Geotechnical Reconnaissance of the 2002 Denali Fault, Alaska, Earthquake

Abstract

The 2002 M7.9 Denali fault earthquake resulted in 340 km of ruptures along three separate faults, causing widespread liquefaction in the fluvial deposits of the alpine valleys of the Alaska Range and eastern lowlands of the Tanana River. Areas affected by liquefaction are largely confined to Holocene alluvial deposits, man-made embankments, and backfills. Liquefaction damage, sparse surrounding the fault rupture in the western region, was abundant and severe on the eastern rivers: the Robertson, Slana, Tok, Chisana, Nabesna and Tanana Rivers. Synthetic seismograms from a kinematic source model suggest that the eastern region of the rupture zone had elevated strong-motion levels due to rupture directivity, supporting observations of elevated geotechnical damage. We use augered soil samples and shear-wave velocity profiles made with a portable apparatus for the spectral analysis of surface waves (SASW) to characterize soil properties and stiffness at liquefaction sites and three trans-Alaska pipeline pump station accelerometer locations. THE DENALI FAULT EARTHQUAKEOn the afternoon of 3 November 2003, a M7.9* earthquake ruptured the Susitna Glacier fault, the Denali fault, and the Totschunda fault. The M7.9 main shock was preceded by a M6.7 foreshock on October 23, 2002 (Figure 1), on a 45-kilometer segment of the Denali fault (Eberhart-Phillips et al. 2003, Wright et al. 2003. The epicenter of the main shock was 22 km east of the foreshock, and consisted of multiple subevents that Eberhart-Phillips et al. (2003) relate to regions of high slip along the three faults. The first sub-event was a M7.2 thrust on 40 km of the previously unknown Susitna Glacier fault (Figure 1). The second sub-event was a result of right-lateral (dextral) rupture on the Denali fault in the vicinity of where the Black Rapids Glacier crosses the TransAlaska Pipeline System (TAPS) and was equivalent to a M7.3. This second subevent produced a large velocity pulse recorded nearby at TAPS Pump Station 10 (station PS10). The peak acceleration and velocity at PS10 were the highest recorded for the earthquake at 0.36g and 114 cm/s (high-pass-filtered values), respectively (Ellsworth et al., this issue; and the town of Mentasta. This dextral offset sub-event on the Denali fault had a large displacement pulse, and was followed by a right step-over zone to the Totschunda fault where surface displacements were sharply diminished over a 76-km portion of the Totshunda fault. The total 340-km surface rupture was unidirectional from west to east. The seismic moment inverted from the geodetic and strong-motion data is estimated to be between M7.8 and M7.9 (Eberhart-Phillips et al. 2003). The overall duration of shaking was about 140 s with individual sub-event displacement pulses having periods of 20-30 s.Because of the remote location of this M7.9 earthquake, the cost to lives and property was remarkably low. There were no fatalities, only one injury, and damage was estimated at approximately $40 million dollars. The fault ruptured beneath the transAla...

Surface Rupture of the 2002 Denali Fault, Alaska, Earthquake and Comparison with Other Strike-Slip Ruptures

“…We also found liquefaction of unconsolidated sediment near the fault trace on valley floors. (See also Kayen et al 2004. )…”

Nonlinear Suppression of High‐Frequency S Waves by the Near‐Field Velocity Pulse With Reference to the 2002 Denali Earthquake

“…Martirosyan et al () give 63.4239°N,−145.7658°E, which is self‐consistent with the datum used by Carver et al () to locate temporary stations. This station was ~3 km north of the surface rupture (Figure ) with strike‐slip displacement of ~5 m (Eberhart‐Phillips et al, ; Hreinsdottir et al, ; Dreger et al, ; Frankel ; Haeussler et al, ; Kayen et al, ; Oglesby et al, ; Wright et al, ; Asano et al, ; Liao and Huang, 2008; Walker & Shearer, ). Rupture approached PS10 from the west‐northwest.…”

“…We mostly apply laterally homogeneous scaling formulations in this paper for brevity and to avoid cloaking simple inferences with onerous mathematics. The structure near PS10 is somewhat laterally heterogeneous (Brocher et al, ; Frankel ; Kayen et al, ). PS10 is near the fault at the side of the flower structure (Figure ) where the deep geological structure may not be horizontally stratified.…”

Control of style-of-faulting on spatial pattern of earthquake-triggered landslides

“…The reasons for this choice are: (a) the terrain characteristics of that area are similar to those in the Wenchuan area; (b) that earthquake has the same magnitude as the 12 May 2008 Wenchuan earthquake; (c) the 300-km-long rupture generated along the Denali fault by that earthquake is, on a regional-scale, similar to the 240-km-long rupture generated along the Yingxiu-Beichuan fault by the 12 May 2008 Wenchuan earthquake; and (d) the landslides triggered by that earthquake were concentrated mostly in a roughly 30-km-wide zone around the surface rupture (Eberhart-Phillips et al 2003;Jibson et al 2006). However, compared with tens of thousands of landslides triggered by the 2008 Wenchuan earthquake, only thousands of landslides were triggered by the 2002 Alaska earthquake (Harp et al 2003;Jibson et al 2004Jibson et al , 2006Kayen et al 2004). In addition, unlike the variable slips along the Yingxiu-Beichuan fault during the 12 May 2008 Wenchuan earthquake, the slips along the Denali fault during the 3 November 2002 were oblique with predominant strike-slip components Dreger et al 2004;Velasco et al 2004).…”