SSA 2015 - Nanometrics Event Schedule - Booth #7


SSA 2015 Poster Paper Presentations
  • "Eartquake Frequency -  Magnitude Distribution and Fractal Dimension in Northern California"

              Wednesday 22 April, Poster # 104 / Exhibit Hall A     

  • "Investigating the Relationship Between Velocity Model Caomplexity and Location Accuracy"

             Thursday 23 April, Poster # 95 / Exhibit Hall A


EGU 2015 Poster Presentation
Not going to the SSA this year?  Please click here to request a copy of the poster papers and case study.


Take a look at our new Technical Resources Site, for papers, publications and more!


Abstract: Earthquake Frequency- Magnitude Dristribution and Fractal Dimension in Northern California

Sepideh Karimi,  Dario Baturan,  Wesley Greig and Andrew Law, Nanometrics Inc, Ottawa, Canada

The aim of this study is to 1) show how the heterogeneity of seismogenic volumes can lead to differences in the fractal dimension (D-value) and the slope of the frequency – magnitude relationship (b-value) and 2) demonstrate the correlation in the temporal variation of these two parameters and the time of large magnitude earthquakes. Local distribution of D and b values, as well as their ratio, reveal information about geological complexity and changes in the stress level in the region of study.

The earthquake catalog of Northern California reported by U.S. Geological Survey is used for this investigation with a focus on recent seismicity. The dataset is recorded by the Northern California Seismic Network (NCSN) which operates 580 stations in northern and central California and acquires data from an additional 159 stations maintained by other institutions. Using the maximum curvature method (Wiemer and Wyss, 2000), we calculate the magnitude of completeness at each grid point in the region of interest and find that it varies from M 0.3 to M > 1.0. Utilizing the correlation integral and spatial distribution of events, we determine earthquake sub-clusters. Along with the analysis of spatial variation of the b and D values, we also evaluate temporal changes of these two parameters and their ratio for different earthquake clusters.  Strong correlation is observed between fluctuation in b and D values, the time of the largest magnitude event and the number of events in successive time windows at each cluster.

The outcome of this research is in good agreement with other published studies for this area (eg; Wyss et al 2004, Okubo & Aki, 1987). Our results confirm that monitoring variations in b and D values as well as their ratio in time and space can be used to effectively identify earthquake clusters and also as a potential large earthquake precursor.


Abstract: Investigation the Relationship between Velocity Model Complexity and Earthquake Location Accuracy

Andrew Law,  Dario Baturan,  and Wesley Greig, Nanometrics Inc, Ottawa, Canada


Earthquake location is one of the primary applications of seismic data and accurate earthquake locations are a prerequisite for a broad range of studies. On large, regional scales with many stations, higher location error is expected and uncertainties or unmodeled heterogeneities in the assumed velocity model often do not have a significant impact on the quality of an earthquake location. However for sparse, local networks such as those used for induced seismicity monitoring, uncertainties in the velocity model can have a large impact on earthquake locations and may be detrimental to the analysis of seismicity in a region. For example accurate computation of event magnitude, a necessity for induced seismicity monitoring networks, requires an accurate event location. In this study, a synthetic experiment is performed to evaluate the effect of velocity model complexity, station density, and errors in the assumed velocity model with respect to the quality of earthquake locations. We construct travel time grids for velocity models of varying complexity and then randomly simulate events to generate travel times and picks for a synthetic catalogue. We locate the simulated events with networks of varying station density and analyze the resultant errors in computed event location. Finally the effect of locating earthquakes using an incorrect velocity model is explored. We find that velocity model complexity plays an important role in the quality of earthquake locations that can be expected for a local network. In particular, the presence of low velocity zones significantly increases location uncertainty for sparse networks, especially in the vertical direction. However increasing the number of stations can reduce some of the uncertainty introduced by velocity model complexity. Our results highlight the importance of accurate velocity structure in earthquake location for local networks.

Apr 10, 2015