Earthquakes and geological structures of the St. Lawrence Rift System - Intraplate Earthquakes - page 86

occurred ( Figure 4.2 ) . Near the epicentre of the earthquake, many Ottawa-Bonnechere
graben faults have been mapped and many have conspicuous topographical signatures
visible in remote-sensing imagery (Bent et al ., 2002 ) . The focal mechanism of the 1935
earthquake (the largest event of the seismic zone) indicates thrust faulting on a moderately
dipping northwest-striking plane, an orientation similar to the graben faults of the area
(Bent, 1996 ) . Similarly, in the Lower St. Lawrence Seismic Zone, SLRS faults are present
and may be reactivated by the current seismic activity.
The rift-seismicity hypothesis has seismic hazard implications. The seismic provisions
in the 2005 National Building Code of Canada use the larger of the ground motions
derived from two seismic source zone models. The first model assumes that the historical
earthquake clusters denote areas that will continue to be active. The second model assumes a
common geological framework for the seismicity clusters (i.e., passive Paleozoic rift faults)
and regroups them into large source zones (Adams et al ., 1995 ) . In areas with historical
damaging earthquakes, the hazard derived from the historical model dominates, whereas in
areas without significant activity but where rift faults are present, the hazard is dominated
by the geologically derived model (Adams and Halchuk, 2003 ) .

4.5 The rift hypothesis and the SLRS: discussion and conclusions

Since the mid 1980s, the rift model has served as the main hypothesis that relates seismicity
and faults along the SLRS. Although the rift model explains many characteristics of the
SLRS seismicity, some questions remain. At the global level, the rift model was based on
a correlation with “extended crust” that was correct for 100% of earthquakes of magnitude
larger than 7.0; 60% between M 6 and 7; and 46% for earthquakes smaller than M 6
(Johnston and Kanter, 1990 ) . Recently, however, Schulte and Mooney ( 2005 ) concluded
that on a global scale the correlation of seismicity within stable continental regions (SCR)
and ancient rifts has been overestimated in the past. They note that several apparently non-
rifted crust areas have experienced multiple large ( M 6.0) events. The rift model defined

from a global perspective was extrapolated to the SLRS where only the 1663 Charlevoix
shock is thought to have exceeded magnitude 7.0 . 2 We note that the historical earthquake
of 1663 has an uncertain epicentre location based on limited intensity information and is
consequently only assumed to be along a rift fault. The rift hypothesis was extrapolated
to smaller earthquakes in the magnitude 5.5 to 6.5 range, with most having occurred
prior to instrumental recording, with uncertain magnitudes, epicentres, and unknown focal
mechanisms. Based on these limitations, we raise questions about the applicability of the
global rift model to SLRS earthquakes, especially for M < 4.5 earthquakes.
Globally, we note that many recent earthquakes do not fit the assumed local seismotec-
tonics model. Therefore, the correlation with any earthquake that occurs within the loose
geographic boundaries of the SLRS with the faults of that system is not necessarily correct.
In the case of the 1988 M 5.9 Saguenay earthquake, for example, despite the focal mech-
anism that did not match the orientation of the most conspicuous graben faults that were

2 Although the 1663 event was not part of the database of Johnston and Kanter ( 1990 ) , which only considered the last 200 years.