Conference PaperPDF Available

The May 2014 West Salt Creek landslide in Mesa County, Colorado


Abstract and Figures

On May 25, 2014, a very large rockslide/rock avalanche (sturzstrom) occurred in rural western Colorado, about 38 mi (61 km) east of Grand Junction. The landslide was initiated by a disturbed 2,900- ft (880 m) wide and 900-ft (280 m) deep block of Green River Formation that slipped along a preexisting listric shear plane of a geologically recent (post-glacial) landslide. The top elevation was 9,500 ft (2,895 m). As the rockmass (composed of shale, marlstone, and oil shale) rotated, the heaved and oversteepend front disaggregated almost instantaneously and the pulverized rockmass “flowed” in discrete rock avalanche surges downvalley for 2.8 mi (4.5 km). The deposits ultimately covered nearly one square mile (2.6 km2) of the West Salt Creek valley. There was smaller precursor landslide activity earlier that morning that blocked an irrigation ditch. Three local men, investigating the blockage that afternoon, were killed by the main catastrophic failure at 5:45pm MST. The assumed location of their truck is buried by 125 ft (38 m) of rock debris. The 660-ft (200 m) wide, valley-constrained toe narrowly missed a producing oil and gas well pad. The debris field was surprisingly dry. No water or mud seeped or fanned out from the steeply sloped (≤40°) edge of the avalanche toe. The rockslide and resultant avalanche moved 39x107 yd3 (3x107 m3) of rock and soil down 2,100 ft (640 m) of elevation and caused a 3-minute seismic wave train and 2.8 magnitude earthquake. The H/L mobility index is 0.14 and angle of reach a very flat 8 degrees. The most rapid avalanche surges overtopped a 280-ft (80 m) high ridgeline at an outside bend on the west side as well as a 180-ft (55-m) high ridgeline on the east side of the valley. Preliminary velocity estimates at these locations, based empirically on the overtopped ridge heights and trim lines above the pre-landslide valley floor, range from 40 to over 85 mph (18 to 38 mps). Approximately half of the volume was incorporated into the rock avalanche. The remainder is the remnant of the rotated block, currently estimated to be 2,900x700x500-feet (880x210x150 m). This highly disturbed and potentially unstable rock remains immediately downslip from the exposed 450-ft (137-m) high headscarp. The back-tilt of the block has formed a large depression below the headscarp that is filling to form a 420 acre-ft (5x105 m3) sag pond. In addition to the long-term instability of the block, additional concerns are the potential for mud/debris flows resulting from potential pond breaches during next year’s spring runoff, as well as mini-tsunamis if retrogressive failures occur at preexisting scarps above the main headscarp to displace the water. From a regional perspective it has caused reevaluation of long run-out ground morphology previously assumed to be coalesced mud flows, the need for LiDAR bare-earth elevation models, and density and siting of oil and gas well pads in the vicinity.
Content may be subject to copyright.
Time to Face the Landslide Hazard Dilemma:
Bridging Science, Policy, Public Safety, and Potential Loss
AEG Professional Forum
FEBRUARY 26th-28th, 2015
University of Washington (UW), Seattle, Washington
1100-H Brandywine Blvd.
Zanesville, Ohio 43701
Phone: 844-331-7867 (toll free) Fax: 740-452-2552303-757-2926
AEG Professional Forum
Time to Face the Landslide Hazard Dilemma:
Bridging Science, Policy, Public Safety, and Potential Loss
Forum Dates: Thursday and Friday, February 26 and 27, 2015
Location: University of Washington, Seattle, 316 South Campus Center
Field Trip: Saturday, February 28, 2015 - Everett and Oso Landslides, WA
AEG and the University of Washington Earth & Space Sciences (UW ESS) welcome you
to Seattle, Washington and what we expect to be an interesting and thought provoking
Professional Forum (Forum). Recent landslide tragedies such as the Oso/Hazel Landslide
in Snohomish County, Washington and in Grand Mesa and Boulder, Colorado, as well as
other damaging landslides in Wyoming and Utah were in the local, regional and national
news a considerable amount in 2014. In some cases, these areas were known to have
had historic/pre-historic landslides, which illustrates a disconnect between the current
state of practice in landslide hazard and risk assessment and its incorporation into land
use and development zoning, and codes and ordinances that serve the best interests of the
public. As we approach the one year anniversary of the Oso/Hazel landslide in March,
we felt the time was right for a Forum of this type and hope we can build some consensus
on what we should be doing collectively and how best to more proactively convey
information to public officials and the public.
This Forum, “Time to Face the Landslide Hazard Dilemma: Bridging Science, Policy,
Public Safety, and Potential Loss”, brings together private and public sector technical
experts on landslide hazards, public officials, and insurance and planning representatives
to present some of the current programs and strategies being employed and to discuss
how we can better protect public health, safety, and infrastructure. Improvements could
include: (1) communication of landslide hazards and risk to the public, (2) coordination
between federal, state, and local agencies and geological surveys and private sector
geologists and geotechnical engineers, and (3) development of more comprehensive
landslide hazards maps and technical information that is understandable and available to
the public and can be incorporated into land use planning and policy.
The main objective of the Forum is to develop strategies for bridging the gap between
science and public interests. The presentations include lessons learned from Oso/Hazel
and other recent landslides; existing landslide hazard assessment programs (domestic and
international); laws and regulations in various states, local jurisdictions, and
internationally; lessons that can be learned from other hazards; and perspectives from
policy makers, and the insurance and real estate industry. The Forum also includes
workshop discussions on developing strategies for national action while recognizing state
and local rights.
Slope processes annually produce damage, destruction, injury, and death in many areas of
the country. While there is no consistent detailed data on annual financial losses
associated with landslides in the United States, the current estimates of the direct costs
are on the order of 10’s to 100’s of millions of dollars and could exceed a billion dollars
when indirect costs are included. The public and political perceptions of the presence,
nature, and potential threats from these processes vary greatly, even in landslide-prone
regions. Regulations addressing potential hazards are non-existent in many areas, but
where regulations have been promulgated; they are generally limited to new construction
and forest practices. Few resources other than post-disaster private donations are
available to financially assist victims affected by slope processes. Insurance policies
generally exclude landslides and other natural disasters, or are very restrictive in
coverage. Federal landslide assistance, for example from FEMA, is restricted to small
grants for homeowners, flood insurance for “mudslides (e.g. mudflows)”, and partial
funding for replacement of public facilities, with specific limitations. Therefore, the
potential for establishing insurance coverage and the potential role we have in supporting
this is another theme we will explore during the Forum.
Another goal of the Forum is to prepare either a standalone AEG publication and/or
special edition Environmental & Engineering Geology Bulletin with papers from the
invited speakers, summaries of the workshop discussions, recommendations for improved
communications between interested parties, and possibly recommended guidance if a
majority consensus can be developed at the conference.
We want this to be an interactive meeting and encourage you to interact and have
discussions during the breaks and ice breaker, and compile questions and ideas and
actively participate in the workshop on Friday afternoon.
We also want to acknowledge our financial sponsors and contributing co-sponsors
outlined in this program for their support!
As the planning committee, we look forward to meeting you and a successful Forum and
field trip!
Jennifer Bauer, Jeff Keaton, Mark Molinari, and Kathy Troost
AEG Professional Forum
Table of Contents
PROGRAM SCHEDULE ..........................................................................i
PROGRAM SPONSORS ............................................................................v
Landslide mapping programs and impacts ............................... 1
State programs and implementation ............................................. 31
Landslide risk, prediction, and local implementation ............ 55
Regulations, Planning and Public Policy ...................................... 77
Where do we go from here? .............................................................. 93
POSTER ABSTRACTS ..............................................................................94
SPEAKER BIOS ..........................................................................................106
ACKNOWLEDGEMENTS ........................................................................113
AEG Professional Forum
Time to Face the Landslide Hazard Dilemma:
Bridging Science, Policy, Public Safety, and Potential Loss
Conference Dates: Thursday and Friday, February 26 and 27, 2015
Location: University of Washington, Seattle, 316 South Campus Center
Field Trip: Saturday, February 28, 2015 - Everett and Oso Landslides, WA
Keynote Speakers
Keynote 1: The Oso, Washington Landslide Overview Lessons Learned and Catalyst for this
Conference; presented by Dr. Joe Wartman, PE, UW, co-leader of the GEER Oso team.
Keynote 2: What is Landslide Hazard? Inventory Maps, Uncertainty, and an Approach to
Meeting Insurance Industry Needs; presented by Dr. Jeffrey Keaton, PG, Amec Foster Wheeler,
co-leader of the GEER Oso team.
Keynote 3: Lessons from the National Earthquake Hazards Reduction Program can be applied to
the National Landslide Hazards Program: A Rational Approach; presented by Mr. Richard Roth,
Jr., Consulting Insurance Actuary.
Keynote 4: Catalyst for the 2nd Day Workshop Saving Lives and Property, Is Science the Easy
Part?; presented by Dr. Dave Montgomery, UW, member of the GEER Oso team and member of
the WA governor’s Oso Landslide Commission.
Day one focuses on current conditions and knowledge. Day two focuses on needed
improvements and what we do next. Day three is a fieldtrip to visit landslides that exemplify the
problems with our current systems. Posters will be displayed in the conference room and time is
allotted for poster presentations. The poster author will be expected to be at their posters for the
break times specified for posters to interact with attendees.
AEG Professional Forum
Day 1 Program
7:00 8:00am Pre-Program Set-Up and Posters Coffee, Tea and Baked Goods
Morning Session: Landslide Mapping Programs and Impacts
8:00 8:15 Introduction and Welcome: Kathy Troost, UW and Mark Molinari,
8:15 9:00 Keynote 1: The Oso, Washington Landslide Overview Lessons Learned
and Catalyst for this Conference - Dr. Joe Wartman, UW and GEER Oso
9:00 9:45 Keynote 2: What is Landslide Hazard? Inventory Maps, Uncertainty, and
an Approach to Meeting Insurance Industry Needs - Dr. Jeffrey Keaton,
Amec Foster Wheeler and GEER OSO team
9:45 10:15 Mid-Morning Break, Refreshments, and Posters
10:15 10:45 Current Status of the U.S. Geological Survey Landslide Hazards Program
Dr. Jonathan Godt, USGS
10:45 11:15 Canadian Technical Landslide Guidelines and Best Practices for Landslide
Professionals Mr. Doug VanDine, VanDine Geological Engineering Ltd.
11:15 11:45 Landslides: the poor stepchild of geological hazards; what we can do and
what other countries are doing to solve this situation Dr. Scott Burns,
Portland State University
11:45 12:15pm U.S. Army Corps of Engineers Reflections on Ground Failures and
Particularly Landslides Dr. Richard Olsen US COE Headquarters
12:15 1:15 Lunch (provided) and Posters
1st Afternoon Session: Example State Programs and Implementation
1:15 1:45 Landslide Hazard and Risk Reduction Process in Oregon Mr. Bill Burns,
1:45 2:15 North Carolina Landslide Mapping Public and Private Programs Ms.
Jennifer Bauer, Appalachian Landslide Consultants
2:15 2:45 Landslides in Kentucky: Inventory, Data Delivery, and Collaboration
Mr. Matt Crawford, Kentucky Geological Survey
2:45 3:15 Landslide inventory in Washington State: the past, present, and future
Mr. Stephen Slaughter, Washington DNR
AEG Professional Forum
3:15 3:30 Mid-Afternoon Break Beverages, Snacks, and Posters
2nd Afternoon Session: Landslide Risk, Prediction and Local Implementation
3:30 4:00 Managing Slope Hazards Along Transportation Infrastructure Mr. Tom
Badger, WSDOT
4:00 4:30 Landslide Modeling and ForecastingRecent Progress by the U.S.
Geological Survey Dr. Rex Baum, USGS
4:30 5:00 Landslide Hazard Mapping in King County Mr. John Bethel, King
County, WA
5:00 5:30 Landslide Management During Property Development in Los Angeles
County Mr. Charles Nestle, County of Los Angeles Department of
Public Works, CA
5:30-6:30 pm Posters/Ice Breaker Hosted Drinks and Appetizers
Day 2 Where do we go from here?
Morning Session: Regulations, Planning and Public Policy Financial Impacts and
How Do We Better Protect Public Safety and Reduce Financial
8:00 8:15 am 2nd Day Introduction and Goals Jeff Keaton and Mark Molinari
8:15 9:00 Keynote 3: Lessons from the National Earthquake Hazards Reduction
Program can be applied to the National Landslide Hazards Program: A
Rational Approach Mr. Richard Roth, Jr., Consulting Insurance Actuary
9:00 9:30 Managing Recognized Hazards: Land Use Planning and Zoning,
Strategies and Public Education/Notification Mr. Mark Molinari,
9:30 10:00 What do planners do with hazard information? Bridging the
science/policy gap Mr. David Sherrard, AICP, Planning Consultant
10:00-10:30 Mid-Morning Break, Refreshments, and Posters
10:30 11:00 How Emergency Management Can Achieve a "Landslide Victory" Related
to Community Preparedness and Public Warning Mr. Mike Chard,
Boulder CO Emergency Management Director
AEG Professional Forum
11:00 11:30 How Can Congress Help? The Potential for Comprehensive Landslide
Policy Ms. Abigail Seadler, American Geosciences Institute (AGI)
11:30 - 12:00 pm Posters
12:00 1:15pm Lunch (provided) and Posters
1:15 1:45 Summary of Key Points from NRC’s Board on Earth Sciences and
Resources Special Workshop: “Landslides and Landslide Risk: Needs for
the Next Decade” Dr. Joe Wartman, PE, UW
1:45 2:30 Keynote 4: Saving Lives and Property, Is Science the Easy Part? Dr.
David Montgomery, UW and Washington Governor’s Commission on Oso
2:30 3:00 Workshop Introduction
3:00 3:30 Mid-Afternoon Break, Beverages, Snacks, and Posters
3:30 5:00 Workshop Where Do We Go From Here? (conveners Kathy Troost and
Jennifer Bauer)
5:00- 5:30 pm Workshop Recap and Closing
AEG Professional Forum Program Sponsors
Financial Sponsors
Ice Breaker -
AEG Washington Section
c/o Chad R. Lukkarila, LEG, PE
AEG Washington State Section Chair
14710 NE 87th Street, Suite 100
Redmond, WA 98052
Thursday Lunch -
Cascade Drilling, LP
John Murnane CPG, LHG
Marketing and Communications
Cascade Drilling, LP
Woodinville, WA
Office: 425-485-8908 x 2107
Cell: 206-396-1558
Friday lunch -
Geobrugg North America, LLC
Geobrugg Geohazard Solutions
Tim Shevlin, PG
Regional Manager, NW USA & Hawaii
tel: 503-423-7258 fax: 505-771-4081
Salem, OR 97302
Student Lunches -
Shannon & Wilson, Inc.
Bill Laprade, LEG, LG, Senior VP
400 N. 34th Street
Seattle WA 98103
Thursday afternoon break -
GeoStabilization International
Bryan Wavra, PE
NW Project Development Engineer
Mobile: 503-999-4187
Washington Section
AEG Professional Forum Program Sponsors
Contributing Sponsors
Canadian Geotechnical Society
AEG Professional Forum Extended Abstracts
Landslide mapping programs and impacts
Keynote 1
The Oso, Washington Landslide Overview
- Lessons Learned and Catalyst for this Conference
Wartman, Joseph, PhD, PE, University of Washington,
Jeffrey Keaton,
Scott Anderson,
Jean Benoît,
John deLaChapelle,
Robert Gilbert,
David R. Montgomery,
This presentation highlights the principal findings of the National Science Foundation
(NSF)-supported Geotechnical Extreme Events Reconnaissance (GEER) Association
scientific research team that performed a field reconnaissance of the Oso Landslide
beginning approximately 8 weeks after its occurrence (Keaton et al. 2014). The
presentation primarily focuses on observations made and data collected at the landslide
site, but additionally reviews regional and local geologic conditions, climatic setting,
eyewitness accounts, local land-use and landslide risk assessment. The principal findings
are summarized as follows.
Impacts and Significance
The Oso Landslide (Figure 1) claimed 43 lives, making it the deadliest landslide disaster
in the history of the continental United States. In addition, it caused significant injuries to
at least 10 people who were struck by the landslide, but survived. The landslide
completely destroyed Steelhead Haven, a community of almost 50 homes, as well as
several residences located off a nearby roadway. The landslide also buried portions of
State Highway 530, resulting in complete closure of this important arterial thoroughfare
for over 2 months, and several more months of reconstruction.
AEG Professional Forum Extended Abstracts
Figure 1: The Oso Landslide (3 May 2014).
Landslide Setting
The Oso Landslide is situated in the North Fork Stillaguamish River valley (Figure 2).
The last glacial advance into the Puget Lowland deposited a thick sequence of sediments
into the valley. The glacially-derived sediments include interbedded layers of clay, silt,
sand, gravel, cobbles, and boulders (Figure 3). The geomorphic evidence in the valley
reveals that the portion of the North Fork Stillaguamish River Valley in the vicinity of
Oso Landslide has experienced multiple large landslides over at least the past six
thousand years. Many of these ancient landslides have similar morphology to the 2014
Oso Landslide, and indeed the Oso Landslide was a reactivation of one of these ancient
landslides. The 2014 Oso Landslide was large (Figure 4), but the other ancient landslides
in the valley are of similar size. There is geomorphic evidence that a landslide that is
even larger than the Oso landslide is located immediately to the west of the Oso
Landslide. This larger landslide similarly ran out across almost the entire North Fork
Stillaguamish River Valley and appears to have pushed the river channel to the south
margin of the valley.
Figure 2: Topography in the vicinity of the Oso Landslide (after Keaton et al. 2014).
History of Landslides at the Oso Site
Multiple episodes of historic movement of the Oso landslide have been described in
several studies dating back to the 1950s. The observed historic activity appears to be
periodic with the modern headscarp episodically advancing headward between 1952 and
AEG Professional Forum Extended Abstracts
2006, but with the main slide mass constrained to approximately the same portion of the
slope where the earlier 2006 landslide failed.
Figure 3: View of Oso Landslide and geologic exposures in scarps (after Keaton et al. 2014).
Figure 4: Elevation differences between 2014 and 2013 LiDAR data. Color keyed to vertical change with
red hues representing decrease in elevation (net erosion) and green hues representing increase in elevation
(net deposition) (after Keaton et al. 2014).
AEG Professional Forum Extended Abstracts
Initiation of the 2014 Oso Landslide
The Oso Landslide initiated on Saturday, 22 March 2014, at approximately 10:37 a.m.
local time on a clear day. Records indicate no significant seismic activity in the days
preceding the landslide and therefore it is unlikely that it had a seismogenic origin.
Instead, it is highly probable that the intense 3-week rainfall that immediately preceded
the event played a major role in triggering the landslide. The intense rainfall in the first
three weeks of March at the nearest rain gauge was determined to be less than the 100-
year event for this period of time, and the previous months in the fall and winter of 2013
and 2014 were relatively dry. Precipitation in the Oso region is highly variable and
analysis of weather radar for the area for the week preceding the landslide indicates that
precipitation at the Oso Landslide was at least 229 mm, suggesting that the precipitation
at the Oso Landslide for March 2014 might have been more than 760 mm.
Beyond the rainfall trigger itself, there are other factors that likely contributed to
destabilization of the landslide mass. These include: (i) alteration of the local
groundwater recharge and hydrogeologic regime due to previous landsliding and,
possibly, land use practices, (ii) weakening and alteration of the landslide mass due to
previous landsliding and other natural geologic processes, and (iii) changes in stress
distribution resulting from removal and deposition of material from earlier landsliding.
Detailed consideration of land use practices (most notably, timber harvesting) was
beyond the scope of our investigation; however, it is known most of the large landslides
in the Stillaguamish River Valley pre-date logging. Given the size and depth of the
landslide, if timber harvest practices did influence on the landslide, it was through
modification of the groundwater recharge regime rather than by any shallow-depth loss of
root mass reinforcement.
Figure 5: Landslide-generated velocity-time histories presented plotted as smoothed, high-pass filtered (<1
Hz) smoothed envelopes. Owing to the smoothing function, the velocity amplitude data is relative rather
than absolute (after Keaton et al. 2014).
AEG Professional Forum Extended Abstracts
Oso Landslide Morphology and Dynamics
During our field reconnaissance we identified six distinctive zones and several subzones
of the landslide mass that are characterized by different geomorphic expression resulting
from different styles of deformation, geologic materials, and vegetation. These reflect the
highly complex nature of the landslide. The seismic recording of the landslide suggest
that the event was marked by multiple episodes of mass movement separated by a few
minutes (Figure 5). This corroborates with our data found during the reconnaissance,
which provides evidence of multiple stages of failure.
Hypothesized Landslide Sequence
Based on the reconnaissance observations, seismic recordings, and other available data,
we hypothesize that the landslide occurred in multiple stages. The first major stage of
movement is interpreted to be a remobilization of the 2006 slide mass and a headward
extension that included part of the forested slope of the ancient landslide. We believe that
the second stage occurred several minutes later in response to the unloading and the
redirection of stresses within the landslide mass.
Landslide Risk Assessment, Management, and Communication
Studies conducted in the decades preceding the Oso Landslide clearly indicated a high
landslide hazard at the site. However, these studies were primarily focused on the impacts
of landslides to the river versus the impacts to people or property. In addition, it does not
appear that there was any publicly communicated understanding that the debris from a
landslide could run-out as much as 1 km, as it did in the 2014 event. Since the 1950s, a
variety of means were considered to manage the risk associated with this slope, ranging
from stabilizing the riverbank to minimize erosion to moving the river channel and
removing development by buying out properties. At the time of the 2014 event, two
means had been employed to manage risk from a landslide: (i) conventional land-use
restrictions implemented by Snohomish County and the Washington Department of
Natural Resources and (ii) riverbank stabilization implemented by the Stillaguamish
Tribe of Indians. Our assessment of the risk for fatalities due to landslides in this portion
of the valley indicates that it is comparable to risks from flooding in other areas in the
United States but relatively high compared to guidelines for landslides in other developed
countries and for large dams in the United States (Figure 6). Currently there are no
national or state guidelines in the United States concerning levels of risk due to natural
landslides that warrant action.
Lessons Learned
Several broader lessons have been learned in this investigation that may benefit others
involved in the study of landslides and the zoning of communities adjacent to sloping
ground and potentially unsafe slopes.
The history and behavior of past landslides and associated colluvial soil masses should
be carefully investigated when mapping areas for zoning purposes. At the Oso Landslide
site, multiple past failures retrogressively moved upslope each time creating new
conditions with increased susceptibility to groundwater infiltration, and preferential
AEG Professional Forum Extended Abstracts
underground seepage pathways, and further weakening the previously failed mass over
time and increased overall volume of potentially unstable landmass.
Figure 6: Rough approximation of risk from landslides in 5-km stretch of North Fork Stillaguamish River
Valley in vicinity of Oso compared with related benchmarks for human safety risk (after Keaton et al.
• The risk of landslides to people and property should be assessed and communicated
clearly and consistently to the public. These assessments should be continuously updated
as new information about slope behavior becomes available and as potential
consequences change due to changes in development or mitigation.
• The ability to implement monitoring and warning systems to reduce the impacts of
landslides to people and property should be considered and advanced.
Methods to identify and delineate potential landslide runout zones should be revisited
and reevaluated.
Advancements in imagery to understand slope behavior should be exploited to the
greatest extent possible. Lidar imagery has proven to be a very useful an valuable tool in
identifying landslide deposits, reconstructing landslide history, and evaluating mass
movements of the current landslide event. This technology has been made feasible over
the last decade or so and still does not cover most of the country. Its availability here, and
its availability at multiple times (2003, 2013, and after the failure in 2014) allows an
understanding of slope and landslide morphology, and thereby hazard and risk, that was
AEG Professional Forum Extended Abstracts
not present prior to 2003 in this valley and is currently not present in most locations.
Additionally, high-resolution aerial photography also is a valuable tool to help delineate
zones within the failed mass and document damages prior to recovery and clean up
Seismological recordings of landslides should be utilized to assist in understanding
failure sequence in terms of the timing of significant movements, especially in large and
complex events. Use of conventional slope stability analysis methods alone may be
insufficient for accurate evaluation of failure mechanisms.
Doppler weather radar should be utilized in providing data regarding precipitation
intensity, amount, and variability estimates at locations of interest that are distant from
established gauges.
Keaton, J. R., Wartman, J., Anderson, S., Benoît, J., deLaChapelle, J., Gilbert, R., and
Montgomery, D. R.: The 22 March 2014 Oso Landslide, Snohomish County,
Washington, GEER report, NSF Geotechnical Extreme Events Reconnaissance,
andslide_Report.pdf (last accessed 4 February 2015)
AEG Professional Forum Extended Abstracts
AEG Professional Forum Extended Abstracts
Keynote 2
What is Landslide Hazard? Inventory Maps, Uncertainty, and an Approach to
Meeting Insurance Industry Needs
Keaton, Jeffrey R., F.ASCE, F.GSA, Amec Foster Wheeler,
Richard J. Roth Jr., Consulting Insurance Actuary,
The objective of this paper is to 1) identify some aspects of landslides and slope processes that
need to be understood in addressing the dilemma among science, policy, public safety, and
potential loss and 2) describe a simple approach to meet the needs of the insurance industry that
also promotes landslide hazard research. Landslides are difficult to characterize in ways that
provide information that can be used by the insurance industry. Why is the insurance industry a
focus and what are its needs? In this context, the insurance industry refers to private insurance as
opposed to government programs; both need the same information, but commercial insurance
products must have viable prices to be available, whereas government programs have different
pricing metrics. Most of the concepts in this paper were introduced by Keaton and Roth (2008,
2009), Roth and Keaton (2010), Haneberg and Keaton (2012), and Keaton and Haneberg (2013).
A condition with the potential for causing an undesirable consequence is a hazard. Fell et al.
(2008a,b) note that the description of landslide hazard should include the location, volume (or
area), classification and velocity of the potential landslides and any resultant detached material,
and the probability of their occurrence within a given period of time. Jelínek et al. (2007) state
that a landslide hazard map ideally indicates the probability of landslides occurring in a given
area at a given time or with a given frequency. A hazard map, however, may be as simple
locations of old landslides to indicate potential instability, or as complex as a quantitative result
incorporating probabilities based on variables such as rainfall thresholds, slope angle, soil type,
and levels of earthquake shaking. The word “risk” sometimes is used when “hazard” is what is
meant. Hazard has units of probability and intensity or size (as in earthquake magnitude and
frequency). Risk has units of dollars and lives (as in losses and injuries or fatalities).
Damage to structures founded on landslide deposits or utilities buried in them results from
displacement, particularly at internal and external boundaries, whereas damage to structures in
trajectories of mobile landslides, debris flows, or rock falls is done by pressure, impact, or
inundation; parameters such as volume, area, or velocity may imply displacement or impact, but
would require interpretation for any possible correlation. Additionally, would average velocity or
maximum velocity be reported? How would velocity be determined or estimated? How would
potential damage be inferred from landslide velocity? In a post-landslide event investigation,
what observations would be made to estimate the range of landslide velocity at the location of a
damaged building or utility line? These are challenging questions for consideration in bridging a
gap to improve understanding and communication.
A landslide inventory map is commonly considered to be a type of landslide hazard map because
it depicts an implicit potential for a future undesirable consequence. Inventory maps identify
locations of past landslide activity with information regarding size and shape. Such maps are
useful for planning and zoning by implying that future slope movements will take place where
past movements have occurred. Inventory maps do not address the probability of a landslide
occurring or the extent of damage that might occur at any location within the inventory map area.
AEG Professional Forum Extended Abstracts
Inventory maps are sometimes considered to be “ground truth”; however, all maps of any kind
have uncertainty, inaccuracies, and errors or omissions (Keaton and Haneberg, 2013).
The two panels in Figure 1 show moderate to high landslide “hazard” areas (left) and earthquake
hazard (right) in the United States. Both of the maps in Figure 1 display hazard, with only an
implicit indication of risk of damage or loss. Similarities between earthquake hazard maps and
landslide hazard maps were recognized by Perkins (1997). The landslide hazard and the
earthquake hazard maps in Figure 1 differ in two fundamental ways: a) hazard intensity and b)
hazard frequency. The landslide hazard map denotes areas where landslides have occurred in the
past (incidence) or where slope and geology conditions suggest that landslides could occur
(susceptibility) without regard to how much movement might occur in the future or how likely
such movement might be. The earthquake hazard map displays the distribution of expected
horizontal earthquake accelerations (hazard intensity) corresponding to a 10% probability of
exceedance in 50 years (hazard frequency), which is ground motion with an average recurrence of
475 years or an annual frequency of 0.002107. Acceleration is useful for engineers to use to
calculate forces in building components. The acceleration map could be converted to a
displacement map by considering a standard mass with a standard frequency of vibration, but that
would be vibratory displacement, rather than permanent displacement.
Figure 1. Left: Areas of moderate to high landslide hazard in the United States from GIS data developed by
Godt (1997). Right: Earthquake hazard in the United States displayed as peak horizontal acceleration
corresponding to 10% exceedance probability in 50 years; data from USGS website
Earthquake Insurance
Earthquake insurance products were developed by insurance companies after seismologists,
geologists, and engineers were able to provide suitable information to insurance actuaries so that
estimates of earthquake shaking could be combined with building vulnerability to create models
of earthquake-induced loss. A wide gap existed between the needs of the insurance industry and
the interests of the science and engineering community. This gap was bridged by an effort that
brought together insurance actuaries, seismologists, geologists, engineers, planners, lawyers, and
politicians to discuss topics such as a) seismology and geology, b) engineering, c) sociology and
public policy, and d) insurance. These discussions allowed earthquake researchers to understand
AEG Professional Forum Extended Abstracts
the needs of the insurance industry. Experience with earthquake damage since insurance was
made available has resulted in improved knowledge about earthquake processes and building
responses. It also has resulted in improvements in the information gathered after earthquake
disasters have occurred. For example, details of building damage are collected so that correlations
can be made among geologic site condition, earthquake ground motion level, and building age,
height, and type. Such information is used to improve earthquake zoning regulations and building
code provisions.
The basic elements of earthquake loss estimation model are:
1) Identification ground motion at a location on an annualized basis in terms of acceleration
including distance attenuation and site amplification effects;
2) Estimation of expected damage to building structures and contents in terms of percentage of
loss including consideration of age, type, number of stories, and commercial usage; and
3) Estimation of expected damage to building structures and contents in terms of dollars of loss
related to interruption of business functions and replacement or repair cost.
Earthquakes are recurring events. The locations in the United States where earthquakes are
expected to occur can be deduced from the hazard map in Figure 1. Early seismic hazard maps for
building codes consisted of zones (0-3) based on incidence of earthquake damage (Figure 2).
Modern seismic hazard maps (Figure 1) display continuous contours of basic ground motion
associated with an annual probability of exceedance (p=0.0021). Supplementary geologic data are
used to account for site conditions, enabling the insurance industry to develop loss models needed
to offer earthquake insurance.
Figure 2. Seismic damage zones from an early building code.
The socio-economic impact of landslides commonly is underestimated, which results in
lack of specific information about damage caused by landslides. Many landslides occur in
response to other natural events, namely rainstorms, floods, and earthquakes, so the
economic impact usually is not considered separately. Hazard maps may incorporate
landslide potential along with other slope processes, such as soil erosion, which is
portrayed as unstable soil hazards. The distribution of damaged buildings may be noted
AEG Professional Forum Extended Abstracts
on maps, but it is common for damage from multiple sources to be combined without
identifying landslide as a cause.
The concept of landslide magnitude or intensity is challenging. Fell et al. (2008a, b)
suggest that landslide volume or area is a measure of landslide magnitude or intensity.
Volume and area are quantitative indices suitable for differentiating land-use planning
zones and comparing with other landslide-prone areas. From a private insurance
perspective, however, some or possibly most landslide damage is caused by displacement
or differential displacement which may or may not be reflected in landslide volume or
area. Debris flows and rock falls operate under different sets of mechanics and boundary
conditions, demonstrating that the landslide problem has complexities that most
geologists and engineers have long recognized. Such complexities result in three
important conditions for landslide hazard and risk management: 1) they make challenging
a single, comprehensive slope-movement hazard and risk scheme; 2) they promote
complicated concepts and terminology that inhibit communication of geologic and
engineering factors with non-technical people; and 3) they attract attention of geologists
and engineers to devote increasing levels of effort in explaining details of slope
movement processes so that professional society conferences typically do not involve
people with broad ranges of education and experience.
Areas of Nil Hazard
Estimating with confidence areas where landslide hazard is essentially nil would be a
significant step toward insuring landslides. Such areas could be depicted on a landslide
hazard map that is similar to the early seismic hazard maps (Figure 2). Areas without
histories of slope movements and without topography, geology, or geomorphology
conducive to slope movements might be Slide Zone 0 (p<<0.01). Areas with histories of
slope movements might be Slide Zone 3 (p>0.01). Slide Zone 1 would be hilly areas
where no landslides have occurred and none are expected based on geotechnical study
(p<0.01). Slide Zone 2 would be hilly areas where no landslides have occurred but slope-
movement susceptibility is real based on geology or geomorphology (p≈0.01).
A hypothetical example landslide hazard map (Figure 3) was developed by Keaton and
Roth (2010) based on real data of various types available in GIS format on the internet.
The example location is an arbitrary area of 2.186 x 105 km2 in the east-central United
States. The early seismic zones for this area are shown on Figure 2, whereas the modern
seismic hazard map details are shown in Figure 1. The landslide incidence and
susceptibility map (Godt, 1997) also is shown in Figure 1. Bedrock geology units and
contours of total precipitation from US Department of Commerce National Oceanic and
Atmospheric Administration for the 500-year, 10-day rainfall also were used in this
Some simple assumptions were made regarding the input data to create the landslide
hazard zones shown in Figure 3. Slide Zone 3 encompasses all areas of moderate and
high landslide incidence and susceptibility, as well as areas with earthquake ground
shaking characterized by peak accelerations ≥15% g that have a 10% probability of
exceedance in 50 years and areas in which precipitation ≥400 mm is generated by the
AEG Professional Forum Extended Abstracts
500-year, 10-day rainfall. Slide Zone 0 corresponds to relatively simple geology and low
topographic relief. Other areas are combined into Slide Zones 1 & 2. Selection of
moderate and high the landslide incidence and susceptibility areas for inclusion in Slide
Zone 3 is logical. Inclusion of threshold values in Slide Zone 3 for earthquake shaking
and precipitation implies that landslides are reasonably likely to occur for process
intensities that exceed these thresholds.
Figure 3. Example of simple landslide hazard zones (Slide Zones 0, 1 & 2, and 3) from Keaton and Roth
Godt et al. (2012) developed a prototype landslide hazard map of the United States at a postal Zip
Code scale based on a simple correlation of ground slope and detailed landslide inventory maps
in five regions. This prototype map was focused on defining two hazard zones: 1) negligible
landslide hazard and 2) non-negligible landslide hazard. Marchesini et al. (2014) evaluated non-
susceptible landslide areas in Italy and recognized that little effort has been made to propose and
test methods to assess where landslides are not expected to occur. They further note that ”This is
surprising, because planners and decision makers are equally, or more interested in knowing
where landslides are not foreseen, or cannot occur in an area, than knowing where susceptibility
is high or very high.”
Actually, it goes beyond a simple interest in knowing where the nil hazard areas are located.
Private insurers might offer policies for properties in Slide Zones 0 and 1 but not in Slide Zone 2
unless damage mitigation measures were implemented and maintained, and then only for certain
types of construction and uses. Policies would not be offered for property in Zone 3 without
comprehensive geotechnical analysis and hazard and risk mitigation. The success of earthquake
AEG Professional Forum Extended Abstracts
insurance suggests that improvements will be made in simple landslide loss models once
landslide insurance becomes a viable product for private insurers.
Next Steps
Earthquakes and hurricanes are hazards for which insurance coverage is available because models
have been developed by commercial companies to support the insurance industry. Similar loss
estimation models have not been developed for landslides because a) the geologists and engineers
do not fully understand the insurance industry needs and b) no market exists for model output
without insurance. Development of a landslide loss model for insurance is more difficult and
complex than development of an earthquake or hurricane loss model:
1) Landslides are secondary events triggered by primary events (earthquakes and rainfall).
2) Landslides are not necessarily recurring events, unlike earthquakes and rainfall.
3) Given a triggering event (earthquake or rainfall), landslides may or may not actually occur.
4) Landslides damage to buildings and land commonly is total, raising the catastrophe potential.
5) Every loss model is based on the estimation of probability distributions of the essential
predictive variables; such distributions for landslide parameters are unavailable currently.
On the other hand, a landslide loss estimation model should be developed because:
1) Landslides are common, causing serious economic loss in many populated places worldwide.
2) The growing population prefers to build on high ground with scenic views, often near cliffs
or steeply sloping areas that may be prone to landslides.
3) Local governments typically restore damaged land, roads, and infrastructure at great cost.
4) Many hilly areas have unfavorable geologic conditions which can be prone to landslides;
unfavorable conditions may be predictable, but land uses may contribute to instability.
5) Hillsides can be improved to mitigate predicted damage to reduce vulnerability.
6) The mechanics of landslides are largely understood.
7) Geospatial systems allow triggering seismic or storm events and landslides to be integrated.
8) Ground movements can be detected when they are minor, possibly permitting mitigation
measures to be implemented to reduce potential losses.
An approach that is similar to earthquake hazard mapping and risk assessment is needed for
landslide hazard mapping and risk assessment. Analogues between earthquake hazard maps and
landslide hazard maps were described by Perkins (1997); he recognized that landslide intensity
required connections to variations in causes (magnitude and distance for earthquake-induced
slides, rainfall amount and rate for rain-induced slides). He proposed that the landslide
community consider developing probabilistic maps of landsliding caused by these two triggering
processes to identify the probability that the threshold of slope movement might occur. Perkins
(1997) believed that probabilistic maps of events that trigger landslides would be simpler to
prepare than maps that depict the response of the slope to the triggering events. Perkins (1997)
also recognized that procedures for defining landslide susceptibility were needed to permit
estimation of the relative likelihood of landslide movement compared to some benchmark site
conditions (topography, geology, groundwater). The landslide susceptibility information would
allow useful distinctions to be made in relative hazard even though detailed quantification of the
landslide mechanics may not be possible.
Perkins (1997) described four elements of a probabilistic earthquake-induced landslide intensity
map: 1) Geographic description of future earthquakes in terms of magnitude and corresponding
annual rate, 2) Definition of landslide intensity, 3) Landslide intensity relative to earthquake
magnitude and distance for a standard site condition, and 4) Definition of site susceptibility. The
intensity definition is critically important for the needs of the insurance industry compared to the
AEG Professional Forum Extended Abstracts
needs of emergency response officials. A large landslide mass that moves a small distance over a
long period of time could substantially damage buildings and buried infrastructure (water, sewer),
but pose little risk of injury or death to people. Precipitation as a trigger and landslide intensity
relative to precipitation duration and rainrate (intensity) for a standard site condition need to be
included in the probabilistic landslide model.
Fell, R.; Corominas, J.; Bonnard, C.; Cascini, L.; Leroi, E.; and Savage, W.Z., 2008a, Guidelines
for landslide susceptibility, hazard and risk zoning for land use planning: Engineering Geology,
Vol. 102, pp. 85-98.
Fell, R.; Corominas, J.; Bonnard, C.; Cascini, L.; Leroi, E.; and Savage, W.Z., 2008b,
Commentary for Guidelines for landslide susceptibility, hazard and risk zoning for land use
planning: Engineering Geology, Vol. 102, pp. 99-111.
Godt, J.W., 1997, Digital Compilation of Landslide Overview Map of the Conterminous United
States: U.S. Geological Survey Open-File Report 97-289, available from the USGS website accessed January 2015.
Godt, J.W.; Coe, J.A.; Baum, R.L.; Highland, L.M.; Keaton, J.R.; and Roth, R.J., Jr., 2012,
Prototype landslide hazard map of the conterminous United States: in Eberhardt, E.; Froese, C.;
Turner, A.K.; and Leroueil, S., eds., Landslides and Engineered Slopes: Protecting Society
through Improved Understanding, Taylor & Francis Group, London, pp. 245-250.
Haneberg, W.C., and Keaton, J.R., 2012, Ground Truth: An Obstacle to Landslide Hazard
Assessment?: Geological Society of America Abstracts with Programs. Vol. 44, No. 7, p.345.
Jelínek, R.; Hervás, J.; and Wood, M., 2007, Risk Mapping of Landslides in New Member States:
European Commission Joint Research Centre Institute for the Protection and Security of the
Citizen, EUR 22950 EN, accessed January 2015 from the JRC website
Keaton, J.R., and Haneberg, W.C., 2013, Landslide inventories and uncertainty associated with
ground truth: in F. Wu and S. Qi, eds., Global View of Engineering Geology and the
Environment, London, Taylor & Francis, pp. 105-110.
Keaton, J.R., and R.J. Roth, Jr., 2008, Mapping landslides for the insurance industry lessons
from earthquakes: in EuroEnGeo 2008, II European Conference of International Association for
Engineering Geology, Madrid, Spain, CD-ROM Proceedings, Paper 063, 6 p.
Keaton, J.R., and R.J. Roth, Jr., 2009, Potential Value of Landslide Inventory Maps from an
Insurance Perspective: Geological Society of America Abstracts with Program, Vol. 41, No. 7, p.
Perkins, D.M., 1997, Landslide hazard maps analogues to probabilistic earthquake ground motion
hazard maps: in D. Cruden & R. Fell, eds, Landslide Risk Assessment, Balkema, Rotterdam, pp.
AEG Professional Forum Extended Abstracts
Marchesini, I.; Ardizzone, F.; Alvioli, M.; Rossi, M.; and Guzzetti, F., 2014, Non-susceptible
landslide areas in Italy and in the Mediterranean region: Natural Hazards and Earth System
Sciences, Vol. 14, pp. 2215-2231.
Roth, R.J., Jr., and Keaton, J.R., 2010, Fundamental Inputs Necessary for Insuring Landslides: in
Williams, A.L.; Pinches, G.M.; Chin, C.Y.; McMorran, T.J.; and Massey, C.I., eds, Geologically
Active: Proceedings of the 11th IAEG Congress, Auckland, New Zealand, 5-10 September 2010,
CRC Press, CD Paper, 5 p.
AEG Professional Forum Extended Abstracts
Current Status of the U.S. Geological Survey Landslide Hazards Program
Godt, Jonathan W., Program Coordinator, USGS Landslide Hazards,
As the population moving into potentially hazardous areas grows, the overall exposure to
landslide impacts rises. Changing land-use patterns and increasing wildfire frequency
also contribute to an increase in the landslide threat. Recent landslide disasters in
Colorado, Washington State, and in other mountainous parts of the world are stark
reminders that landslide science is needed to assist decision makers and response
agencies, assess hazards, and alleviate impacts from landslides. When triggered by heavy
rainfall or earthquake shaking, thousands of landslides can occur, impacting broad
regions. For example, a days-long rainstorm initiated about a thousand debris flows over
a 3430 km2 area of the Colorado Front Range in early September 2013 killing three.
Debris flows and landslides contributed to the extensive damage to the transportation
network and public and property in the area; losses that triggered more than $500 million
in public expenditures for rebuilding and recovery (Coe et al., 2014). Isolated landslides
such as the Oso slide in Washington State may occur without obvious triggers yet travel
long distances with tragic consequences (Iverson et al., 2015). The Oso landslide
devastated a community, claimed the lives of 43, and disrupted travel on a major east-
west route for more than a month.
The Landslide Hazards Program (LHP) is one of six Programs of the U.S. Geological
Survey (USGS) Natural Hazards Mission Area (Holmes et al., 2013). The LHP and its
predecessor have operated since the mid-1970s as a congressionally authorized program
dedicated to the reduction of damage and avoidance of hazards from landslides. The
fiscal-year 2015 budget of the Program was $3.5 million supporting the work of about 20
scientists, technicians, and others. To enhance resilience to landslides, the LHP conducts
targeted research to understand conditions and processes of landslide initiation and
mobility. This understanding is used to develop methods and models for landslide hazard
assessment, develop and deploy systems to monitor threatening landslides, and to
develop methods and tools for landslide early warning and situational awareness.
Program activities are targeted towards the types of landslides that result in human and
economic losses such as those that are highly mobile, those initiated by heavy rainfall,
and those exacerbated by the effects of wildfire. As the only Federal program dedicated
to landslide science, the LHP provides results of investigations for use by private
consultants in geology and geotechnical engineering and by planners and decision makers
at all levels of government and the private sector.
Natural laboratories to improve process understanding
Improving societal resilience to impacts and adapting to changes in landslide frequency
and magnitude require assessments and forecasts of landslide occurrence. However,
diverse styles of landslide movement and gaps in the understanding of mechanical and
hydrological processes driving slope movement prevent precise prediction of landslide
occurrence. The heterogeneous and hidden nature of the subsurface presents additional
challenges. Research to close gaps in process understanding relies on observations of the
AEG Professional Forum Extended Abstracts
hydrological and other environmental conditions that initiate and exacerbate landslide
initiation and motion. Collected using a system of well-characterized natural laboratories
in landslide-prone locations currently operating in Oregon, California, Colorado, North
Carolina, Pennsylvania, and elsewhere, these observations provide a foundation (e.g. Lu
and Godt, 2013) for improved prediction and forecasting (Baum and Kean, this volume).
Post-wildfire debris flow hazard assessment
Garden-variety rainstorms can initiate debris flows from steep hillsides recently burned
by wildfire. The USGS products for situational awareness for post-fire debris flows are
hazard assessments and early-warning criteria. The hazard assessments provide empirical
estimates of debris-flow probability and magnitude (Cannon et al., 2010) and the early
warning criteria identify rainfall amounts above which debris flows can be expected
(Staley et al., 2013). Burned Area Emergency Response (BAER) teams and the
emergency management community use hazard assessments to plan and prepare for post-
fire debris flows, and the National Weather Service (NWS) uses the early-warning
criteria to issue warnings. In response to requests from BAER teams and other partners,
the LHP recently moved to web-based delivery of the post-wildfire hazard assessments
( Information is now provided in
a format that can be readily ingested into geographic information systems and typically
delivered a few days after wildfires have been contained. Previously, hazard assessments
took about one month to complete. This change in delivery procedure represents
substantial improvements in the timeliness and usability of post-fire debris-flow hazard
Response to landslide crises
The LHP played significant roles in supporting the response to recent landslide disasters.
For example, following the Oso landslide, the USGS dispatched a team to assess the
potential for additional landslide activity to affect search, rescue, and recovery
operations, which at their peak, involved hundreds of professionals and volunteers. The
USGS team provided expertise to search operations and assisted Snohomish County and
Washington State in establishing a system to monitor and assess landslide stability in
near real time. Using technology developed in collaboration with the USGS Volcano
Hazards Program and tested at the natural landslide laboratories (Reid et al., 2012), the
system was operated by personnel both onsite and remotely for five weeks, seven days a
week, during daylight hours until active search operations ended on April 28, 2014.
Going forward
A 2004 National Research Council assessment of a USGS strategy to reduce landslide
losses recommends phasing in a greatly expanded program (NRC, 2004). Envisioned as a
partnership among Federal and State agencies, and the academic and private sectors, the
expanded program would initially focus on targeted research with the emphasis shifting
to operational objectives as the science matures. The NRC framework is relevant today
and the need for a broader program, as indicated by recent landslide disasters, has not
AEG Professional Forum Extended Abstracts
In the coming years, the LHP will expend considerable effort to understand two long run
out landslides; the Oso landslide in Washington State and the West Salt Creek landslide
in western Colorado. Because landslides that run out long distances travel at high speed,
their impact can be great. The LHP will also continue to focus on the debris-flows
triggered by the record-breaking September of 2013 rainstorm in the Colorado Front
Range. Documenting and understanding the conditions that lead to such events provide
important data to test models for prediction and forecasting. Data collection to develop
additional early warning criteria to expand post-fire debris-flow warnings beyond
southern California is underway, as are the data collection and model development to
improve the post-fire hazard assessments. Emerging technologies and techniques for
landslide monitoring and imaging promise to improve landslide detection and process
understanding. Using tools such as terrestrial-based LiDAR and InSAR, imaging from
Unmanned Aerial Systems, and geophysical methods in the research setting of a natural
laboratory provides skill and capability necessary for operational activities and crisis
Partnerships at all levels of government and the academic and private sectors will
continue to be essential for the success of USGS efforts to improve resilience and reduce
losses from landslides. Of particular importance are partnerships with State geological
surveys. State surveys provide much of the fundamental geologic data underpinning
landslide hazard assessments and play key roles in translating scientific understanding
into wise decisions.
Cannon, S. H.; Gartner, J. E.; Rupert, M. G.; Michael, J. A.; Rea, A. H.; and Parrett, C.,
2010, Predicting the probability and volume of post wildfire debris flows in the
intermountain western United States: Geological Society America Bulletin, Vol. 122, pp.
Coe, J. A.; Kean, J. W.; Godt, J. W., Baum, R. L.; Gochis, D. J., and Anderson, G. S.,
2014, New insights into debris-flow hazards from an extraordinary event in the Colorado
Front Range: GSA Today, Vol. 24, No. 10, pp. 4-10. doi: 10.1130/GSATG214A.1.
Holmes, R. R.; Jones, L. M.; Eidenshink, J. C.; Godt, J. W.; Kirby, S. H.; Love, J. J.;
Neal, C. A.; Plant, N. G.; Plunkett, M. L.; Weaver, C. S.; Wein, A.; and Perry S. C.,
2013, U.S. Geological Survey Natural Hazards Science Strategy Promoting the Safety,
Security, and Economic Well-Being of the Nation: U.S. Geological Survey Circular 1383-
F, 79 p.,
Iverson, R. M.; George, D. L.; Allstadt, K.; Reid, M. E.; Collins, B. D.; Vallance, J. W.;
Schilling, S. P.; Godt, J. W.; Cannon, C. M.; Magirl, C. S.; Baum, R. L.; Coe, J. A.;
Schulz, W. H.; and Bower, J. B., 2015, Landslide mobility and hazards: implications of
the 2014 Oso disaster: Earth Planetary Science Letters, Vol. 412, pp. 197-208. doi:
AEG Professional Forum Extended Abstracts
Lu, N. and Godt, J. W., 2013, Hillslope Hydrology and Stability: Cambridge University
Press, New York, NY, 458 p.
National Research Council, 2004, Partnerships for Reducing Landslide Risk: Assessment
of the National Landslide Hazards Mitigation Strategy: The National Academies Press,
Washington D.C., 144 p.
Reid, R. E.; LaHusen, R. G.; Baum, R. L.; Kean, J. W.; Schulz, W. H., and Highland, L.
M., 2012, Real-Time Monitoring of Landslides: U.S. Geological Survey Fact Sheet 2012-
3008, 4 p.,
Staley, D. M.; Kean, J. W.; Cannon, S. H.; Laber, J. L.; and Schmidt, K. M., 2013,
Objective definition of rainfall intensity-duration thresholds for the initiation of post-fire
debris flows in southern California, Landslides, Vol. 10, pp. 547-562. doi:
AEG Professional Forum Extended Abstracts
Canadian Technical Landslide Guidelines and Best Practices for Landslide
VanDine, Doug, PE, PGeo (BC), VanDine Geological Engineering Limited, Victoria,
British Columbia, CANADA,
Réjean Couture, PhD, QC),
Peter Bobrowsky, PhD, PGeo (BC),
This presentation introduces the Geological Survey of Canada’s (GSC) initiative to
develop Canadian technical landslide guidelines and best practices for landslide
In 2006, the GSC was involved in a project with the USGS to inform and educate the
non-professional community about landslides. The goal was to write a document that
could be understood easily by the average citizen. The Landslide Handbook (Highland
and Bobrowsky, 2008) was the result. The handbook has been subsequently translated
and published in Mandarin, Japanese, Spanish and Portuguese.
Following the success of that initiative, in 2010, the GSC consulted with landslide
professionals and users of landslide data across Canada. Over 40 meetings were held with
stakeholders representing academia, government, industry and consultants. The results of
those meetings were summarized by Couture (2010). Some Canadian provinces had
landslide guidelines (e.g. Association of Professional Engineers and Geoscientists of
British Columbia, 2010), but no national guidelines existed in Canada in contrast to other
countries (e.g. Australian Geomechanical Society, 2007). Almost all participants who
were consulted agreed that Canadian landslide guidelines were a good idea. All agreed
that the GSC was the appropriate organization to lead the development and publication
Purpose and Organization
The initiative was to develop a document that would provide guidance, best practices and
additional information for Canadian landslide professionals; one that could be easily
incorporated into practice without being a legislated document or otherwise limiting
practice. The intent was to: 1) update, but not duplicate, information; 2) describe ‘what
should be done’, not ‘how to do it’; 3) focus on Canadian issues and needs; 4) be a
resource for all levels of government; and 5) provide a common landslide terminology.
For efficiency, it was decided to publish each chapter ‘on-line’ as a GSC Open File as
soon as it was completed, following appropriate external and internal GSC review.
Accompanying each open file would be a request for review comments from all readers.
It was agreed that the Canadian Technical Landslide Guidelines and Best Practices
related to Landslides: a national initiative for loss reduction should be written by a team
of landslide professionals as a series of chapters under the direction of ‘chapter leads’ and
an overall ‘scientific editor’ and bilingual ‘co-editor’. The entire process would be
overseen and approved by an ‘advisory panel’. D. VanDine (VanDine Geological
Engineering) was retained as scientific editor and J. Lafleur (École Polytechnique de
AEG Professional Forum Extended Abstracts
Montréal) was retained as co-editor. The initial advisory panel was composed of the
Chair, J. Locat (Université Laval), R. Couture (GSC and Federal Government Scientific
Authority), D. Cruden (University of Alberta), C. Froese (Alberta Geological Survey and
Canadian Geotechnical Society (CGS) representative), M. Jakob (BGC Engineering), N.
Morgenstern (University of Alberta), M. Ruel (CN Rail) as well as the scientific editor
and co-editor. In 2012, P. Bobrowsky (GSC) assumed the role of Federal Government
Scientific Authority and M. Porter (BGC Engineering) assumed the role of CGS
After several meetings, the editors and the advisory panel identified various chapters and
chapter leads (Table 1 at end of abstract). The scientific editor, co-editor and chapter
leads were given financial honoraria for their efforts. The remainder of the contributions
were on a voluntary basis.
Status as of early 2015
Between the fall of 2010 and now, the chapter leads and 60 other landslide professionals
researched and wrote their respective chapters. As completed, each chapter was reviewed
by the scientific editor, three reviewers; two external to the GSC; one internal to the
GSC, and the advisory panel. Upon approval and with the required changes, the chapters
were published on-line as GSC Open Files (Table 2 at end of abstract). The chapters can
be downloaded from, and typing ‘Open File XXXX’ (the open
file number). With the publication of each open file, review comments were encouraged
and have been received from both the national the international landslide communities.
One chapter, Terminology, has already been revised.
Summary of Content
Currently (January 2015) 10 of the 11 chapters have been completed and nine have been
published; one chapter should be published shortly; the remaining chapter is being
finalized. The following briefly summarizes the content of each chapter.
Introduction (GSC Open File 6765) sets the stage by introducing landslides in Canada,
then reviews the purpose, scope and limitations of, and the process used to develop,
the subsequent chapters.
Terminology (GSC Open File 6824) provides a glossary of common landslide-related
terms used in Canada, and elsewhere, and sets the standard for common usage in the
subsequent chapters. The revised version of this chapter (GSC Open File 7623)
expands the number of terms and definitions and also provides a reference citation for
each term.
Classification, Description, Causes and Indirect Effects (GSC Open File 7359)
updates Cruden and Varnes (1996) and provides a method to classify and describe
landslides. It also discusses related topics such as landslide size, intensity, travel
angles, causes and indirect effects.
Socio-Economic Significance (GSC Open File 7311) summarizes the socio-economic
significance of Canadian landslides considering both direct and indirect costs. It
reviews the significance of Canadian landslides in the context of other natural hazards
and landslides world-wide. It includes a summary of losses associated with 56 notable
Canadian landslides.
AEG Professional Forum Extended Abstracts
Review of Existing Guidelines (GSC Open File 7058) reviews 35 national and
international guidelines, standards, and best practices-type publications, as they relate
to landslide risk management, landslide zoning, geotechnical assessment, land use
planning, treatment, and codes of responsibilities.
Professional Practice (GSC Open File 6981) summarizes current requirements for
landslide professionals: qualifications, responsibilities, and quality management. It
also addresses the related issues of professional liability and insurance including
property owner’s landslide insurance.
Risk Management (GSC Open File 6996) reviews the steps of an effective landslide
risk management process: initiation, assessment (identification, analysis, and
evaluation), treatment, communication and consultation, and monitoring and review.
Identification and Mapping (GSC Open File 7059) focuses on approaches associated
with: landslide identification; mapping, map elements, components and types of
landslide maps; and field mapping methods, including field description of landslides.
Investigation, Analysis, Monitoring and Treatment (GSC Open File pending) will
briefly review both commonly used methods, as well as introduce the reader to new
methods and techniques that are currently being developed, tested and used.
Risk Evaluation (GSC Open File 7312) examines aspects of landslide risk: individual
vs. societal risk; voluntary and involuntary risk; tolerable vs. acceptable risk;
qualitative and quantitative risk methods; and partial risk. It includes examples of
current landslide tolerable risk criteria across Canada.
Examples of Common Landslide Types (GSC Open File to be published shortly)
provides a number of examples of common Canadian landslide types, and their typical
Going Forward
Eventually the chapters will be revised, updated and compiled into a GSC bulletin-series
publication entitled Canadian technical landslide guidelines and best practices related to
landslides: a national initiative for loss reduction. GSC bulletins typically include
comprehensive final technical reports on topics of both national or broad regional and
local interest and are available for free download.
The authors are grateful for the support provided by Natural Resources Canada (Public Safety
Geosciences Program), and the tremendous contributions made by the chapter leads, contributors
and authors, reviewers, editors, and members of an advisory panel.
AEG Professional Forum Extended Abstracts
Table 1. Chapters and Chapter Leads
Classification, Description, Causes and Indirect
University of Alberta
Socio-Economic Significance
Review of Existing Guidelines
Professional Practice and Insurance Issues
VanDine Geological Engineering
Risk Management
VanDine Geological Engineering
Identification and Mapping
Investigation, Analysis, Monitoring and
Risk Evaluation and Communication
BGC Engineering
Examples of Common Landslide Types
University of British Columbia and
Université Laval
Table 2. Chapters, Authors, Year of Publication and GSC Open Files (Publications Are Available
From Http://Geopub.Nrcan.Gc.Ca/
Open File
R Couture
R Couture
Terminology (revised)
P Bobrowsky and R Couture
Classification, Description, Causes
and Indirect Effects
D Cruden and D VanDine
Socio-Economic Significance
R Guthrie
Review of Existing Guidelines
B Wang, M Ruel, R Couture, P Bobrowsky and
A Blais-Stevens
Professional Practice and Insurance
D VanDine
Risk Management
D VanDine
Identification and Mapping
L Jackson, Jr., P Bobrowsky and A Bichler
Investigation, Analysis, Monitoring
and Treatment
P Bobrowsky, S Bean, M-A Brideau, M Lato, S
McDougall, S Powell, and D VanDine
Risk Evaluation and Communication
M Porter and N Morgenstern
Examples of Common Landslide
O Hungr and J Locat
in press
*not all contributors are listed as authors or co-authors, but they are acknowledged.
AEG Professional Forum Extended Abstracts
Association of Professional Engineers and Geoscientists of BC. 2010. Guidelines for Legislated
Landslide Assessments for Proposed Residential Developments in BC, revised May 2010, 75p.
Australian Geomechanics Society. 2007. A National Landslide Risk Management Framework for
Australia. Australian Geomechanics Society, Australian Geomechanics, 42(1): 1-12.
Bobrowsky, P, and Couture, R. 2014. Landslide Terminology Canadian Technical Guidelines
and Best Practices related to Landslides. Geological Survey of Canada Open File 7623, 68p.
Bobrowsky, P, Bean, S, Brideau, M-A, Lato, M, McDougall, S, Powell, S and VanDine. D.
(pending) Investigation, Analysis, Monitoring and Treatment Canadian Technical Guidelines
and Best Practices related to Landslides. Geological Survey of Canada.
Couture R. 2010. Summary Report - Consultation with potential Canadian users groups and
contributors, National Technical Guidelines on Landslides, internal ESS Geosciences for Public
Safety Program report, 49p.
Couture, R. 2011a. Introduction Canadian Technical Guidelines and Best Practices related to
Landslides: a national initiative for loss reduction. Geological Survey of Canada Open File 6765,
Couture, R. 2011b. Terminology Canadian Technical Guidelines and Best Practices related to
Landslides: a national initiative for loss reduction. Geological Survey of Canada Open File 6824,
Cruden, D and VanDine, D. 2013. Classification, Description, Causes and Effects Canadian
Technical Guidelines and Best Practices related to Landslides: a national initiative for loss
reduction. Geological Survey of Canada Open File 7359, 22p.
Guthrie, R. 2013. Socio-Economic Significance Canadian Technical Guidelines and Best
Practices related to Landslides: a national initiative for loss reduction. Geological Survey of
Canada Open File 7311, 19p.
Highland, L and Bobrowsky, P. 2008. The Landslide Handbook a guide to understanding
landslides. United States Geological Survey Circular 1325, 129p.,
Hungr, O and Locat, J. (in press) Examples of Common Landslide Types Canadian Technical
Guidelines and Best Practices related to Landslides. Geological Survey of Canada.
Jackson, L Jr., Bobrowsky, P, and Bichler, A. 2012 Identification and Mapping Canadian
Technical Guidelines and Best Practices related to Landslides. Geological Survey of Canada
Open File 7059, 33p.
Porter, M and Morgenstern, N. 2013. Risk Evaluation and Communication Canadian Technical
Guidelines and Best Practices related to Landslides. Geological Survey of Canada Open File
7312, 21p.
VanDine, D. 2011. Professional Practice and Insurance Issues Canadian Technical Guidelines
and Best Practices related to Landslides. Geological Survey of Canada Open File 6981, 13p.
AEG Professional Forum Extended Abstracts
VanDine, D. 2012. Risk Management Canadian Technical Guidelines and Best Practices related
to Landslides. Geological Survey of Canada Open File 6996, 8p.
Wang, B, Ruel, M, Couture, R, Bobrowsky, P and Blais-Stevens, A. 2012. Review of Existing
Guidelines Canadian Technical Guidelines and Best Practices related to Landslides. Geological
Survey of Canada GSC Open File 7058, 13p.
AEG Professional Forum Extended Abstracts
Landslides: the poor stepchild of geological hazards; what we can do and what other
countries are doing to solve this situation
Scott Burns, Professor Emeritus, Portland State University, Dept. of Geology;
For years landslides have taken the back seat to earthquakes, tsunamis, floods, and
volcanic eruptions when it comes to attention from decision makers of news
organizations, funding, and emergency management. All major discussions talk about
other geological hazards first. Insurance for landslides is still practically impossible to
secure for homeowners. Emergency managers take care of the other geological hazards
Things have changed in the last year with the new poster child of landslides, the Oso
Landslide (or also called the Highway 530 Landslide), happening in Washington. This
event that killed 43 people on March, 2014 and caused extensive damage, made national
and international headlines. Some reports say that it is the most deadly landslide in US
history. Analysis of the landslide reveals that it happened in an area with extensive past
landslide activity. Questions like, “Could it have been prevented?” and “Should people
have been allowed to live in the area?” arose. National focus on landslides increased as a
result of this one event. We need to take this event and use it to help secure better
mapping of landslides, the production of landslide susceptibility maps, and making
landslide insurance available to homeowners. The time is now.
We now have the ability to make excellent hazard maps of landslides. Using LiDAR
imagery, we can now make excellent landslide inventory maps. We can create landslide
susceptibility maps using GIS software coupled with analysis of past landslides of the
region. These maps can be used by decision makers regarding building on property
prone to landslides. Combining susceptibility maps with maps of infrastructure and
population distributions, one can also make good risk maps. All of these can then be
used by land use planners to make better decisions about land use related to landslides.
The technology is now there; we just need to convince planners to use them.
Another need is the availability of landslide insurance for homes. Insurance is now
available for floods and earthquakes as an add-on to a homeowner’s policy. It is my hope
that we can also have insurance available for landslides. Some countries already have
landslide insurance available to homeowners such as Switzerland and Germany. Others
have all-hazard insurance, like that which is available in New Zealand. Insurance
companies need actuarial information gathered before they can start offering landslide
insurance. Work by the USGS and state geological surveys in recording landslide losses
will help towards this end.
AEG Professional Forum Extended Abstracts
AEG Professional Forum Extended Abstracts
U.S. Army Corps of Engineers Reflections on Ground Failures and Particularly
Olsen, Richard, PhD, P.E. Senior Geotechnical Engineer, U.S. Army Corps of Engineers,
The U.S. Army Corps of Engineers (USACE) has a long history of investigating and
evaluating ground failures, from earth embankment slides to massive landslides in
foreign countries. USACE has been requested by almost every US federal agency as well
as other governments to provide assistance for difficult ground failure issues including
massive landslides. At present USACE has 1000+ geotechnical engineers, geologists,
and support staff working on a wide range of projects. USACE is constantly improving
and educating geo professionals at all levels of expertise and experience with unique
class room classes, web courses, special training, and special mentoring efforts. For
ground failure issues USACE has developed new concepts for emergency based
reconnaissance, field investigations, unique ground failure mechanics, probabilistic risk
based evaluation, etc. USACE has evaluated landslides and ground failures throughout
the world (and for most geologic formations) but our uniqueness is our size we can
quickly bring in experts for field effort to report review. Our national view on ground
failure evaluation is that we have the right experts and can develop large in-house based
groups very quickly having high effectiveness.
This lecture will start by describing USACE history in geosciences, show numerous past
effort examples, and provide an HQ overview of USACE efforts in geosciences
concerning ground failure evaluation and in particular landslide evaluation.
AEG Professional Forum Extended Abstracts
AEG Professional Forum Extended Abstracts
State programs and implementation
Landslide Hazard and Risk Reduction Process in Oregon
Burns, William J., MS, CEG, Oregon Department of Geology and Mineral Industries
Landslides are one of the most widespread and damaging natural and sometimes human-
induced, disasters in Oregon. An examination of statewide direct damage losses from
landslides found average annual estimates over $10 million and over $100 million in
during severe winter storms (Wang and others, 2002). In order for communities in
Oregon to actively work on landslide risk reduction, they must first better understand the
hazard and risk.
Over the last 10 years, the Oregon Department of Geology and Mineral Industries
(DOGAMI) have developed a landslide risk reduction process. The process was
developed with input from many sources including the U.S. Geological Survey Landslide
Program and several U.S. State Geological Surveys. The process includes the following
components (Figure 1):
1. Collect light detection and ranging (lidar) data
2. Complete landslide inventories
3. Create landslide susceptibility maps
4. Create a database of asset data important to the communities
5. Perform regional risk analysis
6. Implement risk reduction actions
Figure 1. Diagram of the landslide risk reduction process
During the last 10 years DOGAMI implemented this process through projects with
communities (cities, counties, other state agencies, and federal agencies) in Oregon. We
have found the key to success of this process is the collaboration with the community
from the beginning of the process all the way through the end.
The first step is to collect LiDAR data throughout the community. In a study presented at
the First North American Landslide conference (2007), Burns (2007) reported the use of
the LiDAR data resulted in the identification of between 3 to 200 times the number of
landslides found with other data sets. The accuracy of the spatial extent of the landslides
identified was also greatly improved with lidar data. DOGAMI is the lead of the Oregon
AEG Professional Forum Extended Abstracts
Lidar Consortium (OLC). The basic OLC strategy for developing data acquisition areas
is to start with a local funding partner and work to enlarge that area by finding additional
partners and adding OLC funds to link areas together into large contiguous blocks
(DOGAMI, 2009).
The next step is to complete a detailed inventory of the existing landslides in the study
area. In order to create the best available and consistent landslide inventories in Oregon,
Burns and Madin (2009) developed the Protocol for Inventory Mapping of Landslide
Deposits from Light Detection and Ranging (lidar) Imagery. Once a landslide inventory
is complete, it is published and then compiled into the Statewide Landslide Information
Database for Oregon (SLIDO)(Burns, 2014). The current version of SLIDO, 3.2 has over
53,000 landslide locations from 351 studies. The database can be viewed and/or
downloaded from the internet at
After the landslide inventory is complete, landslide susceptibility maps can be created. In
order to create consistent susceptibility maps across Oregon, methods were developed for
both shallow and deep landslides (Burns and others, 2012 and Burns and others, 2013).
DOGAMI is currently working on a channelized debris flow susceptibility method and
has plans to work on rock fall in the near future. Figure 2 is a facsimile of a landslide
inventory, shallow landslide susceptibility and deep landslide susceptibility maps of the
NW quarter of the Oregon City quadrangle, which is in the southern portion of the
Portland region.
Figure 2. Suite of landslide maps including landslide inventory, shallow landslide susceptibility and deep
landslide susceptibility maps of the NW quarter of the Oregon City quadrangle (Burns and Madin, 2009;
Burns and others, 2013).
In order to evaluate risk, the assets must be minimally defined and understood. Some of
the basic asset datasets we examine in landslide projects have included the following: 1)
land use and value, 2) building use and value, 3) permanent resident location and count,
4) primary infrastructure [roads, electric transmission, rail, facilities], and 5) essential
facilities [schools, police, fire, hospitals]. Once the hazard and assets are better
AEG Professional Forum Extended Abstracts
understood and data is captured in a geographic information system (GIS), various levels
of risk analysis can be performed. This process can range from simple to complex.
We have been performing two types of risk analysis: exposure (at-risk) and FEMA’s
HAZUS. The exposure analysis is completed using GIS software. Simply put, an asset is
considered to be exposed to the hazard if it is located within a selected hazard zone.
Exposure is determined through a series of spatial and tabular queries between hazard
zones (landslide inventory and susceptibility) and assets (buildings, land, people,
infrastructure, etc.) and reported by the community (city, county, state). In DOGAMI’s
study of Clackamas County, we found 2,885 landslides in the inventory, but without the
additional information, it is difficult for a community to understand if there is risk. In the
same study, we found over 7,000 residents and over 3,000 buildings are located on
existing large deep landslides, which were the additional risk component the community
needed to help prioritize actions (Burns and others, 2013).
The second type of risk analysis we perform is one using Hazus-MH, a risk modeling
software package developed by FEMA, the National Institute of Building Sciences
(NIBS), and other public and private partners (FEMA, 2011). Hazus-MH software can be
used to model a variety of earthquake, flood, and wind probabilistic hazards and/or
hazard event scenarios. Because there is no landslide module, we use the earthquake
module with and without earthquake induced landslide hazards. Then we subtract the 2
outputs so that the earthquake induced landslide damage and losses can be examined
alone. In a study of the City of Astoria (Burns and Mickelson, 2013), DOGAMI found
the loss ratio (ratio of the total asset value to the expected loss value) from a typical
crustal earthquake increased from 12% to 21% when the landslide hazard component is
considered. The effect of co-seismic triggered landslides results in nearly double the
losses caused by earthquake shaking alone (Burns and Mickelson, 2013).
All of the information generated (as outlined above) must then be transferred to the
community usually through a series of meetings and discussions. The next step is to work
on landslide risk reduction. Landslide risk reduction and mitigation can be facilitated by
three primary actions, including: 1) increasing awareness, 2) putting in place appropriate
regulations, and 3) comprehensive planning.
Making everyone aware of the hazard in their area is crucial to help them understand the
associated danger and how they can prepare themselves. One of the main purposes of the
new inventory and susceptibility maps is to help accomplish/promote education
throughout the communities. Once the hazard is understood better, the entity can work on
risk reduction. Fliers can be made available on websites and/or distributed to help
educate individuals of useful activities to reduce landslide risk. Examples of helpful
flyers include Homeowners Guide to Landslides (Burns and others, n.d.) and DOGAMI
fact sheet Landslide hazards in Oregon
Maps produced as part of these projects are suited for use in a landslide ordinance and/or
building code regulation. The maps could also be used in short-and long-term
AEG Professional Forum Extended Abstracts
development planning and comprehensive planning and maintenance planning. Some
planning could result in avoidance in high hazard areas and even buyouts in very high or
life-threatening areas.
Some landslides are hard to mitigate, especially if development has already occurred.
These situations will likely need to involve cooperation from several entities (for
example, city and land owners) as the slides can span entire neighborhoods. To reduce
the likelihood of a slide reactivating, a public awareness campaign should be undertaken
to educate homeowners and land owners about the landslide hazards in their areas and
how to reduce their risk. Residents on mapped landslide areas should participate in a
neighborhood risk reduction program where all affected land owners (private and public)
help reduce the overall risk. Risk reduction measures should include: minimizing
irrigation on slopes, avoiding removing material from the base of slopes, avoiding adding
material or excess water to top of slopes, draining water from surface runoff, down-
spouts, and driveways well away from slope and into storm drains or natural drainages,
and consulting an expert to conduct a site-specific evaluation if considering major
Burns, S. F., Harden, T. M., and Andrew, C. J., [n.d.], Homeowner’s guide to landslides:
recognition, prevention, control, and mitigation: Portland, Oreg., Portland State University, and
Federal Emergency Management Agency Region X, Bothell, Wash., 12 p. Web: http://
Burns, W.J., 2014, Statewide Landslide Information Database for Oregon, release 3.2: Oregon
Department of Geology and Mineral Industries, Web:
Burns, W.J. and Mickelson, K.A., 2013, Landslide Inventory, Susceptibility Maps, and Risk
Analysis for the City of Astoria, Clatsop County, Oregon: Oregon Department of Geology and
Mineral Industries, Open-File Report O-13-05
Burns, W.J., Madin, I.P., Mickelson, K.A., 2012, Protocol for shallow-landslide susceptibility
mapping: Oregon Department of Geology and Mineral Industries, Special Paper 45
Burns, W.J., Madin, I.P., 2009, Landslide Inventory Map of the Northwest Quarter of the Oregon
City Quadrangle, Multnomah County, Oregon, Oregon Department of Geology and Mineral
Industries, IMS-26
Burns, W. J., 2007, Comparison of remote sensing datasets for the establishment of a landslide
mapping protocol in Oregon. AEG Special Publication 23: Vail, Colo., Conference Presentations,
1st North American Landslide Conference
Burns, W.J., Madin, I.P., 2009, Protocol for Inventory Mapping of Landslide Deposits from Light
Detection and Ranging (lidar) Imagery, Oregon Department of Geology and Mineral Industries,
Special Paper 42
Burns, W.J., Mickelson, K.A., Jones, C.B., Pickner, S.G., Hughes, K.L., Sleeter, R., 2013,
Landslide hazard and risk study of northwestern Clackamas County, Oregon: Oregon Department
of Geology and Mineral Industries, Open-File Report O-13-08, 74 map plates
AEG Professional Forum Extended Abstracts
DOGAMI, 2009, OLC Business Plan,
FEMA (Federal Emergency Management Administration), 2011, Hazus®-MH 2.1, Multi-Hazard
loss estimation methodology, software and technical manual documentation. Web: assets/documents/24609?id=5120
Wang, Y., Summers, R.D., and Hofmeister, R.J., 2002, Landslide loss estimation pilot project in
Oregon: Oregon Department of Geology and Mineral Industries Open-File Report O-02-05, 23 p.
AEG Professional Forum Extended Abstracts
AEG Professional Forum Extended Abstracts
North Carolina Landslide Mapping Public and Private Programs
Bauer, Jennifer B., Appalachian Landslide Consultants, PLLC,
Fuemmeler, Stephen J.,
Landslide mapping in the mountains of Western North Carolina has been, and continues
to be an exercise of science, politics, communication, and patience. The first
comprehensive hazard mapping program in NC began with the North Carolina
Geological Survey (NCGS), although limited mapping of landslides and landslide
deposits had been carried out by previous researchers independently (NC Department of
Transportation; Mills, 1998; Michalek, 1968; Otteman, 2001; Pomeroy, 1991; Witt,
2005; and others). State funding for the NCGS program ended in 2011 with 4 of the 19
designated counties completed. Former NCGS geologists formed Appalachian Landslide
Consultants, PLLC (ALC) to continue the mapping efforts on the private side (Figure 1).
ALC has contracted to non-profit organizations to create landslide inventory and
susceptibility maps for high-need watersheds. This paper discusses the key elements of
each mapping program with respect to triggers for the mapping, funding, scale, general
mapping process, and current status. We will also discuss lessons learned, or key
components to a successful program. Portions of this paper have been discussed in
previous publications (Bauer and Fuemmeler, 2014; Bauer, 2012; Bauer et al, 2012).
Figure 1. North Carolina locations where landslide maps are available
NCGS: Macon, Watauga, Buncombe, Henderson. ALC: watersheds in Haywood and Jackson.
AEG Professional Forum Extended Abstracts
North Carolina Geological Survey Landslide Hazard Mapping Program
Program Triggers
In September 2004, back to back tropical systems crossed the mountains of western
North Carolina within two weeks of each other. Intense rainfall from these events
triggered over 400 slope movements, that caused five deaths, destroyed 27 homes, and
disabled transportation throughout the region (Wooten et al, 2008). Prior to 2005, the
NCGS had a staff of two to three working on a geohazards inventory in Gorges State
Park in Transylvania County, NC and the NC segment of the Blue Ridge Parkway
(Wooten et al, 2007; Latham et al., 2009). After the 2004 storms, the NC General
Assembly passed the Hurricane Recovery Act of 2005, authorizing the NCGS to create
county-wide landslide hazard maps. The stated purpose of the landslide hazard maps
were to inform county and state officials, emergency management, and citizens of current
and potential landslide hazards in their locality and to protect public safety.
Two years of the NCGS geohazards program (2003-2005), and two geohazards positions,
were funded through the NC Division of Emergency Management with FEMA hazard
mitigation grants. The six years (2005-2011) of the NCGS Landslide Hazard Mapping
Program (LHMP) were funded through the state General Assembly appropriations in
response to the disaster declaration from the 2004 hurricanes. This new LHMP allowed
for the creation of five geologist positions dedicated to creating landslide inventory and
hazard maps for the 19 counties identified in the disaster declaration.
In 2007 the Safe Artificial Slope Construction Act was introduced in the General
Assembly, proposing a state-wide steep slope development ordinance which linked
regulations to the landslide hazard maps. This ordinance was never approved, but brought
attention to the hazard maps and potential disclosure requirements during real estate
transactions. In 2008, the LHMP was granted recurring funding through state’s
Appropriations Act and continued until 2011. The economic downturn and the change in
leadership (and majority ideologies) in the NC General Assembly led to the end of
appropriated funding for the LHMP. Elimination of the funding terminated the five
designated geologist positions. Media interviews with State representatives indicate that
the cuts were part of a general budget reduction (Johnson, 2011). However, interviews
also allude to a fear of the maps being used for regulations on private land. One state
Representative was quoted as saying “They are a backdoor approach to more
regulations.” This misunderstanding was in spite of the fact that the NCGS is not a
regulatory agency and as such neither promoted nor discouraged use of the maps for
regulatory purposes. The NCGS stressed (and stated on the maps) that the maps do not
substitute for an on-site stability assessment by a qualified geologist or engineer, but
show areas where such assessments are warranted prior to ground-disturbing activities.
Mapping Process
The NCGS LHMP used intense fieldwork combined with the Stability Index Map
(SINMAP) landslide hazard model to determine where natural debris flows were more
likely to initiate and a “debris flow pathways” model (developed in-house) to show areas
AEG Professional Forum Extended Abstracts
that could potentially be impacted by natural debris flows. The details of this mapping
procedure has been discussed in various other publications (Bauer et al, 2012; Wooten et
al, 2007, Wooten et al, 2011) and will not be included here. Instead, we will highlight the
general process, with a focus on the public outreach and communication components of
the LHMP.
Before mapping began for a particular county, the NCGS met with the county planning
board, emergency managers, soil and erosion control staff and other interested county
personnel. Once mapping was underway, NCGS geologists would talk to landowners as
needed to gain access to property. When a draft of the hazard maps was ready, NCGS
invited county personnel and members of the landslide research community to review the
map products, both in the office and in the field. The NCGS presented the final hazard
maps and landslide statistics to the county, their guests, and Emergency Management.
Lastly, the GIS data files were delivered to county GIS staff and a brief workshop
informed them about the data and how to use it.
In addition to the public outreach mentioned above, NCGS staff gave, and continues to
give, many presentations (upon request) to various local governments, community
organizations, homebuilders and real estate organizations, universities, and professional
conferences. From 2005-2011, the NCGS gave over 200 presentations, field trips, and
articles to over 10,000 audience members (Bauer et al, 2012). The breakdown of these
outreach activities given to different groups is shown in Table 1.
Table 1. Outreach activities of NCGS and ALC.
Type of group
Number of
trips by NCGS
Number of
trips by ALC
Elected officials
Elementary school classes
Emergency Management
Real Estate Brokers
Community or Civic Organizations
Environmentally-minded Organizations
Other Scientists and Engineers
Planning boards, Groups, or Councils of
Geoscience Organizations
Since elimination of the funding for landslide mapping, the NCGS has ceased efforts to
inventory landslides for the purpose of hazard mapping. In the case of emergencies, or
requests from local governments, emergency management, or the public, NCGS
geologists will provide guidance on the use of the maps or evaluate landslides on a case-
AEG Professional Forum Extended Abstracts
by-case basis. Through these requests for technical assistance, the NCGS landslide
geodatabase is maintained and updated.
During the course of the program, inventory and hazard maps were completed for Macon,
Watauga, Buncombe, and Henderson Counties. Work in Jackson County was underway
when the program was terminated. PDF copies of the maps are available and GIS data
layers can be ordered on the NCGS website (
information). The hazard maps were available in an online map viewer through the state,
but were taken offline in 2012.
The fate of the NCGS maps has been varied. In Buncombe County, the maps are used as
part of the Hillside Development Standards to identify areas for conservation or to trigger
a Global Stability Analysis of the property by a licensed geotechnical engineer.
Buncombe County includes generalized versions the NCGS slope movement inventory,
SINMAP, and Known and Potential Debris Flow Pathways layer on their online GIS map
viewer ( The Buncombe
County landslide hazard maps are also included in the Buncombe County Multi-hazard
risk tool online map viewer (
hazard-risk-tool). A detailed explanation of the data layers is not included with either of
these online viewers.
In Macon County, the maps were initially available on the County GIS online map
viewer, but were taken down, as controversy over the County’s planning board and
potential steep slope ordinances escalated. The case of Macon County is one where
misunderstandings about the maps and what they showed prevailed. In a 2011 news
article, one member of the County planning board said he didn’t understand how the
maps were made and was hesitant to take them at face value. He said he saw landslide
mapping as “an arbitrary and fruitless endeavor that will do little to actually predict
where a slide might hit in the future” (Johnson, 2011). Some in Macon County were
afraid the maps would devalue property. One real estate broker stated confusion about
disclosing the maps as a material fact during real estate transactions. Currently, the
Macon County maps are available at the County planning office. The Planner and soil
and erosion control officer use them on a case-by-case basis to answer questions by
residents. There are no regulations tied to the maps.
In Watauga and Henderson Counties, the maps are retained at the County Planners’
offices. Watauga County has an “Ordinance To Govern Subdivisions And Multi-Unit
Structures” which includes guidelines for developing on steeper slopes. Henderson
County has permitting requirements based on slope steepness. Neither county uses the
maps for any regulation.
Private-Non-profit Geological Stability Mapping Programs
Program Triggers
One of the most influential triggers of initiating landslide mapping through a private-non-
profit partnership was the desire to keep the mapping going after the NCGS program
ended. A supporter of the landslide mapping living in Haywood County was disappointed
AEG Professional Forum Extended Abstracts
that his county would not be mapped. He was a member of the Board for a well-respected
water quality non-profit organization and saw the benefits the maps could have to protect
water quality. This supporter approached the authors and other members of the NCGS
program, asking if they were interested in mapping Haywood County as a private entity.
Soon after, the authors started Appalachian Landslide Consultants, PLLC (ALC) as a
private consulting firm offering landslide mapping and site-specific evaluations as part of
their services.
Summer of 2011, the non-profit organization began writing grant proposals, and was
awarded a grant on the condition of finding other project partners. The supporter arranged
to speak at a meeting of the County Commissioners, where he received unanimous
support of the idea from the Commissioners of both political parties. This support did not
come with county funding, however. Another non-profit organization became a project
partner and solicited grant funds to provide a match. The majority of the funding came
from private, water quality grant funds, although a small portion was from the federal
Appalachian Regional Commission fund.
Due to the available funds, counties were broken down into sub-watersheds (Figure 1)
and map products were broken into inventory and susceptibility maps. Initial funding
provided in 2012 was for inventory mapping of the Jonathan and Richland Creek
watersheds (84,465 acres) in Haywood County. Subsequent funding in 2013 supported
susceptibility maps for these watersheds. In 2014, inventory mapping in the East Fork of
the Pigeon River in Haywood County, and in the Wayehutta Creek/Tuckaseegee River
watershed in neighboring Jackson County was funded.
Mapping Process
Having seen the damage that misunderstandings can cause to the perception of landslide
mapping, ALC and its project partners re-evaluated all aspects of the mapping program
with a focus on public communication and simplified map products. This new program
encouraged communication with those groups who had shown resistance to the NCGS
program and included members of the Board of Realtors and Home Builders Association
as well as local governments and non-profits as project stakeholders. Meetings with the
stakeholders were held before, during, and after the mapping took place. Separate
meetings with the Board of Realtors and Home Builders Association were also held to
address their specific concerns relating to tax and property values, insurance, real estate
disclosure, and slope ordinances. ALC also created an informational brochure that
explained the project to handout to landowners and other members of the community.
ALC developed an empirical geomorphic debris flow susceptibility model to use in place
of SINMAP. This model is both easier for the general public to understand and captures
more natural landslides in a smaller area than SINMAP. This layer was titled “Where
natural debris flows might start” to make its purpose clear to the user. A model similar to
the potential debris flow pathways model used by the NCGS was carried over to this
program. This layer was titled “Where natural debris flow might go.” A third layer was
added to the hazard map products title “Slope construction caution areas” to highlight
AEG Professional Forum Extended Abstracts
slope steepness over an average 20 degree threshold where the vast majority of failures
on human modified slopes occur.
Education on how to use the maps is still underway. ALC gave a presentation about the
final maps to a joint meeting with the Haywood County Board of Realtors and the Home
Builders Association. The completed maps, as well as a user’s guide, are available on
ALC's website ( A “solutions
manual” is being prepared to help landowners who have slope stability problems on their
property. Public outreach by ALC since 2011 includes 47 presentations and field trips to
around 1090 attendees. The breakdown of these activities is in Table 1.
Sub-watershed scale landslide inventory and susceptibility mapping continues based on
availability of grant funding, which can vary from year to year and funding cycles. To
date, state and local government funding has not been used for these private/non-profit
partnership created landslide maps. In general, public and non-profit support continues
and increases with continued public outreach efforts.
Lessons learned
Through involvement in both the public NCGS program, and private/non-profit ALC
program, we have learned many lessons about the importance of communicating the
science effectively. Political tendencies of the time will affect the style and message of
this communication. We have found that it is best to stress that the maps provide
information, and are non-regulatory. We promote their usefulness in making informed
decisions about one’s own safety, protection of their investments, and protecting the
regions’ water quality. Most import is communicating to the audience “What’s in it for
Public outreach and education for the community and including stakeholders from a
variety of backgrounds is important for heading off misunderstandings about the maps
and what they show. By personally meeting with groups that may be opposed to the idea
of mapping, personal connections are made. These personal connections open the door to
a better understanding of the science and information. When first meeting with the
County Board of Realtors, we met resistance to the mapping. By the end of the inventory
Phase, they wanted the maps available as soon as possible, so they could better serve
their clients.
Another key component to success of landslide mapping is having a community
champion outside of the mappers themselves. This person needs to be someone that
others in the community trust, doesn’t have their own political agenda, and believes in the
results of the mapping. County Emergency managers can fill this role, but can sometimes
be tied by the positions of their County Commissioners. A private citizen or widely-
respected non-profit can fill this role as well.
AEG Professional Forum Extended Abstracts
Finally, not everyone will be in agreement about the usefulness of the maps. Openly
hearing what they have to say, responding respectfully and factually, and moving on is
the most productive way to handle these situations. Lack of responding or emotionally
charged responses can fan the flames of opposition.
Whether mapping one parcel, sub-watershed, one county, or one state, for the mapping to
be effective in saving lives and property, the end users must be aware of the project from
the beginning. Most often we see public interest in landslides peak just after a natural
disaster, but they quickly forget not long after the event. As stakeholders in promoting the
science, benefits, and use of landslide susceptibility maps, we must work to consistently
spread our message the public and those who communicate to them.
Bauer, Jennifer B. 2012. North Carolina Geological Survey Landslide Hazard Mapping Program
What We Learned and How You Can Use It. Assoc Env. & Eng. Geol. Presidential
presentations, unpublished.
Bauer, Jennifer B., Fuemmeler, S.J., Wooten, R.M., Witt, A.C., Gillon, K.A., Douglas, T.J. 2012.
Landslide hazard mapping in North Carolina overview and improvements to the program: in
Eberhardt, E.; Froese, C.; Turner, A.K.; and Leroueil, S., eds., Landslides and Engineered Slopes:
Protecting Society through Improved Understanding, Taylor & Francis Group, London, pp.257-
Bauer, Jennifer B., Stephen J. Fuemmeler. 2014. Geologic Stability Mapping in Haywood
County, NC A case study of a team effort. Association of Environmental & Engineering
Geologists 2014 Annual Meeting Scottsdale, Arizona, Program with Abstracts, p.43.
Johnson, Becky. 2011. Landslide hazard maps axed by state: Risky slopes in Jackson, Haywood
to remain a mystery for now. Smoky Mountain News, June 29, 2011, online edition.
Latham, Rebecca S., Wooten, Richard M. , Cattanach, Bart L, Merschat, Carl E, Bozdog,
George, N., 2009. Rock slope stability analysis along the North Carolina section of the
Blue Ridge Parkway: using a geographic information system (GIS) to integrate site data and
digital geologic maps, 43rd US rock mechanics symposium and 4th US-Canada rock mechanics
symposium, American Rock Mechanics Assoc, , 28 June 1 July 2009, 12p.
Mills, H.H., 1998, Deposits and landforms on the piedmont slopes of Roan, Rich, and Snake
Mountains, Northwestern North Carolina and Northeastern Tennessee: Southeastern Friends of
the Pleistocene 1998 Field Trip Guidebook, p. 50-82.
Michalek, D.D., 1968, Fanlike features and related periglacial phenomena of the southern Blue
Ridge: PhD Dissertation, University of North Carolina at Chapel Hill, 198 p.
Otteman, R.A., 2001, Using GIS to model debris flow susceptibility for the Bent Creek
Experimental Forest near Asheville, North Carolina: Master's Thesis, East Carolina University,
181 p.
AEG Professional Forum Extended Abstracts
Pomeroy, J.S. 1991. Map showing late 1977 debris avalanches southwest of Asheville, western
North Carolina: U.S. Geological Survey Open-File Report 91-334. 25 p. map scale 1:24,000.
Witt, A.C. 2005. Using a GIS (geographic information system) to model slope instability and
debris flow hazards in the French Broad River watershed, North Carolina: Master’s thesis,
North Carolina State University, Raleigh, NC. 165 p.
Wooten, R.M., Latham, R.S., Witt, A.C., Gillon, K.A., Douglas, T.J., Fuemmeler S.J., Bauer,
J.B., Reid J.C. 2007. Landslide hazards and landslide hazard mapping in North Carolina. In:
Schaefer VR, Schuster RL, Turner AK (eds) Conference presentations 1st North American
landslide conference, Vail Colorado. Assoc Environ Eng Geol Special Publication 23:458471.
Wooten, R.M., Gillon, K.A., Witt, A.C., Latham, R.S., Douglas, T.J., Bauer, J.B., Fuemmeler,
S.J. & Lee, L.G. 2008. Geologic, geomorphic, and meteorological aspects of debris flows
triggered by Hurricanes Frances and Ivan during September 2004 in the Southern Appalachian
Mountains of Macon County, North Carolina (southeastern USA): Landslides 5(1): 31-44.
Wooten R.M., Witt A.C., Douglas T.J., Fuemmeler S.J., Bauer J.B., Gillon K.A. & Latham R.S.
2011. Slope movement hazard maps of Henderson County, North Carolina. North Carolina
Geological Survey Digital Data Series, GHMS-5 (DDS-GHMS-5).
AEG Professional Forum Extended Abstracts
Landslides in Kentucky: Inventory, Data Delivery, and Collaboration
Crawford, Matthew M., Kentucky Geological Survey,
Although the exact costs for mitigating landslides in Kentucky are not known, they are
significant. Data collected since 2000 from the Kentucky Transportation Cabinet, who
landslide and rockfall damage and repair along roadways, suggests mitigation costs
exceed $10 million per year (Figure 1). Data collected from 2003 to 2013 shows the
Kentucky Hazard Mitigation Grant Program funded projects that acquired landslide-
damaged homes, or stabilized a landslide affected area, for a total of $5.3 million. Along
the Mississippi River in Hickman, one of the largest landslides in the state affected
numerous buildings and the towns water supply, costing the U.S. Army Corp of
Engineers more than $17 million to stabilize the slope. Hazard mitigation efforts continue
across Kentucky to help citizens facing landslide problems; however, obtaining funding
and implementing these projects can take years. The State and local government agencies
that respond to or document landslides vary, and the data collection and assessment of
landslide activity among these agencies also vary widely. This paper aims to discuss the
tools and research important for assessing landslide hazards, but also advocate the need to
connect research and communication, in order to guide the public and ensure reduction in
risk and losses.
Figure 1. Thin, translational landslide that damaged a road in central Kentucky. Photo by Brandon Nuttall,
Kentucky Geological Survey.
Landslide Inventory
As population grows and development continues in both urban and rural parts of
Kentucky susceptible to landslides, a foundation for assessing landslide hazards is
imperative. Tools must be developed to ensure that these hazards are recognized and
citizens are educated about landslide risks. In 2010, the Kentucky Geological Survey
(KGS) created a landslide inventory, an applied database containing known landside
locations from different sources. The database facilitates adding, querying, and
AEG Professional Forum Extended Abstracts
displaying data that can help land-use planners, transportation personnel, emergency
managers, meteorologists, geotechnical engineers, and the general public. Most planners
or local government officials do not have the resources to properly address landslides.
Providing these agencies access to a landslide inventory in a geospatial context will
complement any geotechnical site analysis, building and zoning regulations, and
residential property assessment needed to minimize slope failure. Networking and
communicating with officials who deal with landslides and are stakeholders in research is
the challenge to support the need for landslide inventories.
Several other state geological surveys participate in landslide inventory projects
( They collect data for displaying and analyzing
landslide information and coordinate efforts to promote the national and local recognition
of landslide hazards. These efforts are vital in order to communicate landslide hazards to
the public, with the goal of reducing risk.
Data Delivery
The KGS created geographic information system files from the inventory database that
were then used to create an online, interactive Landslide Information Map
( The map provides an overall view of
landslide hazards across the state (Figure 2). Querying the landslide locations shows the
source, geographic information, landslide type, material, underlying geology, landslide
dimensions, date of occurrence, and many other attributes, if available (Figure 3). The
map can be used to identify preexisting landslides and serve as a basis for landslide
hazard assessment. Preexisting landslide deposits are susceptible to subsequent failure, so
a landslide inventory map showing known locations in a geologic and geomorphic
context can be used for land-use planning, slope development regulations, and general
hazard awareness.
Figure 2. The KGS online, interactive Landslide Information Map. Red dots and orange polygons are
known landslide locations. Landslide data is available along with many basemap layers and other geologic
data sets. The gray area is a LiDAR derived hillshade layer.
AEG Professional Forum Extended Abstracts
Local officials and the public can use this map as a tool to assess whether or not they
need to look for site-specific features indicating landslide activity and begin the process
of avoiding damage, future movement, remediation, and loss of life. A quick look at the
map may assist in an effective response to a landslide. The public can add to the landslide
inventory database and map by using the KGS Report a Landslide form. Users can go
online and enter basic observational and descriptive information regarding a landslide,
such as county location, route, failure location, movement rate, size, and damage. They
can also use a comments box to address issues not on the form or ask additional
Figure 3. Zoomed-in view of the Landslide Information Map showing landslide locations (red
dots and orange polygons) and areas susceptible to debris flows (arrows). Each feature is able to be queried
to see landslide related attributes.
KGS is also involved in specific landslide research projects. Several ongoing, site-
specific landslide characterization projects collect geologic, geotechnical, and
geophysical data that can be of use to the engineering community, planners, and the
public. We are specifically looking at electrical resistivity as a way to characterize
landslides (Figure 4). Using electrical resistivity can advance landslide hazard
assessments by establishing relationships among resistivity contrasts and the failure zone,
moisture content, suction, and other factors related to slope stability. Delivering research
data in a timely fashion, in the form of reports, maps, or presentations to stakeholders,
can help government agencies and communities understand the threats that landslides
Figure 4. Electrical resistivity profile showing interpreted landslide features. B1 and B3 are
boreholes that helped interpret features.
AEG Professional Forum Extended Abstracts
Collaboration and Mitigation
KGS has collaborated with others in a variety of ways to help with landslide hazard
mitigation. One example was cooperation and communication with Kentucky Emergency
Management during a post-disaster response. In April 2014, a landslide destroyed an
unoccupied home, threatened others, and threatened to damage an adjacent roadway in
northern Kentucky (Figure 5). Kentucky Emergency Management asked KGS to provide
geologic information, assess the hazard, and determine the steps to mitigate the situation.
Geology, soils, landslide type, causes, and subsequent landslide movement were
discussed. We identified landslide features such as scarps, tension cracks, failure surface,
thick toe slopes, and seeps. This collaboration facilitated safety and mitigation decisions,
and also provided valuable scientific information to Emergency Management and the
Figure 5. An unoccupied home destroyed by a landslide, northern Kentucky.
A second example involved collaboration among the University of Kentucky, Kentucky
Emergency Management, and the Hazard Mitigation Grants Program. This
program provides funding to the state for projects to reduce damages and losses for
existing and future disasters. Federal, state, and local agencies form a partnership to
develop the mitigation plan. The Federal Emergency Management Agency (FEMA)
represents 75% of the costs and the remaining 25% must come from state and local
applicant community. KGS was contacted for a post-disaster landslide hazard assessment
of several homes heavily damaged by landslides in eastern Kentucky (Figure 6). We
identified landslide features outside the homes and damage inside the homes. KGS
provided an assessment of the geologic conditions, landslide causes, and subsequent
movement to be used in a FEMA Hazard Mitigation Grant Program report, ultimately
assisting with the cost-benefit analysis for mitigation decisions regarding the landslide.
AEG Professional Forum Extended Abstracts
These types of collaborations are not happening nearly enough. Providing relevant
scientific information to local officials and the public is an important step in reducing loss
of property, possibly loss of life, and educating the public on the risks and long-term
sustainability. Collaborative efforts, even if spur of the moment or informal, help to
publicize the value of science and sets the stage for future mitigation efforts. However,
establishing a standard practice of communication and information sharing between
scientific experts and communities is preferred.
Figure 6. Examination of a landslide-damaged home, eastern Kentucky. Photo by Mike Lynch, Kentucky
Geological Survey.
Future Guidance
In addition to the collaborations with government offices, a majority of the public-service
requests KGS receives involve landslide damage at a private residence or commercial
building. Few of the people making these requests are educated about slope stability and
the features affecting their property. It is likely that they have some familiarity with the
term landslide, but they are unaware that a landslide could actually occur and put their
property at risk. Usually those affected do not know who to contact for assistance. As the
hazards community knows, standard homeowners insurance policies do not cover
landslide damage. Every county and community has different government offices that
respond to landslide problems. It might be the local emergency management office, the
area development district, the conservation district office, the county judge executive, or
a combination of parties. These government agency responders do their best, but by
themselves, they cannot provide the public awareness and policy that might prevent
landslide damage.
Effective communication and proper guidance to the citizens who need to make decisions
is a challenge. Geologic hazards scientists must make efforts to combine sound research
and communication to stay connected to stakeholders invested in the risks that come with
landslides. How can landslide hazard assessments become connected to the public as
standard practice? For example, the National Weather Service should be more involved
AEG Professional Forum Extended Abstracts
with landslide hazard assessment. They provide the watches and warnings for severe
thunderstorms, flash floods, or other events with excessive precipitation often
accompanied by landslides; however, there is usually no mention of landslides in the
advisories. Another example is to develop a protocol of specific roles and actions among
scientific experts and other agencies that are invested in slope stability, either before or
after a landslide event. Connecting to emergency managers, policy makers, developers,
land-use planners, and homeowners who are involved with hillslope development and
stability is a long-term and complicated goal, but one that must be achieved to reduce
landslide losses.
KGS coordinates multiple efforts to assess, communicate, and mitigate landslide hazards.
We developed and maintain a landslide inventory database and interactive landslide
information map that provides known landslide locations in a geologic and geomorphic
context. The map is an excellent tool to communicate landslide hazards to local
communities and officials who make decisions regarding risk. The landslide inventory
and delivery of landslide research data has resulted in collaborations with state and local
government. The State and local government agencies that respond to or document
landslides vary in their approaches to data collection, assessment, documentation, and
mitigation of landslide activity. Effective communication with these stakeholders is
necessary to start to move toward proactive solutions in order to reduce loss, instead of
retroactively focusing on outcomes of events that have already caused damage.
AEG Professional Forum Extended Abstracts
Landslide inventory in Washington State: the past, present, and future
Slaughter, Stephen L., LEG, Washington State Department of Natural Resources,
Division of Geology and Earth Resources,
The Washington Department of Natural Resources (WDNR) has maintained a landslide
GIS database for nearly a decade. The database now contains over 50,000 landslides (Fig.
1) collected from over 200 mapping projects of varying detail, scale, and purpose.
Despite the large number of landslides, nearly 90 percent of the state lacks the detailed
and systematic landslide inventories that are necessary to accurately identify areas of
landslide susceptibility and hazard. Much of the current landslide inventory focuses on
state-managed forestlands; thus there is a significant lack of detailed mapping in urban
and rural areas.
Figure 1. Map of the current landslide inventory for Washington State. The current database contains over
50,000 landslides (purple polygons), yet much of the inventory is in managed forestland and lacks the
detail necessary for landslide susceptibility and hazard maps. Note a lack of landslides adjacent to city
limits (yellow polygons with orange outline). County lines are dashed.
In 2009, the Division of Geology and Earth Resources (DGER) landslide hazard mapping
program (LHMP) was cut and a detailed landslide inventory has not been added to the
database since. Landslides were far removed from the public consciousness in
Washington State until the 2007 and 2009 storms initiated thousands of landslides, many
in remote and rural areas of the state. There were no deaths associated with storm-related
landslides, and property damage was minimal despite the large area of landslide impact.
AEG Professional Forum Extended Abstracts
On March 22, 2014 the SR530 "Oso" landslide killed 43 people and returned landslides
to the consciousness of Washington State. Renewed attention by concerned citizens,
legislators, and other government officials has rekindled the possibility of restoring the
LHMP to pre-2009 staffing levels. This restoration would create a landslide mapping
program capable of tackling landslide inventory, susceptibility, and hazard mapping in
urban and rural neighborhoods across the state of Washington.
The Past
The DGER landslide database is an amalgamation of landslide mapping projects that
cover a range of detail, scale, and purpose. The primary contributors of landslide data to
the inventory are WDNR and forest management companies whose purpose was to map
landslide hazards in watersheds that may have an increased potential of mass wasting
from forest management activities. The mapping programs, called Watershed Analysis
(WA) and Landslide Hazard Zonation (LHZ), both followed systematic mapping
procedures and depended heavily upon analyst interpretation of aerial imagery and other
remote sensing tools to map landslides and quantify hazards. The quality of landslide
inventories from these programs varies, primarily due to the available technology at the
time and the protocols used by each analyst. Basemaps were typically 10-m digital
elevation models (DEMs) or USGS 7.5-minute topographic maps. Analysts would
identify landslides from aerial imagery and topographic maps and draw landslides on the
basemaps. This type of mapping could introduce significant uncertainty in landslide
location, which becomes problematic when newer and more precise imagery (high-
resolution lidar DEM) shows the landslide boundaries to be incorrect.
An additional shortcoming with the database is the “certainty” category, where the
analyst ascribes their confidence that a mapped landslide exists. Instead of a matrix or
other method to quantify the analysts’ confidence of landslide existence, both LHZ and
WA used a qualitative statement to describe the analysts certainty. This method simply
queries the analyst’s confidence in their skills of observation and interpretation and does
little to inspire confidence for users of the data who recognize the actual significance of
the designation of certainty.
The Present
The database has remained largely unmodified since the conclusion of the WA (~2003) and LHZ
(~2009) programs and budget cuts to the LHMP in 2009. One significant change to the database
was to qualitatively describe, for each mapping project, the confidence of landslide data and
location accuracy. Project-level uncertainty was recognized as an issue when we noticed that
some users of the database wrongly attempted to analyze landslide data inappropriate for the data
quality. For instance, major precipitation events in 2007 and 2009 triggered thousands of
landslides and led to updates to the landslide database; however, much of the mapping was
reconnaissance with poor location confidence. The reconnaissance quality of the data was not
effectively communicated in the database, such that those attempting to analyze the inventory did
not understand the limitations of these data.
Our response to more effectively communicate the database limitations was to categorize
the methods used to identify and map landslides for each mapping project. For example,
did the methodology require the analyst to map landslides using multiple years of aerial
images, lidar DEM, and field reconnaissance? Or, did the analyst merely review one set
of aerial images and perform no field reconnaissance? The intent of these new categories
AEG Professional Forum Extended Abstracts
is to reduce potential misinterpretation or unsuitable analysis of the data. Previously, this
information was not easily available to the typical user of the database; however, it
remains to be seen whether these new categories will impact future use of the data.
The landslide database is viewable and downloadable from an online mapping system,
the Washington State Geologic Information Portal ( The
portal displays the landslide inventory, the mapping project areas, and the updated
landslide data.
The Future
After the SR530 “Oso” landslide, there has been a renewed interest in landslide hazard
mapping by concerned citizens, legislators, and other government officials. In the
summer of 2014, county and state geologists met to determine the interest of a statewide
effort to begin a systematic landslide inventory, susceptibility, and hazard mapping
program. The five participating counties are all in favor of the effort; however, funding
and time are both limiting factors. Furthermore, it was agreed that a common landslide
mapping protocol was necessary and that DGER should take the lead in writing the
protocol. This protocol would be the guiding document for all landslide mapping projects
in the state and is intended to ensure that any project, whether performed by private
companies, universities, counties, or DGER, follow a similar format across the state.
DGER would act as the data repository for landslide mapping and perform quality and
consistency checks for data produced by outside partners. DGER would host and
distribute the data via the Washington State Geologic Information Portal
The next step is to write a protocol that describes the process of landslide mapping.
Fortunately, the Oregon Dept. of Geology and Mineral Industries (DOGAMI) has
produced excellent protocols on mapping landslide inventory from lidar (Burns and
Madin, 2009), mapping shallow landslide susceptibility (Burns and others, 2012), and
mapping deep-seated landslide susceptibility (Burns and others, 2013).These protocols
are examples of the direction the county/state partnership has tentatively agreed on and
intend to pursue. However, implementation of these protocols for Washington State is
hindered by the lack of lidar coverage, which is necessary for creating a high-quality
landslide database. Currently the state has approximately 28 percent of the state is
covered by LiDAR, though more than nearly half of the data is low quality and would
likely need to be recollected.
Though the current staffing of the LHMP (one person) cannot sustain any level of
systematic landslide mapping, a portion of the DNRs budget proposal to the 2015 state
legislature is for DGER to restart the LHMP with up to five additional staff. If funded,
these staff will begin mapping landslides following a systematic landslide inventory and
susceptibility protocol developed over the coming months. Landslide mapping priority
will be set for cities and highways across the state and no mapping projects will
commence without first performing significant outreach with communities and
community leaders to ensure the mapping efforts are wanted and the benefits are
understood. There are many benefits to having a renewed landslide mapping program for
the citizens of the state of Washington; however, the time commitment and necessary
AEG Professional Forum Extended Abstracts
tools to complete a credible and accurate tool will require much-needed staffing and
budgetary support for the hazard mapping program for a long term program.
Burns, William J.; Madin, Ian P., 2009, Protocol for inventory mapping of landslide
deposits from light detection and ranging (LIDAR) imagery: Oregon Department of
Geology and Mineral Industries Special Paper 42, 30 p.
Burns, William J.; Madin, Ian P.; Michelson, Katherine A., 2012, Protocol for shallow-
landslide susceptibility mapping: Oregon Department of Geology and Mineral Industries
Special Paper 45, 32 p.
Burns, William J.,; Michelson, Katherine A.; Jones, Cullen B.; Pickner, Sean G.; Hughes,
Kaleena L. B.; Sleeter, Rachel, 2013, Landslide hazard and risk study of northwestern
Clackamas County, Oregon: Oregon Department of Geology and Mineral Industries
Open-File Report O-13-08, 1 DVD [74 plates, GIS data, 38 p. text].
AEG Professional Forum Extended Abstracts
Landslide risk, prediction, and local implementation
Managing Slope Hazards along Transportation Infrastructure
Badger, Thomas C., Washington State Department of Transportation,
The Problem
The initial establishment of transportation routes that accompanied human settlement
often followed valleys and avoided, when possible, traversing steep terrain. In many
cases, modern transportation infrastructure occupies the same corridor that was used
during settlement, inheriting many of the assets and liabilities associated with its location.
Slope hazards are one such liability, where the corridor is sited near or within adverse
terrain and unfavorable geologic conditions. Earthwork required to maintain reasonable
grade and necessary width, especially prior to modern geotechnical practice, has often
exacerbated the frequency, magnitude, and consequence of slope failures. Where
substantial lengths and widespread distribution of transportation corridors are sited
near/within unstable terrain, such as the highway system in Washington State, there is a
high likelihood for frequent and sometimes severe slope-related disruption to the system,
and thus a need for the strategic management of these hazards.
Management of Slope Hazards
Prior to the early 1990s, the Washington State Department of Transportation (WSDOT)
had a primarily reactive approach to managing problem slopes. In 1993, the agency
implemented the Unstable Slope Management System (USMS) to address the more than
3200 slopes along Washington’s 7000-mile-long highway system that are known sources
of rockfall, landslides, debris flows or erosion (Lowell and Morin, 2000; Lowell et al.,
2005). The program includes four main activities:
Inventory development rationally evaluate all known unstable slopes along
WSDOT’s highways utilizing a numerical rating system.
Slope ranking provide a basis to prioritize slopes based on highway functional
class, which would address the most critical highway facilities with the greatest
needs first.
Project scoping provide for early scoping by preparing conceptual designs and
cost estimates for benefit-cost analyses.
Prioritization prioritize projects, statewide, for design and construction based on
expected benefit from the investment, purposely avoiding a “worst-first”
prioritization approach.
Since program inception, WSDOT (with considerable Federal participation) has invested
over $450 million to remediate roughly 250 unstable slopes either through a proactive
benefit-cost programming process, as urgent conditions have arisen, or as part of other
highway improvement projects. In 2007, WSDOT instituted a risk-reduction scaling
program to partially address rockfall-prone slopes that could not meet programming
criteria for long-term hazard remediation. As many of the high-risk unstable slopes along
AEG Professional Forum Extended Abstracts
the most critical highway corridors have now been addressed , the initial programming
criteria has been revised to broaden the range of slopes eligible for programming.
Even with a proactive management program to address slope hazards, emergent needs
will persist. Further, a decades-long funding commitment is required to realize
measurable benefits from such a program (Golder Associates, 2005). As many slope
hazards that threaten transportation infrastructure lie outside the available right-of-way,
the mitigation opportunities can be greatly limited. Adjacent development common to
transportation infrastructure can also eliminate avoidance measures (i.e., realignment) as
a viable mitigation measure.
Golder Associates, Inc., 2005, Review of Washington State Department of Transportation
P-3 unstable slope inventory and prioritization process: Memorandum to WSDOT (dated
12/9/2005), 9 p.
Lowell, S.M. and Morin, P., 2000, Unstable slope management in Washington State:
TR News 207: pp. 11-15.
Lowell, S.M., Moses, L.J., and Badger, T.C., 2005b, Management of slope hazards and
risk along Washington State (U.S.A.) highways. In Hungr et al. (Editors), Proc. Intl.
Conference on Landslide Risk Management, Vancouver, Canada, supplementary volume
AEG Professional Forum Extended Abstracts
Landslide Modeling and ForecastingRecent Progress by the U.S. Geological
Rex L. Baum, U.S. Geological Survey,
Jason W. Kean,
Landslide studies by the U.S. Geological Survey (USGS) are focused on two main
objectives: scientific understanding and forecasting. The first objective is to gain better
understanding of the physical processes involved in landslide initiation and movement.
This objective is largely in support of the second objective, to develop predictive
capabilities to answer the main hazard questions. Answers to the following six questions
are needed to characterize the hazard from landslides: (1) Where will landslides occur?
(2) What kind(s) of landslides will occur? (3) When will landslides occur? (4) How big
will the landslides be? (5) How fast will the landslides travel? (6) How far will the
landslides go? Although these questions are sometimes recast in different terms, such as
frequency or recurrence rather than timing (when), the questions or their variants address
the spatial, physical, and temporal aspects of landslide hazards. Efforts to develop
modeling and forecasting capabilities by the USGS are primarily focused on specific
landslide types that pose a high degree of hazard and show relatively high potential for
Hazard as a function of landslide type
Type and rate of landslide movement affect the degree of danger posed by landslides and
consequently determine the priorities for assessment and forecasting. Landslides that
move slower than 50 mm/s may cause severe property damage, but rarely kill or injure
people (Cruden and Varnes, 1996). Rapid landslides, moving at speeds faster than 50
mm/s, threaten life safety and may cause damage to the built environment.
Consequently, rock falls, debris flows, rapid earth flows, rock avalanches, and debris
avalanches pose the greatest danger to human life. Slow earth slides and earth flows are
common types of landslides affecting residential properties and infrastructure. These
slow landslides typically occur on slopes of 10 degrees or less and are often unrecognized
prior to development. Slow, almost imperceptible movement of these landslides
gradually destroys streets, utilities and buildings. Predictive capabilities must define both
temporal and spatial aspects of the hazard for rapid landslides, but only spatial aspects for
slow landslides.
Factors affecting spatial predictability
Predicting the locations of future landslides is challenging, but landslide inventories,
factor maps, historical landslide databases, and models can provide some constraints on
future landslide locations. Evidence of past landslides, regardless of type, is the best
indicator of potential locations, type, size, and travel distance of future landslides. High
resolution optical imagery from air- and space-borne platforms and high-resolution
topography, such as light detection and ranging (lidar), are useful for identifying and
mapping evidence of past landslides (e.g. Schulz, 2007). Many landslides occur when
landslide deposits are reactivated by strong ground motion, precipitation, snowmelt, or
AEG Professional Forum Extended Abstracts
other factors. Analyses using statistical methods, artificial neural networks, and
deterministic models combined with geospatial databases of historical landslides are
useful in narrowing the possible locations of future landslides (Baum et al., 2014). With
few exceptions, predicting the exact locations of future landslides is usually quite
difficult due to uncertainty about the physical aspects. For example, uncertainty about
the thickness of colluvium on steep hillsides makes it difficult to predict the exact
locations of shallow rainfall-induced landslides. Similarly, uncertainty about fracture
patterns, groundwater flow, past movements and related factors that determine the
stability of rock slopes make it difficult to predict where future rock slides might occur
that could mobilize into rock avalanches.
Factors affecting temporal predictability
Temporal predictability of landslides varies depending on their size, type and the
immediate cause or “trigger” mechanism that induces the landslides. Large landslides are
less common and usually more difficult to predict than abundant small landslides that
occur over a broad region in response to heavy rainfall. Shallow, rainfall-induced
landslides and the debris avalanches and debris flows that mobilize from them probably
have the highest degree of predictability given the relatively close connection between
rainfall infiltration and pore pressure rise and the ready availability of real-time rainfall
measurements and weather forecasts. Erosion-induced debris flows from recently burned
areas are also highly predictable. This results from the close temporal connection
between rainfall and debris-flow occurrence (Staley et al., 2013). Snowmelt-induced
landslides are more difficult to predict due to the difficulty of quantifying the rate of
melting and infiltration, but an air temperature index can constrain the timing of
snowmelt-induced landslides (Chleborad, 1998). Effects of precipitation on landslides
deeper than a few meters are often delayed and result from the cumulative effects of
precipitation over time periods spanning days to years, depending on the local geology,
depth, and site-specific conditions.
Various factors influence the temporal predictability of certain landslide types. The
timing of earthquake-induced landslides is considered unpredictable despite the close
temporal correspondence between landslides and strong ground motion, because the
timing of earthquakes is fairly unpredictable. Timing of landslides induced by human
activities, such as irrigation and excavation, is also usually difficult to predict due to the
dependence of these landslides on site-specific conditions. Challenges in predicting
timing of rock fall include the poor correlation with rainfall, their common occurrence
during earthquakes, and abundant events without an obvious trigger. Recent studies
indicate that temperature variation affects deformation of large slabs in exfoliating
granite (Collins and Stock, 2014). In some areas, rock fall has long been attributed to
freeze-thaw cycles; however the data seem inconclusive. Much remains to be learned
before rock fall timing can be predicted accurately and consistently.
Timing of rock avalanches and large debris avalanches is usually difficult to forecast.
Rock avalanches typically originate on the faces of large, deep rock slides, which may
move as a result of earthquakes, toe erosion, gradual buildup of pore-water pressure,
mining, or other causes. Rock avalanches have been predicted only in rare cases where
AEG Professional Forum Extended Abstracts
detailed, continuous monitoring has been performed (Panikow et al., 2014). Predicting
the timing of large, deep rock slides and debris avalanches (the March 2014 landslide
near Oso, Washington, an example of the latter) is usually difficult due to the complex
relationship between precipitation and pore pressure at the depth of movement. Linkages
between (short- and long-term) heavy precipitation, groundwater recharge, regional
groundwater flow, and deeply seated landslide movement in mountainous areas are
poorly understood for several reasons. Structural and lithological details determine the
pathways of recharge and flow (Hodge and Freeze, 1977). Upslope contributing area is
typically much greater and more complex than in the case of shallow landslides, making
it difficult to discern the time scales for precipitation and surface water to affect flow and
pore-water pressure at depth (Iverson, 2000). Hydrogeologic and geophysical
characterization of these details is difficult and costly.
USGS tools and models under active development
The USGS is actively developing several computer programs for landslide modeling.
The models have been used for published research and assessments and address different
aspects of landslide hazard. The LAHARZ model (Schilling, 2014) runs a series of tools
in ArcGIS1 for assessing hazard (inundation) zones for large volcanic debris flows. It is
based on simple volume-area relationships and can also be applied to non-volcanic rock
and debris avalanches and small debris flows. The program DCLAW (George and
Iverson, 2014) models initiation and movement of debris flows and rock and debris
avalanches. It numerically solves the mathematical equations that describe the physical
processes in these flows and was applied recently to modeling the March 2014 debris
avalanche near Oso (Iverson et al., 2014). The SCOOPS program (Reid et al., 2000)
combines 3-D limit-equilibrium slope stability analysis with an efficient search routine to
assess the potential for large deep landslides in a 3-dimensional digital landscape.
SCOOPS can use results from USGS groundwater models, such as MODFLOW
(Harbaugh et al. 2000), to assess effects of groundwater flow on slope stability. The
program TRIGRS (Baum et al., 2010) is used to analyze shallow, rainfall-induced
landslides. This program solves equations representing rainfall infiltration and slope
stability to produce a series of maps showing evolution of shallow slope stability during a
rainstorm (Fig. 1). Process-based models describing runoff, sediment transport, and
debris-flow initiation in recently burned areas are in the early stages of development.
Examples of spatial forecasting
Researchers around the world have put much effort into modeling and predicting where
landslides are likely to occur (susceptibility) and how far they are likely to travel (runout
or inundation zones). When tested against maps of historical landslides, nearly all
assessments make imperfect predictions of hazard zones that require a probabilistic
interpretation. A few examples by USGS authors illustrate the range of hazard questions
addressed. Harp et al. (2011) used field studies and an empirical model to rank source
areas and a probabilistic model to identify runout zones for rock fall. Predicted locations
of precipitation-induced landslides have been based on analysis of high-resolution
1 Any use of trade, product, or firm names is for descriptive purposes only and does not imply
endorsement by the U.S. Government.
AEG Professional Forum Extended Abstracts
topography (Schulz, 2007) and deterministic models (Godt et al., 2008). Deterministic
models have also been applied to assessing potential for deep landslides on coastal bluffs
(Brien and Reid, 2008). Our methods for rapid assessment of post-fire debris flows are
based on regression analysis (Cannon et al., 2010). These methods have recently been
partially automated so that assessments can be distributed quickly
Figure 1. Maps of the Puget Sound area, Washington, showing time-varying (a) pressure head and (b)
landslide susceptibility indicated by factor of safety (modified from Baum et al., 2010, Fig. 8).
Temporal and spatial forecasting for early warning
Landslide forecasting combines the spatial aspects (where? how far? how big?) of
susceptibility or hazard assessment with the temporal aspect (when?). For reasons
explained previously, this section focuses on precipitation-induced landslides. Due to
uncertainties inherent in rainfall measurements and forecasts, temporal predictions are
best interpreted in a probabilistic framework (forecasts). Landslide forecasting depends
on a combination of monitoring and modeling. Monitoring rainfall and comparing some
measure of cumulative rainfall or rainfall intensity and duration to a threshold is probably
the simplest example of landslide forecasting and is the basis for most operational
landslide warning systems (Guzzetti et al., 2008; Baum and Godt, 2010). Rainfall
thresholds are commonly based on analysis of historical landslide occurrence in relation
to rainfall for a specific geographic area, and threshold exceedance can be related to
probability of landslide occurrence (Chleborad et al., 2008). In recent years,
deterministic models also have been used to develop rainfall thresholds (Godt and
McKenna, 2008; Godt et al., 2008). As a demonstration of using rainfall thresholds, the
AEG Professional Forum Extended Abstracts
USGS tracks rainfall at NWS gages in the Seattle area and compares recent rainfall to
thresholds. Current conditions are posted on a USGS web page
( to give an indication of time-
varying landslide susceptibility. Empirical rainfall thresholds are also used to provide
early warning of debris flows for burned areas in southern California (Staley et al., 2013).
Recent research has demonstrated that hydrologic and geotechnical monitoring can
significantly improve the accuracy of forecasts. Godt et al. (2012) showed that
monitoring of soil water content and pore pressure/matric suction when interpreted with
the aid of models for variably saturated flow and slope stability can accurately identify
times of greatest potential for landslides, including some that are not readily predictable
based on precipitation alone (Fig. 2). Field monitoring in recently burned areas by Kean
et al. (2011) showed that debris flows were usually triggered by brief (<30 min.) periods
of intense rainfall, contrary to previous studies based on less precise timing of debris flow
occurrence. Intensive field monitoring of shallow landslide and burned area debris flows
is also providing insights needed to improve process-based models and their predictive
Recent events, including the September 2013 Colorado Front Range debris flows (Coe et
al., 2014), the highly mobile debris and rock avalanches near Oso, Washington (Iverson
et al., 2015), and near Collbran, Colorado, respectively, and burned area debris flows at
locations around the western United States have highlighted some of the ongoing
challenges in landslide modeling and forecasting. High-resolution topographic data are
needed in most landslide prone areas of the country to aid recognition of past long-runout
landslides and potential sources of future ones. Despite recent advances in models, much
remains to be learned about the initiation and dynamics of debris flows and rock and
debris avalanches. Ongoing field investigations, including large scale mapping and
laboratory testing, at Oso and Collbran, are aimed at clarifying avalanche dynamics and
are expected to lead to new insights that will drive future model improvements.
Additional geotechnical and hydrologic monitoring and modeling studies will be needed
and are planned for continued improvement in forecasting capabilities for rainfall-
induced landslides and burned area debris flows. Increasing demand for early warning is
driving the need for tools, data and methods to more rapidly assess the potential
locations, timing and extent of rainfall-induced landslides and debris flows in both
burned and non-burned areas. Motivated by these needs, along with opportunities to
learn from recent events, the USGS expects to make significant advancements in
landslide science in the coming years.
AEG Professional Forum Extended Abstracts
Figure 2. Slope stability as a function of time based on hydrologic monitoring data and generalized
effective stress (suction stress) concept, (a) hourly and cumulative rainfall, (b) volumetric water content, (c)
pressure head, (d) suction stress, (e) one-dimensional factor of safety (Godt et al., 2012).
Baum, R. L. and Godt, J. W., 2010, Early warning of rainfall-induced shallow landslides and
debris flows in the USA: Landslides, Vol. 7 no. 3, pp. 259272. doi: 10.1007/s10346-009-0177-0
Baum, R. L.; Godt, J. W.; and Savage, W. Z., 2010, Estimating the timing and location of shallow
rainfall-induced landslides using a model for transient, unsaturated infiltration: Journal
Geophysical Research, Earth Surface, Vol. 115, No. F03013, doi:10.1029/2009JF001321
Baum, R. L.; Schulz, W. H.; Brien, D. L.; Burns, W. L.; Reid, M. E.; and Godt, J. W., 2014,
Progress in Regional Landslide Hazard AssessmentExamples from the USA. In Sassa, K.,
Canuti, P., and Yin, Y. (Editors), Landslide Science for a Safer Geo-Environment, Vol. 1, pp. 21
36. Springer, Cham, Switzerland, doi: 10.1007/978-3-319-04999-1_2
AEG Professional Forum Extended Abstracts
Brien, D. L. and Reid, M. E., 2008, Assessing deep-seated landslide susceptibility using 3-D
groundwater and slope-stability analyses, southwestern Seattle, Washington. In Baum, R. L.;
Godt, J. W.; and Highland, L. M., (Editors), Landslides and Engineering Geology of the Seattle,
Washington, Area: Geological Society of America, Reviews in Engineering Geology, Vol. XX.
pp. 83101, doi: 10.1130/2008.4020(05).
Cannon, S. H.; Gartner, J. E.; Rupert, M. G.; Michael, J. A.; Rea, A. H.; and Parrett, C., 2010.
Predicting the probability and volume of postwildfire debris flows in the intermountain western
United States: Geological Society America Bulletin Vol. 122, pp. 127144.
Chleborad, A. F., 1998, Use of Air Temperature Data to Anticipate the Onset of Snowmelt-Season
Landslides: U.S. Geological Survey Open File Report 98-124, 16 p.
Chleborad, A. F.; Baum, R. L.; and Godt, J. W., 2008, A prototype system for forecasting
landslides in the Seattle, Washington, Area, In Baum, R.L., Godt, J.W., and Highland, L.M.,
(Editors), Engineering geology and landslides of the Seattle, Washington, area: Geological
Society of America Reviews in Engineering Geology Vol. XX, pp. 103120, doi:
Coe, J. A.; Kean, J. W.; Godt, J. W.; Baum, R. L.; Jones, E. S.; Gochis, D. J.; and Anderson, G.
S., 2014, New insights into debris-flow hazards from an extraordinary event in the Colorado
Front Range: GSA Today, Vol. 24, no. 10, pp. 410, doi:10.1130/GSATG214A.1
Collins, B. D. and Stock, G. M., 2014, Detecting thermally driven cyclic deformation of an
exfoliation sheet using lidar and radar, 2014 Vertical Geology Conference, Univ. of Lausanne,
Vaud, Switzerland, pp. 179183.
Cruden, D. M. and Varnes, D. J., 1996. Landslide types and processes. In Turner, A. K. and
Schuster, R. L. (Editors), Landslides--Investigation and Mitigation. Washington D.C., National
Academy Press, Transportation Research Board Special Report 247, pp. 3675.
George, D. L. and Iverson, R. M., 2014, A depth-averaged debris-flow model that includes the
effects of evolving dilatancy. II. Numerical predictions and experimental tests, Proceedings of the
Royal Society, Ser. A, Vol. 470, No. 20130820, doi: 10.1098/rspa.2013.0820
Godt, J. W., and McKenna, J. P., 2008, Hydrological response of hillside materials to infiltration:
Implications for shallow landsliding in the Seattle area, In Baum, R. L.; Godt, J. W.; and
Highland, L. M. (Editors), Landslides and Engineering Geology of the Seattle, Washington, area,
Geological Society of America Reviews in Engineering Geology, Vol. XX, pp. 121135, , doi:
Godt, J. W., Schulz, W. H., Baum, R. L., and Savage, W. Z., 2008, Modeling rainfall conditions
for shallow landsliding in Seattle, Washington, In Baum, R. L., Godt, J. W., and Highland, L. M.,
(Editors), Landslides and Engineering Geology of the Seattle, Washington, area: Geological
Society of America Reviews in Engineering Geology, Vol. XX, pp. 137152, , doi:
Godt, J. W.; Sener-Kaya, B.; Lu, N.; and Baum, R. L., 2012, Stability of infinite slopes under
transient partially saturated seepage conditions: Water Resources Research, Vol. 48, No.
AEG Professional Forum Extended Abstracts
Guzzetti, F.; Peruccacci, S.; Rossi, M.; and Stark, C. P., 2008, The rainfall intensityduration
control of shallow landslides and debris flows: an update. Landslides, Vol. 5 pp.317.
Harbaugh, A. W.; Banta, E. R.; Hill, M. C.; and McDonald, M. G., 2000, MODFLOW-2000, the
U.S. Geolocial Survey modular ground-water modeluser guide to modularization concepts and
the ground-water flow process: U.S. Geological Survey Open-File Report 00-92, 121 p.
Harp, E. L.; Dart, R. L.; and Reichenbach, Paola, 2011, Rock fall simulation at Timpanogos Cave
National Monument, American Fork Canyon, Utah, USA: Landslides, Vol. 8, pp. 373379.
Hodge, A. L., and Freeze, R. A., 1977, Groundwater flow systems and slope stability: Canadian
Geotechnical Journal, Vol. 14, pp. 466476.
Iverson, R. M., 2000, Landslide triggering by rain infiltration: Water Resources Research, Vol.
36, pp. 1,8971,910.
Iverson, R. M.; George, D. L.; Allstadt, K.; Reid, M. E.; Collins, B. D.; Vallance, J. W.;
Schilling, S. P.; Godt, J. W.; Cannon, C. M.; Magirl, C. S.; Baum, R. L.; Coe, J. A.; Schulz, W.
H.; and Bower, J. B.; 2015, Landslide mobility and hazards: implications of the 2014 Oso
disaster, Earth and Planetary Science Letters, Vol. 412, pp. 197208, doi:
Kean, J. W.; Staley, D. M.; and Cannon, S. H., 2011, In situ measurements of postfire debris
flows in southern CaliforniaComparisons of the timing and magnitude of 24 debrisflow events
with rainfall and soil moisture conditions: Journal Geophysical Research, Vol. 116, No. F04019,
doi: 10.1029/2011JF002005.
Pankow, K. L.; Moore, J. R.; Hale, J.M.; Koper, K.D.; Kubacki, T.; Whidden, K.M.; and
McCarter, M.K., 2014, Massive landslide at Utah copper mine generates wealth of geophysical
data. GSA Today, Vol. 24, pp. 49, doi:10.1130/GSATG191A.1.
Reid, M. E.; Christian, S. B.; and Brien, D. L., 2000, Gravitational stability of three-dimensional
stratovolcano edifices: Journal Geophysical Research, Vol. 105, No. B3, pp. 60436056.
Schilling, S. P., 2014, Laharz_pyGIS tools for automated mapping of lahar inundation hazard
zones: U.S. Geological Survey Open-File Report 2014-1073, 78 p.,
Schulz, W. H., 2007, Landslide susceptibility revealed by LIDAR imagery and historical records,
Seattle, Washington: Engineering Geology, Vol. 89, pp. 6787.
Staley, D. M.; Kean, J. W.; Cannon, S. H.; Laber, J. L.; and Schmidt, K. M., 2013, Objective
definition of rainfall intensity-duration thresholds for the initiation of post-fire debris flows in
southern California. Landslides, Vol. 10, pp. 547562, doi: 10.1007/s10346-012-0341-9.
AEG Professional Forum Extended Abstracts
Landslide Hazard Mapping in King County
Bethel, John, King County,
Greg Wessel,
Anne Weekes,
Sevin Bilar,
The March 2014 SR-530 landslide in Snohomish County, Washington, focused attention
on landslide hazards in the Pacific Northwest. King County, located immediately south of
Snohomish County, is similar in topography, geology, and climate; and is therefore
subject to similar landslide hazards.
A review of landslide preparedness in King County following the SR530 slide found that
the County lacked up-to-date mapping showing the extent and character of landslide
hazards. Existing landslide mapping in King County was completed in the 1990s by
geologists experienced in landslide recognition and assessment. However, these
practitioners were working with the resources then available, such as U.S. Geological
Survey topographic mapping, stereo aerial photography, and scattered geologic mapping.
This work predated Light Detection and Ranging (LiDAR), a remote sensing technology
that produces high-resolution images of the ground surface. While this earlier mapping
effort successfully identified the general vicinity of unstable slopes, it lacked the
resolution to reliably or precisely delineate areas of instability.
Given the availability of modern imaging and an abundance of recent geologic mapping,
the opportunity existed to create an updated map that significantly improved
understanding of landslide hazards in King County. The present landslide mapping
program was initiated for this purpose and is staffed by County geologists, with oversight
from a committee of geoscientists from government, academia, and the private sector.
Funding constraints have limited the present work to areas within the major river valleys
in King County, along with Vashon and Maury Islands. With potential changes in
funding opportunities this work could be expanded to cover the entire County (Figure 1).
Figure 1. Limits of current landslide mapping project in King County, WA.
AEG Professional Forum Extended Abstracts
This current program is in the early stages with progress to date including identification
of the types of landslide phenomenon active in King County, preliminary mapping of
landslide features identifiable on LiDAR imagery, investigation of appropriate mapping
methodologies for these various landslide phenomenon, and identification of source
information on historical landslides in King County.
Landslides in King County
The degree of danger posed by landslides in King County varies according to differences
in mass movement type, location, size, rate of travel, and possible runout distance.
Evidence of past landsliding, regardless of type, is a known indicator of future landslide
hazard. Critical variables, such as how far landslide debris will travel in a future
occurrence, are much more difficult to identify and predict (Iverson et al., 2015).
Therefore, the first task in our mapping effort was to identify where landslides have
occurred and the types of slope failure found in these locations. Two distinct
physiographic settings, the Cascade Mountains to the east and the Puget Lowland to the
west, are associated with known differences in landslide locations and types within King
Puget Lowland: Most of the population and infrastructure in King County is located in
the Puget Lowland, an area that was overridden by continental glacial ice. Much of this
area bears the clear imprint of its glacial history with a rolling, fluted, low gradient
topography that is stable and shows little evidence of post-glacial modification (Figure
2). This glacial surface is divided by a network of broad troughs, some of which contain
bodies of water (i.e., Lake Washington, Puget Sound), and others are largely filled with
alluvium (i.e., Kent Valley, Snoqualmie Valley). The floors of these troughs have little
relief and are therefore not subject to landslide processes. Landsliding in the Puget
Lowland is concentrated in narrow bands of steep slopes at the topographic transition
between the glaciated upland surface and the trench floors. These occur either along river
and stream valleys, or along coastal bluffs.
Figure 2. A portion of the Cedar River Valley and surrounding uplands.
AEG Professional Forum Extended Abstracts
We have identified several different landslide processes active along these steep
transition zones. These include prehistoric slumps, debris flows, and shallow debris
avalanches. Each of these types of landslide phenomenon requires different approaches to
Prehistoric Slumps
Where fluvial or coastal erosion has created steep, high slopes in glacial sediments, these
slopes sometimes failed in large deep-seated landslides termed slumps (Figure 3a).
Slumps often entail both downslope movement and rotation of the slide material. These
features are common on coastal bluffs and valleys walls; virtually all of these events
failed initially in prehistoric times. Although old features, the failed landslide debris in
slumps is sometimes remobilized in response to unusually high precipitation,
earthquakes, or ill-advised human disturbance. Prehistoric slumps are perhaps the easiest
landslide features to identify because their size and distinctive topography renders them
readily identifiable using LiDAR imaging (Burns and Madin, 2009). Such features
sometimes appear topographically sharp (Figure 3a). Such distinct topography suggests
relatively recent formation or ongoing movement, and such distinct features are easy to
identify and map. However, some slumps may appear more subdued in other cases
making identification and delineation more challenging and less certain (Figure 3b). Such
subdued topography likely reflects an older initial age and/or less active recent
Figure 3. Differences in the clarity of topographic imagery with similar resolution shown in a) slump
feature on the Cedar River, River Mile 17 and b) slump in the Cedar River Valley, River Mile 10.4, King
County, WA.
Debris Slides
Shallow debris slides are the most common type of slope movement in the Puget
Lowland and affect both natural and man-made slopes. These occur when a thin layer of
soil becomes saturated, detaches, and moves downslope (Figure 4). These types of
failures are typically too small to identify using LiDAR-based imagery. Although
comparatively small in scale, these shallow landslides often block transportation
corridors, damage infrastructure, and can be damaging or even deadly if they impact
occupied structures.
AEG Professional Forum Extended Abstracts
There are several approaches to identifying slopes prone to shallow landsliding. These
include modeling using some form of an infinite slope equation (Montgomery et al.,
2001, Harp et al., 2006), applying a geomorphic landscape classification as suggested by
Shultz (2005), or simply applying a threshold based on ground surface slope. The County
is currently evaluating these approaches for King County and has yet to adopt a final
Figure 4. Shallow Debris Slide on West Valley Highway, King County, WA. (Photo credit KC DOT,
Channelized Debris Flows
Debris flows are a third relatively common type of landslide process, often occurring as a
direct result of either a debris slide or reactivation of a prehistoric slump. If debris from
such an upslope failure enters a steep swale, the wet mass may travel down the swale as
thick slurry (Figure 5). Such debris flows can accumulate additional water and sediment
as they flow downslope, often recruited from the bed of the swale. These flows pose a
serious risk, especially when emerging from the steep drainages where they originate and
flow out onto level, potentially occupied areas below.
Figure 5. Debris Flow above 156th Place SE, Cedar River, King County, WA. (Photo credit KC, 2006).
AEG Professional Forum Extended Abstracts
Often debris flows occur repeatedly from the same swale, with recurrence intervals in the
range of decades to centuries. Over time these repeated events deposit sediment at the
mouth of the ravine, where the surface gradient decreases. These deposits take the form
of a truncated cone, identifiable by closely-spaced contours generated from LiDAR-based
digital terrain models (Figure 6). Such debris fans can mark areas at risk of inundation
from debris flows. A challenge with this approach is differentiation between fans formed
as a result of debris flows, and similar looking features formed by fluvial processes.
Figure 6. Debris Fan, Snoqualmie River Valley near Fall City, King County, WA.
Complex Landslides
One type of mass movement may trigger another event creating complex or compound
movements. The 2014 SR-530 landslide is an example of a complex slide exhibiting
multiple styles of movement occurring during a single event. The SR-530 landslide
appears to have failed initially as preexisting landslide debris was remobilized as a debris
flow. This initial movement removed toe support and triggered a massive rotational
slump (Keaton et al., 2014). The 2001 Cedar River landslide, triggered by the magnitude
6.8 Nisqually earthquake was another example of such an occurrence in King county.
The highly visible features characteristic of this type of compound slide are often
associated with large prehistoric slump zones along river valleys and are readily apparent
in LiDAR imagery.
Cascade Mountains: In the Cascade Mountain physiographic province of eastern King
County, the suite of landslide processes reflects the largely bedrock substrate and the
alpine topography. While these areas are less populated than the Puget Lowland, there are
communities present in the mountains, as well as critical transportation and utility
corridors. Like the Puget Lowland, the Cascade Mountains are adjusting to the effects of
recent glaciations. These glaciers originated locally, in the mountains themselves, rather
than flowing from the north as did the continental glaciers that affected the Puget
AEG Professional Forum Extended Abstracts
Bedrock Slumps
Large rotational slumps are also present in the mountainous areas of eastern King
County. Unlike similar features in the lowland, these slumps occur in bedrock (Figure
7). It is likely that many such features occurred due to glacial scour and loss of
buttressing during deglaciation. These features are large enough to be visible and
mapped by interpretation of LiDAR imagery.
Figure 7. Northwest Slope of Taylor Mountain, east of State Route Highway 18, King County WA.
Rockfall (Figure 8) is another locally significant mass-wasting process in alpine areas.
Rockfall entails single or multiple boulders being dislodged and falling from steep cliff
faces. Rockfall can be triggered by earthquakes or severe weather. Infrequently, an entire
rock slope may collapse creating a rock avalanche. Although no rock avalanches have
occurred in King County in historic times, there is clear evidence of such prehistoric
failures. Rockfall hazards can be mapped based on the extent of existing rockfall debris,
or based on models accounting for the height and slope of source area. No mapping
criteria have been selected as yet for this process.
Debris Flows
Debris flows, as described above, are also a common mass-wasting process in alpine
environments, but occurring with greater frequency and energy than in the Puget
Lowland. The mapping approach for alpine debris flows will be similar to that adopted
for the Puget Lowland.
SR Hwy 18
AEG Professional Forum Extended Abstracts
Figure 8.Rockfall; Boulder on Moon Valley Road below Mt. Si, King County, WA.
(Photo credit J. Bethel, 2014)
Runout and Landslide Dams
As the SR-530 tragedy made clear, identifying unstable slopes is not sufficient to
understand the risk posed by landslides. A complete assessment of landslide hazards
should include consideration of the potential runout areas of landslide debris. For some
mapping methodologies, identification of landslide runout is inherent in the mapping
protocol, as in the identification of debris fans or areas subject to rockfall. Other mapping
methodologies may identify only source areas and not the distribution of mobilized
debris. The County intends to evaluate runout potential during this study. However,
recent forensic evaluation of the SR-530 slide has revealed how challenging it is to make
precise runout predictions given that small variations in pre-failure conditions can
produce dramatically different runout distributions (Iverson et al., 2015). Given this
constraint it seems likely that runout predictions are likely to be both generalized and
Where landslide debris is deposited in a river or stream channel, such debris has the
potential to create landslide dams causing flooding upstream with the potential to create
breaching and subsequent damaging flows downstream. Where assessments of landslide
runout suggest a significant potential for landslide dam formation, the potential character
and consequences of such an occurrence will be evaluated.
Our initial work to date has provided us with an inventory of deep-seated slumps and
other larger landslide features whose distinctive topography can be readily identified
using LiDAR imagery. We are investigating methods for identifying and mapping areas
susceptible to other landslide processes in King County. We are in the process of
reviewing the extensive historical records and datasets derived from a variety of sources
that provide spatial, temporal, and geophysical information on active landslides within
AEG Professional Forum Extended Abstracts
the county. These datasets will be used to identify areas of past instability that suggest the
potential for future movement. (Baum et al., 2014).
A final map with supporting data of areas in King County river corridors is scheduled for
release in mid-2016 by the County. This product will be valuable for public education,
land use regulation, comprehensive planning, county operations, and emergency
management. Availability of this information will help educate and inform King County
residents, local municipalities, and county operation managers of the hazards to life and
property posed by active landslides.
Baum, R. L., Schulz, W. H., Brien, D. L., Burns, W. J., Reid, M. E. (2014), Regional
Landslide Hazard AssessmentExamples from the USA; Jonathan W. Godt, in:
Landslide Science for a Safer Geoenvironment, Vol.1: The International Programme on
Landslides (IPL). Editors: Sassa K., Canuti, P., Yin Y., Springer, 515p.
Burns, W.J., Madin, I.P., (2009), Landslide Protocol for Inventory Mapping of Landslide
Deposits from Light Detection and Ranging (Lidar) Imagery, Oregon Department of
Geology and Mineral Industries, Special Paper 42.
Harp, E. L., Michael, J. A. , and Laprade, W.T., (2006), Shallow-landslide hazard map of
Seattle, Washington: , U.S. Geological Survey Open-File Report 2006-1139, 23 p., 1
Iverson, R.M., George, D.L., Allstadt, K., Reid, M.E., Collins, B.D., Vallance, J.W.,
Schilling, S.P., Godt, J.W., Cannon, C.M., Magirl, C.S., Baum, R.L., Coe, J.A., Schulz,
W.H., and Bower, J.B., (2015), Landslide mobility and hazards: implications of the 2014
Oso disaster: Earth and Planetary Science Letters, Elsevier, v. 412, 15 February 2015,
pp. 197-208.
Keaton, J. R., Wartman, J., Anderson, S., Benoît, J., deLaChapelle, J., Gilbert, R.,
Montgomery, D. R., (2014), The 22 March 2014 Oso Landslide, Snohomish County,
Washington. Geotechnical Extreme Events Reconnaissance.
Montgomery, D. R., H. M. Greenberg, W. T. Laprade, W. D. Nashem, (2001), Sliding in
Seattle: Test of a Model of Shallow Landsliding Potential in an Urban Environment,
Water Science and Application, v. 2, p. 59-73.
Schulz, W.H., (2005), Landslide susceptibility estimated from LIDAR mapping and
historical records for Seattle, Washington: U.S. Geological Survey Open-File Report
2005-1405, 13 p., 1 plate.
AEG Professional Forum Extended Abstracts
Landslide Management During Property Development in Los Angeles County,
Nestle, Charles, MS, CEG, County of Los Angeles Department of Public Works,
The County of Los Angeles requires geologic and soils engineering investigations for
proposed development projects that are located within or adjacent to mapped landslides,
or that are suspected of possibly being affected by landslides, or where it is possible that
the development may cause landslides or slope instability.
Political support for the ability to impose these requirements followed the post-war
housing boom which pushed development into the foothill and hillside areas of Los
Angeles County during a period of relatively low rainfall, and prior to the development of
any grading codes. Two periods of relatively high rainfall in the 1950’s resulted in
millions of dollars in damage to private property and infrastructure, which prompted the
first grading ordinance developed by the City of Los Angeles. The County of Los
Angeles followed a few years later with their own grading code and the inclusion of the
requirement for geologic investigations.
Local agencies may add more restrictive amendments to the State Building Code on the
basis of local climate, topography, and geological conditions. The County of Los
Angeles has added robust requirements for geologic investigations as amendments to
their Building Code beginning with their 1962 Code. Two of those code sections provide
the basic authority by which the County geologists may require an investigation by a
State-licensed engineering geologist.
Section 110.2.2
Except as provided in Section 110.2.3, work requiring a building or grading
permit by this Code is not permitted in an area determined by the Building
Official to be subject to hazard from landslide, settlement or slippage.
Section 111
The Building Official may require an engineering geology or soils engineering
report, or both, where in the Building Official's opinion, such reports are essential
for the evaluation of the safety of the site. The engineering geology or soils
engineering report or both shall contain a finding regarding the safety of the site
of the proposed work against hazard from landslide, settlement or slippage and a
finding regarding the effect that the proposed building or grading construction
will have on the geotechnical stability of the area outside of the proposed work.
When an application for building or grading is filed with the Division of Building and
Safety, a determination is made based on specific and subjective criteria as to the need for
geotechnical review. If determined to be needed, the project is referred for geotechnical
review, where a County staff geologist and soils engineer determine if an investigation
will be needed. This determination is based on a physical site inspection; review of
AEG Professional Forum Extended Abstracts
stereo pairs of aerial photographs, proprietary GIS data including satellite imagery, files
for adjacent properties; and local knowledge. If the reviewer has any reason to believe
that slope stability may be an issue, or that geomorphic conditions suggest landsliding of
any vintage may have affected the topography, then a site-specific geologic investigation
will be required.
The following guidelines and requirements address how the County geotechnical
reviewers will approach the review of the consultant’s geotechnical investigation, and
how the County’s policies on development in landslide areas are implemented. These
policies address both proposed subdivisions of land and existing single lots that are
impacted by landslides. Landslides are considered a geologic hazard unless it can be
demonstrated that the landslides have minimum factors of safety for gross static stability
of 1.5 and minimum pseudostatic stability of 1.1. Stability analyses must be supported
with onsite subsurface exploration and appropriate laboratory testing.
Unmitigated landslide hazards that could adversely impact offsite property (outside of the
subdivision or proposed development) do not have to be mitigated when the following
conditions are met:
 An existing landslide crosses an existing property boundary.
 Existing conditions will not be changed, worsened, or otherwise be affected by
the proposed development.
 The hazard does not pose a threat to proposed building areas on the property or
within the proposed subdivision.
The proposed development will not increase the potential for failure of the
Each landslide, including the potential affected area, is contained entirely on one
A landslide and the surrounding affected area inside a proposed subdivision may be
permitted to remain unmitigated if it is determined that it cannot affect any proposed
development, and if it is recorded as a Restricted Use Area, where no construction will be
permitted. Determination of the surrounding affected area (e.g. areas that may be
impacted by headward expansion or runout at the toe) shall be addressed by the
geotechnical consultants and supported with data and analyses.
The cases described below refer to the attached figure at the end of this paper.
Case A
Problem: Proposed lot line crosses landslide.
Solution: Mitigate landslide, or adjust lot line to constrain the landslide on one lot or
provide remediation of the landslide. The landslide hazard cannot be
subdivided. Designate unmitigated landslide hazard and surrounding
affected area as a Restricted Use Area.
AEG Professional Forum Extended Abstracts
Case B
Problem: The designated building area is at the toe of a landslide, which is
contained within the lot boundaries.
Solution: Relative to the building area, the landslide must be mitigated (e.g.,
buttressed or removed) or be demonstrated to be stable and not a threat to
the safety of the proposed structure. Alternatively, a safe building area
may be found elsewhere on the lot not affected by geotechnical or
geologic hazards.
Cases C and G
Problem: Landslides are outside of the subdivision boundary where remediation is
not possible.
Solution: Unless the building pad can be shown to have an adequate setback from
the landslide-affected areas, or the landslide will be stabilized, an alternate
building area is required.
Cases D and E
Problem: A landslide transects existing property boundaries.
Solution: No mitigation is required when a landslide transects an existing property
boundary. However, it must be clearly demonstrated that the proposed
development will not adversely affect the stability of the landslide, or the
landslide must be mitigated.
Cases F and G
Problem: The landslides are either entirely inside or outside of the property
boundary and do not affect the safety of the building area.
Solution: It must be demonstrated that the proposed development will not adversely
affect or contribute to the instability of the landslide, resulting in adverse
effects on adjacent property and relative stability. Otherwise, the landslide
must be mitigated.
Case H
Problem: An existing landslide has been determined to be a hazard and the affected
area could cross to an adjacent lot.
Solution: Landslide must be removed or mitigated.
AEG Professional Forum Extended Abstracts
Case I (not on diagram)
There are some developed areas within the County of Los Angeles where existing
landslides are so large (> ¾ mile) that mitigation is not possible. If the geotechnical
consultants are unable to demonstrate that these landslides are stable, then no permits can
be issued.
AEG Professional Forum Extended Abstracts
Regulations, planning and public policy
Keynote 3
Lessons from the National Earthquake Hazards Reduction Program can be applied to the
National Landslide Hazards Program: A Rational Approach
Roth Jr, Richard J., Consulting Insurance Actuary, Huntington Beach, CA 92646
Jeff Keaton, AMEC Foster Wheeler, Los Angeles, CA,
According to the USGS National Landslide Information Center, landslides are a serious geologic
hazard common to almost every State in the United States. It is estimated that in the United
States, they cause in excess of $1 billion in damages and from about 25 to 50 deaths each year.
See Map 1 below (source: “Landslides 101”)
Map 1.
Globally, landslides cause billions of dollars in damages and thousands of deaths and injuries
each year, particularly in underdeveloped countries.
Historically, the insurance industry in the United States has had an unwillingness to offer
insurance for damage due to natural hazards. The reasons are actuarial: such policies are difficult
to price and the losses are catastrophic (up to thousands of claims at once or none at all). Also,
the demand by property owners for this insurance is low, mainly because the public’s perception
of risk from most natural hazards is low and that government programs will cover their losses.
AEG Professional Forum Extended Abstracts
For these reasons, the common Homeowners insurance policy excludes damage due to “earth
movement.” The primary coverage in a Homeowners policy include fire, burglary and personal
liability, all of which can be priced and happen infrequently at any particular home, but often
enough that the aggregate future losses can be predicted by actuaries based on past loss statistics.
But past loss statistics from natural disasters cannot necessarily be used in the same way to
predict future losses from natural disasters. Computer models based on geophysics must be used.
The rise of the personal computer in the 1980s made possible the development of natural hazards
models, which could be used by insurance companies. Windstorm and earthquake models are
now well developed and are essential underwriting and pricing tools for insurance companies.
Windstorm and earthquake insurance policies are now readily available. The same could be said
for flood insurance policies, but flood insurance is almost entirely offered by the federal
government as a government program.
There is no engineering model specifically focused on landslide insured losses. Therefore, the
property owners can get insurance for any natural hazard, except landslides. With landslide
coverage included, the Homeowners policy would be almost a true “all-risk” policy. To have
complete insurance coverage for “earth movement,” the policy would have to include earthquake
liquefaction and sinkholes as well.
The federal National Earthquake Hazards Reduction Program (NEHRP) can be credited with
funding the geological, seismological and structural engineering research necessary to the
development of the earthquake computer models.
NEHRP started in 1977 to promote the study of earthquakes, with a view toward promoting
mitigation, sound building practices, life safety, and economic recovery. NEHRP is the work of
four primary agencies: FEMA, NIST, NSF and USGS.
The basic input in the earthquake models include the vulnerability of buildings, the soil
conditions, and the seismology of the region. NEHRP sponsored research contributed
significantly to this basic input and promoted the development of improved building code
USGS is known for its work on natural hazards, particularly for its leadership on earthquakes.
USGS also has a well-established program focusing on landslides, with a section on landslides
on its website: This is the USGS Landslide Hazards Program. Besides
advancing public safety, the program investigates past landslides, monitors current sites, and
undertakes research to make accurate landslide hazard maps and forecasts of landslide
The results of this research will be particularly useful to insurance companies. This research
seeks to answer the major questions:
1. When and where will landslides occur?
2. How big will they be?
3. How fast and how far will they move?
4. What areas will the landslides affect or damage?
5. How frequently do landslides occur in a given locality?
AEG Professional Forum Extended Abstracts
USGS is developing software for landslide assessments and modeling, with names like Seismic
Landslide Movement Modeled using Earthquake Records - “SLAMMMER” and TRIGRS which
is a program relating slope stability to rainfall infiltration.
As the Chief Property/Casualty Actuary for the California Insurance Department for twenty
years, I took a strong interest in insuring natural disasters, especially earthquakes. The 1989
Loma Prieta and 1994 Northridge earthquakes occurred while I was there.
Under the authority of the Department, I collected summary damage statistics from all the
insurance companies for both of these events. Under contract with USGS, these data sets were
compared with geologic and seismic data to get a relationship between ground acceleration and
building damage. The results were consistent with published attenuation curves. This kind of
research was important for the development and improvement of earthquake modeling.
Also while I was at the Department, I worked with Karl Steinbrugge, a retired professor from the
University of California Berkeley, on ways to measure the expected insured damage losses that
the insurance industry in California would likely sustain from a one-in-one hundred-year
earthquake event. The result was a detailed questionnaire which every insurance company had to
submit every year.
The insurance companies reported estimated Probable Maximum Loss (PML) amounts by zones
delineated in Map 2.
Map 2.
AEG Professional Forum Extended Abstracts
I summarized the results of these questionnaires and published them every two years. The recent
versions of these reports are available on the Department’s web site.
The publication of the damage statistics from past earthquakes and the projected damage
statistics from future earthquakes predated most of the development of the computer earthquake
models. The annual PML questionnaires played a major role in making insurance companies
aware of the usefulness of loss modeling.
This earthquake related history is applicable to landslides. With the publication of improved
hazard maps, which include identified landslide areas, property owners could take a greater
interest in insurance.
In California, when a house is sold, the buyer must be informed if the house is near a known
active fault zone. With improved landslide hazard maps, the same requirement could be made for
designated landslide zones.
In order to project the expected insured damage losses from a landslide, a trigger (such as
excessive rainfall) causing the landslide must be modeled and a method for estimating the
damage losses must be available, given that the landslide occurs. Preliminary work is now being
done through the National Landslide Hazards Program (such as, the computer program
It is very difficult to get a mortgage on a large commercial building without an assessment of the
earthquake exposure and PML exposure. If the exposure is high enough, earthquake insurance
would be required. The same type of evaluation of the landslide exposure could be required. The
commercial exposure includes damage to the building, damage to contents and business
interruption (loss of sales and disruption of suppliers).
Actually, commercial building owners buy more earthquake insurance than homeowners do.
This could very well be true for landslides, depending on whether commercial buildings are built
in landslide areas.
As the computer landslide loss models become available, the insurance rates would then vary by
specific locations, as well as by commercial use and type of contents. However, commercial
insurance products might not be viable, because policy prices might have to be too high for
property owners to be willing to purchase the insurance. In that case, property owners would
undertake to mitigate the risk through land engineering or purchase only insurance against severe
Landslide loss models, based on the work sponsored by the National Landslide Hazards
Program, would also be of great benefit to regional and local governments, for emergency
response planning and possibly for maintenance programs of agencies responsible for buried
utilities that could leak and contribute to triggering or to the severity of landslide movements.
AEG Professional Forum Extended Abstracts
Managing Recognized Hazards: Land Use Planning and Zoning, Strategies and Public
Molinari, Mark, PG, LEG, AECOM,
Many of the previous presentations at this forum have discussed and shown how various
national, state and local agencies have or are in the process of implementing landslide
assessment programs and, in some instances, how that information is being used for land use
decisions and public information. This presentation is intended to supplement these prior ones
by showing how selected different counties and cities in areas known to have significant
landslide hazards address potential slope hazards in development zoning, ordinances and permit
requirements. In addition, several examples of how local and general information regarding
landslide hazards and mitigation is made available to the general public are presented. The
review is not comprehensive and is based on examples in the western U.S. where the author is
most familiar; however it is likely representative of the type and range of what is implemented in
many other jurisdictions.
Most land use development in the U.S. is regulated by cities and counties via a combination of
zoning, ordinances, permit application requirements and building codes. Requirements regarding
geologic hazards and soil foundation conditions, including landslides, are most commonly
incorporated in development ordinances, permit applications/requirements and building codes.
Some jurisdictions just adopt the state or international building code with or without some
modifications/exemptions; however others have more specific requirements in ordinances and/or
guidance indicating when a more detailed site specific engineering geologic and/or geotechnical
engineering report is required, the qualifications of the responsible engineer/geologist, and the
report content and methods to be used. Most allow the permit application/report reviewer(s),
usually an engineer or engineering geologist, to grant exemptions based on their review of the
information and data provided and/or their experience and knowledge of nearby similar sites.
Several jurisdictions have different requirements depending on the height and/or steepness of the
slope. Adjacent or nearby jurisdictions may have substantially different requirements.
Some of the methods used to manage development on or near slopes and in the vicinity of known
landslides other than engineering methods (e.g. retaining walls and other engineered mitigation
measures) include: avoidance, setbacks from slope crests and toes, Geologic Hazard Abatement
Districts, designated landslide hazard zones or hillslope areas, and incorporation into open space
or recreational areas (e.g. parks and golf courses) within subdivisions/large developments.
Examples of designated landslide hazard zones/areas include portions of the Portuguese Bend
landslide complex in Rancho Palos Verdes and the community of La Conchita, both located in
southern California. Geologic Hazard Abatement Districts (GHADs) are local assessment
districts that can be established in California under state law (Division 17 of the Public
Resources Code, Sections 26500 - 26654) for the purpose of prevention, mitigation, abatement,
or control of geologic hazards, which are defined as "an actual or threatened landslide, land
subsidence, soil erosion, earthquake, or any other natural or unnatural movement of land or
AEG Professional Forum Extended Abstracts
Example Landslide Hazard Zones and GHAD’s
Rancho Palos Verdes has an approximately 1,200-acre Landslide Moratorium Area (LMA)
where there is a prohibition on “the filing, processing, approval or issuance of building, grading
or other permits.” At La Conchita in Ventura County, there is a known risk to residences from
landslides and mudslides from the steep slope bounding the community. Any building permit
application for either a new or increased area of a habitable structure within the LMA requires
the applicant to sign a waiver acknowledging: (1) that the property owner is aware that the parcel
is situated in a designated Geological Hazard Area and (2) that it is their responsibility to obtain
independent expert information regarding the potential risk and making a judgment thereafter
regarding whether to proceed with the permit process. The owner is also required to obtain a
geotechnical investigation and report for review by the County, which can deny the permit
request if it cannot be demonstrated that the hazard can be mitigated.
A proposal to establish a GHAD can be by either a petition signed by owners of at least 10
percent of the real property in the district, or by resolution of a local legislative body. The
proposal must be approved by a local legislative body (i.e. city of county which allows formation
of a GHAD). The proposal must be accompanied by a "plan of control" "which describes in
detail a geologic hazard, its location and the area affected thereby, and a plan for the prevention,
mitigation, abatement, or control thereof". The plan must prepared by a California certified
engineering geologist. The land within a proposed GHAD does not have to be contiguous or
within the same political boundary. The primary requirement is that a GHAD must benefit the
property within it and ensure the health, safety, and welfare of the residents. The property
owners in the district are assessed annual fees to provide the capital needed to implement the
mitigations measures, including maintenance, outlined in the plan of control or pay off bonds if
issued to obtain the capital needed. There have been between almost 40 GHADs or similar
improvement associations formed in California.
Example Setbacks
Setbacks from slope crests and toes are usually a prescribed minimum value or calculated based
on certain criteria. Salt Lake City and North Salt Lake City in Utah are an example where two
adjacent cities have different setback requirements. North Salt Lake City adopts the
International Building Code for soils and foundations and its setback criteria for slope less than
1:1 are shown in Figure 1. Salt Lake City uses different criteria to establish setbacks for the
Permit Area Boundary (Figure 2) and the building structure (Figure 3) based on the slope height.
An example of a prescribed buffer is the City of Bainbridge Island, Washington which has a
standard buffer of known landslide hazard areas “equal to the height of the slope or 50 feet,
whichever is greater, shall be established from all edges of a landslide hazard area” and the edge
of all buildings and structures must be 15 feet from the buffer edge to allow for construction
activity. Where no other reasonable alternative exists, reductions may be allowed based on
criteria specified in the City’s “Development Standards” within the “Geologically hazardous
areas” section of the Municipal Code and the decision of the City engineer.
AEG Professional Forum Extended Abstracts
Figure 1 Setback criteria from International Building Code adopted by North Salt Lake City for slopes less than
1:1 where H is the slope height in feet. Additional criteria are specified for slopes > 1:1.
Figure 2 Setback criteria for the Permit Area Boundary specified in Salt Lake City code 18.28.040: LAND
a = Setback distance at toe in feet
b = Setback at top in feet
H = Height from toe to top of cut/fill slope
PA = Permit Area (Lot Area Excluding Any Undevelopable Area)
Less than 5'
5' to 30'
Over 30'
Note: 1. Additional width may be required for interceptor drain.
Figure 3 - Setback criteria for the building structures specified in Salt Lake City code 18.28.040: LAND
AEG Professional Forum Extended Abstracts
Public Education and Information
While researching the information to support this presentation, it became readily apparent that
there is a wide range of codes, ordinances, and requirements related to development on and near
slopes. The primary source of information used, and probably what most of the public would
use, is the website of the city, county or other jurisdiction.
Most state geological surveys, and particularly those with histories of significant landslides, have
specific pages with general information on landslide hazards and mitigation and usually have
something more specific to the types and areas of landslides within the state. Several prior
presentations discuss ongoing (and aborted) state programs to catalogue and map known
landslides and present the data on static maps and/or interactive GIS-based maps on the state
website. While this information may be of value to individuals depending on the scale and
detail of the mapping, it is probably of most value when it is updated on a fairly regular basis,
and used and specifically referenced by cities and counties in their ordinances an used to define
hazards zones that require more detailed site specific investigation s part of the permitting and
plans approval process.
City and county websites vary significantly in the usability and ease when searching for the
pertinent codes and ordinances related to slopes and landslides and most took several attempts to
find all the pertinent info even when using the “search function” on the city/county website.
Some larger cities and counties have catalogued and mapped known landslides and potentially
hazardous slopes (e.g. Seattle, WA) or have specific maps prepared by the state geological
survey (e.g. Ventura County, CA); however many that I looked at have little in any general
information for the public that is readily accessible and or that directs them to where they can get
Santa Clara County, California is a good example of providing pertinent information to the
general public via the County web site and it has designated and mapped Geologic Hazards
Zones (GHZ) and an associated ordinance (Geologic Ordinance section C12-600-624)that guide
development. The GHZs were produced by combining information from a variety of published
and unpublished sources regarding the location and extent of possible faults, landslides,
compressible soils, dike failure flooding, and liquefaction. The County GHZs identify areas
where available information suggests specific geologic hazards may be present and have been
updated as new information becomes available. In those areas, the ordinance requires that the
owner/applicant submit a geologic report (prepared and signed by a Certified Engineering
Geologist [CEG]) for review by the County Geologist prior to approval of certain applications
for construction.
The Planning Department has several web pages on geologic hazards in the County
x) and the GHZs are mapped on the 1:24,000 US Geological Survey topographic quad maps
which are listed on a separate web page
neMaps.aspx). These maps are easily accessed in pdf form or electronically in GIS and kmz
AEG Professional Forum Extended Abstracts
format via hyperlinks on the web page. There are also direct links to the Geologic Ordinance
(you don’t have to wade through the various codes and ordinances like most cities/counties!).
The City of Seattle also has lots of useful information and links on their website for the
Department of Planning & Development, although it is split between Emergency Management
( and the codes section
for Environmentally Critical Areas
( The
City has also periodically hosted meetings on landslides aimed at the general public on
successive Saturdays at North Seattle and South Seattle community colleges. These include
presentations by an engineer, and engineering geologist and others and there are tables manned
by local volunteers from AEG and the American Society of Engineers who provide information
and answer questions.
The examples noted above and others will be discussed in the presentation.
AEG Professional Forum Extended Abstracts
AEG Professional Forum Extended Abstracts
How Emergency Management Can Achieve a "Landslide Victory" Related to Community
Preparedness and Public Warning
Chard, Mike, Director, Boulder, CO Office of Emergency Management,
During the Colorado Flooding of 2013 debris flows exacerbated flash flooding conditions and also
directly caused the death of one Boulder County resident. The risk of landslides remained long after the
flood waters passed and this This session will look at the lessons learned from the experience, new
programs to address the risk and also look forward as to the future needs of Emergency Managers to
educate residents about the risks, prepare communities based on high risk areas, perform predictive
landslide assessments and ultimately warn residents in enough time to allow for engineering solutions or
AEG Professional Forum Extended Abstracts
AEG Professional Forum Extended Abstracts
How Can Congress Help? The Potential for Comprehensive Landslide Policy
Seadler, Abigail, The American Geosciences Institute,
It often takes a tragedy for society to understand why geoscience is important. The SR 530
landslide in Oso, Washington and the West Salt Creek landslide in Mesa County, Colorado were
no exception. These devastating events catalyzed scientists, engineers, insurers, emergency
managers, the public, and decision makers at all levels to reexamine what we know about
domestic landslide hazards and reinvigorated the push for comprehensive, national landslide
In the wake of the landslides, the American Geosciences Institute through the Hazards Caucus
Alliance, the Association of Environmental & Engineering Geologists, the Association of
American State Geologists, and the American Society for Civil Engineers Geo-Institute worked
with federal and local stakeholders to produce two congressional briefings on landslide hazards.
The briefings addressed how advances in the geosciences can help inform sound decision-
making and attempted to build upon the ideas and recommendations put forth in the U.S.
Geological Survey Circular 1244 (Spiker and Gori, 2003) and subsequent National Research
Council report (National Academies Press, 2004), which assessed strategies for national
landslide hazard mitigation. After the success of the briefings and interest from congressional
offices, the societies decided to see what feasible legislative action could result from the
Discussions with multiple stakeholder groups revealed two potential policy solutions or “Asks.
These included: (1) the creation of an interagency program that would standardize and streamline
efforts to reduce the losses from landslide hazards currently spread among several federal
agencies, and/or (2) the creation of a competitive grant program that would allow state and local
organizations to apply for funds to map local landslide hazards. These concepts were outlined in
the aforementioned reports. So, why hasn’t Congress taken action, and why aren’t these
programs being funded?
An understanding of the inner workings of Congress is crucial to recognize and overcome the
obstacles and potential barriers to passing, and more importantly funding, legislation. The federal
government is comprised of three branches: legislative, executive, and judicial. Congress, or the
legislative branch, serves as the voice of the people and states in the federal government. It is the
role of Congress to pass legislation that is enforced by the executive branch and overseen by the
judicial branch; it is also charged with passing an annual budget to fund the federal government.
Congress is split into two distinct chambers, the Senate and the House of Representatives, which
have their own roles, rules, and cultures. It is within these chambers where even great legislation
gets mired in priorities, timing, and politics.
Obstacles and Barriers to Potential Legislation
Many factors contribute to the ultimate success or demise of a bill. Something as innocuous as timing can
make or break legislation. For example, even though the SR 530 and West Salt Creek landslides happened
in March and May, respectively, the societies were unable to hold the briefings until late June. By the
AEG Professional Forum Extended Abstracts
time they framed the “Ask,” it was July, which gave us limited time to meet with Members of Congress
before they recessed for a month in August. When Members returned in the fall, campaign season was
already underway and everyone on Capitol Hill was thinking about winning midterm elections by a
landslidenot landslide hazards.
As a result, before discussing the tools to hone a message to Congress and to form relationships
with staffers, it is important to understand the ways in which even the best ideas and bills “die,”
or do not pass.
1. DOA: A bill can die before it is even introduced if it does not have enough support.
Before a bill is introduced, the Member of Congress sponsoring the legislation must build
support for his or her bill and curry favor with other Members. Without this key step, bills are
blindly referred to a committee and are likely never to be brought to the Floor.
2. In committee: A bill can die in committee because of leadership priorities and politics.
Once a bill has been introduced, it is referred to a congressional committee that has jurisdiction
over the issue at hand. Once in committee, the bill is reviewed, revised, voted on, andwhen in
the Housereferred to other committees for their input and vote before it is reported to the Floor
for another vote. If a committee Chair does not prioritize the bill it may never pass out of
committee and make it to the Floor for a final vote. Furthermorewhen in the Houseif
multiple committees wish to review a bill it could get stopped in multiple places along the way,
thereby losing momentum, and never reach the Floor.
3. On the Floor: A bill can die on the House or Senate Floor if the majority party
leadership does not place it on the schedule. After a bill has passed out of committee it is
reported to the House or Senate Floor for a vote by the entire chamber. However, before a bill
can be voted on it must be put on the legislative calendar, which is controlled by the Speaker of
the House or Senate Majority Leader. Even if the bill happens to see a vote, it can, of course, still
be voted down. Also, if the Speaker of the House or the Majority Leader in the Senate do not
support the bill, he or she can simply not put it on the calendar.
4. In the other chamber: A bill can die in the opposing chamber where it was not
introduced if the appropriate committee leadership does not prioritize the bill. Once a bill
passes out of one chamber, it must be taken up and considered by the other. For example, if a bill
passes the House Floor it is then sent to the Senate where it is referred to committee again.
Although the bill does not need to be reintroduced, it can still die if the Chair of the designated
committee does not like the bill or simply prioritizes other issues.
5. By presidential veto. The President can veto a bill if he or she does not agree with it.
6. By lack of appropriation: Even if a bill is signed into law it can still “die” if the
appropriations committees do not fund it. This is especially salient in the current fiscal
environment where many in Congress are trying to reduce the deficit and cut discretionary
spending. Every congressional committee except for the Senate and House Appropriations
Committees pass authorizing legislation. Authorizing legislation creates and sets top-level
amounts of money for each program. Appropriating legislation actually designates the funds;
only the Senate and House Appropriations Committees can draft those bills. Therefore, even if a
AEG Professional Forum Extended Abstracts
bill is signed into law, the Chairs of the Appropriations Committees can still choose not to fund
it, or to give it not enough funding, in essence killing the bill.
Other factors that contribute to a bill’s success include its perceived importance and staff
turnover. Landslides are, unfortunately, not a high-priority issue for most congressional offices.
Therefore, it is up to the community to raise the profile of landslides. The community needs to
make Hill offices realize that landslides are just as important as earthquakes, just as fascinating
as volcanoes, and just as terrifying as tsunamis if we don’t do anything about them. Stakeholders
accomplish this through relationships with staffers that help convey the landslides message to
their bosses. This communication proves especially tricky because the turnover rate for
congressional staffers and Congressmen is so high. As a result, stakeholders must constantly
retell their story and re-forge relationships in order to re-build trust and raise the profile of an
issue once again.
How to Hone Your Message:
Congressional offices are incredibly busy. On a given day, a staffer could have back-to-back
meetings on everything from designating a post office to Lou Gehrig’s disease. Therefore, when
tailoring your message to congressional offices, it is important to keep your message on target
and brief and to keep in mind that even though the issue you want to discuss is incredibly
important to you, the staffer may have other priorities.
1. Have a clear “Ask”: Be specific. Groups always ask congressional offices for more money,
so an “Ask” that goes one step further stands out and tells the staffer that you have put some
thought into what you want.
2. Get community buy-in: An “Ask” is made stronger by multiple people or stakeholder groups
having the same one. Reach out to other related groups in your community to see if they will
contact offices with the same message. The more people the better.
3. Relate your “Ask” back to the office: Congressional offices are ultimately beholden to the
people they represent. Therefore, when considering whether an issue is important or not, they
need to know why it is important to them. With landslides, they need to know two things: how do
they impact their constituents—how many peoples’ lives are at risk and what critical
infrastructure is vulnerableand how much do they cost.
4. Be brief: A quick email or phone call with a couple hard statistics goes a long way. Give
staffers the option to contact you for more information.
5. Be patient, but follow up: Oftentimes emails fall through the cracks. If you do not hear back
from the office within a week send a short, polite, follow-up email providing a little more
information and giving them another chance to contact you for more information. Continue to
follow up every month or so with new information on how the issue relates to them.
This process is ultimately about forming a relationship and building trust. It can take years and
multiple Congresses to foster enough support to pass a bill. However, as long as someone in
AEG Professional Forum Extended Abstracts
office is willing to talk about the importance of landslide hazard research, assessment, and
mitigation, we are doing something right.
Spiker, Elliot C; Gori, Paula L. 2003. National Landslide Hazards Mitigation StrategyA
Framework for Loss Reduction: U.S. Geological Survey Circular 1244.
Committee on the Review of the National Landslide Hazards Mitigation Strategy, Board on
Earth Sciences and Resources. 2004. Partnerships for Reducing Landslide Risk: Assessment of
the National Landslide Hazard Mitigation Strategy: National Research Council of the National
Academy of Sciences.
AEG Professional Forum Extended Abstracts
Where do we go from here?
Saving Lives and Property, Is Science the Easy Part?
Montgomery, David R., PhD, University of Washington,
In the aftermath of the tragic Oso/Hazel and other tragic landslide, the Governor of Washington
convened a Commission to make recommendations to reduce landslide impacts in the future.
Some of the recommendations refer to emergency response, preparedness, and communication;
some to communication within the geosciences and the regulating organizations. The following
geoscience-oriented recommendations warrant additional discussion, vetting, and details. What
is missing from the list? Could recommendations like these be the basis for nationwide
program? How would we implement such a program?
1) Develop a statewide landslide hazard mapping and risks assessment program.
2) Establish hazards resilience institute.
3) Conduct landslide investigations.
4) Update state laws and consider innovative development regulations.
5) Develop public awareness initiatives and educational programs.
6) Expand real estate curriculum.
AEG Professional Forum Poster Abstracts
Understanding the critical role that slope angle can play on the behaviour of landslides
Beddoe, Ryley, PhD, Lassonde School of Engineering, York University, Toronto, ON Canada
Typically, one's intuition would be that a steeper slope is going to be more dangerous when it comes to a
landslide than a shallower slope. In fact, results from physical model experiments showed that for
granular soils there is not a linear relationship between slope angle and landslide consequences. Rather it
was found that a more gently inclined slope has a greater capacity for water storage than a steeper slope.
Which in turn can result in shallower slopes being more susceptible to liquefaction. The significance of
these findings is that there are increased consequences for a landslide associated with liquefaction
behaviour - it will travel further and faster resulting in higher risks for people and infrastructure
downslope, as was seen during the Oso landslide. As we move forward in improving land use guidelines
and policies, these scientific results further highlight the critical importance there is for correctly
addressing gently inclined slopes that have an area prone to liquefaction, in order to adequately mitigate
either the triggering or mobility of these high risk flow-slides.
AEG Professional Forum Poster Abstracts
Rigorous Peer Review and Regulatory Oversight Mitigate
Landslide Hazards in the Santa Monica Mountains, Los Angeles County, California
Burger, Karin L.M., PG, CEG, Los Angeles County Department of Public Works, Geotechnical and
Materials Engineering Division, Alhambra, CA, USA;
Within the Santa Monica Mountains there are several large landslide complexes with dense residential
development that are under the oversight of the County of Los Angeles. Residential development within
the Santa Monica Mountains has been occurring since the 1920’s. As accessibility to the area increased,
areas were developed for residential use with much of the development occurring prior to any in-depth
geotechnical oversight or the adoption of geotechnical review requirements in the building code in 1962.
In the process of development, many landslides were subdivided without consideration for the geologic
hazards associated with the properties or impacts this would have for future development.
With the adoption of a geotechnical review process in the building code, consultants had to demonstrate
that properties would be free from landslide, settlement, or slippage and that development would not
adversely impact off site property through the presentations of data and analysis. This led to a wide
range in the quality of geotechnical investigations and reporting. In the late 1960’s and early 1970’s the
USGS performed extensive geologic mapping of several quadrangles in the Santa Monica Mountains.
These geologic maps have been utilized by County geologists as an aid in determining the extent that an
investigation must go to address geologic hazards.
In addition to mapped geotechnical hazards, all of the properties developed within the unincorporated
area of the Santa Monica Mountains are serviced by an on-site wastewater treatment system (System).
The Los Angeles County Health Department’s Environmental Health branch oversees new Systems for
development of vacant land or additions to existing structures that require an upgrade to the existing
System. During this process, a developer must also obtain geotechnical approval of their proposed
improvements, which requires stability assessment of the property with respect to the location of the
proposed System. Due to the nature of the projects, which consist primarily of single-family residences,
funds may be limited for a consultant to perform a truly comprehensive assessment of the whole
property. This often results in the consultant attempting to model the most “conservative” scenario
rather than compiling subsurface data that accurately represent the site.
The County of Los Angeles approaches each project in the Santa Monica Mountains on an individual
basis, but within the context of the local community and expects the same from the geotechnical
consultant. For the past 30 years, the Los Angeles County’s Building and Safety Division has required
the property developer to obtain a geotechnical report addressing the proposed improvements for most
grading and building plans in the area. Development within the Santa Monica Mountains often requires
the consultant to consider off-site conditions that could impact the proposed improvements. The
geotechnical report and improvement plans are then reviewed by a registered engineering geologist and
soils engineer with no guarantee of an approval of the project. Through this rigorous peer review
process and regulatory oversight, the County of Los Angeles has been able to limit property damage and
losses within the Santa Monica Mountains.
AEG Professional Forum Poster Abstracts
Seattle Streets Landslide Hazard Assessment and Management
Dransfield, James S., PE, Amec Foster Wheeler, Bothell, WA, USA,,
Henry Brenniman, LEG, Amec Foster Wheeler, Bothell, WA, USA
Lingjun “Steve” Hou, PE, Seattle Department of Transportation, Seattle, WA, USA, and
Todd Wentworth, PE, Amec Foster Wheeler, Bothell, WA, USA
In 1999, the Seattle Department of Transportation (SDOT) initiated a plan to assess landslide hazards
along arterial streets. Amec Foster Wheeler was hired to identify hazards, evaluate risks, and prioritize
potential mitigation projects that would reduce operation and maintenance costs and improve public
One initial task involved identifying arterial street segments with landslide hazards. Review of landslide
hazard maps, interviews with SDOT staff, and site reconnaissance identified 73 arterial street segments
with potential landslide hazards. Interviews with City stakeholders revealed important factors to use in
prioritizing sites from the standpoint of operation and maintenance. These factors included traffic
volume, current conditions, and emergency vehicle access.
Using weighted decision factors, the relative risk of each street segment was ranked to identify street
segments with the highest priority for mitigation projects. More in-depth geologic reconnaissance and
research was performed on the top one-third (24) highest priority street segments. Evaluations were
conducted for each site to identify the length of affected street, type of hazard, recommended repair
options, and estimated repair costs. The costs and benefits were evaluated for each of these high-priority
sites, and results of the analysis were provided to SDOT with a rational basis to manage and prioritize
landslide risks.
The poster will outline the risk assessment framework and describe the cost-benefit ranking. Landslide
repair projects implemented at several of the high-priority sites over the past 15 years will be
summarized to demonstrate the effectiveness of the program.
AEG Professional Forum Poster Abstracts
Advancing the State of the Art in Landslide Monitoring: Preliminary Results from a
Comparative Analysis of LiDAR-based 3D Landslide Displacement Measurement Methods
Haugen, Benjamin D., Dept. Geol. & Geol. Eng., Colorado School of Mines, Golden, CO, USA
Arjun Aryal, School Ocean & Ear. Sci. & Tech., University of Hawaii, Honolulu, HI, USA
Edwin Nissen, Dept. Geophysics, Colorado School of Mines, Golden, CO, USA
Wendy Zhou, Dept. Geol. & Geol. Eng., Colorado School of Mines, Golden, CO, USA
Landslide professionals may soon favor multi-temporal Light Detection and Ranging (LiDAR) over
traditional “fixed-point” landslide displacement monitoring techniques (e.g., survey monuments,
inclinometers, etc). While traditional methods are accurate and reliable, they require field personnel to
enter hazardous terrain and have limited spatial resolution (i.e., measurements per unit area). Safety
concerns tend to limit the temporal resolution of the monitoring data. Inadequate spatial resolution limits
the fidelity and completeness of 3D kinematic analyses. LiDAR-based methods solve both problems:
data is collected from safe ground, at any time, and the data can provide nearly-continuous displacement
field measurements at resolutions exceeding one measurement per square meter. Multi-temporal LiDAR
can thus be used to fully characterize the 3D displacement field of complexly-deforming landslide
masses without putting field personnel at unnecessary risk. While some questions about LiDAR
methods’ reliability and accuracy have been raised in the professional and research communities, recent
advancements in data processing techniques support the viability of LiDAR for complex 3D landslide
displacement fields. Here we present early findings from a comparative analysis of LIDAR-based 3D
displacement field measurement methods. Our test dataset is from a slow-moving landslide near Granby,
Colorado. LiDAR point clouds were collected in May and August 2014. Displacement field
measurements were compared with contemporaneous survey data on both the landslide body and stable
terrain. While our results indicate some variance in the accuracy of available methods that obliges
additional investigation, but show great promise for the development of an accurate and reliable
methodology for measuring 3D landslide displacement fields using LiDAR point clouds.
AEG Professional Forum Poster Abstracts
The 2012 Johnsons Landing landslide: rethinking landslide risk management in British Columbia
Jordan, Peter, BC Ministry of Forests, Lands and Natural Resource Operations, Nelson, BC, Canada,
On July 12, 2012, a large landslide occurred on the mountainside above Johnsons Landing, a small
community in the West Kootenay region of British Columbia. Four houses were destroyed and two
others damaged, and four people were killed in their homes. The investigation which followed addressed
several aspects of the disaster, including emergency response, risk management, land use and zoning,
hazard identification and warning, and landslide hazard monitoring. The landslide resulted from the
sudden failure of an estimated 320,000 m3 of glacial deposits. It travelled as a debris avalanche for 1.9
km at an average gradient of 16 degrees, at speeds of up to 150 km/hr. A secondary debris flow travelled
an additional 1 km to Kootenay Lake. As a first-time landslide in surficial material, the event was
unusual due to its large size, exceptional mobility, and lack of an obvious cause. The landslide followed
record rainfall and high snowmelt runoff in spring and early summer. The investigation found that the
valley above the failure contained a complex of slow-moving or dormant bedrock failures, which may
have caused deformation in the deep glacial deposits, gradually weakening them to the point of failure.
A challenging question for landslide hazard studies is how such features can be mapped, and how to
identify the few that might result in rapid failure. This exceptional event, along with several other
unusual landslides and floods in 2012 and 2013, prompted the BC government to begin a process re-
examine its policies and programs with respect to landslide risk management.
AEG Professional Forum Poster Abstracts
Defining Areas with Nil Landslide Hazard is a step toward a
Comprehensive Landslide Loss Model
Keaton, Jeffrey R, F.ASCE, F.GSA, Amec Foster Wheeler, Los Angeles, CA, USA,
Richard J. Roth, Jr., Consulting Insurance Actuary, Huntington Beach, CA, USA,
Landslide risk currently is unquantifiable in terms suitable for insurers to use for setting policy prices.
Insurers need probabilistic models of landslide hazards that quantify damage likelihood and extent.
Probabilistic seismic hazard and loss models permit earthquake risk to be identified quickly for any
street address. Early seismic hazard maps had four zones based on incidence of damage. Areas without
topography, geology, or geomorphology conducive to slope movements and with histories of no slope
movement would be Zone 0 (p<<0.01). Areas with slope movement incidence would be Zone 3
(p>>0.01). Zone 1 (p<0.01) would be hilly areas where no landslides have occurred and none are
expected based on analysis. Zone 2 (p≈0.01) would be hilly areas where landslides have not occurred
but are considered susceptible based on geology or geomorphology. Geo-professionals tend to
microzone slopes in Zone 3, but insurers would be helped more by a well-defined Zone 0 which might
allow landslide damage to be included in all-peril policies. Once viable products are available, private
insurers will seek to improve risk models for expanding Zone 0 boundaries and quantifying risk in other
zones. Local governments effectively become insurers by providing recovery and reconstruction funds
following landslide events. Local governments and private insurers need the same type of geoscience for
managing landslide risk. The success of earthquake insurance suggests that improvements will be made
in simple landslide loss models once private insurers have viable landslide insurance products.
AEG Professional Forum Poster Abstracts
Landslide Hazard Inventories and Uncertainty Associated with Ground Truth
Keaton, Jeffrey R, F.ASCE, F.GSA, Amec Foster Wheeler, Los Angeles, CA, USA,
William C. Haneberg, PhD, Fugro GeoConsulting, Inc., Houston, TX, USA,
Forecasts of future landsliding currently rely on "ground truth" inventories to validate and refine
quantitative or qualitative predictive models. Acceptance of inventory maps as deterministic truth,
however, neglects errors and uncertainties inherent in the inventory process. A well-documented
evaluation of landslide inventory maps of a 300-km2 area in northern Italy prepared by three groups of
geomorphologists revealed that landslide polygon positional mismatch between maps was in the 55-65%
range, whereas the positional mismatch was 80% for superposition of all three maps. Additional
discrepancies must exist with landslide inventories that also include classification, volume, activity and
other characteristics. Simply understanding the general nature of the uncertainty and variability
associated with inventory maps is a substantial challenge; however, the basis for quantifying the
uncertainty and variability requires independent reassessment of all aspects of the inventory process or
acceptance of some theoretical a priori value for variance. Evaluation of some hazards (e.g., surface
fault rupture) is done with apparent consensus of location and substantial attention devoted to the
uncertainty associated with estimates of activity (e.g., slip rate). The scale of inventory mapping is
substantially greater for landslides than for active faults because of the number and variety of landslides,
the variable degree of activity within an individual landslide, and the fact that landslides are secondary
hazards triggered by primary processes. Possible improvements might include 1) combining the maps by
several independent inventory teams to produce empirical probabilistic maps, 2) using widely applicable
statistical distributions of landslide size to estimate the numbers of landslides likely to have been missed
in an inventory, and 3) using physics-based watershed scale models of landslide occurrence to identify
areas that may have been overlooked in inventories or which may be susceptible to sliding under future
rare or unprecedented conditions.