Seward, in south-central Alaska, was one of the towns most devastated by the Alaska earthquake of March 27, 1964. The greater part of Seward is built on an alluvial fan-delta near the head of Resurrection Bay on the southeast coast of the Kenai Peninsula. It is one of the few ports in south-central Alaska that is ice free all year, and the town’s economy is almost entirely dependent upon its port facilities.
The Alaska earthquake of March 27, 1964, magnitude approximately 8.3–8.4, began at 6:36 p.m. Its epicenter was in the northern part of the Prince William Sound area; focal depth was 20–50 km.
Strong ground motion at Seward lasted 3–4 minutes. During the shaking, a strip of land 50–400 feet wide along the Seward waterfront, together with docks and other harbor facilities, slid into Resurrection Bay as a result of large-scale submarine landsliding. Fractures ruptured the ground for‘severa1 hundred feet back from the landslide scarps. Additional ground was fractured in the Forest Acres subdivision and on the alluvial floor of the Resurrection River valley; fountaining and sand boils accompanied the ground fracturing. Slide-generated wares, possibly seiche waves, and seismic sea waves crashed onto shore; ware runup was as much as 30 feet above mean lower low water and caused tremendous damage; fire from burning oil tanks added to the destruction. Damage from strong ground motion itself was comparatively minor. Tectonic subsidence of about 3.6 feet resulted in low areas being inundated at high tide.
Thirteen people were killed and five were injured as a result of the earthquake. Eighty-six houses were totally destroyed and 260 were heavily damaged. The harbor facilities were almost completely destroyed, and the entire economic base of the town was wiped out. The total cost to replace the destroyed public and private facilities was estimated at $22 million.
Seward lies on the axis of the Chugach Mountains geosyncline. The main structural trend in the mapped area, where the rocks consist almost entirely of graywacke and phyllite, is from near north to N. 20° E. Beds and cleavage of the rocks commonly dip 70° W. or NW. to near vertical. Locally, the rocks are complexly folded or contorted. So major faults were found in the mapped area, but small faults, shear zones, and joints are common.
Surficial deposits of the area hare been divided for mapping into the following units: drift deposits, alluvial fan deposits, valley alluvium, intertidal deposits, landslide deposits, and artificial fill. Most of these units intergrade and were deposited more or less contemporaneously.
The drift deposits consist chiefly of till that forms moraines along the lower flanks of the Resurrection River valley and up tributary valleys. The till is predominantly silt and sand and lesser amounts of clay-size particles, gravel, cobbles, and boulders. Glacial outwash and stratified ice-contact deposits constitute the remainder of the drift deposits.
Fans and fan-deltas have been deposited at the valley mouths of tributary streams. Some, including the one upon which Seward built, project into Resurrection Bay, and deltaic-type deposits form their distal edges. The larger fans—composed chiefly of loosely compacted and poorly sorted silt, sand, and gravel—form broad aprons having low gradients. The fan deposits range in thickness from about 100 feet to possibly several hundred feet and, at least in some places, lie on a platform of compact drift. Smaller fans at the mouths of several canyons have steep gradients and considerable local relief.
Valley alluvium, deposited chiefly by the Resurrection River, consists mostly of coarse sand and fine to medium gravel. In the axial part of the valley it is probably more than 100 feet thick. Near the head of Resurrection Bay, the alluvium is underlain by at least 75 feet of marine deltaic sediments, which are in turn underlain by 600 or more feet of drift in the deepest part of the bedrock valley.
Beach, deltaic, and estuarine sediments, deposited on intertidal flats at the head of the bay and along far1 margins that extend into the bay, arc mapped as intertidal deposits. They consist mostly of silt, sand, and fine gravel, and lesser amounts of clay-size particles.
The earthquake reactivated old slides and trigged new ones in the mountains. Rock and snow avalanches, debris flows, and creep of talus deposits characterized slide activity on the steeper slops. The Seward waterfront had been extended before the earthquake by adding artificial fill consisting of loose sand and gravel; part of the lagoon area had been filled with refuse. After the earth- quake, fill, consisting of silt and sand dredged from the head of the bay, was pumped onto part of the lagoon area and also on land at the northwest corner of the bay.
Response to the disaster was immediate and decisive. City, State, and Federal agencies, as well as other organizations and individuals, gave unstintingly of their time and facilities. Within a few days, there was temporary restoration of water, sewerage, and electrical facilities.
The U.S. Army Corps of Engineers was authorized to select sites and construct a new dock for the Alaska Railroad, a new small-boat basin, and related facilities. The firm of Shannon and Wilson, Inc., under contract to the Corps of Engineers, investigated subsurface soils extensively to determine the factors responsible for the sliding along the Seward waterfront and to assist in site selection for reconstruction of the destroyed harbor facilities. Borings also made along the Seward waterfront and at the head of the bay, and laboratory tests were conducted on pertinent samples. These studies were augmented by geophysical studies both on land and in the bay. In addition, the Corps of Engineers made shallow borings on the intertidal flats at the head of the bay and performed pile-driving and load tests. Borings also were drilled and test pits were dug in the subdivision of Forest Acres.
Sliding along the Seward waterfront markedly deepened the water along the former shoreline. Post-earthquake slopes of the bay floor immediately offshore also are steeper in places than before the earthquake. The strong ground motion of the earthquake triggered the landsliding, but several factors may have contributed to the magnitude and characteristics of the slides. These factors are: (1) the long duration of strong ground motion, (2) the grain size and texture of the material involved in the sliding, (3) the probability that the finer grained materials liquefied and flowed seaward, and (4) the added load of manmade facilities built on the edge of the shore, Secondary effects of the slides themselves—sudden drawdown of water, followed by the weight of returning waves—also may have contributed to the destruction.
Submarine sliding at the northwest corner of the bay occurred in fine-grained deltaic deposits whose frontal slopes probably were in metastable equilibrium under static conditions. Uplift pressures from aquifers under hydrostatic head, combined with the probable liquefaction characteristics of the sediments when vibrated by strong ground motion, probably caused the material to slide and flow seaward as a heavy slurry.
Under static conditions, no major shoreline or submarine landsliding is expected in the Seward area; in the event of another severe earthquake, however, additional sliding is likely along the Seward waterfront and also in the deltaic deposits at the northwest corner of the bay. Fractured ground in back of the present shoreline along the Seward waterfront is an area of incipient landslides that would be unstable under strong shaking. For this reason the Scientific and Engineering Task Force placed the area in a high-risk classification and recommended no repair, rehabilitation, or new construction in this area involving use of Federal funds; it was further recommended that the area should be reserved for park or other uses that do not involve large congregations of people. The deltaic deposits at the head of the bay probably also would be susceptible to sliding during another large earthquake. This sliding would result in further landward retreat of the present shoreline toward the new railroad dock. Specifications for the new dock, whose seaward end is now approximately 1,100 feet from the back scarp of the subaqueous landslide, require design pro- visions to withstand seismic shock up to certain limits.
Earthquake-induced fracturing of the ground in the subdivision of Forest Acres was confined to the lower part of a broad alluvial fan. There, sewer and water lines were ruptured and the foundations of some homes were heavily damaged. Landsliding, such as occurred along the shoreline of the bay, was not a contributing cause of the fracturing. Two hypotheses are offered to explain the fracturing:
1. Seismic energy was transformed into visible surface waves of such amplitude that the strength of surface layer was exceeded and rupturing occurred; tensional and compressional stresses alternately opened and closed the fractures and forced out water and mud.
2. Compaction by vibration of the fine-grained deposits of the fan caused ground settlement and fracturing; ground water under temporary hydrostatic head was forced to the surface as fountains and carried the finer material with it.
Water waves that crashed onto shore, while shaking was still continuing, were generated chiefly by onshore and offshore landsliding. Waves that overran the shores about 25 minutes after shaking stopped and that continued to arrive for the next several hours are believed to be seismic sea waves (tsunamis) that originated in an uplifted area in the Gulf of Alaska. During the time of seismic sea-wave activity and perhaps preceding it, seiche wares also may have been generated within Resurrection Bay and complicated the wave effects along the shoreline.