A 3-Stage Model for Erosional  Modification of the Emerson Fault (Landers, California Earthquake) Scarp,
June 1992-January 2000

ARROWSMITH, J Ramon

Department of Geology, Arizona State University, Tempe, Arizona 85287, ramon.arrowsmith@asu.edu

RHODES, Dallas D.

Department of Geology and Geography, Georgia Southern University, Statesboro, Georgia 30460, DRhodes@GaSoU.edu
 

A 400-m long segment of the Emerson fault, ruptured during the Landers earthquake, provides insights into the early stages of modification of earthquake surface ruptures, particularly for the portions of fault scarps over which there is runoff.  The initial scarp survey began on July 1, 1992, three days after the earthquake.  Seven subsequent surveys have been made.  With survey errors <10 cm after least squares network adjustments, analyses of the differences among the data sets clearly document reintegration of the drainage network and knickpoint incision.

Weather, drainage basin form, and scarp complexity control the rate and style of modification.  Two large ENSO events (1994-95, 1997-98) have occurred since the earthquake, and the first winter after the earthquake (1992-93) was wetter than either of these.  During 7 years of study, 800-1,000 mm of highly seasonal precipitation have fallen on the area.  The effect of basin morphology is expressed largely through catchment area.  The deepest incision occurs in the largest and steepest drainage basin.

Erosional modification of the scarp occurred in three stages.  During the first, or 'Set-up' stage, the disrupted drainage reestablished definite flow paths.  Where the scarp form was simple, incision occurred across the lip at one or more points on the broad, shallow pre-earthquake channel.  At more complex scarps, especially those with open fractures, runoff was diverted to the subsurface  where pipes and open fractures eventually merged into a dominant path for the flow.  The period of  'Integration' was characterized by the reestablishment of a connected drainage network that frequently involved minor captures where multiple incisions had occurred across the scarp.  Since the networks have been reintegrated, the 'Development' stage has seen the reestablished flow paths extend headward into the basin.  Channel incision into uplifted pre-earthquake channel floors has  propagated 5-15 m upstream with flow concentrated in narrow channels.  Although these stages  were not synchronous in all the monitored drainages, the same sequence did occur.  These early stages of scarp modification are not part of most scarp profile development models.