Abstract
Anchorage, Alaska’s largest city, is about 80 miles west-northwest of the epicenter of the March 27 earthquake. Because of its size, Anchorage bore the brunt of property damage from the quake; it sustained greater losses than all the rest of Alaska combined. Damage was caused by direct seismic vibration, by ground cracks, and by landslides. Direct seismic vibration affected chiefly multistory buildings and buildings having large floor areas, probably because of the long period and large amplitude of the seismic waves reaching Anchorage. Most small buildings were spared. Ground cracks caused capricious damage throughout the Anchorage Lowland. Cracking was mast prevalent near the heads or within landslides but was also widespread elsewhere. Landslides themselves caused the most devastating damage.
Triggering of landslides by the earthquake was related to the physical-engineering properties of the Bootlegger Cove Clay, a glacial estuarine-marine deposit that underlies much of the Anchorage area. The Bootlegger Cove Clay contains zones of low shear strength, high water content, and high sensitivity that failed under the vibratory stress of the earthquake. Shear strength in sensitive zones ranged from less than 0.2 tsf to about 0.5 tsf; sensitivity ranged from about 10 to more than 40. Sensitive zones generally are centered about 10 to 20 feet above sea level, between zones of stiff insensitive clay. Many physical tests by the U.S. Army Corps of Engineers were directed toward analyzing the causes of failure in the Bootlegger Cove Clay and finding possible remedies. Strengths and sensitivities were measured directly in the field by means of vane shear apparatus. A4tterberg limits, natural water contents, triaxial shear, sensitivity, dynamic modulus, consolidation strength, and other properties were measured in the laboratory. Pulsating-load tests simulated earthquake loading.
Most of the destructive landslides in the Anchorage area moved primarily by translation rather than by rotation. Thus, all the highly damaging slides were of a single structural dynamic family despite wide variations in size, appearance, and complexity. They slid on nearly horizontal slip surfaces after loss of strength in the Bootlegger Core Clay. Same failures are attributed to spontaneous liquefaction of sand layers. All translatory slides surmounted flat-topped bluffs bounded marginally by steep slopes facing lower ground. Destructive translatory slides occurred in the downtown area (Fourth Avenue slide and L Street slide), at Government Hill, and at Turnagain Heights. Less destructive slides occurred in many other places-mostly uninhabited or undeveloped areas.
In most translatory slides, damage was greatest in graben areas at the head and in pressure-ridge areas at the toe. Many buildings inside the perimeters of slide blocks were little damaged despite horizontal translations of several feet. The large Turnagain Heights slide, however, was characterized by a complete disintegration and drastic lowering of the prequake land surface. Extensive damage back from the slide, moreover, was caused by countless tension cracks.
An approximation of the depth of failure in the Bootlegger Cove Clay in the various slides may be obtained by using a geometric relationship herein called the "graben rule." Because the cross-sectional area of the graben at the head of the slide approximated the cross-sectional area of the space voided behind the slide block as the block moved outward, the depth of failure was equal to the area of the graben divided by the lateral displacement. This approximation supplements and accords with test data obtained from borings. The graben rule should apply to any translatory slide in whi& flowage of material from the zone of failure has not been excessive.
Geologic evidence indicates that landslides similar to those triggered by the March 27 earthquake have occurred in the Anchorage area at various times in the past.
U.S. Department of the Interior |
U.S. Geological Survey