HighResolution Earthquake Location in the New Madrid Seismic Zone

1
High-Resolution Earthquake Location in the New
Madrid Seismic Zone
T51D-0767
Meredith Dunn, Steve Horton, Heather DeShon, and
Christine Powell mmdunn_at_memphis.edu

• Data Set and Analysis
• 1394 Earthquakes
• 313 Stations
• Events recorded between 2000-2006
• Double Difference Relocations found for 6
  clusters using the program hypoDD (Waldhauser,
  2001).

Purpose Provide improvement of fault plane
resolution through calculation of double
difference earthquake locations in the New Madrid
Seismic Zone (NMSZ).
Figure 6 Locations and relocations of clusters 1
and 2 (see figure 3). Cluster 1 is indicated by
red circles, cluster 2 is indicated by blue
circles. Cross Section is oriented N45E. This
arm is thought to be a vertical, right later
strike slip fault.
Clusters 1 and 2
Figure 1 Tectonic setting of the NMSZ.
Seismicity occurs primarily in the Reelfoot Rift,
a possible failed rift system. Red circles
indicate earthquakes.
Figure 3 Map showing catalog locations with
station coverage, as well as major arms of
seismicity and 6 clusters relocated in this study.
Cluster 5
Figure 7 Locations and relocations of cluster 5
(see figure 3) the northern portion of the
central thrust fault. Dip estimated to be 30-32
degrees for relocations, more shallow than
southern portion of fault (see figure 8). Cross
section is oriented N25W.
4
3
Mississippi Embayment
5
Reelfoot Rift
6
1
2
Cluster 6
Figure 8 Locations and relocations of cluster 6
(see figure 3), the southern portion of the
central thrust fault. Dip estimated to be 35-40
degrees, slightly more shallow than original
locations. Cross section is oriented N25W.

• Study Area
• Most seismically active region east of the Rocky
  Mountains.
• Region has history of damaging earthquakes
  including three M7.0 events in 1811-1812.
• Seismicity occurs within Mississippi Embayment,
  characterized by thick sediment layer at
  surface.

Results and Discussion The NMSZ was divided into
6 clusters based on known tectonics and natural
breaks in seismicity. Waveform cross correlation
was performed on the entire data set. Results
were examined for the NMSZ and for individual
clusters. Clustered relocations were more
computationally stable and provide more small
scale improvements over relocation of the whole
NMSZ. Relocations provide a clear image of the
fault plane, with improvement over original
catalog locations.
Future Work
Figure 2 Original catalog locations contained in
the data set. Color indicates depth of events.
The same color legend is used in figures 4 and 5.
Figure 5 Results reconstructed from relocation
of 6 clusters individually. The depths in several
areas such as the northeast and southwest arm are
different than those in figures 2 and 4.
Figure 4 Results from relocation of entire data
set with no clustering performed. These are to
be compared with results in figure 5.