Landslide Mitigation Action Plan

Final Report
2014
Washington State Department of Transportation (WSDOT)
Rail Division

Contact information:

David Smelser
WSDOT Cascades High Speed Rail Program Manager
Olympia, WA 98504
360-705-6916
[email protected]
Acknowledgements

The Landslide Mitigation Action Plan was compiled by the Washington State Department of
Transportation (WSDOT) with participation from the following entities:

BNSF Railway Company (BNSF)

Sound Transit
Amtrak
National Oceanic Atmospheric Administration (NOAA)
Governors Office of Regulatory Innovation and Assistance (ORIA)
Washington State Department of Ecology (Ecology)
Washington State Department of Natural Resources (DNR)
Snohomish County
City of Everett
City of Mukilteo
City of Shoreline
City of Seattle
City of Edmonds
Town of Woodway
Port of Everett
Executive Summary
Each year, landslides along the Pacific Northwest Rail Corridor interrupt rail service for
passenger and freight trains. High numbers of landslides between Seattle and Everett have been
especially problematic for Sound Transit commuters and Amtrak Cascades passengers. Record
numbers of service interruptions (sum of annulments and disruptions for all passenger trains)
during the 2012-2013 winter season prompted collaboration among Washington State
Department of Transportation (WSDOT), BNSF Railway Company, Sound Transit, Amtrak, and
stakeholders to quantify the landslide-related impacts, identify the primary factors within the
corridor that contribute to landslides, and develop mitigation strategies to reduce the occurrence
and impact of landslides.

WSDOT created the Landslide Mitigation Work Group and convened bi-weekly meetings over a
nine-month period. The mission of the Work Group was to develop short- and long-term
strategies to reduce landslide impacts and improve transportation reliability throughout the
corridor.

Documented landslide impacts for Sound Transit commuters and Amtrak passengers include
direct costs, such as annulments (cancelation of trains), busing customers around the closure
area, loss of ridership; and costs to BNSF for landslide debris cleanup. Indirect costs are also
substantial but harder to quantify, and may include declining ridership due to perceived
unreliability of winter service, devaluation of property values and subsequent loss of tax revenue,
loss of commercial productivity, and increased congestion on roads when rail service is
interrupted.

The majority of landslides that impact the rail line are shallow in depth and are sensitive to well-
established factors and conditions. These factors include heavy or prolonged precipitation during
the rainy season; the steep, high slopes that are prevalent along the corridor; underlying geology
frequently associated with shallow landslides; and poor slope management practices carried out
by adjacent landowners, such as discharging stormwater above or on steep slopes and disposing
of yard, construction and earthen debris onto slopes. Commonly, it is a combination of factors
that converge to start landslides.

Potential strategies to reduce landslide interruptions and impacts were explored by the Work
Group. Strategies were outlined and evaluated for implementation time, complicating factors,
and short, moderate-, and long-term effectiveness to reduce or prevent landslides.

The Work Group recognizes that measurable long-term reduction in landslide-related impacts to
passenger service will require substantial investments in capital improvement projects.
Depending on the financial resources available, as well as factors such as permitting, design, and
construction scheduling, the time required to achieve significant reductions in landslide-related
service interruptions will likely take one or more decades.

i
Key Findings
Short-term, low-cost strategies include:
Develop education and public outreach to engage adjacent landowners to improve slope
management practices.
Continue low-cost mitigation options, such as maintenance of slide fences, ditches and
other drainage facilities.
Provide a drainage improvement incentive, such as reduced permit fees from BNSF to
adjacent landowners (limited duration).
Review landslide data through 2007 and develop landslide maps to be completed during
the fall 2013. Inventory can be used to develop detailed landslide hazard maps to assist
local agencies in the development of land use regulations on steep slopes.

Intermediate strategies include:

Research and implement a landslide potential assessment model to inform decisions
between agencies and provide additional time for contingency planning; model validation
is targeted for the 2014-2015 rainy seasons.
Design and construct up to six projects in high-priority landslide areas from 2013-2016 to
mitigate landslide problems and improve service reliability.

Long-term strategies include:

Continue community education and public outreach.
Develop a permit process for improvements to private residential land adjacent to and/or
above the track area, and identify a funding source or sources to implement
improvements.
Explore solutions for long-term slide debris removal and restoration process, such as
beach nourishment.
Optimize design of containment structures and evaluate effectiveness of stabilization
measures for shallow slope failures.
Develop a management system to prioritize and implement slope stabilization projects.
Consider acquisition of additional right-of-way or long-term maintenance/construction
easements on adjacent property in landslide-prone sections to improve opportunities to
implement best-suited mitigation measures. (Note that this long-term strategy would
require additional funding.)
Explore justification for further public investment, recognizing that a significant increase
in capital investment will be required to significantly reduce landslide-related closures.

ii
Table of Contents
Executive Summary ....................................................................................................... i
Key Findings .......................................................................................................................... ii
Introduction ................................................................................................................... 1
Effect of the Plan.....................................................................................................................2
Work Group Coordination .......................................................................................................2
Roles and Responsibilities .......................................................................................................2
Study Schedule, February-September 2013, and Process .........................................................3
Landslide Impacts ......................................................................................................... 4
Service and Cost Impacts of Landslides ...................................................................................5
Indirect Impacts/Costs .............................................................................................................6
Currently Funded Capital Projects ......................................................................................... 10
Current Practice of Managing Landslide Impacts................................................................... 10
Contributing Factors to Landslides ........................................................................... 11
Climatic ................................................................................................................................ 11
Shape and Condition of Slope ............................................................................................... 12
Human Activity ..................................................................................................................... 14
Implications .......................................................................................................................... 15
Typical Mitigation Strategies...................................................................................... 16
Stabilization (Capital Improvement Projects)......................................................................... 16
Protection .............................................................................................................................. 18
Avoidance ............................................................................................................................. 18
Maintenance and Monitoring ................................................................................................. 18
Proactive Versus Reactive Mitigation Strategies .................................................................... 19
Strategies to Reduce Landslide-Related Interruptions and Impacts ...................... 21
Complicating Factors for Landslide Reduction ...................................................................... 22
Implementation Plan Status ....................................................................................... 25
Short-Term Improvement Strategies ...................................................................................... 25
Intermediate Strategies .......................................................................................................... 26
Long-Term Strategies and Recommendations ........................................................................ 26
References ................................................................................................................... 28
Appendices .................................................................................................................. 29

iii
Introduction
Frequent landslides along the railroad corridor, especially between Seattle and Everett during the
wet winter season pose periodic service interruptions for passengers on the Amtrak Cascades,
Amtrak Long Distance, and Sounder. Landslides result in rail closures and emergency project
activities every year, particularly during the rainy season from October to April. Disruption of
rail service within the Seattle to Everett corridor has been especially problematic, with a record
number of annulled and
disrupted daily passenger
trains (sum of both
Sounder and Amtrak
Cascades trips) due to
landslides in 2013.

At the request of the

Washington State Study Corridor
Secretary of
Transportation, the
Washington State
Department of
Transportation (WSDOT)
initiated a joint work group
effort with BNSF Railway
Company (BNSF),
Amtrak, Sound Transit,
and local jurisdictions and
stakeholders called the
Landslide Mitigation Work
Group. The mission was to
investigate contributing
factors to landslides within
the corridor and determine
a path to solutions.

The Work Group

developed the Landslide Figure 1. Landslide Study Corridor
Mitigation Action Plan to
evaluate causes of
landslides within this 26.6-
mile-long railway corridor
(Figure 1), and form reasonable mitigation strategies to reduce impacts to the traveling public.
The extent of the study area was defined by the high frequency of events. Landslides within the
study corridor are triggered by a combination of factors including climatic/hydrologic factors
(e.g., heavy or prolonged precipitation during the rainy season), geomorphic factors (i.e., steep
topography), geologic conditions and impacts from human activities.

1
Effect of the Plan
The purpose of this Plan:
Document potential improvement strategies.
Identify actions to minimize impacts to traveling public.
Identify recommended actions for measureable improvements in interruptions due to
landslides.

This Plan is not intended to:

Guarantee landslides will not occur in the corridor.
Prevent other government agencies or group members from advocating a particular
improvement.
Provide funding for proposed action strategies.

Work Group Coordination

Rail transportation is dependent on partnerships among government agencies, private industry
and other stakeholders. The Work Group was a cooperative effort with WSDOT, BNSF, Sound
Transit, Amtrak and local jurisdictions/stakeholders within the study corridor, such as the
Washington Department of Natural Resources, Washington Department of Ecology, National
Oceanic Atmospheric Administration, and Governors Office of Regulatory Innovation and
Assistance. Local jurisdictions include the cities of Everett, Mukilteo, Shoreline, Edmonds, and
Seattle; the town of Woodway; and Snohomish County. The group implemented a reasonable
strategy to identify contributing factors to landslides within the corridor, develop conclusions
based on research, and create an implementation plan with recommendations for measurable
improvements to the traveling public.

Roles and Responsibilities

The WSDOT Rail Division sponsors the Amtrak Cascades and its intercity passenger rail
service along the Pacific Northwest Rail Corridor, contracting with Amtrak as the service
provider. Sound Transit and Amtrak contract with BNSF for track use.

In WSDOTs Amtrak operating agreement, Amtrak is responsible for operating the

Amtrak Cascades service.
BNSF and Amtrak notify WSDOT of operational changes.

BNSF owns and maintains the rail rights of way and track structures. BNSF is
responsible for maintaining the railway infrastructure in compliance with Federal
Railroad Administration safety standards. As the owner of the track, BNSF is responsible
for addressing landslides within the BNSF right of way (ROW) only. However, landslide
stabilization projects must often be constructed, at least in part, on property outside of
BNSF-owned ROW to be effective, as a majority of landslide activity in this corridor
originates from above and off BNSF property.

2
 Local agencies within the corridor are responsible for permitting development activities
in geologically hazardous and/or sensitive areas (such as steep or unstable slopes) within
their jurisdictions. This includes, but is not limited to, vegetation management and
implementing development standards, such as building setbacks from steep slopes/bluffs,
defining and communicating stormwater runoff requirements, erosion/sediment control
during construction and communicating seasonal restrictions during the rainy season.

Study Schedule, February-September 2013, and Process

February Develop framework for final product; initiate data collection.

March Data collection and documentation.

April Interim report: data collection; develop action strategies.

May Continue development of action strategies; prioritize action strategies.

June Interim report: immediate action strategies.

July-August Implement immediate action strategies; draft final report.

September Final report; executive summary; recommended solutions.

3
Landslide Impacts
Washington State supports a rail system that is
integral to maintaining our economy,
environment and quality of life. The rail system
provides transportation for freight rail (BNSF),
commuter rail (Sound Transit), intercity
passenger rail (Amtrak Cascades), and long
distance passenger rail (Amtrak).

Washington and Oregon jointly sponsor Amtrak

Cascades, a 467-mile-long regional service that
operates between Eugene, Ore., and Vancouver,
British Columbia (Figure 2). Since 2000 Sound
Transit has been operating a system of express
buses, commuter rail and light rail to provide
faster, more dependable ways to commute within
the counties of Snohomish, King and Pierce.
Sound Transit uses a portion of the BNSF line to
provide daily commuter rail service between
Everett and Seattle.

More than 60 areas along the 467-mile-long

Amtrak Cascades route have been identified as at
risk for landslides. However, the majority of
landslides occur within a 26.6-mile-long corridor,
from north Seattle to Everett along steep coastal
bluffs. Since 1914, more than 900 blocking
landslides have occurred along the Seattle-Everett
rail corridor, with 5.5 miles of quarter-mile
sections experiencing 10 or more blocking
landslides (Appendix A).

Figure 2. Amtrak Cascades Route

4
Service and Cost Impacts of Landslides
Landslides not only present risks to passenger service operations, but also have social and
economic effects. Landslides can destroy or damage residential and commercial developments
and agricultural areas, and negatively affect water quality in rivers, lakes and the Puget Sound.
Increased development in landslide prone areas, deforestation and precipitation can all contribute
to higher landslide activity (Schuster 1996).

Direct costs of landslides, such as repair, replacement or maintenance, are more easily identified
than indirect costs, such as loss of property values, loss of tax revenue, loss of commercial
productivity and adverse effects to water quality (Schuster 1996). The Work Group evaluated
direct and indirect costs of landslides within the study corridor.

Direct Costs
Direct costs include capital improvement projects and maintenance costs, such as debris cleanup
and disposal. In most instances, BNSF must dispose of landslide debris offsite. Since 2008 direct
costs for BNSF, as a result of landslide impacts, are estimated at more than $10 million
(Table 1). This does not include losses associated with freight train delays.

Table 1. BNSF Railway Landslide Related Costs

Year Expenditures
2013* $4,041,000
2012 $2,442,000
2011 $796,000
2010 $2,628,000
2009 $374,000
2008 $110,000
* Data through May 2013.

In addition to BNSF capital improvement projects, WSDOT has provided approximately $6.3
million of federal funding for landslide mitigation efforts, with an additional $92,000 directly
from state funds. These expenditures represent progress on expected project costs budgeted at
$16.1 million in federal dollars and $304,000 in state funds.

The Port of Everett identified direct impacts from landslides in the corridor that included
property damage and interruption of seaport operations. For example, the Port spent significant
money cleaning and repairing stormwater treatment facilities (bioswales) and cleaning a public
access trail and Terminal Avenue due to slide damage (Figure 3). The Port cited difficulty
maintaining compliance with stormwater permit conditions when treatment facilities fill with
landslide debris. Landslide debris that spills across Terminal Avenue also impacts cargo staging
areas, construction projects and access to land needed for operations.

5
Figure 3. Terminal Avenue train car derailed by landslide (photograph courtesy of Port of Everett).

Indirect Impacts/Costs
Indirect costs for Amtrak Cascades and Sounder Commuter Rail within the study corridor
include disruptions to service and subsequent loss of ridership. Record numbers of service
disruptions (total number of cancelled or disrupted passenger trips for Amtrak Cascades and
Sounder) occurred during the 2012-2013 season. Costs to the local communities include direct
loss of property, devaluation of property, higher insurance costs for homeowners along the bluff,
and homeowner costs for repairs and/or prevention.

Local jurisdictions, such as the cities of Mukilteo and Everett, identified commuter disruption,
impacting time lost to the individual, as well as increased roadway congestion. The impact to
property owners can include direct loss of property, but also the expense of repair and/or
construction, permitting costs and emotional impact. Some property owners lose access to their
property, which requires time, money and effort to repair. For property owners without resources
to fix the damages, funding is not available and they are profoundly affected.

In addition, the disruption of rail service from a catastrophic event can greatly impact the local
and regional economy. These impacts affect the private sector and all governmental agencies,
from smaller entities to the state level.

6
Amtrak Cascades
Amtrak Cascades trains have been impacted by landslides since daily intercity passenger rail
service was re-established between Seattle and Vancouver, B.C. in May 1995. Since 2009,
WSDOT maintained detailed data on service impacts resulting from landslides. These service
impacts occur in two ways:

1. Trains are canceled and do not operate over any portion of their scheduled route. These
service impacts are called annulments.
2. Trains operate over a portion of their route, with buses deployed to cover one or more
segments of impacted areas between cities. These service impacts are called disruptions.

Seasonal service impacts from 2009 to 2013 ranged from 20 to 71 annulments, and 27 to
104 seasonal disruptions during the season (October-June) from 2009-2013 (Table 2).

Table 2. Amtrak Cascades Seasonal Annulments and Disruptions from 2009-2013

October June Annulments Disruptions

2012 - 2013 50 81
2011 - 2012 23 31
2010 - 2011 71 104
2009 - 2010 20 27

Calculating the financial impacts during service annulments and disruptions is challenging
because many factors influence a persons decision to ride Amtrak Cascades (ticket prices,
automobile fuel prices and on-time performance of train service). The calculation of financial
impacts is further complicated by the fact that travelers holding tickets when a landslide occurs
will still be transported to their destination by either a bus or a combination of a bus and a train.

A comparison between ridership and revenue data for Amtrak Cascades trains between Seattle
and Everett for the past four seasons showed a precipitous drop (20 to 35 percent for major city
pairs) in ridership and revenue from 2012-2013 (Table 3). While this decline in ridership and
revenues was observed in most of Amtraks national network during April 2013, customers may
have chosen not to ride the trains due to concerns for their safety after Amtraks long-distance
Empire Builder train was partially derailed by a landslide near Everett, specifically on April 7,
2013.

7
Table 3. Amtrak Cascades Trains 510, 513, 516 and 517-Ridership and Revenue 2009-2013

October-June Ridership Revenue

2012 - 2013 143,676 $5,860,420
2011 - 2012 163,207 $6,540,335
2010 - 2011 160,275 $6,052,903
2009 - 2010 162,995 $6,018,360

Sounder Commuter Rail

Sounder Commuter Rail, operated by Sound Transit, started its north line service between
Everett and Seattle in December 2003 with a single daily round trip. Landslides began to
significantly impact Sounder service in the 2005-2006 winter with 10 days of cancelled service
and 40 annulments. All but one winter since 2008-2009 has experienced service disruptions from
landslides with the number of impacts growing as service increased from one to four daily round
trips. In the 2012-2013 winter, 28 days of Sounder rail service were disrupted, resulting in 206
annulments (Table 4).

Table 4. Sounder Commuter Rail Seasonal Annulments, Days Impacted, and Daily Trips
Scheduled from 2003-2013

October June Annulments Days Impacted Daily Trips

Scheduled
2012 2013 206 27.5 8
2011 2012 41 7 8
2010 2011 70 9 8
2009 2010 24 3 8
2008 2009 0 0 8
2007 2008 18 3 6
2006 2007 16 4 4
2005 2006 40 10 4
2004 2005 0 0 2
2003 2004 3 2 2

When Sounder service is cancelled, customers are directed to special bus transportation that
Sound Transit arranges to transport riders to Sounder stations. These buses augment existing bus
service, which are often overloaded from absorbing the additional commuters unable to commute
by rail transit. There are also occasions when limited partial service is offered (i.e., morning or
afternoon train, or a train to one or two stations not impacted by slide activity), rather than
cancelling an entire train. For instance, if landslides occur north of Mukilteo, service may be
possible between Seattle and Edmonds/Mukilteo, but not Everett. On these occasions, the
replacement bus service is only required for customers that travel between Seattle and Everett.

The largest financial impact to Sounder north line service as a result of landslides is lost farebox
revenue from declining ridership. However, quantifying these financial impacts is challenging

8
because it is unknown how many customers chose not to ride Sounder rail after a particular
landslide event has impacted service. Other than the original $368 million provided to BNSF for
the permanent easements and track improvements necessary to meet track-capacity requirements,
as well as station construction, Sound Transit does not have additional capital investments in the
corridor beyond what was approved in the 1996 Sound Move ballot measure. In 2008, voters
approved a second platform and other station access improvements at the Mukilteo facility in the
Sound Transit 2 ballot measure. Additional operating costs are incurred by Sound Transit when
buses are required because of cancelled trains, which can cost several thousand dollars per day.
These costs, however, are offset by the elimination of operating costs from cancelled train trips.

In the 2010 to 2011 season, when there were 70 cancelled trips in a season, average daily
ridership decreased by approximately 10 percent, and it was more than a year before ridership
returned to previous levels. The 2012-2013 season took a particularly heavy toll on Sounder
north ridership, where 206 trips were cancelled, which nearly tripled the earlier high of 70 in the
2010-2011 season (Table 4). Although overall annual growth in Sounder ridership exceeded 10
percent during 2012, Sounder north line ridership was down 7 percent (1,215 average boardings)
in July 2013 from the October 2012 high of 1,304 average daily boardings (Table 5).

Table 5. Sounder Commuter Rail North-line Service

Year Annual Boardings Average Daily Percent Growth on

Boardings Average Daily
Boardings
2013 TBD 1,147* 6%*
2012 307,846 1,144 21%
2011 280,767 946 -9%
2010 303,060 1,024 -5%
2009 319,719 1,080 2%
2008 314,072 1,062 26%
2007 252,299 843 27%
2006 201,299 665 43%
2005 151,773 466 68%
2004 88,903 277
*YTD through June 2013 Data

Amtrak Long-Distance Service

The Amtrak Empire Builder and Coast Starlight trains operate in Washington State with a
terminal in Seattle at King Street Station. Because the landslide activity occurs primarily north of
Seattle, the Empire Builder has experienced more impacts from the landslides than the Coast
Starlight. The Amtrak long-distance train service has been impacted by landslides as long as
service has been in existence.

9
Calculating the financial impacts that occur when there are service annulments and disruptions is
challenging for the same reasons mentioned for Amtrak Cascades (i.e., other factors such as
ticket prices, automobile fuel prices and on-time performance of train service). The table below
compares ridership and revenue data for Amtrak long-distance trains that traveled within
Washington state for the past four seasons. The 2010-2013 time periods were impacted by
outages on the Empire Builder line.

Table 6. Amtrak Long Distance Trains 7, 8, 11, and 14

Ridership and Revenue Data from 2009-2013

October June Ridership Revenue

2012 - 2013 247,259 $29,615,975
2011 - 2012 243,438 $29,007,289
2010 - 2011 218,625 $25,567,097
2009 - 2010 239,832 $25,296,150

Currently Funded Capital Projects

Recently, the Federal Railroad Administration (FRA) awarded $16.1 million to WSDOT to
identify, design and construct slope stabilization improvements. WSDOT and BNSF are
collaborating on environmental and engineering work. These long-term improvements will be in
various stages of design and construction from 2013-2016.

Current Practice of Managing Landslide Impacts

As the owner of the rail corridor, BNSF is ultimately responsible for the operational and
maintenance aspects of the track structure. BNSF routinely inspects and maintains the slopes,
ditches, retaining structures and tracks to minimize impacts to railroad operations when
landslides occur. BNSF also uses an extensive network of slide fences through much of the
corridor. When the wires of a slide fence are severed by landslide debris, an indication is
provided to the BNSF dispatcher and train crews are signaled accordingly. Inspection and
monitoring of the rail corridor between Seattle and Everett is heightened during the rainy season.
When a landslide occurs that blocks one or more tracks (referred to as a blocking event), BNSF
imposes an automatic 48-hour moratorium on passenger rail service through the impacted
segment of the corridor. Alternate bus service is then deployed for riders. Impacts to riders vary,
ranging from longer commutes to missed appointments and work days.

Over the years, BNSF has invested millions of dollars in installing slide fences, building
catchment walls and widening ditches to contain the landslide debris and stabilize the slopes.

10
Contributing Factors to Landslides
The occurrence of a landslide is dependent on a combination of site-specific conditions and
influencing factors. Common factors that contribute to landslides fall into four broad categories:

1. Climatic/hydrologic (rainfall or precipitation)

2. Geomorphic (slope form and conditions - i.e., slope shape, height, steepness, vegetation
and underlying geology)
3. Geologic/geotechnical/hydrogeological (groundwater)
4. Human activity.

Climatic
Climatic factors influencing landslides include the duration
of rainfall events, intensity of rainfall, and type of Antecedent
precipitation (i.e., snow or rain), as well as rainfall conditions
conditions over a period of time (antecedent conditions).
Typically, numerous landslide events are associated with Refers to the amount of
intense and/or prolonged periods of rain (Baum et. al., rainfall that has fallen in
previous weeks, months
2000). Recorded landslides impacting the corridor largely or even years.
occurred during the winter wet season between October
and April. An example of an unusually large, deep-seated
landslide occurred in January 1997 south of Edmonds in the town of Woodway (railroad
milepost 14.80) following a two-week period of heavy precipitation (Figure 4). Some episodes of
widespread landsliding corresponded with storms involving the rapid melting of previously
accumulated snow by wind and warm rain, which is referred to as rain-on-snow storm event.
The landslide cut 50 feet into the property above, passed over the railroad tracks and knocked a
freight train into the Puget Sound.1 Many of the shallow landslides prevalent along the corridor
have occurred during a single storm event involving one or more days of intense rainfall (Baum
et al, 2000).

Figure 4: 1997 Woodway landslide.

1
www.ecy.wa.gov/programs/sea/landslides/show/woodway.html

11
Shape and Condition of Slope
Geomorphic (Slope Form)
The form and condition of a slope can affect its stability. Geomorphic factors affecting slope
form include height and steepness, as well as vegetation and underlying geology. Increased
steepness and slope height generally correlate with reduced stability. Many of the landslide-
prone slopes along the corridors are more than ten stories (100 feet) in height and quite steep
(35-45 degrees slope gradient). This steep orientation exceeds the long-term stability of the
relatively weak sediments that comprise the slopes, and such slopes or segments of slopes are
often referred to as being in an oversteepened condition. Increased slope height and the lack of
vegetative cover, especially conifers, increase the amount of rainfall that reaches the slope
surface. Vegetation generally contributes to how well the near-surface soils hold together and
thus helps resist surface erosion. Bare slopes tend to be more prone to erosion than well-
vegetated slopes. Large trees, however, can also be a detriment to localized slope stability,
where they root on steep slopes underlain by dense soils. For this reason, the presence and type
of vegetation and its contribution or detraction from stability needs to be evaluated on a site-
specific basis by qualified professionals.

Whether water infiltrates into the ground or runs off is influenced by the permeability
(porousness) of the geologic substrate, its degree of saturation (affected by antecedent
conditions) and precipitation intensity. The compact (solid) and fine-grained nature of some of
the underlying geologic units within the corridor limits infiltration and increases the likelihood of
saturating and weakening the near-surface, loosened soils. Within the corridor, this condition
commonly results in the separation and rapid transport of relatively thin, slab-like portions of the
slope, known as debris avalanches. Concentrated surface water runoff within drainages and
swales can further lead to channel-confined slope failures, involving the rapid transport of highly
fluidized debris, known as debris flows. More than 80 percent of the documented landslides
between 1914 and 2001 were shallow landslide types (debris avalanches and debris flows)
(Shannon & Wilson, 2001). Figure 5 illustrates how precipitation and groundwater can influence
the occurrence of deep-seated landslides.

12
Figure 5: This sequence of sketches shows a conceptual process that forms bluffs in the northern Puget Sound area and
causes them to retreat. More permeable soils/sediments sit on top of less permeable sediments. Water run off infiltrates
this upper layer until it meets the lower layer, where water is perched. This causes the soils at this interface to saturate
to the point of failing. Lower soil layer failure removes the support for the upper layers and they also fail (Gerstel et al.
1997).

13
Geologic/Geotechnical/Hydrogeologic (Geology and Groundwater)
The geologic conditions, and engineering (geotechnical) and groundwater (hydrogeologic)
characteristics of the geologic units that compose the slope greatly influence its stability.
Generally, the upper portions of the slopes along the corridor are underlain by a sequence of
glacial sediments deposited in advance, beneath and
during the last continental glaciation (Vashon Stade).
Fine-grained lake sediments that formed in front of and Glaciation
then compacted by the advancing ice sheet typically
underlie the coarse-grained Vashon advance deposits, and Alteration of any part of
the earths surface by
have been referred to as transitional beds (Minard, 1982, passage of a glacier, such
1983, 1985; Yount et al., 1993). These transitional beds as erosion or deposition.
are underlain by a variable sequence of very compact
interglacial deposits (called the Olympia beds and
Whidbey Formation) and older glacial deposits (known as Possession and Double Bluff Drifts),
which typically outcrop in the middle to lower portions of the slope. Of all the geologic units
within the corridor, several are recognized as bad actors over 60 percent of the landslides
reported between 1914 and 2001 originated within the transitional beds or the Whidbey
Formation (Shannon & Wilson, 2001).

Landslides also commonly recur in the same areas. Remobilized landslide debris from previous
landslides was another geologic unit significantly contributing (approximately 13 percent) to
landsliding (Shannon & Wilson, 2001). Baum et al. (2000) noted that roughly two-thirds of the
landslides generated during the winter storms of 1995-96 and 1996-97 initiated within the
bounds of mapped landslide events.

Human Activity
Human activities have repeatedly been observed to be a substantial contributor to landslides
within the corridor. These adverse and widespread activities primarily involve the discharge of
stormwater onto or above slide-prone slopes; the cutting and re-grading of slopes; and the
disposal of yard, construction, and earthen or other debris onto the upper portion of the slope
(Shannon & Wilson, 2001). In addition to these adverse
practices by adjacent landowners, the density of upslope
development, even hundreds of feet behind the top of the Transpiration
slope, has the potential to significantly contribute to
groundwater recharge through more concentrated discharge The evaporation of
of storm water runoff. This in turn has the potential to water from leaves.
adversely impact stability of the slopes along the rail
corridor.

More complex in its relationship to slope stability is the effect of removing vegetation. Rooting
depth and the interception and transpiration potential offered by mature conifers during the
winter wet season can be important contributors to stability. Conversely, the effect of wind on
mature conifers, referred to as windthrow, can disturb the substrate in which they root, resulting
in localized slope instability. For these reasons, the presence and type of vegetation and its

14
contribution or detraction from stability needs to be evaluated on a site-specific basis by
qualified professionals.

Implications
While a landslide on an adjacent slope does not always impact the rail line, about 80 percent of
the documented landslides between 1914 and 2001 generated debris that reached one or both
railroad tracks. Despite the investigation bias of this data (landslides are generally only
investigated when they might affect the tracks), the close proximity of the tracks to the base of
the steep slopes and the very limited area available for debris containment is a primary reason for
the apparent high likelihood of impact to the tracks when a landslide does occur. The volume of
debris, material and transport characteristics [i.e., material composition, velocity, viscosity
(thickness), path of travel, etc.], location of landslide initiation, and the potential to gather
additional material during transport (bulking) further influence the potential for debris run-out
onto the tracks and the extent of impacts.

Given the wide range of potential factors that influence landslide initiation characteristics, it is
virtually impossible to predict the location and impacts of a single event within such a long
landslide-prone corridor. However, of all the potential influencing factors, five factors were
judged by Shannon & Wilson (2001) to be the most differentiating in quantifying risk of
landslide-related impacts to the tracks:

1. Density of slides Number of historic landslides per quarter mile of track.

2. Catchment area Available area between the base of the slope and tracks to contain
debris.
3. Slope height Influences both debris volume and impact/run-out characteristics.
4. Geology Tendency of specific geologic units to experience landslides.
5. Line closures Percentage of total number of landslides per quarter mile of track that
impacts tracks.

Such experience is invaluable for prioritizing where and what type of future mitigation should be
considered when funding for capital improvements is available. There is ongoing research to
develop better understanding of the precise climatic conditions that have a high potential of
generating shallow landslides.

15
Typical Mitigation Strategies
There are four basic strategies to mitigate for a particular landslide:

Stabilization
Protection
Avoidance
Maintenance and monitoring

Only stabilization seeks to counter one or more key failure mechanisms and improve stability of
the slope. The latter three strategies (protection, avoidance, and maintenance and monitoring)
allow slope failure and seek to avoid, protect against or limit the associated impacts. The last
mitigation strategy, maintenance and monitoring, is different than a do-nothing alternative; a
do-nothing alternative is a management approach/decision, not a mitigation strategy.

Stabilization (Capital Improvement Projects)

Typical landslide stabilization measures include grading the unstable portion of the slope to a
lower gradient, construction of rock buttresses and retaining walls, and drainage improvements.
Examples shown below entail grading with slope armoring/buttressing (Figure 6) to address a
large deep-seated landslide at railroad milepost (MP) MP 24.5; and patterned reinforcement of
high-tensile-steel wire mesh that could potentially be used to address the abundant shallow-type
landslides that originate upslope of BNSFs ROW (Figure 7). With the exception of drainage
improvements, stabilization measures are typically moderate to high cost, but provide a long-
term solution with low, long-term maintenance costs. Cessation of adverse human activities by
diverting stormwater away from steep slopes, maintaining appropriate native vegetation, and
properly disposing of debris off-site are also considered measures that would improve stability.

p. 16
Figure 6. Recent slope reinforcement project at rail line MP 24.5 (Photographs courtesy of BNSF).

Figure 7. Slope reinforcement project in Germany involving a steep cut in highly weathered sandstone. Reinforcement
consists of high-tensile-steel wire mesh secured with patterned ground anchors, showing installation and re-vegetation
(photographs courtesy of Geobrugg).

p. 17
Protection
Protection measures for landslides primarily focus on containment and/or diversion of the
moving debris. Such measures include walls, berms, ditches and catchment basins, which can be
low to moderate in cost. However, considerable long-term maintenance costs are often associated
with these measures to clean out and dispose of accumulated debris. BNSF currently employs a
number of timber and steel containment walls (Figure 8).

Figure 8. Debris containment wall along BNSF rail line consisting of steel H piles with precast concrete lagging to
facilitate cleanout (photograph courtesy of BNSF).

Avoidance
Avoidance measures constitute a permanent solution to a landslide hazard. Measures include
realignment away from the slope, relocation of the facility, tunnels and elevated structures that
allow passage of debris beneath the facility. The typically high cost of these measures is offset
by the elimination of further landslide-related maintenance costs and exposure to landslide risk.

Maintenance and Monitoring

Maintenance and monitoring measures may involve proactive cleanout of available catchment
areas, routine observation and assessment of slope conditions, landslide-warning (slide) fences,
monitoring slope and weather instrumentation and preemptive closures. Generally, these
measures are relatively low cost and can be highly effective in reducing public exposure to slide
risk. With the exception of cleaning existing catchment areas, these measures do not reduce the
likelihood of a landslide event or the potential of landslide debris reaching the tracks. Slide
fences are used extensively through the corridor to warn of the potential for debris on the tracks

p. 18
(top of the wall in Figure 8). Another measure employed by BNSF is the passenger rail
moratorium imposed for 48 hours following a blocking event due to a landslide.

Figure 9. Slide fence on top of a wall along the BNSF right of way (photograph courtesy of BNSF).

Selection of the most appropriate mitigation strategies is influenced by many factors that often
have little relationship to the factors contributing to the landslide. Some of these include
available funds, right-of-way/property ownership, required permits, access constraints,
environmental effects and service interruption during construction.

Proactive Versus Reactive Mitigation Strategies

The mitigation strategies above can be implemented reactively or proactively. Reactive
responses are instituted at the time of failure with little to no advanced planning. Expenditures
are made when necessary, and are tailored to address actual conditions. No unnecessary
expenditures are made on slopes that might not otherwise fail and impact the facility within a
reasonable timeframe. However, reactive responses are often required at inconvenient times and
locations, and are generally more costly to construct than when the same work is performed
proactively at a more opportune time. Also, there are often more barriers to designing and
constructing what is most effective and best suited for the site under emergent conditions.
Further, direct and indirect costs/impacts especially those indirect are more difficult to
manage by relying solely on reactive responses. Problems with a reactive management approach
for unstable slope impacts to transportation facilities include high public expectations of the
reliability, convenience and safety of the system (Lowell and Norrish, 2013).

p. 19
Proactive responses, on the other hand, require considerable planning, especially when having to
choose among hundreds of landslide-prone slopes. Some of the benefits of a proactive response
generally include lower costs, better conditions to design and build under, and higher reliability.
With the responsibility of managing many unstable slopes along transportation facilities, several
public transportation departments (including WSDOT) instituted management systems for
proactively identifying, prioritizing, programming, funding and ultimately mitigating these
hazards. It is important to stress that implementation of a proactive management system to
address large numbers of landslide-prone slopes does not relieve the need for reactive responses
or eliminate the potential of further closures. When managing numerous unstable slopes, it is not
possible to predict which slope will fail first or when it will fail. In addition, program
implementation requires long-term commitments, since it can take many years to make necessary
improvements to significantly reduce landslide-related closures on such a landslide-prone
corridor. As an example, in 1974 a rock slope maintenance program was implemented along a
rail corridor in British Columbia involving 750 rock fall sites. In the opinion of the geotechnical
specialist involved since program inception, it took nearly three decades for the program benefits
to become clearly recognizable (WSDOT, 2006).

p. 20
Strategies to Reduce Landslide-Related
Interruptions and Impacts
The work group evaluated potential strategies to reduce landslide interruptions and impacts.
Strategies were outlined and evaluated for implementation time, complicating factors and
effectiveness to reduce or prevent landslides over the short-, moderate- and long-term (Table 7).
Strategies include:

1. Conduct community outreach and education:

Engage adjacent landowners to improve slope management practices.
Develop a public information campaign on best practices.
Construct demonstration projects in coordination with adjacent land owners.
Work with municipalities, Washington Department of Ecology and BNSF to streamline
slope management permit process and provide clear direction on best practices (i.e.,
stormwater, vegetation management).

2. Implement vegetation management program:

Work with adjacent landowners to identify and implement vegetation management plans
in specific areas based on recommendations from geotechnical and vegetation specialists.
Work with adjacent landowners to retain and replant native vegetation where it benefits
slope stabilization.

3. Review feasibility of improving monitoring tools:

Research available systems and tools. Representatives from participating agencies have
discussed whether monitoring tools can be developed.

4. Explore options for long-term debris disposal plan:

Evaluate beach nourishment as an option to remove slide debris. The strategy seeks to
improve near-shore habitat and ecological function, as well as to reduce the amount of
landslide debris to be removed offsite. Provides benefit for salmon restoration efforts
through the restoration of forage fish spawning habitat.
Above strategy requires collaboration with U.S. Army Corps of Engineers (Corps),
Ecology and BNSF for permitting revisions.

5. Continue maintenance and monitoring:

Proactively clean out available catchment areas and drainages.
Continue routine observation and assessment of slope conditions.
Maintain slide fences.

6. Consider acquisition of additional right-of-way or long-term maintenance/construction

easements on adjacent property in landslide-prone sections:

p. 21
 Recognizes difficulty of ensuring long-term implementation and maintenance of best
slope management practices by adjacent landowners, and that adjacent landowners may
lack resources to implement necessary improvements.
Provides opportunity to implement best-suited mitigation measures, but assumes more
responsibility.

7. Develop and maintain an inventory of landslide sites for possible implementation of a public-
domain landslide management program:
Develop inventory and a systematic hazard/risk evaluation (rating), which would be
subsequently used for project scoping and preliminary cost estimating, prioritization
(benefit-cost analysis), programming, design and final construction estimating and plan
development.
Use inventory as the basis for project selection, evaluating and justifying project merit,
long-term management of the problem and measurement of program success.
Maintain a public-domain inventory of landslides, which provides a basis to relate
landslide locations and frequency of occurrence to their associated impacts (e.g.,
annulments, volume of debris, closure duration and direct costs). Data would be
invaluable for implementing a public-domain landslide management system, if deemed
appropriate and justifiable.

8. Capital Improvement Projects:

Increase capital investment in landslide mitigation projects. Measurable long-term
reduction in landslide-related impacts will require a significant increase in expenditure on
capital improvement projects. The time required to significantly reduce landslide-related
service interruptions is likely to require one or more decades, depending on the amount of
financial resources available, permitting, design, and construction scheduling.

Complicating Factors for Landslide Reduction

Developing a plan that measurably reduces landslide-related interruptions to passenger rail
service within the corridor is complicated by the following:
Large Problem Area More than 900 landslides have occurred at hundreds of locations
within the 26.6-mile-long corridor since 1914. Many of the adjacent unstable slopes are
greater than 100 feet high.
Land Ownership Most of the landslides on private property are outside BNSFs control
or responsibility. Many of the landslides are partially due to poor slope management
practices conducted by adjacent landowners.
Limited Right-of-Way (ROW) BNSF has a narrow ROW (about 50 feet upslope of the
tracks) available to contain landslide debris or to construct protection structures.
Construction of slope stabilization measures generally requires work outside of BNSFs
ROW.
Differences in Organizational Priorities/Roles/Responsibilities Sound Transit, Amtrak,
and WSDOT are charged with providing public service, and they do not own and are not
directly responsible for track maintenance. BNSF, as a private corporation, is responsible
for track maintenance and identifying, prioritizing and funding its own capital

p. 22
 improvement projects. Priorities for spending available funds may be different depending
on the (public or private) source of the funds. Landslide origination point is often on
private property outside BNSF right-of-way.
Low Risk Tolerance The risk tolerance for public safety is very low, so closure
decisions will always err toward safety.
Assumption of Responsibility BNSF is responsible for determining safe operating
conditions in their Seattle to Everett corridor. Implementation of some of the proposed
mitigation strategies may involve more shared responsibilities or liabilities between
stakeholders as several strategies are not constrained to State- or BNSF-owned right of
way.
Funding Currently, there is no long-term source of public funds for capital
improvements to proactively address landslide-prone slopes. Determining which, if any,
slopes warrant expenditure for remediation, as well as the type and extent of remedial
work, is the responsibility of BNSF.
Permitting Permitting process and timelines vary between agencies such as Ecology
and the Corps, local jurisdictions, and BNSF.

p. 23
Table 7. Potential Strategies to Reduce Landslide Interruptions and Impacts.

Potential Strategies to Reduce Landslides

Strategies Implementation Time Complicating Factors Benefit
Short-term Intermediate Long-term Low High Low Moderate High
Conduct Education/ Ongoing brochure Land ownership (difficult Potential to reduce
Outreach (drainage developed and to ensure long-term landslide initiation with
improvements/best distributed; public implementation); best slope management
slope management workshops scheduled permitting, funding practices
practices)
Implement Vegetation Specific site to be Land ownership, limited May reduce damage to
Management Program identified and right of way, funding, structures/stabilize
recommendations permitting slopes over time
developed
Improve Monitoring Ongoing validation Organizational Does not prevent or reduce
Tools planned within one priorities/responsibilities landslides, but informs
to two years parties of potential
landslide exposure
Explore Long-Term No current plan in Permit modifications Does not prevent or reduce
Debris Disposal Plan place needed; funding landslides, but has benefit
for salmon recovery efforts
Conduct Maintenance Currently Lower cost than Highly effective in reducing
and Monitoring implemented by capital projects public risk exposure, but
Measures BNSF does not reduce landslides
Construct Capital Ongoing 6 Requires obtaining Funding, prioritization of Reaches goal of long-
Improvement Projects proposed locations funding, planning projects, organizational term stabilization of
(containment funded by WSDOT (prioritization), priorities/responsibilities, slopes and/or
structures/stabilization grants (current designing, limited right-of-way prevention of
projects/realignment funding is permitting, landslides in corridor
projects) $16.1 million) construction
Acquire Additional Funding, land Funding, prioritization of Does not prevent or reduce
Right of Way in ownership areas needed; landslides, but provides
Landslide-Prone organizational opportunity for best slope
Sections priorities/responsibility management practices
Develop Public- Information gathered Funding, land ownership, Does not prevent or reduce
Domain Inventory and for action plan could organizational landslides, but guides
Implement Landslide be used as starting priorities/responsibility capital projects; can be used
Management Program point for program to justify further public
investment

p. 24
Implementation Plan Status
Short-Term Improvement Strategies
Community Outreach and Education
The Landslide Work Group identified the need for increased education and outreach to the
community upslope of the rail corridor. Previous studies of landslides in Seattle, with similar
geology, slope conditions, and urban development have shown that more than 80 percent of
landslides are at least partially related to human influence, including poor slope management
practices (Seattle, 2001). Landowner involvement is essential for prevention of landslides as
these studies indicate that improper vegetation removal, inadequate and/or unmaintained
drainage, cutting or grading slopes and dumping debris on slope edges can cause slope instability
and contribute to landslides.

A brochure was developed and delivered to landowners along the top of the slope through the
study corridor in early 2013. In addition, landslide workshops in the city of Mukilteo and the
development of resources on city websites are in progress. To further investigate public
perception of the landslides, a survey was created to gauge public response to education and
outreach efforts and catalog frequently asked questions and/or concerns.

Drainage Improvement Incentive

Improper or poorly designed drainage systems can contribute to slope instability, such as
drainage pipes which outlet mid-slope. To stabilize slopes, drainage should be brought down to
the bottom of the slope. BNSF owns a drainage system at the bottom of the slope.

BNSF can issue permits to property owners for drainage on its ROW. BNSF is offering an
incentive to upslope residents by waiving permit fees (up to $3,500 per permit) until April 2015
to place approved drainage structures onto BNSF property. Insurance requirements are still in
place.

Maintenance and Monitoring

This lower-cost option is currently employed by BNSF in management of the Seattle to Everett
corridor. BNSF will continue to maintain slide fences, ditches and drainage along their right of
way to minimize impacts to railroad operations.

Corridor Landslide Inventory

A landslide inventory database and maps were compiled by the work group (Appendix A) using
previous studies by Shannon & Wilson (2001 and 2007) with data provided by BNSF. Inventory
maps can be used to identify priority areas for remedial work and to develop detailed landslide
hazard maps to assist local agencies in the development of land use regulations for steep slopes.

p. 25
Intermediate Strategies
Capital Improvement Projects
Capital improvement projects are intended to improve passenger service reliability by reducing
the number and severity of track outages due to slope failures along the corridor. Projects are
intended to prevent and minimize service-disrupting landslides by improving the overall slope
stability and implementing measures, such as walls, to prevent landslide debris from impacting
the tracks.

Six mitigation projects funded by WSDOTs American Recovery and Reinvestment Act
(ARRA) grants are in design and a minimum of three (funding dependent) are planned for
construction between 2013 and 2016 (construction is currently underway on two of the six
projects). The six sites were prioritized based on slide history (high frequency of slides and
service disruptions), geotechnical investigation and constructability as well as budgetary,
schedule and property ownership constraints. Improvements primarily involve removing slide
material, terracing slopes, installing trench drains, installing catchment walls, installing slide
fences and appropriately capturing and directing drainage from adjacent properties.

Development of a Landslide Potential Assessment Model

Work is being done to determine whether the likelihood of a landslide event can be reliably
determined by gathering improved rainfall and soil moisture data, and by improving models used
to monitor slide activity. The accuracy of the model will be assessed using historical and 2013-
2014 data. Work in 2013-2014 will focus on installing additional rain gauges at key locations in
the corridor and working with the U.S. Geological Service (USGS) to update their model;
validation of the model would take place in the 2014-2015 rainy season. This work will be
complemented by efforts to improve slope stability at a number of locations in the corridor.

Long-Term Strategies and Recommendations

Continue community outreach and education efforts to the local communities along the corridor
bluff. Recommendations include:
Update education and outreach materials based on community feedback gained through
brochure survey, workshops and local jurisdiction interaction.
Provide support for community workshops.
Develop a streamlined permit process and funding source to implement drainage
improvements and best slope management practices by landowners along the corridor.

Explore solutions for long-term slide debris removal and restoration of near-shore processes,
such as beach nourishment. Recommendations include:
Cooperatively develop restoration plan and updated permit process with agencies
(Ecology and the Corps), BNSF, and local stakeholders (e.g. Puget Sound Partnership
and Snohomish County Marine Resources Committee). Such solutions may not impact
BNSFs operations or limit BNSFs ability to return its tracks to service under current
regulatory structure.

p. 26
Explore justification for further public investment:
Consider acceptable target level-of-service (how many landslide-related interruptions are
tolerable), recognizing that interruption-free service from landslides is likely not
achievable or affordable.
Estimate order-of-magnitude, long-term improvement (capital) costs.
Evaluate projected cost of impacts against long-term improvement (i.e., capital) costs for
a reasonable lifecycle to justify further public investment.
Distinguish public benefit from private benefit on privately owned infrastructure to
ensure taxpayer dollars are used to benefit Washington State, its businesses and
communities.

The science and structural response of a fluid-like mass impacting a rigid structure, like the
debris containment walls commonly used along the corridor, are not well understood, and current
design methodology is poorly constrained. Similarly, the use of patterned-reinforced wire mesh
to address shallow slope instability has not yet achieved widespread use in North America but is
gaining widespread use in Europe. Research efforts should be undertaken to optimize design of
debris containment structures and evaluate effectiveness of slope stabilization methods for
shallow failures. Recommendations include:
Make design improvements to ensure reliability and optimize design of low-deflection,
debris containment structures;
Evaluate test sections of reinforced mesh to determine suitability for more widespread
application.

If further public investment is deemed worthwhile, a landslide management system should be

implemented and managed by a public agency that is closely coordinating with BNSF to
proactively identify, prioritize, program and fund mitigation projects.

p. 27
References
Baum, R.L., E.L. Harp, and W.A. Hultman, 2000, Map showing recent and historic landslide
activity on coastal bluffs of Puget Sound between Shilshole Bay and Everett, Washington, USGS
MF-2346, 1:24,000 scale.

Gerstel, Wendy J.; Brunengo, Matthew J.; Lingley, William S., Jr.; Logan, Robert L.; Shipman,
Hugh; Walsh, Timothy J., 1997, Puget Sound bluffs--The where, why, and when of landslides
following the holiday 1996/97 storms: Washington Geology, v. 25, no. 1, p. 17-31. [accessed Oct
21, 2013 at http://www.dnr.wa.gov/Publications/ger_washington_geology_1997_v25_no1.pdf.
Lowell, S.M. and N.I. Norrish, 2013, Chapter 18: Rockfall management programs, in Rockfall:
Characterization and Control, eds. A.K. Turner and R.L. Schuster, Transportation Research
Board, pp. 595-608.

Minard, J.P., 1982, Distribution and description of geologic units in the Mukilteo quadrangle,
Washington, USGS MF-1438, 1:24,000 scale.

Minard, J.P., 1983, Geologic map of the Edmonds East and part of the Edmonds West
quadrangles, Washington, USGS MF-1541, 1:24,000 scale.

Minard, J.P., 1985, Geologic map of the Everett 7.5-minute quadrangle, Snohomish County,
Washington, USGS MF-1748, 1:24,000 scale.

Seattle, 2001, Seattle Landslide Study. City of Seattle Department of Planning and Development.
Accessed on-line April 2013:
www.seattle.gov/dpd/cms/groups/pan/@pan/documents/web_informational/dpdp025740.pdf.

Shannon & Wilson, 2001, Final Stability Study Between Milepost 6.5, Line Segment 50, and
Milepost 178.4, Line Segment 37, BNSF/Sound Transit Commuter Rail, Seattle to Everett, WA,
79 p.

Shannon & Wilson, 2007, BNSF/Sound Transit Commuter Rail Slope Stability Study Update
Seattle to Everett, Washington, 61 p.

Shuster, R. 1996, Chapter 2: Socioeconomic significance of landslides, in Landslides:

Investigation and Mitigation, Special Report 247, eds. A.K. Turner and R.L. Schuster,
Transportation Research Board, pp. 12-31.

Washington State Department of Transportation, 2006, Unstable Slopes on I-90 Snoqualmie

Pass: Reassessment and Recommendations, report to Governor Gregoire, 121 p.

Yount, J.C., J.P. Minard, and G.R. Dembroff, 1993, Geologic map of the surficial deposits in the
Seattle 30 by 60 quadrangle, Washington, USGS OF 93-233, 1:100,000 scale.

p. 28