I-82 Prosser Landslide Investigation Benton City, WA

The scope of this technical report is to establish the mechanisms by which the eastbound lanes of Interstate 82 at mile post (MP) 91.9 near Benton City continue to deform. Within the Washington State Department of Transportation (WSDOT), the area is known as the Prosser Landslide and has been an ongoing concern since the 1980s. Results from previous technical investigations have been conflicted or inconclusive as to whether landslide movement persists beneath or through the shear key-buttress or that pavement distress is related to swelling of a clay-rich unit that underlies the slope and interstate. For this report, the following steps were taken. First, I conducted a desk review of archived reports, memos, data, and drill logs from the original construction of I-82 and previous geotechnical investigations commissioned by WSDOT. Findings of this desk review are reported in Part III. Second, WSDOT drillers drilled two new boreholes at the Prosser Landslide site above the buttress and instrumentation was installed within the boreholes. Borehole logs produced from the 2013 drilling can be found in Appendix A of this report. Material retrieved from the suspected failure zone during drilling was tested at the WSDOT Materials Lab by WSDOT personnel for its mechanical properties including Atterberg limits, grain-size analysis, and residual shear strength (Appendix B). Samples were also analyzed for mineral content using X -ray powder diffraction (XRD). These data and observations are reported in Part III and Appendix C. Finally, using drill logs produced by WSDOT from the latest drilling and from historic drilling campaigns, I constructed a 2-D geologic model of the landslide site. This model is the basis for slope stability analysis reported in Part IV and Appendix D. This study concludes that the deformation observed in the eastbound lanes of I-82 could be the result of continued landslide movement, despite previous remediation efforts.

Shallow Landslide Risk Assessment of the Westerly Bluffs of the 7.5-minute Freeland Quadrangle, Whidbey Island, Washington

Landslides often occur on slopes rendered unstable by underlying geology, geomorphology, hydrology, weather-climate, slope modifications, or deforestation. Unfortunately, humans commonly exacerbate such unstable conditions through careless or imprudent development practices. Due to local geology, geography, and climatic conditions, Puget Sound of western Washington State is especially landslide-prone. Despite this known issue, detailed analyses of landslide risks for specific communities are few. This study aims to classify areas of high landslide risk on the westerly bluffs of the 7.5 minute Freeland quadrangle based on a combined approach: mapping using LiDAR imagery and the Landform Remote Identification Model (LRIM) to identify landslides, and implementation of the Shallow Slope Stability Model (SHALSTAB) to establish a landslide exceedance probability. The objective is to produce a risk assessment from two shallow landslide scenarios: (1) minimum bluff setback and runout and (2) maximum bluff setback and runout. A simple risk equation that takes into account the probability of hazard occurrence with physical and economic vulnerability (van Westen, 2004) was applied to both scenarios. Results indicate an possible total loss as much as $32.6b from shallow landslides, given a setback of 12 m and a runout of 235 m.

Using Terrestrial LiDAR to Monitor Erosion within the Gold Basin Landslide Complex, Verlot, WA

In 2014 the United States Forest Service closed the Gold Basin Campground of western Washington in an effort to protect the public from unstable hillslopes directly adjacent to the campground. The Gold Basin Landslide Complex (GBLC) is actively eroding via block fall, dry ravel, and debris flows, which contribute sediment into the South Fork of the Stillaguamish River. This sediment diminishes the salmonid population within the South Fork of the Stillaguamish River by reducing habitable spawning grounds, which is a big concern to the Stillaguamish Tribe of Indians. In this investigation, I quantified patterns of degradation and total volume of sediment erosion from the middle lobe of the GBLC over the period of July 2015 through January 2016 using terrestrial (ground-based) LiDAR (TLS). I characterized site specific stratigraphy and geomorphic processes, and laid the groundwork for future, long-term monitoring of this site. Results of this investigation determined that ~ 4,800m3 of sediment was eroded from the middle lobe of the GBLC during the 6 month study period (July 2015 – January 2016). This erosion likely occurred from debris flows, raveling of poorly sorted sand and gravel deposits and block failures of high plasticity silts and clays, and/or other mass wasting mechanisms. The generalized stratigraphic sequence in the GBLC consists of alternating massive beds of sand and gravel with silts and clays. The low permeability of these silts and clays provide a perfect venue for groundwater to percolate, as I observed during field investigations, which likely contributes to the active instability of the hillslopes. Continued monitoring and mapping of this complex will lead to viable information that could help both the United States Forest Service and the Stillaguamish Tribe.

The Ledgewood-Bonair Landslide toe and where did it go? A case study in littoral sediment budgets from a deep-seated landslide in Island County, Washington

On the morning of March 27th, 2013, a small portion of a much larger landslide complex failed on the western shoreline of central Whidbey Island, Island County, Washington. This landslide, known as the Ledgewood-Bonair Landslide (LB Landslide), mobilized as much as 150,000 cubic meters of unconsolidated glacial sediment onto the coastline of the Puget Sound (Slaughter et al., 2013, Geotechnical Engineering Services, 2013). This study aims to determine how sediment from the Ledgewood-Bonair Landslide has acted on the adjacent beaches 400 meters to the north and south, and specifically to evaluate the volume of sediment contributed by the slide to adjacent beaches, how persistent bluff-derived accretion has been on adjacent beaches, and how intertidal grain sizes changed as a result of the bluff-derived sediment, LiDAR imagery from 2013 and 2014 were differenced and compared to beach profile data and grain size photography. Volume change results indicate that of the 41,850 cubic meters of sediment eroded at the toe of the landslide, 8.9 percent was redeposited on adjacent beaches within 1 year of the landslide. Of this 8.9 percent, 6.3 percent ended up on the north beach and 2.6 percent ended up on the south beach. Because the landslide deposit was primarily sands, silts, and clays, it is reasonable to assume that the remaining 91.1 percent of the sediment eroded from the landslide toe was carried out into the waters of the Puget Sound. Over the course of the two-year study, measurable accretion is apparent up to 150 meters north and 100 meters south of the landslide complex. Profile data also suggests that the most significant elevation changes occurred within the first two and half months since the landslides occurrence. The dominant surficial grain size of the beach soon after the landslide was coarse-sand; in the years following the landslide, 150 meters north of the toe the beach sediment became finer while 100 meters south of the toe the beach sediment became coarser. Overall, the LB Landslide has affected beach profile and grain size only locally, within 150 meters of the landslide toe.

A comparison of landslide inventories along the shoreline of Steamboat Island Peninsula, Thurston County, WA: office based protocol vs. field based protocol

The mountain ranges and coastlines of Washington State have steep slopes, and they are susceptible to landslides triggered by intense rainstorms, rapid snow melts, earthquakes, and rivers and waves removing slope stability. Over a 30-year timespan (1984-2014 and includes State Route (SR) 530), a total of 28 deep-seated landslides caused 300 million dollars of damage and 45 deaths (DGER, 2015). During that same timeframe, ten storm events triggered shallow landslides and debris flows across the state, resulting in nine deaths (DGER, 2015). The loss of 43 people, due to the SR 530 complex reactivating and moving at a rate and distance unexpected to residents, highlighted the need for an inventory of the stateís landslides. With only 13% of the state mapped (Lombardo et al., 2015), the intention of this statewide inventory is to communicate hazards to citizens and decision makers. In order to compile an accurate and consistent landslide inventory, Washington needs to adopt a graphic information system (GIS) based mapping protocol. A mapping protocol provides consistency for measuring and recording information about landslides, including such information as the type of landslide, the material involved, and the size of the movement. The state of Oregon shares similar landslide problems as Washington, and it created a GIS-based mapping protocol designed to inform its residents, while also saving money and reducing costly hours in the field (Burns and Madin, 2009). In order to determine if the Oregon Department of Geology and Mineral Industries (DOGAMI) protocol, developed by Burns and Madin (2009), could serve as the basis for establishing Washingtonís protocol, I used the office-based DOGAMI protocol to map landslides along a 40-50 km (25-30 mile) shoreline in Thurston County, Washington. I then compared my results to the field-based landslide inventory created in 2009 by the Washington Division of Geology and Earth Resources (DGER) along this same shoreline. If the landslide area I mapped reasonably equaled the area of the DGER (2009) inventory, I would consider the DOGAMI protocol useful for Washington, too. Utilizing 1m resolution lidar flown for Thurston County in 2011 and a GIS platform, I mapped 36 landslide deposits and scarp flanks, covering a total area of 879,530 m2 (9,467,160 ft2). I also found 48 recent events within these deposits. With an exception of two slides, all of the movements occurred within the last fifty years. Along this same coastline, the DGER (2009) recorded 159 individual landslides and complexes, for a total area of 3,256,570 m2 (35,053,400 ft2). At a first glance it appears the DGER (2009) effort found a larger total number and total area of landslides. However, in addition to their field inventory, they digitized landslides previously mapped by other researchers, and they did not field confirm these landslides, which cover a total area of 2,093,860 m2 (22,538,150 ft2) (DGER, 2009). With this questionable landslide area removed and the toes and underwater landslides accounted for because I did not have a bathymetry dataset, my results are within 6,580 m2 (70,840 ft2) of the DGERís results. This similarity shows that the DOGAMI protocol provides a consistent and accurate approach to creating a landslide inventory. With a few additional modifications, I recommend that Washington State adopts the DOGAMI protocol. Acquiring additional 1m lidar and adopting a modified DOGAMI protocol poises the DGER to map the remaining 87% of the state, with an ultimate goal of informing citizens and decision makers of the locations and frequencies of landslide hazards on a user-friendly GIS platform.

Geology and landslide geomorphology of the Burpee Hills, Skagit County, Washington, USA

Landforms within the Skagit Valley record a complex history of land evolution from Late Pleistocene to the present. Late Pleistocene glacial deposits and subsequent incision by the Skagit River formed the Burpee Hills terrace. The Burpee Hills comprises an approximately 205-m-thick sequence of sediments, including glacio-lacustrine silts and clays, overlain by sandy advance outwash and capped by coarse till, creating a sediment-mantled landscape where mass wasting occurs in the form of debris flows and deep-seated landslides (Heller, 1980; Skagit County, 2014). Landslide probability and location are necessary metrics for informing citizens and policy makers of the frequency of natural hazards. Remote geomorphometric analysis of the site area using airborne LiDAR combined with field investigation provide the information to determine relative ages of landslide deposits, to classify geologic units involved, and to interpret the recent hillslope evolution. Thirty-two percent of the 28-km2 Burpee Hills landform has been mapped as landslide deposits. Eighty-five percent of the south-facing slope is mapped as landslide deposits. The mapped landslides occur predominantly within the advance outwash deposits (Qgav), this glacial unit has a slope angle ranging from 27 to 36 degrees. Quantifying surface roughness as a function of standard deviation of slope provides a relative age of landslide deposits, laying the groundwork for frequency analysis of landslides on the slopes of the Burpee Hills. The south-facing slopes are predominately affected by deep-seated landslides as a result of Skagit River erosion patterns within the floodplain. The slopes eroded at the toe by the Skagit River have the highest roughness coefficients, suggesting that areas with more frequent disturbance at the toe are more prone to sliding or remobilization. Future work including radiocarbon dating and hydrologic-cycle investigations will provide a more accurate timeline of the Burpee Hills hillslope evolution, and better information for emergency management and planners in the future.

Mapping Whidbey Island and Possession Sound landslide susceptibility using an additive linear model

Senior thesis written for Oceanography 445

Sources and transport of surface sediment and metals in B.C. fjord, Tahsis Inlet

Senior thesis written for Oceanography 445

Vasa Creek Channel Conditions and Slope Stability, Bellevue, Washington

The focus of this report is is the channel conditions at Vasa Creek, Bellevue, Washington, with regard to kokanee habitat and slope stability. This required a geomorphic and geologic assessment of the stream and riparian corridor along Vasa Creek. I focused my efforts in a 720m study-reach just south of I-90 in which City of Bellevue had no information. My assessment is divided into 3 categories: channel morphology, geology, and landslide hazards. I described the channel morphology by determining the gradient of the channel, longitudinal and cross-channel geometries, grain size distribution, embeddedness observations, type of channel reaches present, and the locations of significant in-channel woody-debris, landslides, scarps, landslide debris, and erosional features. This was done by conducting a longitudinal survey, 7 cross-channel surveys, pebble counts, and visual observations with the aid of a GPS device for mapping. I completed my geological assessment using both field observations and borehole data provided by GeoMapNW. Borehole data provided logs of the subsurface material at specific locations. In the field, I interpreted local geology using material in the channel as well as exposures in the adjacent slope. I completed the landslide hazard assessment using GIS methods supplemented by field observations. GIS methods included the use of aerial LiDAR to discern slope values and locations of features. Features of interest include the locations of scarps, landslides, landslide debris, and erosional features which were observed in the field. I classified 4 slope classes using ArcMap10 along with the locations of previously mapped landslides, scarps, and landslide debris. I describe the risk of slope failure according to the Washington Administration Code definition of critical areas (WAC 365-190-120 6a-i). My results are presented in the form of a map suite containing a channel morphology map, geology map, and landslide hazard map. The channel is a free-formed alluvial plane-bed reach with infrequent step-pools with riffles associated with landslide debris that chokes the channel. Overall I found that there is not the potential for kokanee habitat due flashy behavior (sudden high flow events), landslide inundation, and a lack of favorable conditions within the channel. The updated geologic map displays advance outwash deposits and alluvium present within the study-reach, as opposed to exposures of the Blakeley Formation along with other corrections from borehole data interpretations. The landslide hazard map shows that there are areas at high risk for slope failure along the channel that should be looked into further.

The Mineralogy and Strength Characteristics of Selected Glaciolacustrine Clays in the Puget Sound Region

The combination of rainy climate, glaciolacustrine clays, and steep topography of the Puget Lowland creates slope stability issues for the regional population. Several glaciolacustrine deposits of laminated silt and clay of different ages contribute to the likelihood of slope failure. The glaciolacustrine deposits are generally wet, range in thickness from absent to >30m, and consist of laminated silt and clay with sand interbeds at the tops and bottoms, sandy laminae throughout the deposits, occasional dropstones and shear zones. The glaciolacustrine deposits destabilize slopes by 1) impeding groundwater flow percolating through overlying glacial outwash sediments, 2) having sandy laminae that lower strength by increasing pore pressure during wet seasons, and 3) increasing the potential for block-style failure because of secondary groundwater pathways such as laminae and vertical fractures. Eight clay samples from six known landslide deposits were analyzed in this study for their mineralogy, clay fraction and strength characteristics. The mineralogy was determined using X-ray Diffractometry (XRD) which revealed an identical mineralogic suite among all eight samples consisting of chlorite, illite and smectite. Nonclay minerals appearing in the X-ray diffractogram include amphibole and plagioclase after removal of abundant quartz grains. Hydrometer tests yielded clay-size fraction percentages of the samples ranging from 10% to 90%, and ring shear tests showed that the angle of residual shear resistance (phi_r) ranged from 11° to 31°. Atterberg limits of the samples were found to have liquid limits ranging from 33 to 83, with plastic limits ranging from 25 to 35 and plasticity indices ranging from 6 to 48. The results of the hydrometer and residual shear strength tests suggest that phi_r varies inversely with the clay-size fraction, but that this relationship was not consistent among all eight samples. The nature of the XRD analysis only revealed the identity of the clay minerals present in the samples, and provided no quantitative information. Thus, the extent to which the mineralogy influenced the strength variability among the samples cannot be determined given that the mineral assemblages are identical. Additional samples from different locations within each deposit along with quantitative compositional analyses would be necessary to properly account for the observed strength variability.

Sources and transport of surface sediment and metals in B.C. fjord, Tahsis and Zeballos Inlets

Senior thesis written for Oceanography 445