ABSTRACT Seattle’s geologic record begins with Eocene deposition of fluvial arkosic sandstone and associated volcanic rocks of the Puget Group, perhaps during a time of regional strike-slip faulting, followed by late Eocene and Oligocene marine deposition of the Blakeley Formation in the Cascadia forearc. Older Quaternary deposits are locally exposed. Most of the city is underlain by up to 100 m of glacial drift deposited during the Vashon stade of Fraser glaciation, 18–15 cal k.y. B.P. Vashon Drift includes lacustrine clay and silt of the Lawton Clay, lacustrine and fluvial sand of the Esperance Sand, and concrete-like Vashon till. Mappable till is absent over much of the area of the Vashon Drift. Peak local ice thickness was 900 m. Isostatic response to this brief ice loading was significant. Upon deglaciation, global ice-equivalent sea level was about −100 m and local RSL (relative sea level) was 15–20 m, suggesting a total isostatic depression of ~115–120 m at Seattle. Subsequent rapid rebound outstripped global sea-level rise to result in a newly recognized marine low-stand shoreline at −50 m. The Seattle fault is a north-verging thrust or reverse fault with ~7.5 km of throw. Conglomeratic Miocene strata may record initiation of shortening. Field relations indicate that fault geometry has evolved through three phases. At present, the north-verging master fault is blind, whereas several surface-rupturing faults above the master fault are south verging. The 900–930 A.D. Restoration Point earthquake raised a 5 km × 35 km (or larger) area as much as 7 m. The marine low-stand shoreline is offset by a similar amount, thus there has been only one such earthquake in the last ~11 k.y. Geomorphology is largely glacial: an outwash plain decorated with ice-molded flutes and large, anastomosing tunnel valleys carved by water flowing beneath the ice sheet. Euro-Americans initially settled here because of landscape features formed by uplift in the Restoration Point earthquake. But steep slopes and tide flats were not conducive to commerce: starting in the 1890s and ending in the 1920s, extensive regrading removed hills, decreased slopes, and filled low areas. In steep slopes the glacial stratigraphy is prone to landslides when saturated by unusually wet winters. Seismic hazards comprise moderately large (M 7) earthquakes in the Benioff zone 50 km and more beneath the city, demi-millennial M 9 events on the subduction zone to the west, and infrequent local crustal earthquakes (M 7) that are likely to be devastating because of their proximity. Seismic shaking and consequent liquefaction are of particular concern in Pioneer Square, SoDo, and lower Duwamish neighborhoods, which are largely built on unengineered fill that was placed over estuarine mud. Debris from past Mount Rainier lahars has reached the lower Duwamish valley and a future large lahar could pose a sedimentation hazard.

Single-step, laser-fusion 40 Ar/ 39 Ar ages of single muscovite grains with an automated micro-extraction system is a precise and relatively rapid way of analyzing large numbers of grains. This study used >500 muscovite grains from a late Wisconsinan sandy loess from eastern Long Island, New York, in order to evaluate the potential of Ar-Ar ages of single grain muscovite as a provenance tool for loess. The samples for dating were from a 2.7 m core of sediments from a small kettle hole in Wildwood State Park on the north shore of Long Island. These eolian deposits consist of a bimodal distribution of poorly sorted medium silt and medium sand that are buff colored, homogeneous, and unstratified. Long Island is a good place to test this approach, because the 40 Ar/ 39 Ar and K/Ar ages for muscovite in the potential bedrock sources to the north in New England vary systematically from ca. 450 Ma in the west to ca. 200 Ma in the east. The majority of muscovite ages in the loess range from 250 to 400 Ma, and muscovite age populations along the core show a change in proportion of muscovite input from the different provenances in New England. The results of this study confirm that using 40 Ar/ 39 Ar ages of a large number of single muscovite grains is a good method for examining the provenance of muscovite in loess, and thus understanding processes that produce loess.