NCEDC News/Blog

Double Difference Locations Now Available

Categories:   Northern California Seismic System (NCSS)  |  Magnitude  |  Data Availability  |  Earthquake Catalogue  |  Catalogs  |  Earthquake Locations

June 2, 2014 

The NCEDC is pleased to announce the availability of double difference (DD) hypocenter solutions from 1984 to the present for earthquakes recorded by the Northern California Seismic System (NCSS). The NCSS catalog can be searched via http://ncedc.org/ncedc/catalog-search.html by selecting "NCSS Double-Difference Catalog (1984-Present)" from the "Input Catalog" drop-down box at the top of the web page. Generation, archiving, and searching of the DD catalog at the NCEDC is made possible through the efforts of Felix Waldhauser, Pete Lombard, Stephane Zuzlewski, Paul Friberg, and Allan Walter.

The NCSS double-difference catalog is generated by 2 separate processes. The "base" catalog was computed at the Lamont-Doherty Earth Observatory (LDEO) by Felix Waldhauser and David Schaff from a simultaneous re-analysis of 27 years (1984-2011) of digital waveforms available at the NCEDC. Waveform cross correlation (CC) and DD methods are used to improve the resolution in the existing Hypoinverse earthquake catalog by up to three orders of magnitude. The base catalog is based on 1.7 billion CC differential time measurements with correlation coefficients, Cf>=0.7 from all correlated pairs of events that are separated by less than 5 km (Schaff and Waldhauser, 2005). These data, combined with 11.4 million P arrival times from the NCSS catalog, were inverted for the precise relative locations of 512,059 events using the hypoDD algorithm (Waldhauser and Schaff, 2008). Information on the base catalog is available at http://www.ldeo.columbia.edu/~felixw/NCAeqDD.

The "real-time" catalog is computed automatically using the Real Time Double Difference (DD-RT) method (Waldhauser, 2009). The DD-RT process relocates new earthquakes relative to the DD base catalog by using differential times between the new event and its neighboring (reference) events in the base catalog. Information on DD-RT is available http://ddrt.ldeo.columbia.edu. The Northern California DD-RT catalog also continues to be computed in near real-time at LDEO and made available at http:/ddrt.ldeo.columbia.edu. While the base catalogs at the NCEDC and LDEO are identical, the DD-RT solutions computed by the NCSS and LDEO systems are not guaranteed to be identical due to differences in computing environments and when waveforms are retrieved by the two systems. Tests indicate that the differences between solutions computed by the two DD-RT systems are below the noise level, with standard deviations of a few tens of meters/several milliseconds.

Users should understand the following limitations and caveats when using the DD catalog.

  • The base catalog will be periodically updated, resulting in a complete replacement of the entire DD catalog. Information on updates will be posted to the NCEDC blog and on the Lamont DDRT server at http://ddrt.ldeo.columbia.edu when that occurs.
  • It is not possible to compute a DD solution for every earthquake location in the NCEDC catalog. Users requiring the most complete catalog should select the "USGS NCSN Catalog (1967 - Present)" instead of the DD catalog.
  • The magnitudes reported with the DD catalog are identical to those reported in the "USGS NCSN Catalog (1967 - Present)". For most cases the small differences in hypocenter coordinates resulting from the DD method are unlikely to appreciably change event magnitudes. However, for poorly constrained earthquakes (e.g., offshore earthquakes), the DD hypocenters can differ by several kilometers from the NCSN catalog, and in such cases the reported magnitude is unreliable.
  • DD solutions have not been computed prior to 1984, as there are no digital waveforms available. We are considering computing solutions from only P arrival time data before 1984.
  • DD solutions are not reviewed by data analysts. Consequently, the status is set to "A" (automatic), even though the Hypoinverse solution is marked as "F" (finalized).
  • For users selecting the "Hypoinverse" output format, note that the following hypocenter solution values have no corresponding values in the DD catalog
    • Number of P & S times with final weights greater than 0.1.
    • Maximum azimuthal gap, degrees.
    • Distance to nearest station (km).
    • RMS travel time residual.
    • Azimuth of largest principal error (deg E of N).
    • Dip of largest principal error (deg).
    • Size of largest principal error (km).
    • Azimuth of intermediate principal error.
    • Dip of intermediate principal error.
    • Size of intermediate principal error (km).
    • Size of smallest principal error (km).
    • Number of S times with weights greater than 0.1.
  • The following error estimates are available:
    • Bootstrap relative location errors at the 95% confidence level (base catalog).
    • Formal least-squares errors at the 95% confidence level (real-time catalog).

References

Schaff, D.P. and F. Waldhauser, Waveform Cross-Correlation-Based Differential Travel-Time Measurements at the Northern California Seismic Network, Bull. Seism. Soc. Am., 95, 2446-2461, 2005.
Waldhauser, F. and D.P. Schaff, Large-scale relocation of two decades of Northern California seismicity using cross-correlation and double-difference methods, J. Geophys. Res.,113, B08311, doi:10.1029/2007JB005479, 2008.
Waldhauser, F., Near-real-time double-difference event location using long-term seismic archives, with application to Northern California, Bull. Seism. Soc. Am., 99, 2736-2848, doi:10.1785/0120080294, 2009.

The NCEDC is a joint project of the University of California, Berkeley and the USGS, with facilities located at the University of California, Berkeley. Please send us email at ncedcinfo {AT} ncedc {DOT} org or fill out our comments form if you have questions or comments about these changes.

NCEDC News

Categories

View All Posts By Category

View Posts By Date