Abstract The city of Seattle, Washington State, lies within the Puget Sound Lowland, an elongate structural and topographic basin between the Cascade Range and Olympic Mountains. The area has been impacted by repeated glaciation in the past 2.4 m.y. and crustal deformation related to the Cascadia subduction zone. The present landscape largely results from those repeated cycles of glacial scouring and deposition and tectonic activity, subsequently modified by landsliding, stream erosion and deposition, and human activity. The last glacier to override the area, the Vashon-age glacier of the Fraser glaciation, reached the Seattle area ca. 14,500 14 C yr B.P. (17,400 cal yr B.P.) and had retreated from the area by ca. 13,650 14 C yr B.P. (16,400 cal yr B.P.). The Seattle area sits atop a complex and incomplete succession of glacial and nonglacial deposits that extends below sea level and overlies an irregular bedrock surface. These subsurface materials show spatial lithologic variability, are truncated by many unconformities, and are deformed by gentle folds and faults. Sediments that predate the last glacial–interglacial cycle are exposed where erosion has sliced into the upland, notably along the shorelines of Puget Sound and Lake Washington, along the Duwamish River valley, and along Holocene streams. The city of Seattle straddles the Seattle uplift, the Seattle fault zone, and the Seattle basin, three major bedrock structures that reflect north-south crustal shortening in the Puget Lowland. Tertiary bedrock is exposed in isolated locations in south Seattle on the Seattle uplift, and then it drops to 550 m below ground under the north half of the city in the Seattle basin. The 6-km-wide Seattle fault zone runs west to east across the south part of the city. A young strand of the Seattle fault last moved ~1100 yr ago. Seattle has also been shaken by subduction-zone earthquakes on the Cascadia subduction zone and deep earthquakes within the subducting plate. Certain postglacial deposits in Seattle are prone to liquefaction from earthquakes of sufficient size and duration. The landforms and near-surface deposits that cover much of the Seattle area record a brief period in the geologic history of the region. Upland till plains in many areas are cut by recessional meltwater channels and modern river channels. Till plains display north-south drumlins with long axes oriented in the ice-flow direction. Glacially overridden deposits underlie the drumlins and most of the uplands, whereas loosely consolidated postglacial deposits fill deep valleys and recessional meltwater channels. Ice-contact deposits are found in isolated locations across the uplands and along the margins of the uplands, and outwash deposits line upland recessional channels. Soft organic-rich deposits fill former lakes and bogs. A preliminary geologic map of Seattle was published in 1962 that is only now being replaced by a detailed geologic map. The new map utilizes a data set of 35,000 geotechnical boreholes, geomorphic analyses of light detection and ranging (LIDAR), new field mapping, excavation observations, geochronology, and integration with other geologic and geophysical information. Findings of the new mapping and recent research include recognition of Possession- and Whidbey-age deposits in Seattle, recognition that ~50% of the large drumlins are cored with pre-Vashon deposits and 50% with Vashon deposits, and that numerous unconformities are present in the subsurface. Paleotopographic surfaces display 500 m (1600 feet) of relief. The surficial deposits of Seattle can be grouped into the following categories to exemplify the distribution of geologic materials across the city: postglacial deposits 16%, late glacial deposits 12%, Vashon glacial deposits 60%, pre-Vashon deposits 9%, and bedrock 3%. of these, 49% are considered fine-grained deposits, 19% are considered intermediate or interbedded deposits, and 32% are considered coarse-grained deposits. These percentages include only the primary geologic units and not the overlying fill and colluvial deposits.

Abstract Seattle, Washington, has one of the most progressive landslide ordinances and mitigation programs in the United States. The necessity for this is driven by concentrated winter precipitation, steep slopes, and glacial soils that are susceptible to instability. Early in the European development, engineers recognized the vulnerability of Seattle hillsides to landslides, particularly when the hillsides were disturbed. In this paper, we trace the history of mapping of landslide-prone ground in Seattle, started by Miller ( 1973 ), detailed by Tubbs ( 1974, 1975 ), and most recently catalogued by Shannon & Wilson ( 2000, 2003 ). Owing to the relative homogeneity of geology (glacial and other Pleistocene nonglacial soils), landsliding consists mainly of debris landslides, debris avalanches, and slumps. Debris avalanches that engage the shallow colluvial soils are by far the most common. Deep-seated slumps are not as common but can encompass a large area and affect many property owners. Several studies agree that ~80% of the landslides include one or more human influences. To limit the amount of damage that landslides cause to private properties and City of Seattle infrastructure, rules were promulgated by the city in 1984. These rules have been revised three times, and they now serve as an example for the rest of the state of Washington. Following destructive, widespread slope instability in 1997 and the Shannon & Wilson studies in 2000 and 2003 , Seattle Public Utilities instituted a state-of-the-industry landslide mitigation program that has already paid dividends for the city.

Abstract A comparison of 1719 differential global positioning system (GPS) measurements with a 1 m LIDAR (light detection and ranging) digital elevation model (DEM) covering West Seattle shows that DEM elevation errors range from −4.88 m to +3.32 m. The errors are spatially correlated with a semivariogram range of 40 m, an unclustered mean of −0.11 m, and an unclustered standard deviation of ±0.75 m. Although there are statistically significant correlations between elevation error and elevation, slope angle, and topographic roughness, the relationships are weak and have little explanatory power. Monte Carlo simulations of slope angle, static factor of safety, Newmark yield acceleration, and log Newmark displacement show that elevation errors of the magnitude reported here can have significant effects on derivative calculations. The standard deviation of simulated slope angles increases from ~±2° to ±3° as the true slope angle approaches zero. Errors in slope angles calculated from the LIDAR DEM are smaller than those previously reported for a conventional 10 m DEM covering the same area, but the decrease is not proportional to the decrease in DEM grid spacing. The influence of elevation errors on static factor of safety errors is strongly dependent upon the slope angle and decreases significantly as the slope angle increases. Effects on Newmark yield acceleration and log Newmark displacement are not as profound but are still large enough to impart significant uncertainty into calculated results. Therefore, slope angle errors should be considered to be as significant as geotechnical parameter and pore-water pressure uncertainties when performing slope-stability calculations based on high-resolution LIDAR DEMs.

Abstract Landslides, particularly debris flows, have long been a significant cause of damage and destruction to people and property in the Puget Sound region. Following the years of 1996 and 1997, the Federal Emergency Management Agency designated Seattle as a “Project Impact” city with the goal of encouraging the city to become more disaster resistant to landslides and other natural hazards. A major recommendation of the Project Impact council was that the city and the U.S. Geological Survey collaborate to produce a landslide hazard map. An exceptional data set archived by the city containing more than 100 yr of landslide data from severe storm events allowed comparison of actual landslide locations with those predicted by slope-stability modeling. We used an infinite-slope analysis, which models slope segments as rigid friction blocks, to estimate the susceptibility of slopes to debris flows, which are water-laden slurries that can form from shallow failures of soil and weathered bedrock and can travel at high velocities down steep slopes. Data used for the analysis consisted of a digital slope map derived from recent light detection and ranging (LiDAR) imagery of Seattle, recent digital geologic mapping of the city, and shear-strength test data for the geologic units found in the surrounding area. The combination of these data layers within a geographic information system (GIS) platform allowed us to create a shallow landslide hazard map for Seattle.

Abstract In Seattle, Washington, deep-seated landslides on bluffs along Puget Sound have historically caused extensive damage to land and structures. These large failures are controlled by three-dimensional (3-D) variations in strength and pore-water pressures. We assess the slope stability of part of southwestern Seattle using a 3-D limit-equilibrium analysis coupled with a 3-D groundwater flow model. Our analyses use a high-resolution digital elevation model (DEM) combined with assignment of strength and hydraulic properties based on geologic units. The hydrogeology of the Seattle area consists of a layer of permeable glacial outwash sand that overlies less permeable glacial lacustrine silty clay. Using a 3-D groundwater model, MODFLOW-2000, we simulate a water table above the less permeable units and calibrate the model to observed conditions. The simulated pore-pressure distribution is then used in a 3-D slope-stability analysis, SCOOPS, to quantify the stability of the coastal bluffs. For wet winter conditions, our analyses predict that the least stable areas are steep hillslopes above Puget Sound, where pore pressures are elevated in the outwash sand. Groundwater flow converges in coastal reentrants, resulting in elevated pore pressures and destabilization of slopes. Regions predicted to be least stable include the areas in or adjacent to three mapped historically active deep-seated landslides. The results of our 3-D analyses differ significantly from a slope map or results from one-dimensional (1-D) analyses.

Abstract Empirical rainfall thresholds and related information form the basis of a prototype system for forecasting landslides in the Seattle area. The forecasts are tied to four alert levels, and a decision tree guides the use of thresholds to determine the appropriate level. From analysis of historical landslide data, we developed a formula for a cumulative rainfall threshold (CT), P 3 = 88.9 − 0.67 P 15 , defined by rainfall amounts in millimeters during consecutive 3 d (72 h) periods, P 3 , and the 15 d (360 h) period before P 3 , P 15 . The variable CT captures more than 90% of historical events of three or more landslides in 1 d and 3 d periods recorded from 1978 to 2003. However, the low probability of landslide occurrence on a day when the CT is exceeded at one or more rain gauges (8.4%) justifies a low-level of alert for possible landslide occurrence, but it does trigger more vigilant monitoring of rainfall and soil wetness. Exceedance of a rainfall intensity-duration threshold I = 82.73 D −1.13 , for intensity, I (mm/hr), and duration, D (hr), corresponds to a higher probability of landslide occurrence (30%) and forms the basis for issuing warnings of impending, widespread occurrence of landslides. Information about the area of exceedance and soil wetness can be used to increase the certainty of landslide forecasts (probabilities as great as 71%). Automated analysis of real-time rainfall and subsurface water data and digital quantitative precipitation forecasts are needed to fully implement a warning system based on the two thresholds.

Abstract The temporal forecasting of landslide hazard has typically relied on empirical relations between rainfall characteristics and landslide occurrence to identify conditions that may cause shallow landslides. Here, we describe an alternate, deterministic approach to define rainfall thresholds for landslide occurrence in the Seattle, Washington, area. This approach combines an infinite slope-stability model with a variably saturated flow model to determine the rainfall intensity and duration that leads to shallow failure of hillside colluvium. We examine the influence of variation in particle-size distribution on the unsaturated hydraulic properties of the colluvium by performing capillary-rise tests on glacial outwash sand and three experimental soils with increasing amounts of fine-grained material. Observations of pore-water response to rainfall collected as part of a program to monitor the near-surface hydrology of steep coastal bluffs along Puget Sound were used to test the numerical model results and in an inverse modeling procedure to determine the in situ hydraulic properties. Modeling results are given in terms of a destabilizing rainfall intensity and duration, and comparisons with empirical observations of landslide occurrence and triggering rainfall indicate that the modeling approach may be useful for forecasting landslide occurrence.

Abstract We describe the results from an application of a distributed, transient infiltration–slope-stability model for an 18 km 2 area of southwestern Seattle, Washington, USA. The model (TRIGRS) combines an infinite slope-stability calculation and an analytic, one-dimensional solution for pore-pressure diffusion in a soil layer of finite depth in response to time-varying rainfall. The transient solution for pore-pressure response can be superposed on any steady-state groundwater-flow field that is consistent with model assumptions. Applied over digital topography, the model computes a factor of safety for each grid cell at any time during a rainstorm. Input variables may vary from cell to cell, and the rainfall rate can vary in both space and time. For Seattle, topographic slope derived from an airborne laser swath mapping (ALSM)–based 3 m digital elevation model (DEM), maps of soil and water-table depths derived from geotechnical borings, and hourly rainfall intensities were used as model inputs. Material strength and hydraulic properties used in the model were determined from field and laboratory measurements, and a tension-saturated initial condition was assumed. Results are given in terms of a destabilizing intensity and duration of rainfall, and they were evaluated by comparing the locations of 212 historical landslides with the area mapped as potentially unstable. Because the equations of groundwater flow are explicitly solved with respect to time, the results from TRIGRS simulations can be portrayed quantitatively to assess the potential landslide hazard based on rainfall conditions.

Abstract Over the past four decades, ongoing deformation of an 18-m-thick peat deposit within the flat-lying Mercer Slough has resulted in damaging deflections, and near-collapse in three cases, of pile-supported Interstate 90 bridges and a major water line on the east side of the slough. The peat is partially underlain by a dense sand unit, which includes a highly pressurized aquifer that produces artesian flow 1–2.5 m above the ground surface. Inclinometers on the east side of the slough show the peat flowing toward the structures and then apparently directed west along the interstate centerline. Large displacements recorded in several inclinometers near the center of the slough suggest a length of deforming peat that approaches 600 m, which is likely initiating retrogressively. Potential causal mechanisms include poor engineering characteristics of the peat, presence of high hydrostatic pressure transmitted within and beneath the peat, seasonal water-level variations of Lake Washington and induced hydraulic gradients within the peat, dredging of the Mercer Slough channel, puncturing of the underlying aquifer by numerous pile foundations, and fill placement along the eastern margin of the slough. The peat is flowing around the pile/shaft foundations; however, excessive lateral loads are still being applied to the foundations in a poorly understood and unpredictable manner. The most severe deflections have occurred in the outermost structures where the peat is primarily flowing transverse to the structures.

Abstract In September 1998, a geotechnical investigation was initiated by the Washington State Department of Transportation (WSDOT) to assess stability of a slope in the vicinity of Ross Point, located along the State Route 166 (SR 166) corridor. The roadway lies at the base of a marine bluff composed of Pleistocene glacial and interglacial deposits, and it has a history of roadway closures associated with landslides and embankment failures. As part of the investigation, geotechnical test borings were drilled to define subsurface conditions, and inclinometer casing and open-standpipe piezometers were installed to monitor slope movement and groundwater levels. All drilling was completed and inclinometer and piezometer installations initialized by 30 December 1998. The Ross Point landslide occurred on 29 January 1999. This landslide is unique in that the subsurface conditions had been thoroughly characterized in a geotechnical investigation prior to failure, and rainfall and groundwater levels were being frequently monitored at the time of the failure. Although slope failures are common in the landslide sequence of Esperance Sand overlying Lawton Clay, monitoring of rainfall and groundwater levels prior to and during landslide failure is unique. The groundwater levels monitored within the Ross Point landslide mass showed a rapid response to intense rainfall events, whereas the groundwater response measured in boreholes located outside of the failure was subdued. Excessive pore-water pressures rapidly developed within the landslide mass following significant rainfall events. This rapid increase in pore-water pressure, coupled with high antecedent groundwater levels, is considered to be the critical factor in triggering the Ross Point landslide. Wood sampled from the lowest nonglacial sediments exposed at Ross Point yielded a 14 C date of 40,570 ± 700 yr B.P. and are considered to be the Olympia beds.

Abstract This volume brings together case studies and summary papers describing the application of state-of-the-art engineering geologic methods to landslide hazard analysis for the Seattle, Washington, area. An introductory chapter provides a thorough description of the Quaternary and bedrock geology of Seattle. Nine additional chapters review the history of landslide mapping in Seattle, present case studies of individual landslides, describe the results of spatial assessments of landslide hazard, discuss hydrologic controls on landsliding, and outline an early warning system for rainfall-induced landslides.

During a one-month period in early 2001, El Salvador experienced two devastating earthquakes. On 13 January, a M-7.7 earthquake centered ∼40 km off the southern coast in the Pacific Ocean caused widespread damage and fatalities throughout much of the country. The earthquake triggered thousands of landslides that were broadly scattered across the southern half of the country. The most damaging landslide, a rapidly moving mass of ∼130,000 m 3 , occurred in the Las Colinas neighborhood of Santa Tecla, where ∼585 people were killed. Another large landslide (∼750,000 m 3 ) near the city of San Vicente blocked the Pan-American Highway for several weeks. One month later, on 13 February, a M-6.6 earthquake occurred ∼40 km east-southeast of San Salvador and triggered additional thousands of landslides in the area east of Lake Ilopango. The landslides were concentrated in a 2500 km 2 area and were particularly abundant in areas underlain by thick deposits of poorly consolidated, late Pleistocene and Holocene Tierra Blanca rhyolitic tephras erupted from Ilopango caldera. Most of the triggered landslides were relatively small, shallow failures, but two large landslides occurred that blocked the El Desagüe River and the Jiboa River. The two earthquakes triggered similar types of landslides, but the distribution of triggered landslides differed because of different earthquake source parameters. The large-magnitude, deep, offshore earthquake triggered broadly scattered landslides over a large region, whereas the shallow, moderate-magnitude earthquake centered within the country triggered a much smaller, denser concentration of landslides. These results are significant in the context of seismic-hazard mitigation for various earthquake scenarios.

Abstract The Alani-Paty landslide has damaged streets, utilities, and homes built on a debris apron in Honolulu, Oahu, Hawaii. Failure of weathered, crudely stratified, highly plastic, debris-apron deposits has created several similar landslides in southeastern Oahu. The Alani-Paty landslide affects about 60 residential lots. It is about 300 m long, 160 m wide, 7-10 m thick, and consists of two main kinematic elements that are separated by a right-lateral shear zone. One element has moved about 4 m, mainly by translation, down a slope of about 12°, and the adjacent element has moved about 3 m down a slope of 9°. Longitudinal stretching in the upslope third and shortening in the downslope two-thirds characterize deformation in each element; landslides in Ohio, Utah, and Colorado have deformed similarly. Smectite-rich clay layers within the deposits are medium to stiff, and measured angles of residual friction range from 6° to 11° with cohesion intercepts less than 12.5 kPa. Saturated hydraulic conductivity within the landslide decreases with depth; below the slip surface, the hydraulic conductivity increases. Rainfall infiltrates at the ground surface, percolates downward and perches on the zone of low hydraulic conductivity near the slip surface, keeping the slide mostly saturated year round. The main body of the landslide moves during rainy periods, when the ten-day average rainfall exceeds 25 mm/day and the pore-water pressures in the upslope quarter of the landslide increase 10-30 kPa. Pore pressure increases within the landslide occur 1-2 days following the onset of rainfall and result from infiltration of rainfall and runoff; after materials above the perennial water table become saturated, downward propagating pressure waves triggered by bursts of intense rainfall produce further, short-lived increases in pore pressure. This elevated pore pressure at the slip surface triggers movement. The ground-water response in the upslope quarter of the landslide is relatively rapid compared to responses in other landslides described in the literature.