access tool, SeismiQuery, will be implemented at the SCEDC in early 2004.

 

If you're interested in searching for basic event information such as date, time, location, depth and magnitude, this information is available via our catalog search page.

 

Q: Is it possible to obtain non-event data (background signal)?
A: The SCEDC continuous-data archives consist of TriNet digital broadband stations from late 1999 to present. For a list of stations available in the archive for a given time, connect to the stand-alone version of STP and type "HELP STA" at the "STP>" prompt. From the web version of STP, clicking on the "Stations" button will bring up a list of stations that were active as of "Time On" (after specifying the time window).

 

Q: What networks does the SCEDC have data for?

A: The SCEDC has data from the following networks:

 

AZ  ANZA Regional Network
BK  Berkeley Digital Seismic Network (BDSN)
CE  California Strong Motion Instrumentation Program
CI  Caltech/USGS Regional Seismic Network (SCSN/TriNet)
GO  USGS Golden Network LA Los Angeles Basin Seismic Network
LI  Laser Interferometer Gravitational-Wave experiment (LIGO)
MX  Mexican National Seismic Network
NC  USGS Northern California Regional Network
NN  Western Great Basin/Eastern Sierra Nevada
NP  United States National Strong-Motion Network
NR  NARS Array SN Southern Great Basin Network
UG  Unocal Geothermal Northern California Network
WR  California Division of Water Resources
ZY  Various Portable Networks in Southern California

 

Historic networks have been incorporated into currently operating networks. For instance, TERRAscope (TS) stations are now under the CI network.

 

Q: What format and on what medium is the data stored?

A: All waveform data is archived in mSEED format on mirrored network appliances with high-capacity (1.6 terabyte) magnetic-disk RAIDs. These systems are built with PC-technology components, using 16 120-Gbyte IDE disks, hot-swappable disk trays, two RAID controllers, dual redundant power supplies and a Linux operating system. The system is configured over a private gigabit network that connects to the two Data Center servers and spans between the Seismological Lab and the USGS. To ensure data integrity, each RAID disk system constantly checks itself against its twin and verifies file integrity using 128-bit MD5 file checksums that are stored separate from the system. The final level of data protection is a Sony AIT-3 tape backup of the files.

 

The parametric data is archived on two identical Oracle 9i replicated databases. These instances run on separate computers in different locations in order to provide backup in case of a computer failure and to allow downtime during maintenance without interrupting user access. The databases are kept synchronized through two-way replication which update every 4 seconds. Data from the real-time acquisition system are pushed to the SCEDC database via one-way snapshot replication so data is available in near real-time.

 

Q: How do I find what stations were recording data during a specific event?

A: You would use the EAVAIL command in STP to list the available triggered waveform archive entries for that event by its event ID.

 

Q: How do I retrieve data from stations that are a certain distance from the epicenter of an event?

A: This is currently not possible, but we are working to make this kind of search possible in the future.

 

Q: What is a location code?

A: The location code is a two character code that, when used in conjunction with the other data specifiers, uniquely identifies a data stream. Location code is used to distinguish between multiple sensors with identical station and SEED channel names. The use of location codes enables the distinction between multiple same-name channels from different, co-located sensors at one site.

 

Q: How do I know if I'm running the latest version of the STP client?

A: A message saying "You are currently running the latest version of STP" will appear when you start STP. If you are not running the latest version, you will get a message telling you that your version of STP is out of date and instructions on how to download the latest version.

 

Q: Why do some of the events have 8-digit event IDs, but most of them have 7-digit event IDs?

A: For current data, the number of digits has to do with which of the real-time servers generated the event ID. There are two real-time servers that mirror one another, so that if the one designated as primary goes down or needs maintenance, the other machine running in parallel can assume the primary role quickly and easily.

 

Q: How long does it take for an event to be "Reviewed by a Seismologist"?

A: There's no hard and fast rule about when the most recent data will have been checked. In times of low seismicity, analysts will have reviewed a day's automated data by the end of that workday. Mondays, or the first workday after an extended holiday, will often be an exception, since the automated system will have generated a few days' backlog to analyze. Swarms, major quakes, and other increases in seismicity will also slow the process down.

 

Analysis by a human will weed out junk "events" such as microwave glitches the automated system has mistaken for a quake; make minor adjustments to magnitude and location; and the like. Some events are checked more than once--larger quakes (generally, those greater than M3.0 and/or those that have been felt by the public) typically are given a cursory review by the duty seismologist to verify that the data look credible; later, the event undergoes more thorough analysis.

 

As long as there have been no major quakes in southern California, you're usually pretty safe in assuming that week-old data have been checked. After a major quake, though, all bets are off! The analysts will typically pick out the events of greatest magnitude, saving smaller quakes (e.g., less than M3.0 or M4.0) for when things calm down.

 

Q: Why do you have duplicate Catalog entries for some events? Why do you show two events in your catalog for the San Simeon event? Which one is correct, and which do you consider to be the SCEDC official San Simeon earthquake? Why is one listed as a teleseism, and one as a regional?

 

From STP:

Event ID Type T0 Lat Long Depth Mag Mag Type

9966485 ts 2003/12/22,19:15:56.000 35.7000 -121.1000 8.00 6.50 w 0.0 9966449 re 2003/12/22,19:15:56.760 35.7085 -121.1043 7.01 6.54 l 0.5

# Number of events= 2

 

A: This is our standard operating procedure for large events. The San Simeon earthquake occurred in northern California, so it is considered a regional event for us in southern California. The magnitude/location solution that was calculated using our methodology at Caltech is the regional (re) solution and the teleseismic (ts) solution was supplied by the National Earthquake Information Center (NEIC). The authoritative solution for this event is the ts version which was provided to the NEIC by the Northern California Seismic Network and is available as a local event in the NCEDC catalog.

 

The waveforms associated with the ts version have a much longer time window than we typically archive for local events. For teleseismic events a distance-magnitude window is used to determine how much data to save. We archive 2000 seconds of waveform data for a M>=6 event at 30 degrees and 3000 seconds for M>=7 anywhere. For larger local events (M > 3.5), the entire array is recorded for ~200 seconds.

 

We would also have two versions if it were in our authoritative region e.g., the Northridge event has our local solution as well as a teleseismic version.

 

Q: What is a Location Code? How does the SCEDC use it?

A: The SCEDC uniquely describes the seismograms we archive and distribute using the FDSN (Federation of Digital Seismic Network) system, which includes the following four fields:

  • Network (2 characters)
  • Station (3-5 characters)
  • SEED Channel (3 characters; see SEED Reference Manual Appendix A)
  • Location Code (2 characters)

e.g., NN.SSS.CCC.LL

 

The FDSN standard has always included the Location Code field, but it was not used by the SCSN until recently. The SCEDC will now use the Location Code field to uniquely identify SCSN data streams.

 

Location Code is used to distinguish between multiple seismograms with identical station and channel names. For example, a station equipped with both STS-1 and STS-2 broadband high gain seismometers would produce two data streams with the same net.sta.cha identification. Also, the SCSN uses orientation codes of [1,2,3] for data channels from downhole sensors and [Z,N,E] for traditionally-oriented surface channels. Without Location Codes, we cannot have multiple downhole sensor packages without changing the station or channel names on the second to n-th downhole sensors.

 

Currently, the default value for SCSN Location Codes is blank i.e., that field contains two blank spaces. For instances where a different Location Code is necessary to uniquely describe a data stream, the SCSN will follow the SEED convention of allowed characters (A-Z, 0-9, space) and identify streams with "01" for the first non-unique stream, "02" for the second, etc. Users (or their software) should not assume that Location Codes have a meaning; the SCSN will not use this field to encode information like emplacement depth, preferred channel, sensor type, orientation etc. However, the full SCNL description can be used as a unique key into complete descriptive information about the characteristics of the data stream.

 

SCEDC users should be aware that if you do not specify a Location Code in your data request, the Data Center will provide all seismograms that match that net.sta.channel, so you may receive multiple seismograms where you only expect one. For ASCII output where Location Code is a whitespace-delimited field, a blank-blank Location Code will be assigned "--" and when parsing ASCII input, "--" should be interpreted as 2 blanks.

 

The naming convention for seismograms will be to only include the Location Code in the filename if it is something other than the default value of blank-blank:

 

Triggered Waveforms
Now: 14176696.CI.USB.HLE.sac
Future: 14176696.CI.USB.HLE.sac and 14176696.CI.USB.HLE.01.sac (if there is an 01 Loc Code)

 

Continuous Waveforms
Now: 20050831000000.CI.USB.HLE.sac
Future: 20050831000000.CI.USB.HLE.sac and 20050831000000.CI.USB.HLE.01.sac (if there is an 01 Loc Code)

 

top


Fault Questions

Q: Why don't you put the faults on which the earthquake occurred on the recent earthquakes reports and index maps?

A: For most of the earthquakes you see on the map every day--those that are too small to cause surface rupture--we don't know what faults are involved. Many earthquakes occur on small, unnamed subsidiary faults located near larger, well-known faults. When we locate an earthquake, there is always uncertainty (a hundred meters or more) about the exact location of its hypocenter (the point within the earth where an earthquake rupture starts). The epicenter, the projection of the hypocenter to the surface, not only has this same uncertainty, but it has very little relevance--earthquakes happen many kilometers beneath the surface, and along surfaces, not just at points.

 

Our automated text reports used to include a line that gave the distance between a quake's hypocenter and the nearest fault. This wasn't because we were drawing a correlation between the two, but rather, for reference (around here, we tend to have the southern California fault map memorized, like many people know freeway layouts). However, we eliminated that line because people erroneously concluded that the fault mentioned was the fault on which the earthquake had occurred.

 

Q: How can I obtain Southern California GIS fault information?

A: The GIS fault information available through our site is an adaptation of our clickable fault map, created by The Redlands Institute, University of Redlands. This is available here or through the anonymous ftp server scec.gps.caltech.edu at: /pub/maps/redlands.

 

top


Station Questions

Q: How do I get station response information?

A: Station response information is available in dataless SEED format at: www.data.scec.org/ftp/stations. You can use STP's CHAN command with the -l flag to retrieve a long listing of channel information which includes simple response information. For a quick list of channels and other information, consult the station list at: http://www.data.scec.org/cgi-bin/stationdb/dig_responselist.cgi

 

Q: What stations have borehole data?

A: Southern California borehole stations include the following: 400K CFS DJJB GR2 JAB JGB LBW1 LBW2 MIS RSB RSS USB WNS For channels and other information, consult the station list at http://www.data.scec.org/cgi-bin/stationdb/dig_responselist.cgi

 

top


Website Quesitons

Q: Why did you change your website?

A: In redesigning the SCEDC website, we made an effort to:

  • Improve the visual appeal of the website. Improve navigation of the site to allow users to get to our most popular webpages with the fewest mouse-clicks.
  • Increase the amount of useful documentation describing data we provide, methods used, metadata and information on accessing the data and tools we provide.
  • Improve our communication and outreach in spite of our limited human resources, as well as solicit input from the user community on how to improve our operation.
  • Collect better metrics on Data Center usage and publish these statistics on a regular basis.

Please send your questions, comments, suggestions or any other feedback on the new website to:

 

Q: Can we reproduce photos or images from your site?

A: You're welcome to use our material, provided (a) you give us credit for it, and (b) you're not using it for commercial purposes. Where possible, we prefer that people link to our site, provided it's clear to the user that (s)he is leaving your site and entering ours.

 

If these are images/graphs/maps we created, you may use them with proper crediting. However, if you're interested in photographs, it is necessary for you to get permission from the photographers. Photographer credits appear after the photo captions. We cannot grant permission to download the photos because we don't hold the rights to them -- we obtained permission to use the photos years ago. For photographs, you might want to try contacting the Earthquake Engineering Research Institute, which sells collections of earthquake-damage images. See samples of their Northridge set.

 

top


Station Questions

Q: How do I get station response information?

A: Station response information is available in dataless SEED format. You can use STP's CHAN command with the -l flag to retrieve a long listing of channel information which includes simple response information. For a quick list of channels and other information, consult the station list.

 

Q: What stations have borehole data?

A: Southern California borehole stations include the following: 400K CFS DJJB GR2 JAB JGB LBW1 LBW2 MIS RSB RSS USB WNS For channels and other information, consult the station.

 

top


Website Quesitons

Q: Why did you change your website?

In redesigning the SCEDC website, we made an effort to:

  • Improve the visual appeal of the website.
  • Improve navigation of the site to allow users to get to our most popular webpages with the fewest mouse-clicks.
  • Increase the amount of useful documentation describing data we provide, methods used, metadata and information on accessing the data and tools we provide.
  • Improve our communication and outreach in spite of our limited human resources, as well as solicit input from the user community on how to improve our operation.
  • Collect better metrics on Data Center usage and publish these statistics on a regular basis. Please send your questions, comments, suggestions or any other feedback on the new website to: .

 

Q: Can we reproduce photos or images from your site?

A: You're welcome to use our material, provided (a) you give us credit for it, and (b) you're not using it for commercial purposes. Where possible, we prefer that people link to our site, provided it's clear to the user that (s)he is leaving your site and entering ours. If these are images/graphs/maps we created, you may use them with proper crediting. However, if you're interested in photographs, it is necessary for you to get permission from the photographers. Photographer credits appear after the photo captions. We cannot grant permission to download the photos because we don't hold the rights to them -- we obtained permission to use the photos years ago. For photographs, you might want to try contacting the Earthquake Engineering Research Institute, which sells collections of earthquake-damage images. See samples of their Northridge set.

 

top


General SCEDC Questions

Q: What is the difference between the SCEDC and SCEC?

A: The Southern California Earthquake Data Center (SCEDC) operates at the Seismological Laboratory at Caltech and is the primary archive of earthquake data for southern California. The Southern California Earthquake Center (SCEC) is headquartered at the University of Southern California and is a community of scientists and specialists who actively coordinate research on earthquake hazards. The Data Center is a central resource of SCEC and is supported by SCEC, but we are a separate entity. It is confusing, particularly because our names are very similar (the difference is the data!), and our URL www.data.scec.org is within the www.scec.org domain.

 

Q: What happens when a major event happens during non-business hours?

A: The Duty Seismologist reviews larger quakes shortly after they happen; she/he makes sure the automated location and magnitude are credible. This involves a scan of waveform data from the seismic stations nearest the quake's epicenter, and typically takes only a few minutes more. (You can view waveform data similar to what the duty seismologist uses by clicking the "waveforms" link near the bottom of the event report.)

 

Subsequently, an analyst will look over all the data that were captured for the event. The waveform data come from hundreds of stations that each have multiple channels (e.g., north-south, east-west, up-down). For a quake of substantial magnitude, full analysis can take over an hour. Some larger quakes will be reviewed more than once and/or by multiple analysts. Sometimes, the magnitude calculated after analysis goes up. Sometimes it goes down. Interestingly, nobody has ever written to ask why a magnitude calculation has gone up. When one goes down, though, people inevitably wonder why.

 

Q: Can you email me when there is an earthquake?

A: We don't have an email notification service, but the US Geological Survey offers one here.

 

Q: Will my home survive an earthquake?

A: The Southern California Earthquake Data Center does not provide safety evaluations. We are not able to provide specific and personalized recommendations for home safety because we do not have expertise in this area. If you want a professional opinion regarding site-specific questions, you'll need to confer with an architect or engineering geologist.

 

top