Reply to Y. Kagan's Comment on "The Gutenberg-Richter of Characteristic Earthquake Model, Which is it?

Steven G. Wesnousky

Published February 1996, SCEC Contribution #251

There are two principal aspects to Kagan's comments.
He first asserts that decisions made in analysis of the data
may bias the results and interpretation I put forth in Wesnousky
(1994, herein referred to as W94). He then provides
his own statistical model to argue that seismicity along the
major strike-slip faults of southern California is adequately
described by the Gutenberg-Richter distribution Log n = a
- bM, where n = number of events of magnitude M and a
and b are empirical constants. I will initially consider the
specific concerns Kagan voices toward the analysis and
show that none lead to a significant change in the concluding
observations, interpretations, or caveats originally stated in
W94. Subsequently, I show that Kagan's statistical model is
flawed when examined in the context of fault mechanics.
The magnitudes of earthquakes his model requires to occur
on specific faults violate empirical scaling laws that relate
fault rupture length, coseismic slip, and earthquake magnitude.

Citation
Wesnousky, S. G. (1996). Reply to Y. Kagan's Comment on "The Gutenberg-Richter of Characteristic Earthquake Model, Which is it?. Bulletin of the Seismological Society of America, 86(1), 286-291.